University of Groningen

Left ventricular dilatation and neurohumoral activation as arrhythmogenic factors inmyocardial infarctionDambrink, Jan Hendrik Everwijn

IMPORTANT NOTE: You are advised to consult the publisher's version (publisher's PDF) if you wish to cite fromit. Please check the document version below.

Document VersionPublisher's PDF, also known as Version of record

Publication date:1995

Link to publication in University of Groningen/UMCG research database

Citation for published version (APA):Dambrink, J. H. E. (1995). Left ventricular dilatation and neurohumoral activation as arrhythmogenic factorsin myocardial infarction: results from the Captopril And Thrombolysis Study. s.n.

CopyrightOther than for strictly personal use, it is not permitted to download or to forward/distribute the text or part of it without the consent of theauthor(s) and/or copyright holder(s), unless the work is under an open content license (like Creative Commons).

The publication may also be distributed here under the terms of Article 25fa of the Dutch Copyright Act, indicated by the “Taverne” license.More information can be found on the University of Groningen website: https://www.rug.nl/library/open-access/self-archiving-pure/taverne-amendment.

Take-down policyIf you believe that this document breaches copyright please contact us providing details, and we will remove access to the work immediatelyand investigate your claim.

Downloaded from the University of Groningen/UMCG research database (Pure): http://www.rug.nl/research/portal. For technical reasons thenumber of authors shown on this cover page is limited to 10 maximum.

Download date: 10-06-2022

LEFT VENTRICULAR DILATATION AND NEUROHUMORALACTIVATION AS ARRHYTHMOGENIC FACTORS

IN MYOCARDIAL INFARCTION

Results from the Captopril And Thrombolysis Study

ISBN 90-9008889-X

1995 J-H.E. DambrinkAll rights are reserved. No part of this publication may be reproduced, stored ina retrieval system, or transmitted in any form or by any means, mechanically orby photocopying, recording or otherwise, without the written permission of theauthor.

Printed by: Drukkerij Elinkwijk B.V., Utrecht.

RIJKSUNIVERSITEIT GRONINGEN

LEFT VENTRICULAR DILATATION AND NEUROHUMORAL

ACTIVATION AS ARRHYTHMOGENIC FACTORS

IN MYOCARDIAL INFARCTION

Results from the Captopril And Thrombolysis Study

PROEFSCHRIFT

ter verkrijging van het doctoraat in de Geneeskunde

aan de Rijksuniversiteit Groningen

op gezag van de Rector Magnificus Dr F. van der Woude

in het openbaar te verdedigen op maandag 11 december 1995

des namiddags te 4.00 uur

door

Jan Hendrik Everwijn Dambrink

geboren op 23 april 1963 te Utrecht

Promotores: prof. dr K.I. Lie

prof. dr W.H. van Gilst

Co-promotor: dr J.H. Kingma

Promotiecommissie: prof. dr M.J. Janse

prof. P.A. Poole-Wilson, MD, FRCP

prof. dr C.A. Visser

Financial support by Bristol-Meyers Squibb B.V. and by the CATSfoundation for the publication of this thesis is gratefully acknowledged.

Aan mijn ouders

CONTENTS

Chapter 1. Introduction

§ 1. General introduction 1

§ 2. Ventricular arrhythmias after thrombolytic therapy for acutemyocardial infarction

2.1. Early ventricular arrhythmias: role of reperfusion 32.2. Late ventricular arrhythmias after thrombolytic therapy 9

§ 3. Left ventricular remodeling and ventricular arrhythmias aftermyocardial infarction: importance of left ventricular dilatation

3.1. Left ventricular remodeling after myocardial infarction 123.2. Association of left ventricular remodeling with arrhythmogenesis 133.3. Mechanisms of arrhythmogenesis in left ventricular dilatation 163.4. Effect of thrombolytic therapy on left ventricular dilatation

and related ventricular arrhythmias 17

§ 4. Neurohumoral activation as an etiologic factor for early and lateventricular arrhythmias

4.1. Sympathetic nervous system activity and ventricular arrhythmiasafter acute myocardial infarction 19

4.2. Renin-angiotensin-aldosterone system activation and ventriculararrhythmias after acute myocardial infarction 22

4.3. Effect of thrombolytic therapy on neurohumoral activation 23

§ 5. Interrelation between neurohumoral activation and left ventriculardilatation 25

§ 6. Rationale for the use of ACE inhibition during thrombolytic therapy

6.1. Effect on early ventricular arrhythmias 266.2. Effect on late ventricular arrhythmias 29

§ 7. Aims of the thesis 37

Chapter 2

Left ventricular dilatation and high-grade ventricular arrhythmias in thefirst year after myocardial infarction 53J Cardiac Failure 1994;1:3-11

Chapter 3

Association of left ventricular remodeling and nonuniform electrical recovery expressed by nondipolar QRST integral map patterns insurvivors of a first anterior myocardial infarction 69Eur Heart J 1993;14(suppl):24 (abstract)Circulation 1995;92:300-310

Chapter 4

Relation between left ventricular dilatation and QRS duration after afirst anterior myocardial infarction 97Eur Heart J 1994;15(suppl):494 (abstract)Submitted

Chapter 5

Norepinephrine levels early after thrombolytic therapy: association withangiographic findings and ventricular arrhythmias 105Submitted

Chapter 6

Association between reduced heart rate variability and left ventriculardilatation in patients with a first anterior myocardial infarction 121Circulation 1993;88:I-107 (abstract)Br Heart J 1994;72:514-520

Chapter 7

Acute intervention with captopril during thrombolysis in patients withfirst anterior myocardial infarction. Results from the Captopril andThrombolysis Study (CATS) 139Eur Heart J 1994;15:898-907

Chapter 8

Which patient benefits from early angiotensin-converting enzymeinhibition after myocardial infarction? Results of one-year serialechocardiographical follow-up from the Captopril and ThrombolysisStudy (CATS) 161Circulation 1995;92:I-24 (abstract)Submitted

Chapter 9

Effects of captopril on early and late arrhythmic events in patients withthrombolytic therapy for a first anterior myocardial infarction 185

Chapter 10

Summary and concluding remarks 207

Hoofdstuk 11

Samenvatting en conclusies 215

Dankwoord 223

Appendix 226

Curriculum vitae 229

1

CHAPTER 1. Introduction

§ 1. General introduction

In 1859, Einbrodt demonstrated that vagal nerve stimulation could preventventricular fibrillation in the canine ventricle.1 This finding was not appreciateduntil Kent et al.2 reproduced this experiment more than a century later. Other in-vestigators showed that the antifibrillatory effect of vagal nerve stimulationcould be attributed to an opposing effect on sympathetic neural input.3 Thisconcept proved highly valuable in the setting of acute myocardial infarction;treatment with beta-receptor blockade appeared to reduce the occurrence ofventricular fibrillation and sudden cardiac death.4 However, once treatment withthrombolytic agents was applied, the additional effect of beta-blockade on mor-tality5 or early ventricular arrhythmias6 was less clear.

In the early 1980s, van Gilst et al.7 described a clear-cut reduction in durationand occurrence of ventricular fibrillation during reperfusion when the angio-tensin-converting enzyme (ACE) inhibitor captopril was added to a Langendorffperfused rat heart, a model of ischemia and reperfusion. This effect was paral-leled by a significant reduction of catecholamine overflow. Subsequent experi-ments8,9 strongly suggested that this blunted neurohumoral response to reperfu-sion played an important role in the reduction of reperfusion-related ventriculararrhythmias. In addition, ACE inhibition proved to reduce myocardial injury,another potentially anti-arrhythmogenic mechanism. Based on these results,obtained both in vitro and in vivo, the Captopril And Thrombolysis Study(CATS) was designed to investigate the effects of captopril in humans duringthrombolytic therapy for acute myocardial infarction.

By that time, it had become clear that changes in morphology of the heart af-ter myocardial infarction, also known as remodeling, could have important im-plications for the occurrence of late ventricular arrhythmias. White et al.10 dem-onstrated the powerful predictive value of left ventricular dilatation assessed 6weeks after acute myocardial infarction for the occurrence of sudden cardiacdeath. Since modulating effects of captopril on left ventricular remodeling hadbeen described,11 an indirect effect on late ventricular arrhythmias could be an-ticipated. Serial echocardiography, which was part of the CATS protocol, al-lowed a more detailed evaluation of this possibly arrhythmogenic factor up toone year after myocardial infarction.

In this thesis, the main results of the CATS study are described with specialinterest in the effects of ACE inhibition on the occurrence of early and latepostinfarction ventricular arrhythmias. In addition, the associations between

CHAPTER 1

2

dilatation of the heart, activation of neurohumoral systems, and ventricular ar-rhythmias are evaluated in detail. Finally, noninvasive techniques, includingbody surface mapping and signal-averaged electrocardiography, are used toidentify possible underlying electrophysiological mechanisms of the relationbetween dilatation and ventricular arrhythmias.

§ 2. Ventricular arrhythmias after thrombolytic therapy for acutemyocardial infarction

In experimental research, ventricular arrhythmias are usually divided intoacute, delayed, and chronic ventricular arrhythmias,12 each with different char-acteristics and mechanisms. In the acute phase, i.e., up to 2-4 hours after onsetof myocardial infarction, ischemia is the predominant arrhythmogenic factor.Clinical arrhythmias equivalent to acute ventricular arrhythmias in the experi-

Definitions used in this thesis

Ventricular arrhythmias include ventricular fibrillation (VF), ventriculartachycardia (VT), accelerated idioventricular rhythm (AIVR), uniform-,multiform-, and paired ventricular premature beats (VPBs), VPBs in b i-geminy, and sudden cardiac death.

Early ventricular arrhythmias include ventricular arrhythmias occurringup to 48 hours after the onset of symptoms of acute myocardial infar c-tion;late ventricular arrhythmias refers to ventricular arrhythmias occurringlater than 48 hours after the onset of myocardial infarction; the observ a-tion period in CATS was limited to one year.

Ventricular tachycardia is defined as a repetition of three or more VPBswith a rate excee ding 100 beats/min.

Sustained ventricular tachycardia includes VT lasting longer than 30seconds or shorter when leading to hemodynamic collapse;nonsustained ventricular tachycardia indicates VT with a durationshorter than 30 seconds, without a hemodynamic collapse.

AIVR is defined as a repetition of three or more uniform ventricular prem a-ture beats with a rate of less than 100 beats/min.Pairs (paired VPBs) refers to a repetition of two ventricular premature beats.High-grade ventricular arrhythmias include ventricular arrhythmias fromLown classes 4A and 4B (‘high Lown grade’) which are paired VPBs andsustained or nonsustained VT.Complex ventricular arrhythmias are defined as > 10 VPBs/hour and/or

INTRODUCTION

3

mental setting usually occur before patients reach the hospital. These arrhyth-mias are an important cause of sudden cardiac death,13 and consist mainly ofventricular fibrillation (VF). After the acute phase, the incidence of ventricularfibrillation drops, and other ventricular arrhythmias emerge, including ventricu-lar tachycardia (VT) and accelerated idio-ventricular rhythm (AIVR).14 These ar-rhythmias occur up to 24-48 hours after the onset of myocardial infarction andare equivalent to the delayed ventricular arrhythmias found in the experimentalsetting, in which abnormal automaticity is the predominant electrophysiologicmechanism. Arrhythmias in this period also include the so-called reperfusion ar-rhythmias. Finally, late in-hospital ventricular arrhythmias and ventricular ar-rhythmias occurring after discharge are the clinical counterparts of chronic ven-tricular arrhythmias in the experimental setting, in which reentry is the predomi-nant mechanism.

In this thesis, early ventricular arrhythmias are defined as ventricular ar-rhythmias occurring up to 48 hours after the onset of symptoms (‘CCU phase’).Late ventricular arrhythmias (‘late hospital phase’ and ‘chronic phase’) are de-fined as ventricular arrhythmias emerging after this period up to one year aftermyocardial infarction.15

2.1. Early ventricular arrhythmias: role of reperfusion

When thrombolysis was considered for use in patients with acute myocardialinfarction, there was serious concern about the possible occurrence of so-calledreperfusion arrhythmias. This concern was based on numerous experimentalstudies which had shown a high incidence of VF when blood flow was restoredto a previously ischemic part of the myocardium. Already in 1934, Tennant andWiggers16 found VF in 10 out of 14 dogs upon reperfusion by release of a liga-ture of the left anterior descending coronary artery. Subsequent experimentalanimal studies in dogs and other species have also reported a high incidence ofVF after sudden restoration of anterograde blood flow.7,17-20 By contrast, in-termittent occlusion17 or gradual reperfusion21 both seem to result in a lower in-cidence of ventricular arrhythmias. In addition, the type of arrhythmia emergingupon reperfusion is largely dependent on the duration of the preceding ischemicepisode.18-20 Battle et al.18 observed an increase in the incidence of VT and VFwhen the duration of ischemia was prolonged from 3-6 minutes to 30-45 min-utes. However, Balke et al.19 observed a 67% reduction of ventricular arrhyth-mias when the ischemic period was extended from 30 to 60 minutes. De Graeffet al.20 noted only AIVR after a 60-minute occlusion period in their closed-chestpig model. The incidence of reperfusion-induced VF appears to be maximal af-ter 20-30 minutes of ischemia, when both reversibly and irreversibly damaged

CHAPTER 1

4

cells are present.22 After this period, life-threatening reperfusion arrhythmias areless likely to occur23 because the majority of cells in the infarcted region be-come necrotic.

Underlying mechanisms

The mechanisms leading to reperfusion arrhythmias are not fully understood.Rapid changes in potassium concentrations, Pco2 and accumulation of calciumand lysophosphoglycerides next to increased adrenergic stimulation all appearto play an important role.22 The electrophysiological mechanism of these ar-rhythmias seems primarily nonreentrant in origin. In a study in the cat heart,Pogwizd et al.24 demonstrated that the initiating mechanism was nonreentrant in75% of reperfusion-induced VT. In addition, these arrhythmias were maintainedby a nonreentrant mechanism in 61% of cases. Likewise, during acceleration ofsustained VT before transition to VF, nonreentrant mechanisms were operative.

Abnormal automaticity is especially noted in case of delayed reperfusion ar-rhythmias occurring several minutes after reperfusion.25 There is no direct evi-dence for triggered activity as a mechanism of reperfusion arrhythmias. How-ever, the effect of calcium-channel blockers on reperfusion arrhythmias26 sug-gests a possible role of delayed afterdepolarizations (DADs). Delayed afterdepo-larizations are especially seen in situations characterized by calcium overloadand may result in triggered activity. Oscillations of the intracellular calciumconcentration cause a transient inward current, which, if threshold is reached,can trigger one or more premature beats. Similarly, there is no conclusive evi-dence that early afterdepolarizations (EADs) play an important role in the occur-rence of reperfusion arrhythmias. Early afterdepolarizations occur during the re-polarization phase and are also capable of triggering ventricular arrhythmias ifthreshold is reached. Still, some investigators have reported the appearance ofEADs upon reperfusion. Priori et al.27 reported a high incidence of EADs (54%)in cats upon reperfusion after a 10-minute period of ischemia. In this study,EADs were associated with the occurrence of reperfusion arrhythmias in 62% ofcases.

In conclusion, most reperfusion arrhythmias are induced by a nonreentrantmechanism. This mechanism appears to be abnormal automaticity in most cases,although triggered activity, induced by DADs or EADs, may play a role.

INTRODUCTION

5

Incidence of reperfusion arrhythmias in man after the administration of thrombolytic therapy

In the early 1980s, intracoronary thrombolysis was introduced as a new ther-apy for acute myocardial infarction. In a number of randomized and nonran-domized studies it was demonstrated that this therapy could safely be applied inhumans, and that the incidence of life-threatening ventricular arrhythmias waslow.28-32 In the study by Ganz et al.,28 frequent ventricular premature beats, bi-geminy, and AIVR were observed after infusion of intracoronary streptokinase.Only one patient required electrical cardioversion for VT; the average time fromonset of symptoms to reperfusion was 4 hours. Goldberg et al.33 studied the in-cidence of reperfusion arrhythmias more extensively. In their study, reperfusionarrhythmias (mainly AIVR) were found in 82% of patients with successfulthrombolysis. No VF was found in the first hours after restoration of flow. Itshould be noted that all of these patients were pretreated with lidocaine. The in-cidence of VF after thrombolytic therapy has been studied in a number of majorthrombolysis studies.34-38 Although these studies did not show a significant dif-ference individually, an analysis of pooled data (Table 1.1) reveals a 13% re-duction in the incidence of VF after thrombolytic therapy. In most of thesestudies, no distinction was made between VF during the acute phase of myo-cardial infarction or during the remaining in-hospital period. In a recent meta-analysis, the occurrence of early VF (up to 24 hours after myocardial infarction)in a number of major trials was evaluated separately. There proved to be nosignificant difference in the incidence of VF between patients treated withthrombolytic agents or placebo (about 3%) during this early period.39

Table 1.1. Incidence of in-hospital VF in 5 major thrombolysis trials

Study Incidence of VF Incidence of VF RR (95% CI) (treatment group) (placebo group)

GISSI-135 388/5860 (6.6%) 439/5852 (7.5%) 0.88 (0.76 - 1.01)ISIS-234

ASSET36

AIMS37

ISAM38

370/8592 (4.3%) 94/2516 (3.7%) 41/624 (6.6%) 37/859 (4.3%)

425/8595 (4.9%) 116/2495 (4.6%) 46/634 (7.3%) 46/882 (5.2%)

0.87 (0.76 - 1.00) 0.80 (0.62 - 1.05) 0.91 (0.61 - 1.37) 0.83 (0.54 - 1.26)

Total 930/18451 (5.0%) 1072/18458 (5.8%) 0.87 (0.80 - 0.95)*

CI indicates confidence interval; RR, risk ratio; VF, ventricular fibrillation. * Mantel-Haenzel risk ratio.

CHAPTER 1

6

As stated previously, the incidence of VF is especially high when reperfusionoccurs after a relatively short episode of ischemia. Therefore, the effect ofthrombolytic therapy administered shortly after onset of symptoms is of particu-lar interest. Recently, the European Myocardial Infarction Project (EMIP) in-vestigators40 studied the incidence of VF in 2750 patients receiving anistreplase130 minutes after the onset of symptoms and before hospital admission. Ven-tricular fibrillation was significantly more frequent in patients receiving anistre-plase (2.5% vs 1.6%). The authors suggested that this higher incidence of VFmay have been related to early reperfusion. This finding has been disputed byothers,41 who stated that VF related to cardiogenic shock or differences in con-comitant medication (e.g., beta-blockers) may have obscured the relation be-tween thrombolytic therapy and the incidence of VF. Other investigators usingvery early thrombolytic therapy within 1 or 2 hours after onset of symptoms42,43

have reported an incidence of VF similar to the placebo group of EMIP (1% orless). Thus, the increased incidence of VF after very early administration ofthrombolytic therapy remains controversial.

One of the previously mentioned major thrombolysis trials, the IntravenousStreptokinase in Acute Myocardial infarction (ISAM) trial38 also reported on theincidence of ventricular arrhythmias other than VF in 1741 patients early afterthrombolytic therapy. In this study, patients given streptokinase had more fre-quent ventricular premature beats (30.5% vs 19.4%, VPBs > 10/hour) and

Table 1.2. Incidence of early ventricular arrhymthmias after thrombolytic therapyusing Holter monitoring

N Duration(hrs)

VF (%) VT (%) AIVR (%)

Miller et al.46 39 12 5 90 90Cercek et al.45

Gore et al.49

Hackett et al.50

Six et al.48

Zehender et al.51

Gressin et al.47

Heidbüchel et al.6

CATS

456738403040244190

241.52.6≤ 424242412

0316410332

85 6502680807554

90 7322377887159

All (range) 733 - 3.6 (0-16) 60 (6-90) 60 (7-90)

AIVR indicates accelerated idioventricular rhythm; CATS, Captopril And ThrombolysisStudy; VF, ventricular fibrillation; VT, ventricular tachycardia.

INTRODUCTION

7

AIVR (5.4% vs 2.4%) than patients in the placebo group in the first 3 hours af-ter thrombolytic therapy. The incidence of VT (3.8% vs 3.5%) was not signifi-cantly different. In the Western Washington intravenous streptokinase trial,44 VTwas more frequent in the first 14 days after thrombolytic therapy (11.0% vs4.5%). The incidence of early VT was not reported separately. Since continuousHolter monitoring was not performed in the above-mentioned studies, the inci-dence of ventricular arrhythmias may have been underestimated. A number ofstudies have described the incidence of ventricular arrhythmias in the setting ofthrombolytic therapy using continuous Holter monitoring6,45-51 (Table 1.2).There are considerable differences in the incidence of early ventricular ar-rhythmias compared to studies from the prethrombolytic era. In an extensivereview, Bigger et al.15 estimated the incidence of various forms of postinfarctionventricular arrhythmias before the introduction of thrombolytic therapy. In a to-tal of 1443 patients, 10% had AIVR in the in-hospital period (range of 8% to46% in various studies), compared to 60% (range 7% - 90%) in 733 patientswith thrombolytic therapy (Table 1.2). Similarly, Bigger et al.15 described a VTincidence of 18.5% in 3698 patients (range 6% - 40%), monitored continuouslyduring their stay in the CCU, compared to a 60% VT incidence (range 6% -90%) in the 733 patients who underwent thrombolytic therapy. Two other stud-ies have investigated the incidence of ventricular arrhythmias after thrombolytictherapy in a placebo-controlled design using 24-hour Holter monitoring. Wilcoxet al.52 found a significant increase in the incidence of ventricular prematurebeats, pairs, and VT in patients treated with alteplase within 5 hours aftersymptom onset. Alexopoulos et al.53 found no difference in the incidence ofventricular arrhythmias after streptokinase or placebo during 24-hour Holtermonitoring. However, when patients with a duration of symptoms less than orequal to 6 hours were considered separately, a higher incidence was seen in pa-tients treated with streptokinase. Therefore, the higher incidence of ventriculararrhythmias (except VF) after thrombolytic therapy is more apparent when thistreatment is applied early after the onset of symptoms.

In some of the studies investigating the incidence of ventricular arrhythmiasafter myocardial infarction, the relevance of a patent infarct-related artery afterthrombolysis has been addressed. Zehender et al.51 found nonsustained VT in95% of 22 patients with successful thrombolysis compared to 38% of eight pa-tients with no reperfusion (p < 0.01). In addition, AIVR was also more frequentin those with a patent infarct-related artery 120 minutes after thrombolytic ther-apy (82% vs 63%). The observation that AIVR may be related to successfulreperfusion had been described previously.33,54 Heidbüchel et al.6 found no dif-ference in the incidence of nonsustained VT (77% vs 79%) or AIVR (74% vs69%) in patients with or without successful reperfusion. However, since angiog-

CHAPTER 1

8

raphy was not performed until 10-14 days after myocardial infarction, reocclu-sion during this interval may have disturbed the relation between reperfusionand ventricular arrhythmias. Similar to the study by Zehender et al.,51 all pa-tients with VF that had angiography (3/6) showed an occluded infarct-relatedartery. Data from the Thrombolysis In Myocardial Infarction (TIMI) Phase IIdatabase55 on 2546 patients have confirmed that sustained VT and VF afterthrombolytic therapy are more frequent in patients with an occluded infarct-related artery.

In conclusion, the initial concern that thrombolytic therapy would producelife-threatening ventricular arrhythmias upon reperfusion has not become real-ity. Although the effect of thrombolysis on VF very early after the onset ofsymptoms (e.g., within 1 hour) is still controversial, thrombolytic therapy ad-ministered within 6 hours does not increase the incidence of VF up to 24 hours.Moreover, thrombolysis appears to actually reduce the incidence of VF duringthe remaining in-hospital period. Studies using coronary angiography have sug-gested that early VF and sustained VT are related to an occluded infarct-relatedartery, and thus unsuccessful reperfusion. Conversely, nonsustained VT andAIVR do occur more frequently after thrombolytic therapy, especially within 6hours after onset of treatment. It has been suggested that these arrhythmias, andparticularly AIVR, are predictors of successful thrombolysis. However, thesesigns of reperfusion are not very sensitive.48

Time course of early ventricular arrhythmias

In a natural history study, Campbell et al.14 described the incidence and timecourse of ventricular arrhythmias in the first 12 hours after myocardial infarc-tion. Three hours after the onset of myocardial infarction, the incidence of VFdropped, while other arrhythmias, including ventricular premature beats andVT, increased in incidence. These data are in agreement with another prethrom-bolytic study by Northover et al.,56 who found the prevalence of VT to increaseup to 12 hours after the onset of symptoms. A similar distribution of ventriculararrhythmias was observed in one of the major thrombolysis trials, the Anglo-Scandinavian Study of Early Thrombolysis (ASSET).52 Compared to placebo,paired ventricular premature beats and VT during Holter monitoring were morefrequent in the 12-hour period after infarction in patients treated with alteplase.After this period, the incidence dropped gradually in both alteplase- and pla-cebo-treated patients, and a difference between these two groups could nolonger be discerned.

INTRODUCTION

9

2.2. Late ventricular arrhythmias after thrombolytic therapy

The incidence of ventricular arrhythmias rapidly declines after the acutephase of myocardial infarction (Figure 1.1). As opposed to early ventricular ar-rhythmias, in which myocardial ischemia and reperfusion are major etiologicfactors, infarct size, remodeling of the ventricle,10 and neurohumoral activation57

become more important in the cause of late ventricular arrhythmias. Sincethrombolytic therapy has a beneficial influence on these factors,58-61 a reductionof late ventricular arrhythmias may be expected.

Late ventricular tachycardia

Figure 1.1. The incidence of various forms of ventricular arrhythmias during Holtermonitoring in CATS is depicted. A rapid decrease in the percentage of patients with ven-tricular arrhythmias is observed after the early phase of myocardial infarction. VPBs indi-cates ventricular premature beats; AIVR , accelerated idioventricular rhythm, VT, ven-tricular tachycardia.

CHAPTER 1

10

Most studies describing late ventricular arrhythmias have assessed their inci-dence in the period shortly before hospital discharge, with special interest in riskassessment of future arrhythmic events. In a review from the 1970s, VT was re-ported in 3% - 14% of myocardial infarction patients before hospital dis-charge.15 In patients participating in the Multi-Center Post Infarction Program,62

VT was reported in 11% of all patients. In this prethrombolytic study, predis-charge VT during Holter monitoring was a strong independent predictor of ar-rhythmic events during patient follow up. In a large study where all patients re-ceived thrombolytic therapy, the Gruppo Italiano per lo Studio della Streptochi-naisi nell’Infarto myocardico (GISSI-2) trial,63 the incidence of nonsustained VTbefore discharge was 7%. A direct comparison of VT in patients who did or didnot receive thrombolytic therapy during the late in-hospital period was per-formed by Turitto et al.64 These investigators found nonsustained VT in 13% ofpatients with thrombolytic therapy and in 11% of patients without thrombolytictherapy during 24-hour Holter monitoring (difference not significant). Only onepatient, not treated with thrombolysis, experienced sustained VT. In conclusion,a considerable variance in the incidence of predischarge VT is observed, bothwith and without thrombolytic therapy, and no unambiguous treatment effectcan be discerned. In addition, it has been suggested that VT in this period isprimarily related to left ventricular dysfunction and relatively independent of thepresence or absence of reperfusion.65

Studies that describe the incidence of VT after hospital discharge are scarce.In a review by Willems,66 the incidence of late VT varies between 1% and 19%.The recorded incidence proves strongly dependent on the duration of Holtermonitoring; a recording time of less than 24 hours rapidly reduces the numberof episodes of VT recorded. Furthermore, when the incidence is estimated onlyon the basis of a patient history, especially nonsustained VT may not add to theincidence rate since it can easily occur without symptoms.

Programmed electrical stimulation may be helpful to identify patients at riskof late VT. Kersschot et al.67 investigated the induction of ventricular arrhyth-mias in patients randomly assigned to streptokinase therapy or conservativetreatment. Induction of sustained monomorphic VT 26 days after myocardial in-farction occurred more frequently in patients treated conservatively, affecting all15 control patients (100%). By contrast, just 10 of 21 patients (48%) treatedwith streptokinase were inducible after a maximum of 3 extra stimuli (p <0.001). Seventeen patients in the streptokinase group (81%) showed early reper-fusion versus no patients in the control group. Bourke et al.68 also described asignificant reduction in the number of patients that were electrically inducibleafter thrombolytic therapy. Similar to the study by Kersschot et al.,67 this finding

INTRODUCTION

11

did not reveal a lower incidence of ventricular arrhythmias during follow-upcare.

Sudden cardiac death

In prethrombolytic studies, an incidence of sudden death of around 10% hasbeen reported in the first 6 months after myocardial infarction.15 Most majorthrombolysis trials have provided information only about the overall cardiacmortality late after myocardial infarction. However, in the APSAC InterventionMortality Study (AIMS)37 sudden cardiac death was reported separately: 25 pa-tients (4.0%) died suddenly in the treated group versus 40 patients (6.3%) in theplacebo group up to one year after myocardial infarction (RR 0.64, 95% CI:0.39-1.03). In the GISSI-2 study,69 in which all patients received thrombolytictherapy, 84 of 8676 patients (1%) died suddenly in the first 6 months after myo-cardial infarction. Therefore, there are indications that not only total mortalitybut also sudden cardiac death may be reduced after thrombolytic therapy.

In conclusion, the application of thrombolytic therapy has beneficially alteredthe natural history of myocardial infarction with respect to the occurrence oflife-threatening ventricular arrhythmias. However, ventricular arrhythmias stillare an important cause of death after myocardial infarction and it remains ofgreat importance to assess the clinical profile of the patient at risk for arrhythmicevents. In the following paragraphs attention will be focused on two importantrisk factors: left ventricular dilatation and neurohumoral activation.

§ 3. Left ventricular remodeling and ventricular arrhythmias aftermyocardial infarction: importance of left ventricular dilatation

In the last decade, many research efforts have been directed at the under-standing of left ventricular remodeling after acute myocardial infarction. Leftventricular remodeling is defined as changes in the topography of both the in-farcted and noninfarcted regions of the ventricle.70 It has become clear that thisprocess has important prognostic implications10 given the fact that there are nowstudies available that demonstrate the beneficial effect of modulation of thisprocess.71 In most studies, emphasis is placed on the functional and anatomicalchanges that result in remodeling of the left ventricle. These include infarct ex-pansion, left ventricular hypertrophy, and left ventricular dilatation. However,there is increasing experimental evidence available that these anatomical

CHAPTER 1

12

changes are paralleled by changes in electrophysiologic properties. Amongthese are increased dispersion in refractoriness, delayed conduction, and the ap-pearance of afterdepolarizations which may explain the high incidence of ven-tricular arrhythmias in patients with progressive remodeling of the left ventricle.

3.1. Left ventricular remodeling after myocardial infarction

The process of left ventricular remodeling begins immediately after the onsetof acute myocardial infarction. Within seconds of a coronary artery occlusion,wall motion abnormalities occur and an overall increase in left ventricular di-mensions can be observed.72 This early form of left ventricular dilatation com-pensates functionally for the locally damaged myocardium and usually results ina normalized stroke volume73 (‘functional dilatation’). During the followingdays a local inflammatory reaction and edema is observed in the infarcted area.Subsequently, over a period of weeks to months, proliferation of fibroblasts oc-curs and the number of collagen fibers increases. This is paralleled by resorptionof necrotic material, and thus scar tissue is generated.70 Before completion ofscar formation, the infarcted area is very vulnerable to straining forces. Dilata-tion and thinning of this area can easily occur due to slippage of myocytes.74

This part of the remodeling process is usually referred to as infarct expansion.75

Infarct expansion occurs on top of early functional dilatation and is the maindeterminant of left ventricular dilatation in the first few days after myocardial in-farction.76 After this period, dilatation of the noninfarcted area gradually be-comes more important. Dilatation of the noninfarcted area appears to be primar-ily dependent on the size of the infarcted area77 and preceding infarct expan-sion.78 Early left ventricular dilatation, thinning of the infarcted ventricular wall,and increased filling pressures all contribute to an increase in wall stress, whichin turn may further promote left ventricular dilatation (‘dilatation begets dilata-tion’) even when scar formation is completed. Pfeffer et al.11 demonstrated anadditional 30% increase in end-diastolic volume 3 months after completion ofscar tissue formation in rats.

Besides dilatation of infarcted and noninfarcted areas, left ventricular hyper-trophy is an integral part of the remodeling process. This can be considered acompensatory mechanism for the loss of functional myocardium. According toLa Place’s law, an increase in wall thickness would result in a reduction of walltension and subsequently prevent further dilatation. However, in experimentalstudies it has been observed that if 40% or more of the left ventricle is infarcted,hypertrophy as a compensatory mechanism fails and progressive dilatation willoccur.79 A critical infarct size above which dilatation can not be compensatedfor has not been established in patients.

INTRODUCTION

13

3.2. Association of left ventricular remodeling with arrhythmogenesis

Patients with progressive remodeling of the left ventricle, especially thosewith symptoms of heart failure, are characterized by a high incidence of ven-tricular arrhythmias.80 Sudden death rates from 4% to 86% have been reported,with most studies citing an incidence of around 50%.81 An association betweenthe process of left ventricular remodeling and the occurrence of ventricular ar-rhythmias can therefore be assumed.

Early phases of remodeling: scar tissue formation and infarct expansion

Early after the onset of the remodeling process, relations between anatomicalchanges and arrhythmogenesis emerge (Table 1.3). An anatomical substrate forventricular arrhythmias may be created even before completion of scar tissueformation. This anatomical substrate usually arises at the rim of the infarctedarea and consists of islands of viable and nonviable myocardium, creating re-gional differences in conduction velocity and refractory period.82 De Bakker etal.83 recently demonstrated that slow conduction in these areas is not caused bya reduced conduction velocity, but by a ‘zig-zag’ course of activation. In 50%of all postinfarction patients without spontaneously occurring sustained ven-tricular arrhythmias, a sustained VT may be inducible 3 weeks after the onset ofmyocardial infarction.84 However, sustained ventricular arrhythmias only occurin a minority of patients. Apparently, next to an anatomical substrate and ade-quate triggering (e.g., by a ventricular extrasystole), additional factors that pro-duce changes in activation and/or repolarization characteristics such as myo-cardial ischemia, increase in wall stress, electrolyte imbalance, or neurohumoralactivation are needed to produce a sustained ventricular arrhythmia.

Another factor that may promote reentry is distension of the infarcted area, orso-called infarct expansion. When infarct expansion is progressive, aneurysmformation can occur in the early phase of left ventricular remodeling.85 In pa-tients with early left ventricular aneurysm formation, mortality is high with ahigh percentage of sudden deaths. Meizlish et al.86 reported a mortality rate of61% in 18 patients in the first year after myocardial infarction with 55% of thesedeaths classified as sudden. Because of its relation to early infarct expansion,these authors referred to the early appearance of a left ventricular aneurysm as‘expaneurysm’. Not all ventricular aneurysms lead to life-threatening ventriculararrhythmias. The presence or absence of late potentials, a marker of a possible

CHAPTER 1

14

substrate for reentry, is a powerful predictor of these arrhythmias during follow-up care.87,88 Increased wall stress at the border of the aneurysm is another factorthat may promote arrhythmogenesis.89

In addition, there are indications that beta-receptors are up-regulated in thetransition zone between aneurysm and normal myocardium.90 This may causelocal differences in effects of circulating catecholamines, resulting in an in-creased dispersion in refractoriness.91 After thrombolytic therapy, propensitytowards development of ventricular arrhythmias in patients with a left ventricu-lar aneurysm is clearly reduced. Sager et al.92 studied 32 patients with a leftventricular aneurysm half of which received thrombolytic therapy. Thirteendays after myocardial infarction, VT could be induced in 88% of those withoutthrombolytic therapy versus only 8% of those in the thrombolysis group. Inaddition, 50% of the patients who did not receive thrombolytic therapy diedsuddenly or had a sustained VT, whereas no arrhythmic events occurred in thethrombolysis group. Thus, it appears that electrophysiologic stability is in-creased after thrombolytic therapy in patients with left ventricular aneurysm.

Later phases of remodeling: left ventricular hypertrophy

Little is known about the role of compensatory hypertrophy in the genesis ofventricular arrhythmias after myocardial infarction. Most information on the re-

Table 1.3 Anatomical remodeling and corresponding arrhythmogenic factors

Phase of remodeling Arrhythmogenic factor Possible mechanism

Scar tissue formation anatomical substrate reentry in presence ofmodulating factors

Infarct expansion(aneurysm formation)

LV hypertrophy

LV dilatation

local stretchbeta-receptors ↑

fibrosiscoronary reserve ↓calcium overload

stretchslippage, uncoupling

SIAs, EADs, DADsdispersion ↑#

slow conduction#

ischemia#

DADs

SIAs , EADs, DADsslow conduction#

sympathetic dysfunction dispersion ↑#

DADs indicates delayed afterdepolarizations; EADs, early afterdepolarizations; SIAs,stretch-induced afterdepolarizations; LV, left ventricular. # factors promoting reentry.

INTRODUCTION

15

lation between ventricular hypertrophy and ventricular arrhythmias has beenobtained from human patients or animals with hypertension.93-96 From thesestudies it is clear that left ventricular hypertrophy is a potent arrhythmogenicfactor. Data from the Framingham study indicate that patients with ventricularhypertrophy have a five-fold increase in risk of sudden cardiac death.94 The un-derlying mechanisms of the increased incidence of ventricular arrhythmias inthe setting of ventricular hypertrophy is probably multifactorial. Myocardialischemia induced by a limited coronary reserve, reentry promoted by interstitialfibrosis, and increased sympathetic activity all add to an accelerated process ofarrhythmogenesis.95 In patients with left ventricular hypertrophy and inducibleventricular arrhythmias the incidence of late potentials is high, suggesting animportant role of delayed conduction as a mechanism for these arrhythmias.96

On a molecular level, it has been shown that the buffering capability of intracel-lular calcium is reduced in hypertrophied myocytes. Therefore calcium homeo-stasis is fragile, and an increased calcium load could generate oscillatory electri-cal currents (such as DADs) which may lead to ventricular arrhythmias to oc-cur.97,98 In the case of myocardial infarction, left ventricular hypertrophy oftencoexists with left ventricular dilatation and myocardial ischemia. These threefactors all interact and augment each other in causing ventricular arrhythmias.99

Since left ventricular hypertrophy and dilatation occur simultaneously as part ofthe remodeling process, it may be difficult to assess their individual contribu-tions.

Although the role of ventricular hypertrophy in the occurrence of ventriculararrhythmias after myocardial infarction is not yet fully established, it may be ofmajor importance considering its arrhythmogenic potential in patients with hy-pertension.

Left ventricular dilatation

The residual left ventricular function after myocardial infarction is largelydetermined by the degree of left ventricular dilatation. Traditionally, left ven-tricular function has been expressed as the ejection fraction. Results from theMulticenter Postinfarction study100 show that ejection fraction is a powerfulpredictor of mortality during the first two years after myocardial infarction. Inaddition, its predictive value for the occurrence of ventricular arrhythmias isalso well established.101 However, ejection fraction may not be a very sensitivemeasure of left ventricular dilatation. Considerable dilatation of the left ventriclecan occur while the ejection fraction remains relatively unchanged. White etal.10 compared the prognostic value of ejection fraction and end-systolic volumeassessed by angiography in 605 patients 4-8 weeks after acute myocardial in-

CHAPTER 1

16

farction. They found that end-systolic volume was a more powerful predictor ofdeath after myocardial infarction than ejection fraction. In addition, this variablewas the most powerful independent predictor of deaths that occurred suddenly,suggesting an important role for left ventricular dilatation in the genesis of life-threatening ventricular arrhythmias. After the introduction of thrombolytic ther-apy, ejection fraction has still remained an important determinant of arrhythmicevents.102 It has been suggested that its predictive value is independent of the re-sult of thrombolytic therapy.65 The prognostic importance of left ventriculardilatation in this setting is less well known.

3.3. Mechanisms of arrhythmogenesis in left ventricular dilatation

Although the importance of left ventricular dilatation for the occurrence ofventricular arrhythmias has been well recognized, knowledge of the underlyingmechanisms of these arrhythmias is incomplete. However, it is clear that the ar-rhythmogenic effect of dilatation is multifactorial in origin, and includes me-chanical, electrical, and neurohumoral components. There are indications thatspecialized stretch-activated membrane channels can induce a depolarizing cur-rent during diastole, leading to stretch-induced afterdepolarizations (SIAs)which may trigger one or more ectopic beats.103 Recently, the existence ofstretch-induced arrhythmias was demonstrated in the isolated canine left ventri-cle.104 In this study by Hansen et al.,104 a sudden increase in diastolic volume(induced by a intracavitary balloon) induced frequent premature ventricularcontractions, occasional ventricular couplets, and nonsustained VT. In the iso-lated Langendorff-perfused rabbit heart, Franz et al.89 showed that both gradualand sudden increases of left ventricular volume induced a reduction of themembrane potential. However, premature ventricular contractions were espe-cially seen in the case of a sudden volume increase.

Next to SIAs and subsequent ventricular arrhythmias, effects on duration ofrepolarization have been described in terms of an increase in volume. Reiter etal.105 described the electrophysiological effects of acute ventricular dilatation inthe isolated rabbit heart. They found that an increase in left ventricular volumedid not affect pacing threshold or conduction velocity, but instead reduced theeffective refractory period in a heterogeneous manner thereby increasing thetemporal dispersion in recovery. In the infarcted ventricle, Calkins et al.106 noteda more pronounced reduction of the refractory period in infarcted compared tononinfarcted sites of the ventricle when volume load was applied. This in-creased dispersion in refractoriness was paralleled by a conversion from nonin-ducible to inducible VT in 4 of 8 ventricles when volume loading was applied.In humans, effects of volume and/or pressure loading on repolarization charac-

INTRODUCTION

17

teristics have been demonstrated during balloon angioplasty,107 performance ofthe Vasalva maneuver,108 and after weaning from extracorporal life support.109

In these studies, acute changes of left ventricular loading were accompanied bya reduction of the refractory period and increased dispersion in refractoriness.109

In addition, the appearance of EADs capable of inducing triggered activity wasnoted.107,108 In these human studies DADs were not observed. However in ani-mal studies, an increase in the amplitude of DADs has been observed when ap-plying stretch to myocardial tissue.110

Most of the above-mentioned studies have described electrophysiologicalchanges in reaction to sudden changes in volume or pressure load in noninfarc-ted ventricles. Whether these changes persist after a longer period of increasedloading and also apply in the setting of acute myocardial infarction remains un-known. In the latter case, slowly progressive left ventricular dilatation, charac-terized by cell slippage, may lead to cellular uncoupling,111 which in turn mayinduce slow conduction and increased dispersion in refractoriness, possiblyadding to the development of late ventricular arrhythmias.

3.4. Effect of thrombolytic therapy on left ventricular dilatation and relatedventricular arrhythmias

Thrombolytic therapy interferes in the relation between remodeling and ven-tricular arrhythmias on several levels. Infarct size is reduced after successfulthrombolysis,58 thus reducing the probability of an anatomical substrate forventricular arrhythmias to be formed. In addition, experimental studies haveshown that reperfusion several hours after coronary occlusion reduces infarctexpansion independent of infarct size.112,113 Accelerated healing of the infarctedarea has been proposed as the most likely underlying mechanism.112 This proc-ess may explain the reduced incidence of ventricular aneurysm after throm-bolytic therapy.85 In addition, less compensatory hypertrophy will occur, andareas of the ventricle showing delayed conduction will be limited. Together witha reduction of infarct size, this finding may explain the lower incidence of latepotentials in patients with successful reperfusion.114-117 Furthermore, left ven-tricular dilatation is reduced when thrombolytic therapy results in a patent in-farct-related artery.60,118

In summary, the process of left ventricular remodeling has several aspectswhich may promote the occurrence of ventricular arrhythmias after myocardialinfarction. Left ventricular dilatation is of particular interest because this aspectof remodeling is clearly related to the occurrence of life-threatening postinfarc-tion ventricular arrhythmias. The underlying electrophysiological mechanisms

CHAPTER 1

18

of this relation are not fully elucidated, but may include delayed conduction anddispersion in refractoriness. The reduced incidence of ventricular arrhythmiasafter thrombolytic therapy may be explained, at least in part, by its modulatingeffects on left ventricular remodeling.

§ 4. Neurohumoral activation as an etiologic factor for early and late ventricular arrhythmias

Neurohumoral activation after myocardial infarction includes increased activ-ity of the sympathetic nervous system and the renin-angiotensin-aldosterone(RAA) system. Under normal circumstances, these systems are balanced by

Hours-20 0 20 40 60 80 100 120

600

700

800

900

1000

1100

1200

1300

1400

0

1

2

3

4

5

6

7

8

Norepinephrine

pg/ml U/l.h

Plasma renin activity

Figure 1.2 . Neurohumoral activation, quantified by norepinephrine levels and plasmarenin activity, is shown during the first few days after myocardial infarction in patients al-located to placebo in CATS. A clear-cut reduction of norepinephrine levels is seen duringthe first 24 hours, whereas plasma renin activity does not increase until after this period.

INTRODUCTION

19

parasympathetic activity and atrial natriuretic peptide. In the setting of acutemyocardial infarction this balance is lost and a relative increase of sympathetictone and activation of the RAA system is observed both of which may have ar-rhythmogenic effects.119,120 Increased sympathetic activity is usually seen up to24-48 hours after infarction, whereas activation of the (systemic) RAA system isnot observed until after this period (Figure 1.2).

4.1. Sympathetic nervous system activity and ventricular arrhythmias afteracute myocardial infarction

The role of increased sympathetic activity in the cause of ventricular ar-rhythmias is well established.121,122 An increase in sympathetic tone can promoteventricular arrhythmias by inducing hypokalemia,123 increasing heart rate, andby enlarging ischemic areas.124 Electrophysiological effects include an increasein firing rate in the case of abnormal automaticity, and the induction of triggeredactivity by augmenting EADs and DADs.125 It is generally accepted that theelectrophysiological effects of beta-adrenergic stimulation are predominantlymediated through cyclic adenosine monophosphate (cAMP).126 These effectsinclude augmentation of the plateau phase and an increase in the rate of repo-larization resulting in a shorter action potential duration. Reentry can occur dueto the heterogeneous distribution of sympathetic nerve fibers.127 Under normalcircumstances an increase in sympathetic tone does not lead to changes in thedistribution of refractory periods. However, in diseased (e.g., infarcted) ventri-cles regional dysfunction of sympathetic nerves can lead to increased dispersionin refractoriness. Calkins et al.128 recently reported on 11 patients with life-threatening ventricular arrhythmias referred for placement of an implantable de-fibrillator. The pattern of sympathetic innervation was assessed using scintigra-phy with 11C-hydroxyephedrine; regional refractory periods were measured in-traoperatively. The investigators showed a clear correlation between regionalsympathetic dysfunction and ventricular refractoriness. This may suggest thatdifferences in refractory period could increase in the case of sympathetic stimu-lation, leading to a higher probability of successful reentry as a mechanism forventricular arrhythmias.

Increased sympathetic activity early after myocardial infarction

In experimental models it has been shown that norepinephrine levels are ele-vated within one minute of coronary artery occlusion.129 The role of norepi-nephrine in the incidence of ventricular arrhythmias is supported by studies thatdemonstrate a lower threshold for VF upon direct stimulation of sympathetic

CHAPTER 1

20

nerves in the setting of myocardial ischemia.3 In addition, complete abolition ofventricular arrhythmias has been shown after denervation of the heart beforeligation of a coronary artery.130,131 In humans, activation of the sympatheticnervous system has also been observed early after the onset of myocardial in-farction.132,133 The extent and duration to which sympathetic activity is increasedhas been found to relate to infarct size and degree of left ventricular dysfunc-tion. In a study by Benedict et al.,134 epinephrine and norepinephrine levels re-turned to normal by the third day in patients with uncomplicated myocardial in-farction. However, in patients with cardiogenic shock these levels did not nor-malize. Sigurdsson et al.135 observed in 55 patients participating in the secondCOoperative North Scandinavian ENalapril SUrvival Study (CONSENSUS II)that catecholamines returned to normal within a few days in patients with un-complicated myocardial infarction, but remained elevated in patients with signsof congestive heart failure. In the Survival And Ventricular Enlargement(SAVE) study,136 which investigated patients with left ventricular dysfunctionbut no overt heart failure after myocardial infarction, a wide variety of sympa-thetic activity was found. Some patients with a low ejection fraction who re-quired use of diuretics showed no neurohumoral activation, whereas other pa-tients with a relatively preserved ejection fraction and no use of diuretics did.Still, ejection fraction proved an independent predictor of norepinephrine andepinephrine levels before hospital discharge despite these exceptions. Elevatedcatecholamines shortly after myocardial infarction appear to have prognosticvalue for cardiac events during follow up. In 12 patients with acute myocardialinfarction, Karlsberg et al.133 found that all patients with peak epinephrine val-ues > 1000 pg/ml died during 18 months of follow up, while survivors had val-ues < 1000 pg/ml. In addition, Sigurdsson et al.135 found significantly highernorepinephrine levels in the 6 out of 55 patients (11%) who died compared tosurvivors during 6 months of follow-up care (909 pg/ml vs 545 pg/ml). Theauthors attributed the difference in norepinephrine mainly to a difference in theextent of myocardial damage.

Heart rate variability early after myocardial infarction

Heart rate or heart period variability represents a measure of the degree ofmodulation of autonomic tone, in particular of parasympathetic activity.137 Itscomponents are expressed as variables from the time domain (e.g., standarddeviation of all RR intervals recorded in 24 hours)138 or frequency domain(high-and low-frequency components, obtained by spectral analysis).139 Al-though the exact underlying connection to activity of the autonomic nervoussystem has not been elucidated,140 evidence is available that reduced heart rate

INTRODUCTION

21

variability represents sympathovagal imbalance and has a clear value in predict-ing mortality138 and arrhythmic events57,141,142 after myocardial infarction. Ca-solo et al.143 found a reduced heart rate variability, measured on the second orthird day after myocardial infarction, in patients with a large enzymatic infarctsize, reduced left ventricular function, and Killip heart failure class ratingshigher than I (for Killip classification, see box on this page). In contrast, patientswith non-Q-wave infarction and those treated with rtPA showed a significantlyhigher heart rate variability. In this study, six patients died within 20 days ofwhom four died suddenly. These patients showed a significantly lower heart ratevariability compared to survivors. The contribution of reperfusion after throm-bolytic therapy was not clear, although patients treated with rtPA demonstrateda higher heart rate variability.

In conclusion, increased sympathetic activity in the early stages after myo-cardial infarction is associated with an adverse prognosis during follow up, dueto an increased incidence of arrhythmic events.

Increased sympathetic activity late after myocardial infarction

As stated previously, catecholamine levels return to normal after a few daysin uncomplicated myocardial infarction. However, in patients with reduced leftventricular function and clinical signs of heart failure, sympathetic activity re-mains elevated. Sigurdsson et al.135 measured norepinephrine levels up to one

month after myocardial infarction and found significantly higher levels in pa-tients with congestive heart failure. The difference in norepinephrine levels,compared to patients without heart failure, was further augmented when head-

KILLIP CLASSIFICATION

I. Absence of third heart sound. Absence of rales over lungfields.

II. Rales over < 50% of the lung fields or third heart sound pr e-sent.

III. Rales over ≥ 50% of the lung fields. Frequently pulmonaryedema.

IV. Shock.

CHAPTER 1

22

up tilt testing was applied. The authors attributed this effect to altered cardiovas-cular control mechanisms or hypovolemia due to diuretic therapy in these pa-tients. The site of myocardial infarction also appears to affect the pattern of re-covery of autonomic function. Flapan et al.144 assessed heart rate variability in20 patients with anterior or inferior myocardial infarction. They found that heartrate variability was relatively preserved in patients with inferior myocardial in-farction, whereas a reduction of heart rate variability up to 6 weeks after admis-sion was observed in patients with anterior myocardial infarction. Other studieshave confirmed this prolonged sympathovagal imbalance in patients with ante-rior myocardial infarction.145,146

Most studies assessing autonomic function by use of heart rate variability doso before hospital discharge to risk-stratify patients for life-threatening ar-rhythmias during follow-up care. In a large study of more than 400 patients,Farrell et al.57 showed that a reduced heart rate variability before discharge hadimportant value in predicting arrhythmic events independent of residual leftventricular function. These authors suggested that the latter may be explainedby selective destruction of autonomic nerves without significant loss of leftventricular function in some patients. This situation could lead to sympathova-gal imbalance with a reduced threshold for VF as a possible consequence.

Although persistent activation of the sympathetic nervous system is mostlyencountered in patients with left ventricular dysfunction and/or signs of conges-tive heart failure, sympathovagal imbalance may also persist in some patientswithout cardiac failure. This may explain, at least in part, the occurrence of life-threatening ventricular arrhythmias in patients with relatively preserved leftventricular function.

4.2. Renin-angiotensin-aldosterone system activation and ventricular ar-rhythmias after acute myocardial infarction

Although the RAA system is not well known for its arrhythmogenic effects,there are some indications that an activated RAA system has electrophysiologiceffects that may promote the occurrence of ventricular arrhythmias. De Langenet al.150 observed a shortening in sinus rhythm cycle length and refractory pe-riod upon infusion of angiotensin II in healthy pigs. In addition, the same inves-tigators were able to induce sustained VT after angiotensin II infusion in 5 of 9previously noninducible pigs 2 weeks after myocardial infarction. Fleetwood etal.151 demonstrated in the isolated rat heart that the duration of VF upon reper-fusion was significantly reduced by a selective angiotensin II receptor antago-nist. The authors suggested that VF was maintained by locally-produced angio-

INTRODUCTION

23

tensin II which could explain the beneficial effects of ACE inhibitors on reper-fusion VF.

Local renin-angiotensin system in cardiac tissue

There is now convincing evidence available for the existence of a local renin-angiotensin system in the heart152 which may explain the findings mentioned inthe previous paragraph. Besides direct electrophysiological effects, locally-produced angiotensin II interferes with the release of norepinephrine from sym-pathetic nerve endings. Angiotensin II promotes presynaptic norepinephrine re-lease, blocks presynaptic norepinephrine reuptake, increases catecholaminesynthesis, and potentiates postsynaptic actions of norepinephrine.153 This activ-ity may further enhance the propensity towards ventricular arrhythmias bymechanisms described in paragraph 4.1. In addition, angiotensin II has impor-tant vasoconstrictive properties which could expand the ischemic area in thecase of acute myocardial infarction.20 Ertl et al.154 demonstrated that collateralflow was increased by ACE inhibition during coronary artery occlusion, result-ing in a smaller infarct size in 21 anesthetized dogs. However, a bradykinin-mediated mechanism could not be excluded in this study. Angiotensin II alsohas properties as a local growth factor, and may be partly responsible for thecompensatory left ventricular hypertrophy observed after myocardial infarc-tion.155 Left ventricular hypertrophy, in turn, is a well-known risk factor forventricular arrhythmias (paragraph 3.2). It has been shown recently that thepropensity for ventricular hypertrophy may be mediated by genetic disposition.Schunkert et al.156 found a clearly-increased risk for left ventricular hypertrophyin normotensive patients who were homozygous for the deletion (D) allele ofthe ACE gene. This genotype is associated with high levels of ACE and is fre-quently present in patients with coronary artery disease.157 In addition, patientswith this genotype are highly prone to left ventricular dilatation in the first yearafter myocardial infarction.158 This result may stem from the form of hypertro-phy that results in lengthening of myocytes.159 Considering these factors, an in-creased incidence of ventricular arrhythmias may be anticipated in patients thatare homozygous for the D allele of ACE.

4.3. Effect of thrombolytic therapy on neurohumoral activation

Effect on sympathetic activity

In animal studies, an increase in norepinephrine overflow is observed uponreperfusion of a previously ligated coronary artery.9 Depending on the duration

CHAPTER 1

24

of preceding ischemia, this finding is paralleled by various types of reperfusionarrhythmias23 (see paragraph 1). In humans, data on norepinephrine levels earlyafter thrombolytic therapy for acute myocardial infarction is limited. Sigurdssonet al.135 assessed norepinephrine levels within 24 hours after the onset of symp-toms of acute myocardial infarction in 34 patients. There was no difference innorepinephrine levels between the 18 patients who were treated and the 16 pa-tients who were not treated with intravenous streptokinase. However, patientsnot receiving thrombolytic therapy were included significantly later after the on-set of symptoms, which may have disturbed the relation between reperfusionand norepinephrine levels. Moreover, angiographic data were not available inthis study. Zabel et al.147 measured heart rate variability within the first hour ofthrombolytic therapy. They found a significant increase in parasympathetic ac-tivity in patients with successful thrombolysis compared to those without suc-cessful thrombolytic therapy. Other studies describing the effect of thrombolytictherapy on heart rate variability late after myocardial infarction have shownsimilar results.61,148 In addition, Pedretti et al.61 showed that this effect on auto-nomic tone was paralleled by a decrease in inducible and spontaneously-occurring ventricular arrhythmias. This result implies that a considerable dis-crepancy exists between experimental and clinical findings. The release of no-repinephrine upon reperfusion in animal experiments has been attributed to anumber of factors that lead to depolarization of sympathetic nerve endings, in-cluding hyperkalemia, acidosis, hypoxia, and formation of membrane-activemetabolites.8 This local release of norepinephrine may be small in comparisonto the systemic release of neurohormones caused by hemodynamic stress duringacute myocardial infarction. Therefore in the case of successful thrombolysis,the hemodynamic benefit of reperfusion may outweigh local overflow of nore-pinephrine, resulting in lower overall systemic levels of this neurohormone. Ithas been shown that thrombolysis prevents early left ventricular dilatation,149 acondition which would otherwise require an increase in sympathetic activity tomaintain a sufficient stroke volume. This conclusion is supported by the find-ings of Zabel et al.,147 who showed significantly lower heart rates in patientswith a patent infarct-related artery during the first hour after thrombolytic ther-apy (70 ± 12 beats/minute vs 80 ± 13 beats/minute, P=0.003).

Effect on the RAA system

Data on the effects of thrombolytic therapy on the RAA system are limited.Thrombolytic therapy may prevent activation of the systemic RAA system bylimiting infarct size58 and preventing early and late left ventricular dilatation.149

However, Nabel et al.160 still found elevated renin and angiotensin II levels up to

INTRODUCTION

25

7 days after thrombolytic therapy in 29 patients, 28 (97%) of whom had a patentinfarct-related artery (successful PTCA for an occluded infarct-related arterywas performed in eight patients). Thus, even when successful, reperfusion cannot fully prevent activation of the RAA system after acute myocardial infarction.

In summary, both sympathovagal imbalance and activation of the RAA sys-tem may increase the propensity towards ventricular arrhythmias after acutemyocardial infarction. A relative increase of sympathetic activity reduces thethreshold for ventricular arrhythmias by inducing hypokalemia, increasing heartrate, and augmenting EADs and DADs. Sympathetic activity usually returns tonormal after a few days, but remains increased in patients with reduced leftventricular function and/or signs of heart failure. However, locally-damagedautonomic nerve fibers may also result in more dispersion in refractoriness inpatients with preserved left ventricular function, hereby increasing the risk ofventricular arrhythmias. Studies using heart rate variability have shown thatthrombolytic therapy reduces sympathovagal imbalance both early and late aftermyocardial infarction, paralleled by a reduction of inducible and spontaneouslyoccurring ventricular arrhythmias. RAA activation after myocardial infarctionmay may also increase the incidence of ventricular arrhythmias, primarilythrough the effects of angiotensin II. This compound has been demonstrated toreduce the refractory period and increase heart rate. In addition, the well-knownvasoconstrictive effect of angiotensin II may also increase oxygen demand andthus promote ischemia, a powerful arrhythmogenic factor. Finally, long lastingRAA activation leads to left ventricular hypertrophy, which in turn is associatedwith an increased risk of ventricular arrhythmias. Recent evidence suggests thatgenetic predisposition plays a role in the development of left ventricular hyper-trophy. Thrombolytic therapy may reduce, but not fully prevent activation of theRAA system after myocardial infarction.

§ 5. Interrelation between neurohumoral activation and left ventricularremodeling

Within seconds after the occlusion of a coronary artery by use of a PTCAcatheter, wall motion abnormalities can be observed, even before the appear-ance of electrocardiographic abnormalities or anginal complaints.72 In addition,increased plasma levels of norepinephrine can be detected within 1 minute ofcoronary artery occlusion.129 Thus it appears that both activation of the sympa-thetic nervous system and left ventricular dysfunction commence rapidly after

CHAPTER 1

26

the onset of coronary occlusion. After 20-30 minutes, ischemia becomes irre-versible, and myocardial infarction is completed in approximately 3-6 hours.161

In patients with small infarcts without clinical signs of heart failure, sympatheticactivity gradually returns to normal in the following few days.132 In contrast, ac-tivation of the RAA system remains present even in limited infarctions.160 In pa-tients with large infarcts (e.g., in patients with congestive heart failure) the sym-pathetic nervous system remains activated alongside an activated RAA sys-tem,132 presumably to maintain cardiac output. However, both systems increasewall stress of the left ventricle by various mechanisms and may thus accelerateremodeling and left ventricular dilatation in particular resulting in a poor out-come.133,134 In intermediate-sized infarctions, sympathetic activity may initiallynormalize but increase later when left ventricular remodeling has led to a re-duced left ventricular function. Alternatively, neurohumoral activation maypersist and add to infarct expansion and subsequent left ventricular dilatation.The degree of neurohumoral activation during the first days to weeks may becrucial for the resulting degree of left ventricular dilatation since the infarctedarea is particularly sensitive to straining forces during this period.

Interaction on the cellular level: consequences for cardiac electrophysiology

Beau and Saffitz162 recently reported on regional sympathetic activity in fail-ing human hearts, obtained after cardiac transplantation. They showed that theuptake of labeled norepinephrine in dilated ventricles was five-fold lower insubendocardial regions compared to subepicardial regions. The investigatorspostulated that a reduced uptake of norepinephrine in subepicardial regionswould result in local down-regulation of beta-adrenergic receptors, which couldlead to increased dispersion in refractoriness. Indeed, (regionally but not trans-murally) increased dispersion in refractoriness, quantified by increased QT dis-persion, has been reported in patients with chronic heart failure.163 In this par-ticular study, this finding did not translate in an increased incidence of ventricu-lar arrhythmias. However, the heterogeneous distribution of sympathetic activityappears to be an attractive additional explanation for the high incidence of ven-tricular arrhythmias in patients with left ventricular dilatation.10

§ 6. Rationale for the use of ACE inhibition during thrombolytic therapy

Although no direct anti-arrhythmic effects of ACE inhibitors have been de-scribed,164 an indirect reduction of ventricular arrhythmias may be anticipatedwhen these compounds are used in the setting of acute myocardial infarction. Ingeneral, blunting of neurohumoral activation, limitation of myocardial damage,

INTRODUCTION

27

and subsequent remodeling of the left ventricle may well result in a reduction ofventricular arrhythmias after acute myocardial infarction.

6.1. Effect on early ventricular arrhythmias

Experimental evidence

The importance of neurohumoral activation for the occurrence of ventriculararrhythmias in the early phases of coronary occlusion and reperfusion has beenrecognized for many years. In 1859 Einbrodt demonstrated that vagal nervestimulation could protect the canine ventricle against VF.1 In addition, later ex-periments showed that cardiac denervation can prevent the occurrence of VFupon reperfusion.130,131

More recently, Sheridan et al.165 found that the release of norepinephrine wasmediated by locally-produced angiotensin II. These findings raised the interestin determining the role of the renin-angiotensin system in the genesis of reper-fusion arrhythmias. In the early 1980s van Gilst et al.7 studied the effect of cap-topril on reperfusion arrhythmias in the isolated rat heart. They found a clear-cutreduction in the incidence and duration of VF which appeared to follow a dose-response relation.166 This change was paralleled by a significant reduction ofmyocardial injury, quantified by purine loss, and outflow of catecholamines.However, the authors suggested that this result flowed from an angiotensin II-independent mechanism, since angiotensin II was not retrieved from the coro-nary effluent. Another study by the same group demonstrated that the anti-arrhythmic effects of captopril were completely abolished by indomethacinwhich suggested a possible prostaglandin-dependent mechanism.8 ProstacyclinPGI2 is known to inhibit the release of norepinephrine167 which could explainthese findings. In the same study, enalapril did not appear to affect the occur-rence of ventricular arrhythmias. However, other investigators have shown thecardioprotective effects of enalapril168 and ramipril,169 both ACE inhibitors with-out a sulfhydryl moiety. Grover et al.170 compared seven ACE inhibitors withdifferent affinities for the local RAA system and agents with and without a sulf-hydryl moiety. These authors found that the cardioprotective effects of ACE in-hibitors were related to the presence of a sulfhydryl moiety and not to the de-gree of local ACE inhibition. Potentially beneficial effects of a sulfhydryl moi-ety may include scavenging of free radicals during reperfusion171 and coronarydilatation.172

Clinical evidence

CHAPTER 1

28

In humans, the effects of ACE inhibition on early ventricular arrhythmias af-ter myocardial infarction, with or without thrombolytic therapy, have been in-vestigated in a limited number of studies. Ambrozioni et al.173 investigated theeffect of zofenopril therapy started within 24 hours after the onset of symptomsin patients not eligible for thrombolytic therapy in the Survival of MyocardialInfarction Long-term survival Evaluation (SMILE) pilot study (Table 1.4). Noeffect on the incidence of ventricular arrhythmias was reported. Ray et al.174

studied the effects of captopril therapy administered 6-24 hours (mean of 15)after the onset of symptoms of acute myocardial infarction. None of these pa-tients received thrombolytic therapy. There was no difference in the incidenceof ventricular arrhythmias during hospitalization. In addition, no effect on levelsof catecholamines was seen.

The investigators did observe a reduction of left ventricular dilatation during13 months of follow-up care. In the pilot study of the fourth International Studyof Infarct Survival (ISIS-4) Pilipis et al.175 investigated the effects of captopriland mononitrate started a mean of 13 hours after the onset of symptoms therapyon the incidence of ventricular arrhythmias. The number of ventricular prema-ture beats and VT during 48-hour Holter monitoring was reduced in patientstreated with captopril or mononitrate. However, only the effect of mononitrateon premature beats reached statistical significance. In this pilot study, 92 of the100 study patients received thrombolytic therapy. Bussmann et al.176 studied theeffect of captopril on infarct size and ventricular arrhythmias in a placebo-controlled study of 46 patients with acute myocardial infarction. Thrombolytictherapy was given in 23 out of 46 patients (50%). Captopril was administeredintravenously 2-18 hours (mean of 10) after the onset of symptoms. In thetreated group, there was a significant reduction in the number of ventricular

Table 1.4. Effect of ACE inhibitors on early ventricular arrhythmias after acute myocardial infarction

Author Compound N Start (h) Thrombolysis

Ambrosioni220 zofenopril 204 < 24 noneRay174

Pipilis175

Bussman176

Kingma178

captoprilcaptoprilcaptoprilcaptopril

991004624

15 13 10 3

none92%50%

100%Di Pasquale177 captopril 72 ≤ 4 100%

AIVR indicates accelerated idioventricular rhythm; FU, follow up; h, hours; i.v., intravenous;

INTRODUCTION

29

premature beats in 48-hour Holter recordings. In addition, seven patients in theplacebo group had VF in the acute phase, compared to none in the treatedgroup. Di Pasquale et al.177 used a different study design. These investigatorscompared in a total of 72 patients the effect of captopril administered 15 min-utes before thrombolytic therapy with captopril administration 3-4 days afterurokinase infusion. A significant reduction in early ventricular arrhythmias, ob-served within 2 hours after the onset of thrombolytic therapy, and predischargeventricular arrhythmias was seen in the early treatment group compared to thelate treatment group. From these studies it appears that a greater effect fromcaptopril therapy can be expected when it is started earlier. The findings of DiPasquale et al.177 are especially interesting in this respect.

The only other study describing the effect of captopril on the occurrence ofventricular arrhythmias during thrombolysis was the CATS pilot study.178 In thisstudy, a low dose of captopril administered either orally or intravenously wasgiven during thrombolysis. The incidence of nonsustained VT during the first 4hours after onset of therapy was reduced in the orally-treated group but not inthe intravenously-treated group. In the orally-treated group, this effect wasparalleled by a significant reduction in plasma norepinephrine levels. In con-trast, norepinephrine levels appeared to increase in the intravenously-treatedgroup. This finding may have been induced by a serious drop in blood pressureobserved in the latter group. The investigators attributed this effect to an inter-action of intravenous captopril with streptokinase which both are known to in-tervene with the bradykinin metabolism resulting in a bradykinin-induced hy-potensive response. Significantly more hypotension was also observed in theCONSENSUS II study in which patients were treated with intravenous enalaprilwithin 24 hours after myocardial infarction.179 The CONSENSUS II investiga-tors suggested that this early hypotensive reaction was responsible for the ex-cess mortality in the active-treatment group. Therefore, when ACE inhibition isapplied soon after myocardial infarction with or without thrombolytic therapy,the oral route of administration is clearly preferable.

CHAPTER 1

30

6.2. Effect on late ventricular arrhythmias

Cardiac electrophysiology - results of programmed stimulation

As stated earlier, ACE inhibitors do not have direct anti-arrhythmic proper-ties. However, effects on cardiac electrophysiology have been described.Kingma et al.180 investigated the effects of ACE inhibition on the inducibility ofventricular arrhythmias in pigs. Induction of VT was prevented in all six previ-ously inducible pigs one week after myocardial infarction. There was a trendtowards a reduction of the refractory period (253 ms vs 228 ms, not significant).The authors attributed the anti-arrhythmic effects of captopril to an indirectblocking effect on norepinephrine release mediated by angiotensin II. In anotherstudy by the same group,150 angiotensin II was infused in five healthy pigs. Incontrast to the previous study, a significant decrease in refractory period (18%)was observed. Furthermore, angiotensin II was administered in 17 pigs withmyocardial infarction produced by coronary artery occlusion using a ballooncatheter. Of nine previously noninducible animals, VT was inducible in fiveanimals after infusion of angiotensin II and an additional increase in spontane-ously occurring ventricular arrhythmias was observed. Of eight inducible pigs,five animals were no longer inducible after captopril infusion. The authors at-tributed this effect of captopril to the prevention of the deleterious effects ofangiotensin II which had caused a reduction of refractoriness in parallel to anincrease in heart rate and blood pressure.

Another pathway by which ACE inhibitors may reduce ventricular arrhyth-mias is via an increase in bradykinin levels. Since ACE is identical to kininase

(Table 1.4 continued)

Protocol FU (h) Early VAs

ineligible for thrombolysis 24 no effectsmall infarcts excludedsuspected infarctioni.v. captoprili.v. or oral captopril

4848244

no effectVT ↓ (NS) and AIVR ↓

VPBs ↓ VF ↓VT ↓ (NS)

randomization early vs late 2 (VF + VT + VPBs) ↓

NS, not significant; VAs, ventricular arrhythmias.

INTRODUCTION

31

II, the enzyme responsible for bradykinin breakdown,181 ACE inhibition willalso lead to an increase in bradykinin levels. In six pigs with inducible sustainedVT after myocardial infarction, Tobé et al.182 infused bradykinin to study itselectrophysiological effects. No effect on the refractory period was seen. How-ever, after bradykinin infusion, 4 of 6 pigs were no longer inducible. Theauthors suggested that a significant drop in blood pressure may have reducedwall stress which in turn could have prevented inducibility. This electrophysi-ological effect of a reduction in wall stress has been previously described.106

Only one small study has investigated the electrophysiological effects of ACEinhibition in humans. In eight patients with previous myocardial infarction andinducible VT, Bashir et al.183 serially studied the electrophysiological effects ofcaptopril and the combination of hydralazine and isosorbide dinitrate. The meanejection fraction of these patients was 24%. A significant increase in refractoryperiod was measured after captopril therapy but not after hydralazine-nitratetherapy. However, no effect on inducibility was found. The difference betweencaptopril and hydralazine-nitrate therapy was explained by a difference in sym-pathetic activity. After hydralazine-nitrate administration, an increase in heartrate from 68 beats/minute to 81 beats/minute was seen, whereas heart rate aftercaptopril therapy was unchanged. The beneficial effect of unloading of the heartwas probably counteracted by increased sympathetic activity during hydra-lazine-nitrate treatment, whereas this increase was blocked during captopriltreatment.

Modifying the anatomical substrate - modulation of the remodeling process

Besides effects on cardiac electrophysiology caused by acute or subacutechanges in loading conditions and modulation of sympathetic activity, other ef-fects of ACE inhibitors may result in a reduction of late ventricular arrhythmias.For instance, a reduction in infarct size may reduce the probability of an ana-tomical substrate for ventricular arrhythmias to be formed, thus reducing thelikelihood of the occurrence of late ventricular arrhythmias.82 There are severalstudies indicating that ACE inhibitors can reduce infarct size. Ertl et al.154 pro-vided evidence that this result could be explained by a beneficial effect oncoronary flow. In addition, Martorana et al.184 showed that the reduction in in-farct size can be abolished by a bradykinin antagonist. Tobé et al.185 furtherdemonstrated the importance of bradykinin in limiting infarct size. Infusion ofbradykinin in pigs with a myocardial infarction produced by a balloon catheterin the left coronary artery resulted in significantly lower creatine kinase levelsand a reduction in late potentials after 2 weeks (compared to a saline-treated

CHAPTER 1

32

group). In addition, this finding was associated with a trend towards a reductionin inducible sustained VT. After thrombolytic therapy, ACE inhibitors contain-ing sulfhydryl groups may theoretically have additional benefit by scavengingfree radicals which may lead to less reperfusion damage.171 However, the con-tribution of reperfusion injury to total infarct size is still being questioned.186

As discussed in paragraph 3.2, left ventricular remodeling after myocardialinfarction can represent an important factor in the cause of ventricular arrhyth-mias. At present, numerous experimental11,187 and clinical studies71,188-191 with orwithout previous thrombolytic therapy,160 have shown the modulating effect ofACE inhibitors on left ventricular dilatation. To date, data on the effects of ACEinhibition applied during thrombolytic therapy on left ventricular dilatation islimited.177,178 In addition, experimental studies have shown that early captopriltreatment can prevent left ventricular hypertrophy after myocardial infarction.192

Finally, there are indications that a reduction in left ventricular hypertrophy byACE inhibition can contribute to a reduction of life-threatening ventricular ar-rhythmias.193

Neurohumoral activation and electrolytes

From studies investigating patients with heart failure it has become clear thatnorepinephrine levels are reduced during chronic treatment with ACE inhibi-tors.194,195 In addition, recent studies have demonstrated that ACE inhibition can

Table 1.5. Effect of ACE inhibitors on late ventricular arrhythmias in patients with heart failure

Author Compound N Selection Protocol Follow up Effect on VAs

Cleland 1984194 captopril 14 NYHA 3-4 double-blind, crossover 2 months VPBs, pairs, VT ↓Cleland 1985195 enalapril 20 NYHA 2-4 double-blind, crossover 1,5 months VPBs ↓Webster200 enalapril 19 NYHA 2-3 double-blind 3 months VPBs, pairs, VT ↓Captopril-Digoxin201

captopril 300 NYHA 2 double-blind 6 months VAs ↓ vs digoxin

De Graeff202 captopril 12 NYHA 3-4 open design 3 months no effect on VAsramipril

Cocchieri203 captopril 22 NYHA 1-2 double-blind 3 months no effect on VAsBechler-Lisinska204 captopril 50 NYHA 3-4 open design 1 month VPBs, VT ↓Kleber205 captopril 93 NYHA 1-3 double-blind 32 months no effect on VPBsPomini206 enalapril 30 NYHA 3-4 open 2 months VPBs, pairs, VT ↓Poquet207 captopril 47 NYHA 2-3 open 1 month VPBs, pairs, VT =Gurlek208 enalapril 24 NYHA 3 double-blind 4 weeks VPBs, pairs, VT ↓Adgey209 lisinopril 156 NYHA 3-4 blinded data, pre-post 24 weeks VT ↓

enalapril EF < 35% treatment

EF indicates ejection fraction; NYHA, New York Heart Association class. Other abbrevations see Table 1.4.

CHAPTER 1

2

Table 1.6. Effect of ACE inhibitors on late ventricular arrhythmias and sudden death

Author Compound N Patient selection Protocol Follow up Effect on VAs

Sudden deathCONSENSUS210

Fonarow211

Newman212

SOLVD I213

MHFT205

VHeFT II214

SOLVD II217

SAVE218

SMILE220

CONSENSUSII17

9

Late VAsSøgaard221

enalaprilcaptoprilcaptoprilenalaprilcaptopril

enalaprilenalaprilcaptopril

zofenoprilenalapril

captopril

253117105

2569170

80442282231

15566090

58

NYHA 4NYHA 3-4NYHA 2-3NYHA 2-3NYHA 2

EF < 45%EF ≤ 35%

AMI,EF ≤ 40%

< 24 h AMI< 24 h AMI

AMI,EF ≤ 45%

vs placebovs hydralazine-ISDN

vs placebovs placebovs placebo

vs hydralazine-ISDNvs placebovs placebo

vs placebovs placebo

vs placebo

6 months8 months3 months41 months2.7 years

2 years37 months42 months

6 weeks6 months

6 months

no effect on SCDSCD ↓ 76% (23 - 92)SCD ↓ 87% (2 - 98)SCD ↓ 10% (NS)no effect on SCD

SCD ↓ 35% (6 - 55)SCD ↓ 7% (NS)SCD ↓ 19% (NS)

SCD ↓ 63% (NS)no effect on SCD

(VPB, pairs, VT) ↓

SAVE Holter219 captopril 553 AMI,EF ≤ 40% vs placebo 2 years VT ↓ at 1 year

AMI indicates acute myocardial infarction; SCD, sudden cardiac death; ISDN, isosorbide dinitrate. Other abbrevations see Table1.4.

INTRODUCTION

35

improve baroreflex sensitivity196 and heart rate variability197 after acute myo-cardial infarction. Both measures indicate that the relative dominance of sympa-thetic activity is reduced, which may contribute to a reduction of ventricular ar-rhythmias after myocardial infarction. Another potentially beneficial effect ofACE inhibitors lies in the modulation of low potassium levels. Hypokalemia isan important arrhythmogenic factor after myocardial infarction198 and is associ-ated with an increased incidence of VF. Underlying mechanisms include slowconduction by hyperpolarization, increased automaticity by increase in the rateof phase 4 depolarization, and induction of afterdepolarizations by reduction ofthe sodium/calcium exchange mechanism.199 An increase in potassium levels asa result of ACE inhibition may prevent ventricular arrhythmias provoked bythese mechanisms. This may be of special importance in the acute phase ofmyocardial infarction, when potassium levels are low due to stress and/or diu-retic treatment.

Evidence for reduction of late postinfarction arrhythmias in humans

Most evidence concerning the effect of ACE inhibitors on late ventricular ar-rhythmias is obtained from patients with heart failure, many of whom have leftventricular dysfunction caused by one or more previous myocardial infarc-tion(s).194,195,200-209 Table 1.5 lists a number of studies investigating the effect ofACE inhibition on ventricular arrhythmias in patients with heart failure. Most ofthese studies were relatively small and used Holter monitoring to detect ven-tricular arrhythmias. The ACE inhibitor most frequently used was captopril orenalapril. Most studies had a double-blind, placebo-controlled design, but someopen studies have also been reported. The follow-up period varied from 4weeks to 32 months. Patient selection, protocol, or duration of follow up did notseem to influence the effect of ACE inhibition on the incidence of ventricular ar-rhythmias. The majority of these studies showed some reduction of ventriculararrhythmias. However, other studies showing no effect may not have been pub-lished.

The effect of ACE inhibitors on sudden cardiac death may also add to thediscussion of the anti-arrhythmic effect of these compounds. A number of trialsthat investigated the effect of ACE inhibitors on total mortality and sudden car-diac death are listed in Table 1.6. The first large trial that studied the effects ofACE inhibition on mortality in patients with serious heart failure (NYHA classIV) was the CONSENSUS study210 (for NYHA classification, see box on op-posite page). This study demonstrated a beneficial effect of enalapril on totalmortality, but not on sudden cardiac death. Fonarow et al.211 compared the ef-fects of Hydralazine combined with isosorbide dinitrate with Captopril in 117

CHAPTER 1

36

patients evaluated for cardiac transplantation (the Hy-C trial). Despite similarhemodynamic effects, sudden cardiac death occurred more often in the hydra-lazine-isosorbide dinitrate group (28% vs 7%). Newman et al.212 described aclear reduction in mortality in 105 patients with moderate to severe heart failuretreated with captopril or placebo. Of the 11 patients that died, 8 died suddenlyand thus an effect on arrhythmogenic death was suspected. In contrast, theStudies Of Left Ventricular Dysfunction (SOLVD) investigators,213 who investi-gated patients with an ejection fraction of less than or equal to 35% and signs ofheart failure (treatment arm, SOLVD I), showed an effect of enalapril on totalmortality compared to placebo but not on sudden cardiac death. Similar to thestudies by Fonarow et al. and Newman et al., most of the patients in this studywere in NYHA heart failure classes II and III. The Munich mild Heart FailureTrial (MHFT), investigating 170 patients with only mild signs of heart failure(NYHA class II), found no difference in sudden death after 2.7 years of followup between patients treated with captopril or placebo. In the second Veteransadministration cooperative vasodilator Heart Failure Trial (VHeFT II),214 Cohnet al. compared the effects of hydralazine-isosorbide dinitrate therapy withenalapril therapy in patients with a reduced ejection fraction (< 45%). Most pa-tients were in NYHA class II or III. A significant reduction of sudden cardiacdeath was found in the enalapril group. In addition, this finding was paralleledby a reduction in VT detected during 4-hour to 8-hour Holter monitoring afterthree months and after one and two years.215 It is important to note that the fre-quency of VT in the hydralazine-isorbide dinitrate group was similar to the inci-dence of VT in the placebo group of the VHeFT I study,216 which had identicalentry criteria as VHeFT I. In SOLVD II (prevention arm),217 patients with a re-duced ejection fraction but no overt heart failure were investigated. In thisstudy, 80% of the patients had had a myocardial infarction before study entry.

INTRODUCTION

37

A small, not statistically significant reduction of sudden cardiac death was ob-served in patients treated with enalapril. The SAVE study218 also investigatedpatients with a reduced ejection fraction (40% or less) and no symptoms of heartfailure (NYHA class I). However, in contrast to SOLVD II, where patients withrecent myocardial infarction were excluded, SAVE patients were randomized 3-16 days after acute myocardial infarction. Although sudden death was reducedby 19% (similar to the effect on total mortality), this finding did not reach statis-tical significance. The SAVE investigators did observe a reduction of VT duringHolter monitoring at one year, but not at two years, in patients treated with cap-topril.219 In the SMILE study,220 the effect of zofenopril on mortality was inves-tigated in patients with anterior wall myocardial infarction who were not eligiblefor thrombolytic therapy. Patients were treated within 24 hours after the onset ofsymptoms. After six weeks, four patients had died suddenly in the zofenopril

group, compared to 11 in the placebo group (p-value not significant). TheSMILE investigators did observe a significant reduction of total mortality. InCONSENSUS II179 more than 6000 patients with acute myocardial infarctionwere randomly assigned to i.v. enalapril or placebo therapy within 24 hours ofsymptom onset. Of all patients, 56% received thrombolytic therapy. There wasno difference in the incidence of sudden cardiac death after 6 months (2.8% vs2.9%, in the treated vs the nontreated group, respectively). Finally, in a recentstudy, Søgaard et al.221 described the incidence of ventricular arrhythmias dur-ing 24-hour Holter monitoring in 56 patients with a reduced ejection fraction (≤45%) but no overt heart failure after acute myocardial infarction. During sixmonths of follow up, an increase in ventricular arrhythmias in the placebo group

NEW YORK HEART ASSOCIATION FUNCTIONAL CLASS

I. Patients without limitation of physical activity. Ordinary activitydoes not cause undo fatigue, dyspnea or angina pain.

II. Patients with slight (IIS) or moderate (IIM) limitation of physicalactivity who are comfortable at rest.

III. Patients with marked limitation of physical activity who arecomfortable at rest.

IV. Patients with inability to carry on any physical activitywithout discomfort. Symptoms may be present at rest.

CHAPTER 1

38

was observed, compared to a decrease in the captopril group resulting in a sig-nificant difference at 180 days.

In summary, most studies investigating the effects of ACE inhibition in pa-tients with heart failure do report on a beneficial effect on either or both suddencardiac death and ventricular arrhythmias. However, this effect does not alwaysreach statistical significance. This finding may result from the use of differentdefinitions of sudden cardiac death or from differences in study design and pa-tient selection. In patients with recent myocardial infarction, a reduced incidenceof ventricular arrhythmias has only been reported in patients with significant leftventricular dysfunction. In patients with relatively preserved left ventricularfunction, there has been no report of a reduction of ventricular arrhythmias.

§ 7. Aims of the thesis

In Figure 1.3, the issues that are addressed in this thesis are depicted. In thefollowing chapter, chapter 2, the association between dilatation of the left ven-tricle and the occurrence of ventricular arrhythmias is described. In chapters 3and 4, the possible underlying electrophysiological mechanisms of this associa-tion are investigated using body surface mapping (chapter 3) and signal-averaged electrocardiography (chapter 4). In chapter 5, the role of early sympa-thetic activity, quantified by norepinephrine levels, in the occurrence of earlyand late ventricular arrhythmias is discussed. In the same chapter, the effect ofsuccessful reperfusion on norepinephrine levels and ventricular arrhythmias isaddressed.

INTRODUCTION

39

In chapter 6, the interaction of neurohumoral activation, quantified as heartrate variability, and left ventricular dilatation is described. In chapter 7, weevaluate the effects of ACE inhibition on dilatation, neurohumoral activation,and ventricular arrhythmias up to three months after myocardial infarction. Inchapter 8, focus is on the effects of ACE inhibition on left ventricular dilatationand congestive heart failure up to 12 months after myocardial infarction. Inchapter 9, the effects of ACE inhibition on ventricular arrhythmias requiringtreatment is described. Finally, in chapter 10 our findings in CATS are summa-rized and commented.

In brief, the aims of this thesis were to address the following questions:

1) does ACE inhibition, administered during thrombolysis, reduce ventriculararrhythmias early after myocardial infarction, and if so, which mechanisms are

Figure 1.3. The associations investigated in this thesis are depicted. For explanation, seetext.

CHAPTER 1

40

operative ?2) What is the role of left ventricular dilatation and neurohumoral activation inthe cause of ventricular arrhythmias after thrombolytic therapy, and, in the caseof left ventricular dilatation, to what extent are slow conduction and dispersionin refractoriness explanatory mechanisms ?3) Can ACE inhibition modify these arrhythmogenic factors, and does this resultin a reduction of late ventricular arrhythmias ?

References

1. Einbrodt: Akademie der Wissenschaften (Wien). Sitsungsberichte 1859;38:345-359.

2. Kent KM, Smith ER, Redwood DR, Epstein SE: Electrical stability of acutelyischemic myocardium: influences of heart rate and vagal stimulation. Circulation1973;47:291-298.

3. Lown B, Verrier RL: Neural activity and ventricular fibrillation. N Engl J Med1976;294:1165-1170.

4. Hjalmarson A: International beta-blocker review in acute and postmyocardial i n-farction. Am J Cardiol 1988;61:26B-29B.

5. Roberts R, Rogers WJ, Mueller HS, Lambrew CT, Diver DJ, Smith HC, WillersonJT, Knatterud GL, Forman S, Passamani E, et al: Immediate versus deferred beta-blockade following thrombolytic therapy in patients with acute myocardial infar c-tion. Results of the Thrombolysis in Myocardial Infarction (TIMI) II-B Study. Cir-culation 1991;83:422-437.

6. Heidbüchel H, Tack J, Vanneste L, Ballet A, Ector H, Van de Werf F: Significanceof arrhythmias during the first 24 hours of acute myocardial infarction treated withalteplase and effect of early administration of a beta-blocker or a bradycardiac agenton their incidence. Circulation 1994;89:1051-1059.

7. van Gilst WH, de Graeff PA, Kingma JH, Wesseling H, de Langen CDJ: Captoprilreduces purine loss and reperfusion arrhythmias in the rat heart after coronary a r-tery occlusion. Eur J Pharmacol 1984;100:113-117.

8. van Gilst WH, de Graeff PA, Wesseling H, de Langen CDJ: Reduction of reperf u-sion arrhythmias in the ischemic isolated rat heart by angiotensin converting e n-zyme inhibitors: a comparison of captopril, enalapril and HOE 498. J CardiovascPharmacol 1986;8:722-728.

9. de Graeff PA, van Gilst WH, Bel K, de Langen CDJ, Kingma JH, Wesseling: Co n-centration-dependent protection by captopril against myocardial damage duringischemia and reperfusion in a closed chest pig model. J Cardiovasc Pharmacol1987;9(Suppl. 2):S37-S42.

10. White HD, Norris RM, Brown MA, Brandt PWT, Whitlock RML, Wildt CJ: Leftventricular end-systolic volume as the major determinant of survival after recoveryfrom myocardial infarction. Circulation 1987;76:44-51.

INTRODUCTION

41

11. Pfeffer JM, Pfeffer MA, Brauwnwald E: Influence of chronic captopril therapy onthe infarcted left ventricle of the rat. Circ Res 1985;57:84-95.

12. Wit AL, Janse MJ: The ventricular arrhythmias of ischemia and infarction. Electro-physiological mechanisms, 1st ed. Mount Kisco, NY, Futura Publishing Company,Inc, 1993.

13. Myerburg RJ, Kessler KM, Bassett AL, Castellanos A: A biological approach tosudden cardiac death: structure, function and cause. Am J Cardiol 1989;63:1512-1516.

14. Campbell RW, Murray A, Julian DG: Ventri cular arrhythmias in first 12 hours ofacute myocardial infarction. Natural history study. Br Heart J 1981;46:351-357.

15. Bigger JT Jr, Dresdale FJ, Heissenbuttel RH, Weld FM, Wit AL: Ventricular a r-rhythmias in ischemic heart disease: mechanism, prevalence, significance, and ma n-agement. Progress in Cardiovascular Diseases 1977;19:255-300.

16. Tennant R, Wiggers CJ: The effect of coronary occlusion on myocardial contra c-tion. Am J Physiol 1935;112:351-361.

17. Sewell WH, Koth DR, Huggins CE: Ventricular f ibrillation in dogs after suddenreturn of flow to the coronary artery. Surgery 1955;38:1050-1053.

18. Battle WE, Naimi S, Avitall B, Brilla AH, Banas JS Jr, Bete JM, Levine HJ: Distin c-tive time course of ventricular vulnerability to fibrillation during and after releaseof coronary ligation. Am J Cardiol 1974;34:42-47.

19. Balke CW, Kaplinsky E, Michelson EL, Naito M, Dreifus LS: Reperfusion ventric u-lar tachyarrhythmias: correlation with antecedent coronary artery occlusion tac h-yarrhythmias and duration of myocardial ischemia. Am Heart J 1981;101:449-456.

20. de Graeff PA, de Langen CDJ, van Gilst WH, Bel K, Scholtens E, Kingma JH, We s-seling H: Protective effects of captopril against ischemia/reperfusion-induced ve n-tricular arrhythmias in vitro and in vivo. Am J Med 1988;84(Suppl. 3A):67-74.

21. Yamazaki S, Fujibayashi Y, Rajagopalan RE, Meerbaum S, Corday E: Effects ofstaged versus sudden reperfusion after coronary occlusion in the dog. J Am CollCardiol 1986;7:564-572.

22. Corr PB, Witkowski FX: Pote ntial electrophysiologic mechanisms responsible fordysrhythmias associated with reperfusion of ischemic myocardium. Circulation1983;68(Suppl. I):I-16-I-24.

23. Manning AS, Hearse DJ: Reperfusion-induced arrhythmias: mechanisms and pr e-vention. J Mol Cell Cardiol 1984;16:497-518.

24. Pogwizd SM, Corr PB: Electrophysiologic mechanisms underlying arrhythmias dueto reperfusion of ischemic myocardium. Circulation 1987;76:404-426.

25. Kaplinsky E, Ogawa S, Michelson EL, Dreifus LS: Instantaneous and delayed ven-tricular arrhythmias after reperfusion of acutely ischemic myocardium: evidencefor multiple mechanisms. Circulation 1981;63:333-340.

26. van Gilst: Reperfusion arrhythmias: mechanisms and prevention [Dissertation].Rijksuniversiteit, Groningen, 1986, pp. 177-210.

27. Priori SG, Mantica M, Napolitano C, Schwartz PJ: Early afterdepolarizations i n-duced in vivo by reperfusion of ischemic myocardium. A possible mechanism forreperfusion a rrhythmias. Circulation 1990;81:1911-1920.

CHAPTER 1

42

28. Ganz W, Buchbinder N, Marcus H, Mondkar A, O'Connor L, Maddahi J, Berman D,Charuzi Y, Beeder C, Peter T, Shah PK, Shell W: Intracoronary thrombolysis inevolving myocardial infarction. Herz 1981;6:37-43.

29. Rentrop P, Blanke H, Karsch KR, Kaiser H, Kostering H, Leitz K: Selective intr a-coronary thrombolysis in acute myocardial infarction and unstable angina pectoris.Circulation 1981;63:307-317.

30. Smalling RW, Fuentes F, Matthews MW, Freund GC, Hicks CH, Reduto LA,Walker WE, Sterling RP, Gould KL: Sustained improvement in left ventricularfunction and mortality by intracoronary streptokinase administration duringevolving myocardial infarction. Circulation 1983;68:131-138.

31. Markis JE, Malagold M, Parker JA, Silverman KJ, Barry WH, Als AV, Paulin S,Grossman W, Braunwald E: Myocardial salvage after intracoronary thrombolysiswith streptokinase in acute myocardial infarction. N Engl J Med 1981;305:777-782.

32. Anderson JL, Marshall HW, Bray BE, Lutz JR, Frederick PR, Yanowitz FG, DatzFL, Klausner SC, Hagan AD: A randomized trial of intracoronary streptokinase inthe treatment of acute myocardial infarction. N Engl J Med 1983;308:1312-1318.

33. Goldberg S, Greenspon AJ, Urban PL, Muza B, Berger B, Walinsky P, Maroko PR:Reperfusion arrhythmia: a marker of restoration of antegrade flow during intr a-coronary thrombolysis for acute myocardial infarction. Am Heart J 1983;105:26-32.

34. ISIS-2 (Second International Study of Infarct Survival Group) CollaborativeGroup: Randomized trial of intravenous streptokinase, oral aspirin, both, or neitheramong 17,187 cases of suspected acute myocardial infarction: ISIS-2. Lancet1988;ii:349-360.

35. Gruppo per lo studio della streptochinasi nell'infarto miocardico (GISSI): Effe c-tiveness of intravenous thrombolytic treatment in acute myocardial infarction. Lan-cet 1986;i:397-402.

36. Wilcox RG, von der Lippe G, Olsson CG, Jensen G, Skene AM, Hampton JR: Trialof tissue plasminogen activator for mortality reduction in acute myocardial infar c-tion. Anglo-Scandinavian Study of Early Thrombolysis (ASSET). Lancet1988;2:525-530.

37. AIMS Trial Study Group: Long-term effects of intravenous anistreplase in acutemyocardial infarction: final report of the AIMS study. Lancet 1990;335:427-431.

38. The ISAM study group: A prospective trial of intravenous streptokinase in acutemyocardial infarction (I.S.A.M.). N Engl J Med 1986;314:1465-1471.

39. Solomon SD, Ridker PM, Antman EM: Ventricular arrhythmias in trials of thro m-bolytic therapy for acute myocardial infarction. A meta-analysis. Circulation1993;88:2575-2581.

40. The European Myocardial Infarction Project Group: Prehospital thrombolytic the r-apy in patients with suspected acute myocardial infarction. N Engl J Med1993;329:383-389.

41. Solomon SD, Antman E: Prehospital thrombolytic therapy for myocardial infar c-tion [letter]. N Engl J Med 1994;330:291.

INTRODUCTION

43

42. Arntz HR, Stern R, Linderer T, Schroder R: Efficiency of a physician-operatedmobile intensive care unit for prehospital thrombolysis in acute myocardial infar c-tion. Am J Cardiol 1992;70:417-420.

43. Weaver WD, Cerqueira M, Hallstrom AP, Litwin PE, Martin JS, Kudenchuk PJ, Ei s-enberg M: Prehospital-initiated vs hospital-initiated thrombolytic therapy. TheMyocardial Infar ction Triage and Intervention Trial. JAMA 1993;270:1211-1216.

44. Kennedy JW, Martin GV, Davis KB, Maynard C, Stadius M, Sheehan FH, RitchieJL: The Western Washington Intravenous Streptokinase in Acute Myocardial I n-farction Rando mized Trial. Circulation 1988;77:345-352.

45. Cercek B, Lew AS, Laramee P, Shah PK, Peter TC, Gan z W: Time course and cha r-acteristics of ventricular arrhythmias after reperfusion in acute myocardial infar c-tion. Am J Cardiol 1987;60:214-218.

46. Miller FC, Krucoff MW, Satler LF, Green CE, Fletcher RD, Del Negro AA, PearleDL, Kent KM, Rackley CE: Ventricular arrhythmias during reperfusion. Am Heart J1986;112:928-932.

47. Gressin V, Louvard Y, Pezzano M, Lardoux H: Holter recording of ventricular a r-rhythmias during intravenous thrombolysis for acute myocardial infarction. Am JCardiol 1992;69:152-159.

48. Six AJ, Louwerenburg JH, Kingma JH, Robles de Medina EO, van Hemel NM:Predictive value of ventricular arrhythmias for patency of the infarct-related cor o-nary artery after thrombolytic therapy. Br Heart J 1991;66:143-146.

49. Gore JM, Ball SP, Corrao JM, Goldberg RJ: Arrhythmas in the assessment of cor o-nary artery reperfusion following thrombolytic therapy. Chest 1988;94:727-730.

50. Hackett D, McKenna W, Davies G, Maseri A: Reperfusion arrhythmias are rareduring acute myocardial infarction and thrombolysis in man. Int J Cardiol1990;29:205-213.

51. Zehender M, Utzolino S, Furtwangler A, Kasper W, Meinertz T, Just H: Time courseand interrelation of reperfusion-induced ST changes and ventricular arrhythmias inacute myocardial infarction. Am J Cardiol 1991;68:1138-1142.

52. Wilcox RG, Eastgate J, Harrison E, Skene AM: Ventricular arrhythmias duringtreatment with alteplase (recombinant tissue plasminogen activator) in suspectedacute myocardial infarction. Br Heart J 1991;65:4-8.

53. Alexopoulos D , Collins R, Adamopoulos S, Peto R, Sleight P: Holter monitoring ofventricular arrhythmias in a randomised, controlled study of intravenous strepto k-inase in acute myocardial infarction. Br Heart J 1991;65:9-13.

54. Gorgels AP, Vos MA, Letsch IS, Verschuuren EA, Bar FW, Janssen JH, Wellens HJ:Usefulness of the accelerated idioventricular rhythm as a marker for myocardial n e-crosis and reperfusion during thrombolytic therapy in acute myocardial infarction.Am J Cardiol 1988;61:231-235.

55. Berger PB, Ruocc o NA, Ryan TJ, Frederick MM, Podrid PJ: Incidence and signif i-cance of ventricular tachycardia and fibrillation in the absence of hypotension orheart failure in acute myocardial infarction treated with recombinant tissue-typeplasminogen activator: results from the Thrombolysis in Myocardial Infarction(TIMI) Phase II trial. J Am Coll Cardiol 1993;22:1773-1779.

CHAPTER 1

44

56. Northover BJ: Ventricular tachycardia during the first 72 hours after acute my o-cardial infarction. Cardiology 1982;69:149-156.

57. Farrell TG, Bashir Y, Cripps T, Malik M, Poloniecki J, Bennett D, Ward DE, CammJ: Risk stratification for arrhythmic events in postinfarction patients based on heartrate variability, ambulatory electrocardiographic variables and the signal-averagedelectrocardiogram. J Am Coll Cardiol 1991;18:687-697.

58. van der Laarse A, Kerkhof PL, Vermeer F, Serruys PW, Hermens WT, VerheugtFW, Bar FW, Krauss XH, van der Wall EE, Simoons ML: Relation between infarctsize and left ventricular performance assessed in patients with first acute myocardialinfarction randomized to intracoronary thrombolytic therapy or to conventionaltreatment. Am J Cardiol 1988;61:1-7.

59. Marino P, Zanolla L, Zardini P, on behalf of GISSI: Effect of streptokinase on leftventricular remodeling and function after myocardial infarction: The GISSI(Gruppo Italiano per lo studio della Streptochinaisi nell'Infarto Miocardico) Trial. JAm Coll Cardiol 1989;14:1149-1158.

60. Jeremy RW, Hackworthy RA, Bautovich G, Hutton BF, Harris PJ: Infarct arteryperfusion and changes in left ventricular volume in the month after acute myoca r-dial infarction. J Am Coll Cardiol 1987;9:989-995.

61. Pedretti RF, Colombo E, Sarzi Braga S, Caru B: Effect of thrombolysis on heart ratevariability and life-threatening ventricular arrhythmias in survivors of acute my o-cardial infarction. J Am Coll Cardiol 1994;23:19-26.

62. Bigger JT Jr, Fleiss JL, Rolnitzky LM: Prevalence, characteristics and significanceof ventricular tachycardia detected by 24-hour continuous electrocardiographic r e-cordings in the late hospital phase of acute myocardial infarction. Am J Cardiol1986;58:1151-1160.

63. Maggioni AP, Zuanetti G, Franzosi MG, Rovelli F, Santoro E, Staszewsky L,Tavazzi L, Tognoni G, on behalf of GISSI-2 investigators: Prevalence and progno s-tic significance of ventricular arrhythmias after acute myocardial infarction in thefibrinolytic era. GISSI-2 r esults. Circulation 1993;87:312-322.

64. Turitto G, Risa AL, Zanchi E, Prati PL: The signal-averaged electrocardiogram andventricular arrhythmias after thrombolysis for acute myocardial infarction. J AmColl Cardiol 1990;15:1270-1276.

65. Marino P, Nidasio G, Golia G, Franzosi MG, Maggioni AP, Santoro E, Santoro L,Zardini P: Frequency of predischarge ventricular arrhythmias in postmyocardial i n-farction patients depends on residual left ventricular pump performance and is i n-dependent of the occurrence of acute reperfusion. The GISSI-2 Investigators. J AmColl Cardiol 1994;23:290-295.

66. Willems AR: Ventricular tachycardia or ventricular fibrillation late after myocar-dial infarction [Dissertation]. Universiteit van Amsterdam, 1990, pp. 23-25.

67. Kersschot IE, Brugada P, Ramentol M, Zehender M, Waldecker B, Stevenson WG,Geibel A, De Zwaan C, Wellens HJ: Effects of early reperfusion in acute myocardialinfarction on arrhythmias induced by programmed stimulation: a prospective, ra n-domized study. J Am Coll Cardiol 1986;7:1234-1242.

INTRODUCTION

45

68. Bourke JP, Young AA, Richards DA, Uther JB: Reduction in incidence of inducibleventricular tachycardia after myocardial infarction by treatment with streptokinaseduring infarct evolution. J Am Coll Cardiol 1990;16:1703-1710.

69. Volpi A, De Vita C, Franzosi MG, Geraci E, Maggioni AP, Mauri F, Negri E, Sa n-toro E, Tavazzi L, Tognoni G: Determinants of 6-month mortality in survivors ofmyocardial infarction after thrombolysis. Results of the GISSI-2 data base. The Adhoc Working Group of the Gruppo Italiano per lo Studio della Sopravvivenzanell'Infarto Miocardico (GISSI)-2 Data Base. Circulation 1993;88:416-429.

70. Pfeffer MA, Brauwnwald E: Ventricular remodeling after myocardial infarction.Experimental observations and clinical implications. Circulation 1990;81:1161-1172.

71. St. John Sutton M, Pfeffer MA, Plappert T, Rouleau JL, Moye LA, Dagenais GR,Lamas GA, Klein M, Sussex B, Goldman S, et al: Quantitative two-dimensionalechocardiographic measurements are major predictors of adverse cardiovascularevents after acute myocardial infarction. The protective effects of captopril. Circu-lation 1994;89:68-75.

72. Visser CA, David GK, Meltzer RS, Kan G, Koolen JJ, Dunning AJ: Two-dimensional echocardiography during PTCA. Am Heart J 1986;111:1035-1041.

73. Gaudron P, Eilles C, Ertl G, Kochsiek K: Early remodelling of the left ventricle inpatients with myocardial infarction. Eur Heart J 1990;11(Suppl. B):139-146.

74. Weisman HF, Bush DE, Mannisi JA, Weisfeldt ML, Healy B: Cellular mechanismsof myocardial infarct expansion. Circulation 1985;57:186-201.

75. Weisman HF, Healy B: Myocardial infarct expansio n, infarct extension and rei n-farction: Pathophysiologic concepts. Prog Cardiovasc Dis 1987;307:73-110.

76. Erlebacher JA, Weiss JL, Weisfeldt ML, Bulkley BH: Early dilatation of the infar c-ted segment in acute transmural myocardial infarction: Role of infarct expansion inacute left ventricular enlargement. J Am Coll Cardiol 1984;4:201-208.

77. Fletcher PJ, Pfeffer JM, Pfeffer MA, Brauwnwald E: Left ventricular diastolic pre s-sure-volume relations in rats with healed myocardial infarction. Circ Res1981;49:618-626.

78. Erlebacher JA, Weiss JL, Eaton LW, Kallman C, Weisfeldt ML, Bulkley BH: Lateeffects of acute infarct dilatation on heart size: A two dimensional echoca r-diographic study. Am J Physiol 1982;49:1120-1126.

79. Anversa P, Olivetti G, Capasso JM: Cellular basis of ventricular remodeling aftermyocardial infarction. Am J Cardiol 1991;68:7D-16D.

80. Parmley WW, Chatterjee K: Congestive heart failure and arrhythmias: an overview.Am J Cardiol 1986;57:34B-37B.

81. Francis GS: Development of arrhythm ias in the patient with congestive heart fai l-ure: pathophysiology, prevalence and prognosis. Am J Cardiol 1986;57:3B-7B.

82. Breithardt G, Borggrefe M, Martinez-Rubio A, Budde T: Pathophysiologicalmechanisms of ventricular tachyarrhythmias. Eur Heart J 1989;10(Suppl. E):9-18.

83. de Bakker JM, van Capelle FJ, Janse MJ, Tasseron S, Vermeulen JT, de Jonge N,Lahpor JR: Slow conduction in the infarcted human heart. 'Zigzag' course of activ a-tion. Circulation 1993;88:915-926.

CHAPTER 1

46

84. Brugada P, Andries EW: Early postmyocardial infarction ventricular arrhythmias.Cardiovasc Clin 1992;22:165-180.

85. Ba'albaki HA, Clements SD: Left ventricular aneurysm: a review. Clin Cardiol1989;12:5-13.

86. Meizlish JL, Berger HJ, Plankey M, Errico D, Levy W, Zaret BL: Functional leftventricular aneurysm formation after acute anterior transmural myocardial infar c-tion: Incidence, natural history, and prognostic implications. N Engl J Med1984;311:1001-1006.

87. Rozanski JJ, Mortara D, Myerburg RJ, Castellanos A: Body surface d etection of de-layed depolarizations in patients with recurrent ventricular tachycardia and leftventricular ane urysm. Circulation 1981;63:1172-1178.

88. Kulakowski P, Dluzniewski A, Budaj A, Ceremuzynski L: Relationship betweensignal-averaged electrocardiography and dangerous ventricular arrhythmias in p a-tients with left ventricular aneurysm after myocardial infarction. Eur Heart J1991;12:1170-1175.

89. Franz MR, Cima R, Wang D, Profitt D, Kurtz R: Electrophysiological effects ofmyocardial stretch and mechanical determinants of stretch-activated arrhythmias.Circulation 1992;86:968-978.

90. Levett JM, McGrath LB, Bianchi JM: Beta receptor derangement in postinfarctionleft ventricular perianeurysm tissue. Clin Cardiol 1991;14:909-912.

91. Han J, Moe GK: Nonuniform recovery of exitability in ventricular muscle. Circ Res1964;14:44-60.

92. Sager PT, Perlmutter RA, Rosenfeld LE, McPherson CA, Wackers FJ, Batsford WP:Electrophysiologic effects of thrombolytic therapy in patients with a transmuralanterior myocardial infarction complicated by left ventricular aneurysm formation.J Am Coll Cardiol 1988;12:19-24.

93. McLenachan JM, Dargie HJ: A review of rhythm disorders in cardiac hypertrophy.Am J Cardiol 1990;65:42G-44G.

94. Dunn FG, Pringle SD: Sudden cardiac death, ventricular arrhythmias and hype r-tensive left ventricular hypertrophy. J Hypertens 1993;11:1003-1010.

95. Messerli FH: Left ventricular hypertrophy, arterial hypertension and sudden death. JHypertens 1990;8(suppl 7):S181-S185.

96. Vester EG, Kuhls S, Ochiulet-Vester J, Vogt M, Strauer BE: Electrophysiologicaland therapeutic implications of cardiac arrhythmias in hypertension. Eur Heart J1992;13(suppl D):70-81.

97. Moalic JM, Charlemagne D, Mansier P, Chevalier B, Swynghedauw B: Cardiac h y-pertrophy and failure--a disease of adaptation. Modifications in membrane proteinsprovide a molecular basis for arrhythmogenicity. Circulation 1993;87(5Suppl):IV21-IV26.

98. Cosin Aguilar J, Hernandiz Martinez A, Andres Conejos F: Mechanisms of ventri cu-lar arrhythmias in the presence of pathological hypertrophy. Eur Heart J1993;14(suppl J):65-70.

99. Bayes de Luna A, Vinolas Prat X, Guindo J: Ventricular arrhythmias in left ve n-tricular hypertrophy and heart failure. Eur Heart J 1993;14 Suppl J:62-64.

INTRODUCTION

47

100. Multicentre Postinfarction Research Group: Risk stratification and survival aftermyocardial infarction. N Engl J Med 1983;309:331-336.

101. Gottlieb SH, Ouyang P, Gottlieb SO: Death after acute myocardial infarction: inte r-relation between left ventricular dysfunction, arrhythmias and ischemia. Am J Car-diol 1988;61:7B-12B.

102. McClements BM, Adgey AA: Value of signal-averaged electrocardiography, rad i-onuclide ventriculography, Holter monitoring and clinical variables for predictionof arrhythmic events in survivors of acute myocardial infarction in the thrombolyticera. J Am Coll Cardiol 1993;21:1419-1427.

103. Lab MJ: Contraction excitation feedback in myocardium: physiological basis andclinical relevance. Circ Res 1982;50:757.

104. Hansen DE, Craig CS, Hondeghem LM: Stretch-induced arrhythmias in the isolatedcanine ventricle. Circulation 1990;81:1094-1105.

105. Reiter MJ, Synhorst DP, Mann DE: Electrophysiological effects of acute ventriculardilatation in the isolated rabbit heart. Circulation 1988;62:554-562.

106. Calkins H, Maughan WL, Weisman HF, Sugiura S, Sagawa K, Levine JH: Effect ofacute volume load on refractoriness and arrhythmia development in isolated,chronically infar cted canine hearts. Circulation 1989;79:687-697.

107. Levine JH, Guarnieri T, Kadish AH, White RI, Calkins H, Kan JS: Changes in my o-cardial repolarization in patients undergoing balloon valvuloplasty for congenitalpulmonary stenosis: evidence for contraction-excitation feedback in humans. Circu-lation 1988;77:70-77.

108. Taggart P, Sutton P, John R, Lab M, Swanton H: Monophasic action potential r e-cordings during acute changes in ventricular loading induced by the Valsalva m a-noeuvre. Br Heart J 1992;67:221-229.

109. Taggart P, Sutton PM, Treasure T, Lab M, O'Brien W, Runnal ls M, Swanton RH,Emanuel RW: Monophasic action potentials at discontinuation of cardiopulmonarybypass: evidence for contraction-excitation feedback in man. Circulation1988;77:1266-1275.

110. Ferrier GR: The effects of tension on acetylstrophanthidin-induced transient dep o-larizations and aftercontractions in canine myocardial and purkinje tissues. Circ Res1976;38:156-162.

111. Saffitz JE, Corr PB, Sobel BE: Arrhythmogenesis and ventricular dysfunction aftermyocardial infarction. Is anomalous cellular coupling the elusive link? Circulation1993;87:1742-1745.

112. Hochman JS, Choo H: Limitation of of myocardial infarct expansion by reperfusionindependent of myocardial salvage. Circulation 1987;75:299-306.

113. Force T, Kemper A, Leavitt M, Parisi AF: Acute reduction in functional infarct e x-pansion with late coronary reperfusion: assessment with quantative two-dimensional echocard iography. J Am Coll Cardiol 1988;11:192-200.

114. Pedretti R, Laporta A, Etro MD, Gementi A, Bonelli R, Anza C, Colombo E,Maslowsky F, Santoro F, Caru B: Influence of thrombolysis on signal-averagedelectrocardiogram and late arrhythmic events after acute myocardial infarction. AmJ Cardiol 1992;69:866-872.

CHAPTER 1

48

115. Vatterott PJ, Hammill SC, Bailey KR, Wiltgen CM, Gersh BJ: Late potentials onsignal-averaged electrocardiograms and patency of the infarct related artery in su r-vivors of acute myocardial infarction. J Am Coll Cardiol 1991;17:330-337.

116. Gang ES, Lew AS, Hong M, Wang FZ, Siebert CA, Peter T: Decreased incidence ofventricular late potentials after successful thrombolytic therapy for acute myoca r-dial infarction. N Engl J Med 1989;321:712-716.

117. Eldar M, Leor J, Hod H, Rotstein Z, Truman S, Kaplinsky E, Abboud S: Effect ofthrombolysis on the evolution of late potentials within 10 days of infarction. BrHeart J 1990;63:273-276.

118. Popovic AD, Neskovic AN, Babic R, Obradovic V, Bozinovic L, Marinkovic J, LeeJ-C, Tan M, Thomas JD: Independent impact of thrombolytic therapy and vesselpatency on left ventricular dilation after myocardial infarction. Serial echoca r-diographic follow-up. Circulation 1994;90:800-807.

119. Cryer PE: Physiology and pathophysiology of the human sympathoadrenal neur o-endocrine system. N Engl J Med 1980;303:436-444.

120. Greenwald L, Becker RC: Expanding the paradigm of the renin-angiotensin systemand angiotensin-converting enzyme inhibitors. Am Heart J 1994;128:997-1009.

121. Binah O, Rosen MR: Mechanisms of ventricular arrhythmias. Circulation1992;85(1 Suppl):I25-I31.

122. Waldo AL, Wit AL: Mechani sms of cardiac arrhythmias. Lancet 1993;341:1189-1193.

123. Nordrehaug JE, Johannessen KA, von der Lippe G, Myking OL: Circulatingcatecholamine and potassium concentrations early in acute myocardial infarction:effect of intervention with timolol. Am Heart J 1985;110:944-948.

124. Wit AL, Janse MJ: The ventricular arrhythmias of ischemia and infarction. Electro-physiological mechanisms, 1st ed. Mount Kisco, NY, Futura Publishing Company,Inc, 1993, p. 257.

125. Podrid PJ, Fuchs T, Candinas R: Role of the sy mpathetic nervous system in thegenesis of ventricular arrhythmia. Circulation 1990;82:I-103-I-113.

126. Lindemann JP, Watanabe AM: Sympathetic control of cardiac electrical activity, inZipes DP, Jalife J (eds): Cardiac electrophysiology. From cell to bedside, 1st ed.Philadelphia, W.B. Saunders Company, 1990, pp. 277-283.

127. Abildskov JA: Neural mechanisms involved in the regulation of ventricular rep o-larization. Eur Heart J 1985;6 Suppl D:31-39.

128. Calkins H, Allman K, Bolling S, Kirsch M, Wieland D, Morady F, Schwaiger M:Correlation between scintigraphic evidence of regional sympathetic neuronal dy s-function and ventricular refractoriness in the human heart. Circulation1993;88:172-179.

129. Karlsberg RP, Penkoske PA, Cryer PE, Corr PB, Roberts R: Rapid activation of thesympathetic nervous system following coronary artery occlusion: relationship to i n-farct size, site, and haemodynamic impact. Cardiovasc Res 1979;13:523-531.

130. Schaal SF, Wallace AG, Sealy WC: Protective influence of cardiac denerva tion an-gainst arrhythmias of myocardial infarction. Cardiovasc Res 1969;3:241-244.

INTRODUCTION

49

131. Ebert PA, Vanderbeek RB, Allgood RJ, Sabiston Jr DC: Effect of chronic cardiacdenervation on arrhythmias after coronary artery ligation. Cardiovasc Res1970;4:141-147.

132. McAlpine HM, Morton JJ, Leckie B, Rumley A, Gillen G, Dargie HJ: Neuroend o-crine activation after acute myocardial infarction. Br Heart J 1988;60:117-124.

133. Karlsberg RP, Cryer PE, Roberts R: Serial plasma catecholamine response early inthe course of clinical acute myocardial infarction: relationship to infarct extent andmortality. Am Heart J 1981;102:24-29.

134. Benedict CR, Grahame-Smith DG: Plasma adrenaline and noradrenaline concentr a-tions and dopamine-beta-hydroxylase activity in myocardial infarction with andwithout cardiogenic shock. Br Heart J 1979;42:214-220.

135. Sigurdsson A, Held P, Swedberg K: Short- and long-term neurohormonal activ a-tion following acute myocardial infarction. Am Heart J 1993;126:1068-1076.

136. Rouleau JL, de Champla in J, Klein M, Bichet D, Moyé L, Packer M, Dagenais GR,Sussex B, Arnold JM, Sestier F, Parker JO, McEwan P, Bernstein V, Cuddy E, L a-mas G, Gottlieb S, McCans J, Nadeau C, Delage F, Hamm P, Pfeffer MA: Activationof neurohumoral systems in postinfarction left ventricular dysfunction. J Am CollCardiol 1993;22:390-398.

137. Malik M, Camm AJ: Components of heart rate variability--what they really meanand what we really measure. Am J Cardiol 1993;72:821-822.

138. Kleiger RE, Miller JP, Bigger JT, Moss AJ: Dec reased heart rate variability and itsassociation with increased mortality after acute myocardial infarction. Am J Cardiol1987;59:256-262.

139. Malliani A, Pagani M, Lombardi F, Cerutti S: Cardiovascular neural regulation e x-plored in the frequency domain. Circulation 1991;84:482-492.

140. Malik M, Camm AJ: Heart rate variability: from facts to fancies. J Am Coll Cardiol1993;22:566-568.

141. Odemuyiwa O, Malik M, Farrell T, Bashir Y, Staunton A, Poloniecki J, Camm J:Multifactorial prediction of arrhythmic events after myocardial infarction. Comb i-nation of heart rate variability and left ventricular ejection fraction with other var i-ables. PACE 1991;14:1986-1991.

142. Odemuyiwa O, Malik M, Farrell T, Bashir Y, Poloniecki J, Camm J: Comparison ofthe predictive characteristics of heart rate variability index and left ventricular eje c-tion fraction for all-cause mortality, arrhythmic events and sudden death after acutemyocardial infar ction. Am J Cardiol 1991;68:434-439.

143. Casolo GC, Stroder P, Signorini C, C alzolari F, Zucchini M, Balli E, Sulla A,Lazzerini S: Heart rate variability during the acute phase of myocardial infarction.Circulation 1992;85:2073-2079.

144. Flapan AD, Wright RA, Nolan J, Neilson JM, Ewing DJ: Differing patterns of ca r-diac parasympathetic activity and their evolution in selected patients with a firstmyocardial infarction. J Am Coll Cardiol 1993;21:926-931.

145. McAreavey D, Neilson JMM, Ewing DJ, Russell DC: Cardiac parasympathetic a c-tivity during the early hours of acute myocardial infarction. Br Heart J1989;62:165-170.

CHAPTER 1

50

146. Pilipis A, Flather M, Ormerod O, Sleight P: Heart rate variability in acute myoca r-dial infarction and its association with infarct site and clinical course. Am J Cardiol1991;67:1137-1139.

147. Zabel M, Klingenheben T, Hohnloser SH: Changes in autonomic tone followingthrombolytic therapy for acute myocardial infarction: assessment by analysis ofheart rate var iability. Journal of Cardiovascular Electrophysiology 1994;5:211-218.

148. Hermosillo AG, Dorado M, Casa nova JM, Ponce de Leon S, Cossio J, KersenovichS, Colin L, Iturralde P: Influence of infarct-related artery patency on the indexes ofparasympathetic activity and prevalence of late potentials in survivors of acutemyocardial infar ction. J Am Coll Cardiol 1993;22:695-706.

149. Golia G, Marino P, Rametta F, Nidasio GP, Prioli MA, Anselmi M, Destro G,Zardini P: Reperfusion reduces left ventricular dilatation by preventing infarct e x-pansion in the acute and chronic phases of myocardial infarction. Am Heart J1994;127:499-509.

150. de Langen CD, de Graeff PA, van Gilst WH, Bel KJ, Kingma JH, Wesseling H: E f-fects of angiotensin II and captopril on inducible sustained ventricular tachycardiatwo weeks after myocardial infarction in the pig. J Cardiovasc Pharmacol1989;13:186-191.

151. Fleetwood G, Boutinet S, Meier M, Wood JM: Involvement of the renin-angiotensinsystem in ischemic damage and reperfusion arrhythmias in the isolated perfused ratheart. J Cardiovasc Pharmacol 1991;17:351-356.

152. Lee MA, Bohm M, P aul M, Ganten D: Tissue renin-angiotensin systems. Their rolein cardiovascular disease. Circulation 1993;87(5 Suppl):IV7-IV13.

153. Ball SG: The sympathetic nervous system and converting enzyme inhibition. JCardiovasc Pharmacol 1989;13 Suppl 3:S17-S21.

154. Ertl G, Kloner RA, Alexander RW, Braunwald E: Limitation of experimental infarctsize by an angiotensin-converting enzyme inhibitor. Circulation 1982;65:40-48.

155. Francis GS, McDonald KM, Cohn JN: Neurohumoral activation in preclinical heartfailure. Remodeling and the potential for intervention. Circulation 1993;87(5Suppl):IV90-IV96.

156. Schunkert H, Hense HW, Holmer SR, Stender M, Perz S, Keil U, Lorell BH, Rie g-ger GA: Association between a deletion polymorphism of the angiotensin-converting-enzyme gene and left ventricular hypertrophy. N Engl J Med1994;330:1634-1638.

157. Cambien F, Poirier O, Lecerf L, Evans A, Cambou J-P, Arveiler D, Luc G, Bard J-M,Bara L, Ricard S, Tiret L, Amouyel P, Alhenc-Gelas F, Soubrier F: Deletion pol y-morphism in the gene for angiotensin-converting enzyme is a potent risk factor formyocardial infar ction. Nature 1992;359:641-644.

158. Pinto YM, van Gilst WH, Kingma JH, Schunkert H: The deletion type allele of theangiotensin converting enzyme gene is associated with progressive ventriculardilatation after anterior myocardial infarction. J Am Coll Cardiol 1995:in press.

159. Anversa P, Olivetti G, Meggs LG, Sonnenblick EH, Capasso JM: Cardiac anatomyand ventricular loading after myocardial infarction. Circulation 1993;87(supplVII):VII22-VII27.

INTRODUCTION

51

160. Nabel EG, Topol EJ, Galeana A, Ellis SG, Bates ER, Werns SW, Walton JA, MullerDW, Schwaiger M, Pitt B: A randomised, placebo-controlled trial of combinedearly intravenous captopril and recombinant tissue-type plasminogen activator the r-apy in acute myocardial infarction. J Am Coll Cardiol 1991;17:467-473.

161. Brauwnwald E, Pfeffer MA: Ventricular enlargement and remodeling followingacute myocardial infarction: mechanisms and management. Am J Cardiol1991;68:1D-6D.

162. Beau SL, Saffitz JE: Transmural heterogeneity of norepinephrine uptake in failinghuman hearts. J Am Coll Cardiol 1994;23:579-585.

163. Davey PP, Bateman J, Mulligan IP, Forfar C, Barlow C, Hart G: QT interval dispe r-sion in chronic heart failure and left ventricular hypertrophy: relation to autonomicnervous system and Holter tape abnormalities. Br Heart J 1994;71:268-273.

164. Hemsworth PD, Pallandi RT, Campbell TJ: Cardiac electrophysiological actions ofcaptopril: lack of direct antiarrhythmic effects. Br J Pharmacol 1989;98:192-196.

165. Sheridan DJ, Culling W, Penny WJ: Electrophysiological disturbances associatedwith acute myocardial infarction. Eur Heart J 1986;7 Suppl B:11-18.

166. de Graeff PA, van Gilst WH, de Langen CDJ, Kingma JH, Wesseling H: Concentr a-tion-dependent protection by captopril against ischemia-reperfusion injury in theisolated rat heart. Arch Int Pharmacodyn Ther 1986;280:181-193.

167. Kahn MT, Malik KU: Modulation by prostaglandins of the release of 3H nora d-renaline evoked by potassium and nerve stimulation in the isolated rat heart. Eur JPharmacol 1982;78:213-218.

168. Li K, Chen X: Protective effects of captopril and enalapril on myocardial ischemiaand reperfusion damage of rat. J Mol Cell Cardiol 1987;19:909-915.

169. Linz W, Scholkens BA, Han YF: Beneficial effects of the converting enzyme inhib i-tor, ramipril, in ischemic rat hearts. J Cardiovasc Pharmacol 1986;8 Suppl 10:S91-S99.

170. Grover GJ, Sleph PG, Dzwonczyk S, Wang P, Fung W, Tobias D, Cushman DW:Effects of different angiotensin-converting enzyme (ACE) inhibitors on ischemicisolated rat hearts: relationship between cardiac ACE inhibition and cardioprote c-tion. J Pharmacol Exp Ther 1991;257:919-929.

171. Westlin W, Mullane K: Does captopril attenuate reperfusion-induced myocardialdysfunction by scavaging free radicals? Circulation 1988;77(Suppl. I):I-30-I-39.

172. van Gilst WH, van Wijngaarden J, Scholtens E, de Graeff PA, de Langen CDJ, We s-seling H: Captopril-induced increase in coronary flow: an SH-dependent effect onarachidonic acid metabolism ? J Cardiovasc Pharmacol 1987;9(Suppl 2):S31-S36.

173. Ambrosioni E, Borghi C, Magnani B: Early treatment of acute myocardial infar c-tion with angiotensin-converting enzyme inhibition: safety considerations. SMILEpilot study working party. Am J Cardiol 1991;68:101D-110D.

174. Ray SG, Pye M, Oldroyd KG, Christie J, Connelly DT, Northridge DB, Ford I,Morton JJ, Dargie HJ, Cobbe SM: Early treatment with captopril after acute my o-cardial infarction. Br Heart J 1993;69:215-222.

175. Pipilis A, Flather M, Collins R, Hargreaves A, Kolettis T, Boon N, Foster C, A p-pleby P, Sleight P: Effects on ventricular arrhythmias of oral captopril and of oral

CHAPTER 1

52

mononitrate started early in acute myocardial infarction: results of a randomisedplacebo controlled trial. Br Heart J 1993;69:161-165.

176. Bussmann WD, Micke G, Hildenbrand R, Klepzig H Jr: Captopril bei akutemHerzinfarkt: Einfluss auf Infarctgrösse und Rhythmusstörungen. Dtsch MedWochenschr 1992;117:651-657.

177. Di Pasquale P, Barone G, Pater na S, Cannizzaro S, Giubilato A: Efficacy of capt o-pril before thrombolysis in acute myocardial infarction: preliminary findings.Drugs Exp Clin Res 1990;16:581-589.

178. Kingma JH, van Gilst WH, de Graeff PA, Louwerenburg HW, Six AJ, Wesseling H:Captopril during thrombolysis in acute myocardial infarction: feasibility, toleranceand beneficial neurohumoral effects, in MacGregor GA, Sever PS (eds): Currentadvances in ACE inhibition, 1st ed. London, Churchill Livingstone, 1989, p. 301.

179. Swedberg K, Held P, Kjekshus J, Rasmussen K, Ryden L, Wedel H, on behalf of theCONSENSUS II study group: Effects of early administration of enalapril on mo r-tality in patients with acute myocardial infarction. N Engl J Med 1992;327:678-684.

180. Kingma JH, de Graeff PA, van Gilst WH, van Binsbergen E, de Langen CD, Wesse l-ing H: Effects of intravenous captopril on inducible sustained ventricular tachyca r-dia one week after experimental infarction in the anaesthetized pig. Postgrad Med J1986;62 Suppl 1:159-163.

181. Regoli D, Barabe J: Pharmacology of bradykinin and related kinins. Pharmacol Rev1980;32:1-46.

182. Tobe TJ, de Langen CD, Tio RA, Bel KJ, Mook PH, Wesseling H: Effects ofbradykinin on inducible sustained ventricular tachycardia two weeks after myoca r-dial infarction in pigs. J Cardiovasc Pharmacol 1991;17:701-706.

183. Bashir Y, Sneddon JF, O'Nunain S, Paul VE, Gibson S, Ward DE, Camm AJ: Co m-parative electrophysiological effects of captopril or hydralazine combined with n i-trate in patients with left ventricular dysfunction and inducible ventricular tach y-cardia. Br Heart J 1992;67:355-360.

184. Martorana PA, Kettenbach B, Breipohl G, Linz W, Scholkens BA: Reduction of i n-farct size by local angiotensin-converting enzyme inhibition is abolished by abradykinin antagonist. Eur J Pharmacol 1990;182:395-396.

185. Tobe TJ, de Langen CD, Tio RA, Bel KJ, Mook PH, Wesseling H: In vivo effect ofbradykinin during ischemia and reperfusion: improved electrical stability twoweeks after myocardial infarction in the pig. J Cardiovasc Pharmacol 1991;17:600-607.

186. Kloner RA: Does reperfusion injury exist in humans? J Am Coll Cardiol1993;21:537-545.

187. Pfeffer MA, Pfeffer JM, Steinberg C, Finn P: Survival after an experimental my o-cardial infarction: Benificial effects of long term therapy with captopril. Circula-tion 1985;72:406-412.

188. Pfeffer MA, Lamas GA, Vaughan DE, Parisi AF, Brauwnwald E: Effect of captoprilon progressive ventricular dilatation after anterior myocardial infarction. N Engl JMed 1988;319:80-86.

INTRODUCTION

53

189. Sharpe N, Smith H, Murphy J, Greaves S, Hart H, Gamble G: Early prevention ofleft ventricular dysfunction after myocardial infarction with angiotensin-converting-enzyme inh ibition. Lancet 1991;337:872-876.

190. Bonaduce D, Petretta M, Arrichiello P, Conforti G, Montemurro MV, Attisano T,Bianchi V, Morgano G: Effects of captopril treatment on left ventricular remode l-ing and function after anterior myocardial infarction: comparison with digitalis. JAm Coll Cardiol 1992;19:858-863.

191. Bonarjee VV, Carstensen S, Caidahl K, Nilsen DW, Edner M, Berning J: Attenu a-tion of left ventricular dilatation after acute myocardial infarction by early initi a-tion of enalapril therapy. CONSENSUS II Multi-Echo Study Group. Am J Cardiol1993;72:1004-1009.

192. Pinto YM, van Wijngaarden J, van Gilst WH, de Graeff PA, Wesseling H: The e f-fects of short- and long-term treatment with an ACE inhibitor in rats with myoca r-dial infarction. Basic Res Cardiol 1991;86 Suppl 1:165-172.

193. Mendoza I, Moleiro F, Pulido M, Marques J, Roderiguez A, Condado J, Acosta J:Hypertension, left ventricular hypertrophy and sudden death. LVH regression andsuppression of ventricular arrhythmias with ACE inhibitiors. J Am Coll Cardiol1992;19:386A.

194. Cleland JG, Dargie HJ, Hodsman GP, Ball SG, Robertson JI, Morton JJ, East BW,Robertson I, Murray GD, Gillen G: Captopril in heart failure. A double blind co n-trolled trial. Br Heart J 1984;52:530-535.

195. Cleland JGF, Dargie HJ, Ball SG, Gillen G, Hodsman GP, Morton JJ, East BW, Ro b-ertson I, Ford I, Robertson JIS: Effects of enalapril in heart failure: a double blindstudy of effects on exercise performance, renal function, hormones, and metabolicstate. Br Heart J 1985;54:305-312.

196. Bonaduce D, Petretta M, Morgano G, Attisano T, Bianchi V, Arrichiello P, RotondiF, Condorelli M: Effects of converting enzyme inhibition on baroreflex sensitivityin patients with myocardial infarction. J Am Coll Cardiol 1992;20:587-593.

197. Bonaduce D, Marciano F, Petretta M, Migaux ML, Morgano G, Bianchi V, SalemmeL, Valva G, Condorelli M: Effects of converting enzyme inhibition on heart periodvariability in patients with acute myocardial infarction. Circulation 1994;90:108-113.

198. Nordrehaug JE, Johannessen KA, von der Lippe G: Serum potassium concentrationas a risk factor of ventricular arrhythmias early in acute myocardial infarction. Cir-culation 1985;71:645-649.

199. Campbell RW, Higham D, Adams P, Murray A: Potassium: its relevance for a r-rhythmias complicating acute myocardial infarction. J Cardiovasc Pharmacol1987;10 Suppl 2:S25-S28.

200. Webster MWI, Fitzpatrick A, Nicholls G, Ikram H, Wells JE: Effect of enalapril onventricular arrhythmias in congestive heart failure. Am J Cardiol 1985;56:566-569.

201. The Captopril-Digoxin Multicenter research group: Comparative effects of therapywith captopril and digoxin in patients with mild to moderate heart failure. JAMA1988;259:539-544.

CHAPTER 1

54

202. de Graeff PA, Kingma JH, Viersma JW, Wesseling H, Lie KI: Acute and chronic e f-fects of ramipril and captopril in congestive heart failure. Int J Cardiol 1989;23:59-67.

203. Cocchieri M, Alunni GF, Del-Favero A, Fortunati F, Bardelli G, Capponi EA, RegiL, Boschetti E: Comparative effects of ibopamine and captopril in mild congestiveheart failure. Focus on the long-term effects of inodilation on ventricular arrhyt h-mias. Cardiology 1990;77 Suppl 5:36-42.

204. Bechler-Lisinska J, Cholewa M, Gorski L, Markiewicz K: [Effect of vasodilatoragents on the character and incidence of cardiac arrhythmia in chronic heart fai l-ure]. Z Gesamte Inn Med 1990;45:71-76.

205. Kleber FX, Niemoller L, Doering W: Impact of converting enzyme inhibition onprogression of chronic heart failure: Results of the Munich Mild Heart Failure Trial.Br Heart J 1992;67:289-296.

206. Pomini G, Gribaldo R, Rugna A, Lupia M, Molfese G, Carenza P: Reduction ofcomplex ventricular arrhythmias after enalapril treatment in patients with advancedstable heart fai lure. G Ital Cardiol 1991;21:59-65.

207. Poquet F, Ferguson J, Rouleau JL: The antiarrhythmic effect of the ACE inhibi torcaptopril in patients with congestive heart failure largely is due to its potassiumsparing effects. Can J Cardiol 1992;8:589-595.

208. Gurlek A, Erol C, Basesme E: Antiarrhythmic effect of converting enzyme inhib i-tors in congestive heart failure. Int J Cardiol 1994;43:315-318.

209. Adgey J, Clarke M, Raza A, Meddis D: Study of the safety and efficacy of ACE i n-hibitors and their effects on 24-hour electrocardiographic monitoring in the trea t-ment of moderate-to-severe heart failure: an interim analysis. Am J Cardiol1992;70:142C-144C.

210. The CONSENSUS trial study group: Effects of enalapril on mortality in severecongestive heart failure. N Engl J Med 1987;316:1429-1435.

211. Fonarow GC, Chelimsky-Fallick C, Stevenson LW, Luu M, Hamilton MA,Moriguchi JD, Tillisch JH, Walden JA, Albanese E: Effect of direct vasodilationwith hydralazine versus angiotensin-converting enzyme inhibition with captopril onmortality in advanced heart failure: the Hy-C trial. J Am Coll Cardiol 1992;19:842-850.

212. Newman TJ, Maskin CS, Dennick LG, Meyer JH, Hallows BG, Cooper WH: Effectsof captopril on survival in patients with heart failure. Am J Med 1988;84:140-144.

213. The SOLVD investigators: Effect of enalapril on survival in patients with reducedleft ventricular ejection fractions and congestive heart failure. N Engl J Med1991;325:293-302.

214. Cohn JN, Johnson G, Ziesche S, Cobb F, Francis G, Tristani F, Smith R, DunkmanWB, Loeb H, Wong M, et al: A comparison of enalapril with hydralazine-isosorbidedinitrate in the treatment of chronic congestive heart failure. N Engl J Med1991;325:303-310.

215. Fletcher RD, Cintron GB, Johnson G, Orndorff J, Carson P, Cohn JN: Enalapril d e-creases prevalence of ventricular tachycardia in patients with chronic congestiveheart failure. The V-HeFT II VA Cooperative Studies Group. Circulation1993;87(6 Suppl):VI49-VI55.

INTRODUCTION

55

216. Cohn JN, Archibald DG, Ziesche S, Franciosa JA, Harston WE, Tristani FE, Dun k-man WB, Jacobs W, Francis GS, Flohr KH, et al: Effect of vasodilator therapy onmortality in chronic congestive heart failure. Results of a Veterans AdministrationCooperative Study. N Engl J Med 1986;314:1547-1552.

217. The SOLVD investigators: Effect of enalapril on mortality and the development ofheart failure in asymptomatic patients with reduced left ventricular ejection fra c-tions. N Engl J Med 1992;327:685-691.

218. Pfeffer MA, Brauwnwald E, Moyé LA, Basta L, Brown EJ, Cuddy TE, Davis BR,Geltman EM, Goldman S, Flaker GC, Klein M, Lamas GA, Packer M, Rouleau J,Rouleau JL, Rutherford J, Wertheimer JH, Hawkins CM, on behalf of the SAVE i n-vestigators: Effect of captopril on mortality and morbidity in patients with left ve n-tricular dysfunction after myocardial infarction. Results of the Survival and Ve n-tricular Enlargement Trial. N Engl J Med 1992;327:669-677.

219. Packer M, Rouleau J-L, Moyé LA, Rouleau JR, Bernstein V, Cuddy TE, Lewis S,Sussex SA, Sestier F, Goldman S, Jacobson K, Lamas G, McCans J, Randall OS,Wertheimer JH, Davis BR, Brauwnwald E, Pfeffer MA: Effect of captopril on ve n-tricular arrhythmias and sudden death in patients with left ventricular dysfunctionafter myocardial infarction: SAVE trial. J Am Coll Cardiol 1993;21:130A.

220. Ambrosioni E, Borghi C, Magnani B, for the Survival of Myocardial InfarctionLong-Term Evaluation (SMILE) Study Investigators: The effect of the angiotensin-converting enzyme inhibitor zofenopril on mortality and morbidity after anteriormyocardial infarction. N Engl J Med 1995;332:80-85.

221. Søgaard P, Gotzsche CO, Ravkilde J, Norgaard A, Thyges en K: Ventricular a r-rhythmias in the acute and chronic phases after acute myocardial infarction. Effectof intervention with captopril. Circulation 1994;90:101-107.

CHAPTER 2

Left ventricular dilatation and high-grade ventriculararrhythmias in the first year after myocardial infarction

Jan-Henk E. Dambrink,1 MD; Willem P. Beukema,1 MD;Wiek H. van Gilst,2 PhD; Kathinka H. Peels,3 MD; K.I. Lie,4 MD;

and J. Herre Kingma,1,2 MD for the CATS investigators5

1Department of Cardiology, St Antonius Hospital, Nieuwegein2Department of Clinical Pharmacology, University of Groningen, Groningen3Department of Cardiology, Catharina Hospital, Eindhoven4Department of Cardiology, Groningen University Hospital, Groningen5see Appendix

Published in:J Cardiac Failure 1994;1:3-11

DILATATION AND VENTRICULAR ARRHYTHMIAS

55

Abstract

Background. Progressive left ventricular dilatation is an important determi-nant of prognosis after myocardial infarction. The association of this processwith the occurrence of ventricular arrhythmias is less well established.Methods and Results. Of 153 patients with a first anterior myocardial infarc-tion treated with thrombolytic therapy 34 (22%) had high-grade ventricular ar-rhythmias (Lown classes 4A and B) during Holter monitoring after one year.Patients with high-grade ventricular arrhythmias had a larger end-systolic vol-ume (38 ± 12 vs 25 ± 11 ml/m2, p < 0.001) at hospital discharge and more leftventricular dilatation (10 ± 18 vs 1 ± 9 ml/m2, p=0.011) during follow up. Bothan increased end-systolic volume at discharge and subsequent dilatation provedindependent predictors of high-grade ventricular arrhythmias. Six patients diedsuddenly during the first 12 months after myocardial infarction. Four of thesepatients had an enlarged end-systolic volume (LVESVI > 22 ml/m2) at dis-charge, and the three patients who died suddenly after three months showed asignificant increase in end-systolic volume from discharge to three months com-pared to survivors (16 ± 6 vs 2 ± 9, p=0.008).Conclusion. Left ventricular remodeling after myocardial infarction is an inde-pendent predictor for the occurrence of ventricular arrhythmias late after myo-cardial infarction.

Introduction

Left ventricular remodeling after myocardial infarction is characterized bydilatation of the infarcted and noninfarcted regions of the ventricle.1 These mor-phological changes are frequently associated with the occurrence of heart fail-ure2 and sudden death.3 Although the anatomical changes after myocardial in-farction are well described,4-8 little is known about accompanying electrophysi-ological changes in the left ventricle. Recently, it has been demonstrated that in-creased loading and subsequent dilatation of the heart can lead to an increase inventricular arrhythmias.9,10 This may be explained by increased dispersion in re-fractoriness11-14 and early afterdepolarizations,12,14 caused by an increase inventricular wall stress (contraction-excitation feedback).15

Although clinical evidence is available that increased left ventricular dimen-sions lead to a higher incidence of life-threatening ventricular arrhythmias aftermyocardial infarction,3 the relation between left ventricular dilatation and theoccurrence of ventricular arrhythmias is less well established.

CHAPTER 2

56

In the present study, we investigated the relation between left ventriculardilatation and ventricular arrhythmias during the first year after anterior myo-cardial infarction.

Methods

Patients. This study was an ancillary investigation of the Captopril AndThrombolysis Study (CATS), in which the effect of captopril treatment, startedduring thrombolysis, was evaluated in patients with a first anterior myocardialinfarction.16 Informed consent was obtained by witnessed oral consent, laterconfirmed by written informed consent following the acute phase of myocardial

Table 2.1. Baseline characteristics of all patients, patients with Holter monitoringafter 12 months, and patients who died suddenly in CATS

All CATS Patients with Patients who diedpatients Holter at 1 year suddenly

N 298 153 6 Age (years) Male (%) Smoker (%) Hypertension (%) Diabetes mellitus (%)

Time to thrombolysis (h)

≥ 2-vessel disease (%) LAD occluded (%)

α-HBDH Q72 (U/l) LVESVI (ml/m2) LVEDVI (ml/m2)

59 ± 107563229

3.5 ± 1.3

3522

1277 ± 100725 ± 1056 ± 13

58 ± 98158228

3.4 ± 1.3

3818

1149 ± 86524 ± 9

54 ± 12

63 ± 68350170

3.6 ± 2.2

6733

1797 ± 98631 ± 1055 ± 14

LVEF (%) 55 ± 10 56 ± 9 44 ± 11*

α-HBDH Q72 indicates cumulative alpha-hydroxy butyrate dehydrogenase over the first72 hours after myocardial infarction; CATS, Captopril And Thrombolysis Study; h, hours;LAD, left anterior descending artery; LVEF, left ventricular ejection fraction; LVEDVI,left ventricular end-diastolic volume indexed for body surface area; LVESVI, left ve n-tricular end-systolic volume indexed for body surface area. Cornary angiograms weremade 38 ± 78 days after infarction. *Significantly lower than all CATS patients (p=0.011)and patients with Holter monitoring (p=0.008).

DILATATION AND VENTRICULAR ARRHYTHMIAS

57

infarction. Main endpoints included left ventricular volumes, neurohumoral ac-tivation and ventricular arrhythmias. In CATS, 298 patients were included in 12hospitals in The Netherlands. The study was approved by the Institutional Re-view Board of all participating hospitals. Selection criteria included a typicalhistory of chest pain consistent with myocardial infarction with onset of symp-toms no longer than 6 hours before admission, and ECG criteria for acute ante-rior myocardial infarction including at least 1 mm ST segment elevation in leadsI and aVL and/or 2 mm ST segment elevation in two or more precordial leads ofthe 12-lead electrocardiogram, consistent with anterolateral, anteroseptal and/oranterior wall infarction.

Patients had to be eligible for thrombolytic therapy. Exclusion criteria in-cluded presence of a prior myocardial infarction, left bundle branch block and

0

10

20

30

40

50

60

One yearPredischargeEntry

LVESVI (ml/m2)

*

** **

**

Figure 2.1. Left ventricular end-systolic volume (LVESVI) of patients with- (open cir-cles) and without (filled circles) high-grade ventricular arrhythmias during Holtermonitoring at 12 months. LVESVI is already higher upon admission in patients withhigh-grade arrhythmias. In these patients, there is a clear increase in end-systolic vol-ume in the first year of follow up, whereas in patients without high-grade arrhythmiasthere is no significant increase in LVESVI. *p < 0.05, **p < 0.005.

CHAPTER 2

58

severe heart failure (Killip class III or IV). Patients were followed for 12 monthsafter myocardial infarction. Sudden death was defined as death within one hourof symptoms, but also included unwitnessed death in patients who were other-wise stable.

Holter recording. The occurrence of ventricular arrhythmias was assessed bytwo-channel 12-hour Holter monitoring upon admission, and two-channel 24-hour Holter monitoring (Reynolds Medical Tracker recorder) before dischargeand three and 12 months after myocardial infarction. Analysis was performedwith help of a Reynolds Medical Pathfinder 3 analysis system. Tapes were ana-lyzed for the presence of uniform and multiform premature ventricular beats,

0

20

40

60

80

100in LVESVI (ml/m2)

4 quartiles of change

Group IV

Group III

Group II

Group I

I II III IVI II III IVI II III IVI II III IV

****

arrhythmiasHigh-grade

(%)

12 monthsAdmission 3 monthsDischarge

Figure 2.2. The percentage of patients with high-grade ventricular arrhythmias in 4 quarti-les with different levels of left ventricular dilatation is displayed at hospital admission, be-fore discharge and three and 12 months after myocardial infarction. The levels of dilata-tion, expressed as changes in end-systolic volume indexed for body surface area, are asfollows: group I, decrease in LVESVI of more than 4 ml/m2, group II, decrease in LVESVIof 4 ml/m2 to an increase of less than 1 ml/m2, group III, increase in LVESVI of 1 ml/m2 toless than 8 ml/m2, group IV, increase in LVESVI of 8 ml/m2 and more. *p=0.003, **p < 0.001(Chi-square for linear trend).

DILATATION AND VENTRICULAR ARRHYTHMIAS

59

pairs of premature ventricular beats and ventricular tachycardia. Pairs were de-fined as a repetition of two ventricular beats with a maximum interval of 0.6 s.Ventricular tachycardia was defined as a repetition of 3 or more ventricularbeats with a rate exceeding 100 beats per minute. Paired ventricular prematurebeats and ventricular tachycardia (Lown classes 4A and 4B) were defined ashigh-grade ventricular arrhythmias; all other ventricular arrhythmias were clas-sified as low-grade. The presence of high-grade ventricular arrhythmias duringHolter monitoring after one year was chosen as the primary end point for thelogistic regression analysis.

Echocardiography. Echocardiograms were made within 24 hours after ad-mission, at day 3, before discharge and three and 12 months after myocardialinfarction. Left ventricular end-systolic and end-diastolic volumes were calcu-lated from a two- and four-chamber view using the modified biplane Simpson’srule.17 From these volume measurements the ejection fraction was calculated.Measurements were made off-line from end-diastolic and end-systolic still-frames using a Dataview Microsonics cardiac analysis system (Nova Microson-ics). Left ventricular volumes were indexed for body surface area. Furthermore,regional wall motion abnormalities were evaluated using the wall motion scorerecommended by the American Society of Echocardiography.17 In this scoringsystem the left ventricle is divided into 16 segments, scoring each segment as 1for normokinesia, 2 for hypokinesia, 3 for akinesia, 4 for dyskinesia and 5 foran aneurysmal segment. A wall motion score index was computed as the sum ofscores of all segments divided by the number of segments evaluated. Twelveevaluable segments were considered a minimum to reliably assess the wall mo-tion score.

Infarct size. Enzymatic infarct size was estimated by cumulative alpha-hydroxybutyrate dehydrogenase (α-HBDH) values over the first 72 hours aftermyocardial infarction. This method is not influenced by the presence or absenceof reperfusion.18

Statistical analysis. Results are presented as mean ± standard deviation. Dif-ferences in end-systolic volume between groups were examined using analysisof variance and the modified t-test according to the Bonferroni method. TheChi-squared test was used for discrete data. Logistic regression analysis wasused to identify independent relations between baseline characteristics and theoccurrence of high-grade ventricular arrhythmias.19 Calculations were madeusing SPSS/PC+ v 4.0 software. A two-sided p-value of less than 0.05 wasconsidered statistically significant.

CHAPTER 2

60

Results

Of all 298 patients included in CATS, 65 (21%) did not complete a follow-upperiod of 12 months due to withdrawal of consent, death, or other clinicalevents. In 34 patients (11%) Holter monitoring was not performed due to un-specified reasons. Of 199 patients with Holter monitoring available, 46 tapeswere of inadequate quality (15%). Holter monitoring after one year was avail-able in 153 patients (51%).

Table 2.2. Characteristics of patients with and without high-grade ventriculararrhythmias after 12 months

Arrhythmias (N) No arrhythmias (N) p-value

Demographics Age (years) MaleMyocardial injury α-HBDH Q72 (U/l) LVEF (%) Wall Motion ScoreExtent of CAD ≥ 2-vessel disease LAD occludedLV remodeling LVESVI discharge(ml/m2) LVESVI one year (ml/m2) Dilatation (ml/m2) LV aneurysm (%)Neurohumoral activ a-tion Heart rate variabilityEctopic activity VPBs > 10/hourMedication Beta-blockers ACE inhibitors

60.8 ± 8.8 (34)26/34 (76%)

1499 ± 928 (27)48 ± 9 (17)

2.07 ± 0.31(27)

11/23 (48%)7/23 (30%)

38 ± 12 (17)43 ± 17 (19)10 ± 18 (14)4/30 (13%)

27.1 ± 5.6 (20)

10/24 (42%)

9/34 (27%)12/34 (35%)

57.5 ± 9.6 (119)97/119 (82%)

1066 ± 832 (113)57 ± 9 ( 84)

1.74 ± 0.44 (113)

23/67 (34%)9/67 (13%)

25 ± 11 (84)26 ± 14 (77)1 ± 9 (60)

11/108 (10%)

30.0 ± 11.0 (77)

6/97 (6%)

44/119 (37%)55/119 (46%)

0.0700.683

0.019< 0.001< 0.001

0.3670.110

< 0.001< 0.001

0.0110.624

0.108

< 0.001

0.3520.349

Digoxin 6/34 (18%) 6/119 ( 5%) 0.040

ACE indicates angiotensin converting enzyme; CAD, coronary artery disease; VPBs, ve n-tricular premature beats. Other abbreviations as in T able 2.1.

DILATATION AND VENTRICULAR ARRHYTHMIAS

61

Table 2.1 shows baseline characteristics after hospital admission of all CATSpatients, those with Holter monitoring after 12 months, and patients who diedsuddenly during the first year of follow up. There were no differences in charac-teristics between all CATS patients and patients with Holter monitoring after 1year. However, patients who died suddenly showed a tendency to larger end-systolic volumes, resulting in a significantly lower ejection fraction compared tothe other patient groups.

Echocardiographic data of patients with and without high-grade ventriculararrhythmias during Holter monitoring (Lown classes 4A and B) after 12 monthswere investigated. In Figure 2.1, the mean end-systolic volume during follow upin patients with and without high-grade ventricular arrhythmias during Holterrecording after one year is shown. In patients with high-grade ventricular ar-rhythmias, a higher end-systolic volume was present from admission. In addi-tion, left ventricular dilatation was prominent in this group, whereas end-systolicvolume remained relatively unchanged in patients without high-grade arrhyth-mias. In Figure 2.2, the percentage of patients with high-grade ventricular ar-rhythmias divided in 4 quartiles of left ventricular dilatation are shown. Athospital admission, there is little or no difference in the percentage of patientswith high-grade arrhythmias between these groups. However, at hospital dis-charge, the prevalence of arrhythmias has decreased more in those withoutdilatation (groups I and II) compared to those with dilatation (groups III andIV). During follow up, this difference in prevalence of arrhythmias persists, andat 12 months the percentage of patients with high-grade arrhythmias in patientgroup IV is still about 50%, whereas in patients with less or no dilatation, this isabout 10 percent.

In Table 2.2, differences in characteristics in patients with and without high-grade ventricular arrhythmias are listed. In addition to a higher end-systolic vol-ume and progressive dilatation, patients with high-grade ventricular arrhythmias

Table 2.3. Independent predictors of high grade ventricular arrhythmias (logisticregression analysis)

B SE p OR (95%-CI)

Dilatation > 8 ml/m2 1.6361 0.8252 0.0474 5.14 (1.02 - 25.9) LVESVI > 35 ml/m2 1.8387 0.8234 0.0256 6.29 (1.25 - 31.6) VPBs > 10 /hour 1.9890 0.9027 0.0276 7.31 (1.25 - 42.9) Constant -3.3344 0.7662

B indicates regression coefficient; CI, confidence intervals; OR, odds ratio; SE, standarderror. Other abbreviations as in Table 2.1 and 2.2.

CHAPTER 2

62

are characterized by a larger enzymatic infarct size, a lower ejection fraction,relatively more frequent ventricular premature beats and more use of digoxin.

Logistic regression analysis was used to assess the relative importance of end-systolic volume and dilatation after discharge for the occurrence of arrhythmiasafter 12 months. First, five baseline variables strongly associated with the occur-rence of high-grade ventricular arrhythmias after one year were selected(enzymatic infarct size, ejection fraction, wall motion score, end-systolic vol-ume, VPBs during Holter monitoring; see also Table 2.2). These variables werethen converted into binary variables with the 75th percentile as the cut-off point.These variables were then entered in a forward stepwise logistic regressionmodel to identify the variables that provided independent information. Thisanalysis showed that end-systolic volume > 35 ml/m2 and > 10 VPBs/hour werethe only independent variables. After this, a second analysis was performed. Weentered end-systolic volume and VPBs > 10/hour as independent predictors ofventricular arrhythmias, and then entered left ventricular dilatation (increase inend-systolic volume). The results of this analysis are given in Table 2.3. Leftventricular dilatation proved to increase the risk for high grade arrhythmias in-dependently. In addition, this analysis was repeated using end-diastolic volumein stead of end-systolic volume as a determinant of arrhythmias. End-diastolicvolume also proved an independent predictor of high-grade ventricular ar-rhythmias, but further increase in end-diastolic volume did not provide addi-tional independent information.

Table 2.4. Patients who died suddenly in the Captopril And Thrombolysis Study

Patient Time of death Cumulative Ejection End-systolicnumber after AMI α-HBDH fraction volume at

discharge (days) (U/l) (%) (ml/m2)

1 277 2359 37 462 171 2969 53 233 171 543 51 184 58 2153 37 465 26 2129 42 436 277 631 77 14mean 163 ± 106 1797 ± 986 50 ± 15 32 ± 15all - 1277 ± 1007 55 ± 10 27 ± 12p-value - 0.212 0.237 0.320

AMI indicates acute myocardial infarction; α-HBDH, alpha-hydroxybutyrate dehydrog e-nase.

DILATATION AND VENTRICULAR ARRHYTHMIAS

63

In Table 2.4, echocardiographic data of the six patients who died suddenlyare shown. At discharge, patients 1,2,4 and 5 had an enlarged end-systolic vol-ume (> 22 ml/m2). However, all patients who survived the first 3 months butdied suddenly afterwards also had an enlarged end-systolic volume (patients 1,3and 6). Left ventricular dilatation from discharge to 3 months in these patientswas clearly more pronounced than dilatation in the whole study population (16± 6 vs 2 ± 9, p=0.008).

Discussion

Previous studies describing the relation between end-systolic volume3 or leftventricular dysfunction20 and ventricular arrhythmias are angiographic and radi-onuclide studies dating from the prethrombolytic era. The present study dem-onstrates the relation between left ventricular remodeling, quantified by changesin echocardiographically assessed end-systolic volume, and the occurrence ofventricular arrhythmias late after thrombolytic therapy in patients with a firstanterior myocardial infarction. Patients with high-grade ventricular arrhythmiasduring Holter monitoring after 12 months showed a larger end-systolic volumeat baseline and progressive left ventricular dilatation during follow up. Patientswho died suddenly in CATS were all characterized by progressive left ventricu-lar dilatation after discharge. Although an increased end-systolic volume 4-8weeks after myocardial infarction has previously been identified as a powerfulrisk factor for life-threatening arrhythmias,3 the importance of an increased end-

(Table 2.4 continued)

End-systolic Increase in Aneurysm Wall motion Patientvolume after

3 monthsend-systolic

volumeat discharge score index at

dischargenumber

(ml/m2) (ml/m2) (+/-) (WMSI)

66 21 + 2.69 1 - - + 2.56 237 18 - 2.06 3 - - - 1.63 4 - - - 2.44 523 9 - 1.25 6

42 ± 22 16 ± 6 33% 2.13 ± 0.48 mean29 ± 15 2 ± 9 11% 1.83 ± 0.43 all0.141 0.008 0.310 0.135 p-value

CHAPTER 2

64

systolic volume at discharge and progressive left ventricular dilatation duringfollow up has not been appreciated before.

Previous studies

White et al.3 first demonstrated the importance of an increased end-systolicvolume measured 4-8 weeks after myocardial infarction as a risk factor for theoccurrence of life-threatening ventricular arrhythmias i.e. sudden cardiac death.In this study, which evaluated 70 cases of sudden death in a population of 605postinfarction patients after a follow-up period of more than 10 years, end-systolic volume was superior to ejection fraction in predicting survival. In addi-tion, end-systolic volume proved more important than the extent of coronaryartery disease. Patients in this study did not receive thrombolytic therapy. In thepresent study, all patients received streptokinase for a first anterior myocardialinfarction. Although follow up was considerably shorter, and the presence ofhigh-grade ventricular arrhythmias during Holter recording was used as a pri-mary endpoint, end-systolic volume at discharge was also found a better predic-tor of ventricular arrhythmias than ejection fraction. In addition, it was shownthat in patients with these arrhythmias left ventricular dilatation was progressive,and dilatation after discharge also proved an independent predictor of ventricu-lar arrhythmias. Primary myocardial damage, reflected by end-systolic volume at discharge,and subsequent remodeling, objectivated by an increase in end-systolic volumeafter discharge, both independently increase the risk for late ventricular ar-rhythmias.

Finally, our present study confirmed the importance of the number of ven-tricular premature beats during Holter monitoring as a risk factor for ventriculararrhythmias, which has been demonstrated both with21 and without22,23 throm-bolytic therapy.

Underlying mechanisms

Increased loading conditions of the heart can lead to changes in repolariza-tion characteristics,11-13 usually resulting in a shortening of the refractory period.It has been shown, that these changes in repolarization are not uniformly dis-tributed over the ventricle, and therefore dispersion in refractoriness will in-crease.13 Increased dispersion in repolarization may cause a conduction delaysufficient for the development of a sustained ventricular tachyarrhythmia, evenwithout the presence of an anatomical substrate.24,25

DILATATION AND VENTRICULAR ARRHYTHMIAS

65

Early afterdepolarizations have also been recorded during increased loadingconditions of the heart.12,14 Early afterdepolarizations may facilitate the occur-rence of ventricular arrhythmias by inducing triggered activity. Finally, de-rangement of the three-dimensional structure of the heart, characterized by hy-pertrophy, fibrosis and cell slippage, may lead to anomalous cellular coupling,also resulting in delayed conduction and subsequent arrhythmias.26

Effect of ACE inhibition

Although a slightly larger proportion of patients without high-grade ventricu-lar arrhythmias used ACE inhibitors compared to patients with these arrhythmias(46% vs 35%), there was no clear indication in this study that ACE inhibitionreduced the prevalence of ventricular arrhythmias. However, the main results ofCATS indicate that nonsustained ventricular tachycardia was significantly re-duced by captopril in the acute phase of myocardial infarction. This was paral-leled by a significant reduction in norepinephrine levels, but a difference in leftventricular volumes could not be observed.27 Other studies investigating the ef-fect of ACE inhibition on the occurrence of ventricular arrhythmias after myo-cardial infarction do not allow definite conclusions.28-30 In patients with conges-tive heart failure, the evidence is more consistent.31-33 It appears that, if ACEinhibitors do reduce the incidence of ventricular arrhythmias after myocardialinfarction, this effect seems more pronounced in patients with extensive leftventricular remodeling resulting in heart failure.

Clinical relevance

Although anti-arrhythmic drugs have shown to reduce postinfarction ven-tricular arrhythmias, this did not result in a reduction of sudden cardiac death.34

Since left ventricular dilatation is an important risk factor for the occurrence ofsudden death,3 modulation of this process may prove a more appropriate ap-proach in reducing sudden death after myocardial infarction. ACE inhibitionafter myocardial infarction does improve survival.35 Part of this effect has beenattributed to attenuation of left ventricular dilatation. It is very likely that a re-duction in life-threatening ventricular arrhythmias related to this process alsoplays an important role. Therefore, after myocardial infarction, interventions di-rected at attenuating left ventricular dilatation may also reduce life-threateningventricular arrhythmias.Limitations

CHAPTER 2

66

In this study, only patients with a first anterior myocardial infarction treatedwith streptokinase were investigated. Nonetheless, within this population abroad clinical spectrum of patients was included, from aborted infarctions tolarge transmural infarctions with or without left ventricular dyskinesia and aneu-rysm formation. Therefore, the relation between dilatation and ventricular ar-rhythmias could be studied intensively, but these results should be extrapolatedwith caution to patients with different infarct sites or multiple infarcts.

Conclusions

After thrombolytic therapy, increased end-systolic volume at discharge isstrongly associated with the occurrence of high-grade ventricular arrhythmias.Furthermore, progressive dilatation after discharge also contributes independ-ently to the occurrence of these arrhythmias. Therefore, assessment of end-systolic volume, not only before discharge but also during the follow-up period,may help to detect patients at risk for late arrhythmic events.

References

1. Pfeffer MA, Brauwnwald E: Ventricular remodeling after myocardial infarction.Experimental observations and clinical implications. Circulation1990;81:1161-1172.

2. Kostuk WJ, Kazamias TM, Gander MP, Simon AL, Ross J: Left ventricular size aftermyocardial infarction: Serial changes and their prognostic significance. Circulation1973;47:1174-1179.

3. White HD, Norris RM, Brown MA, Brandt PWT, Whitlock RML, Wildt CJ: Leftventricular end -systolic volume as the major determinant of survival after recoveryfrom myocardial i nfarction. Circulation 1987;76:44-51.

4. Fletcher PJ, Pfeffer JM, Pfeffer MA, Brauwnwald E: Left ventricular diastolic pre s-sure-volume relations in rats with healed myocardial infarction. Circ Res1981;49:618-626.

5. Mitchell GF, Lamas GA, Vaughan DE, Pfeffer MA: Left ventricular remodeling inthe year after first anterior myocardial infarction: a quantitative analysis of co n-tractile segment lengths and ventricular shape. J Am Coll Cardiol1992;19:1136-1144.

6. Weisman HF, Bush DE, Mannisi JA, Weisfeldt ML, Healy B: Cellular mechanismsof myocardial infarct expansion. Circulation 1985;57:186-201.

7. Anversa P, Olivetti G, Capasso JM: Cellular basis of ventricular remodeling aftermyocardial infarction. Am J Cardiol 1991;68:7D-16D.

8. Gaudron P, Eilles C, Kugler I, Ertl G: Progressive left ventricular dysfunction andremodeling after myocardial infarction. Circulation 1993;87:755-763.

DILATATION AND VENTRICULAR ARRHYTHMIAS

67

9. Hansen DE, Craig CS, Hondeghem LM: Stretch -induced arrhythmias in the isolatedcanine ventricle. Circulation 1990;81:1094-1105.

10. Franz MR, Cima R, Wang D, Profitt D, Kurtz R: Electrophysiological effects ofmyocardial stretch and mechanical determinants of stretch -activated arrhythmias.Circulation 1992;86:968-978.

11. Lerman BB, Burkhoff D, Yue DT, Franz MR, Sagawa K: Mechanoelectrical fee d-back: independent role of preload and contractility in modulation of canine ve n-tricular excitability. J Clin Invest 1985;76:1843 -1850.

12. Levine JH, Guarnieri T, Kadish AH, White RI, Calkins H, Kan JS: Changes in my o-cardial repolarization in patients undergoing balloon valvuloplasty for congenitalpulmonary stenosis: evidence for contraction -excitation feedback in humans. Circu-lation 1988;77:70-77.

13. Taggart P, Sutton PM, Treasure T, Lab M, O'Brien W, Runnalls M, Swanton RH,Emanuel RW: Monophasic action potentials at discontinuation of cardiopulmonarybypass: evidence for contraction -excitation feedback in man. Circulation1988;77:1266-1275.

14. Taggart P, Sutton P, John R, Lab M, Swanton H: Monophasic action potential r e-cordings during acute changes in ventricular loading induced by the Valsalva m a-noeuvre. Br Heart J 1992;67:221-229.

15. Taggart P, Sutton P, Lab M: Inter action between ventricular loading and repolaris a-tion: relevance to arrhythmogenesis. Br Heart J 1992;67:213-215.

16. van Gilst WH, Kingma JH, for the CATS investigators group: Early interventionwith angiotensin -converting enzyme inhibitors during thrombolytic therapy inacute myocardial infarction: Rationale and design of Captopril and ThrombolysisStudy. Am J Cardiol 1991;68:111D -115D.

17. Schiller NB, Shah PM, Crawford M, DeMaria A, Devereux R, Feigenbaum H, Gu t-gesell H, Reichek N, Sahn D, Schnittger I, Silverman NH, Tajik AJ: Recommend a-tions for quantitation of the left ventricle by two -dimensional echocardiography.American Society of Echocardiography Committee on Standards, Subcommittee onQuantitation of Two -Dimensional Ech ocardiograms. J Am Soc Echocardiogr1989;2:358-367.

18. van der Laarse A, Kerkhof PL, Vermeer F, Serruys PW, Hermens WT, VerheugtFW, Bar FW, Krauss XH, van der Wall EE, Simoons ML: Relation between infarctsize and left ventricular performance assessed in patients with first acute myocardialinfarction randomized to intracoronary thrombolytic therapy or to conventionaltreatment. Am J Cardiol 1988;61:1-7.

19. Altman DG: Relation between several variables. In: Practical statistics for medicalresearch. 1st ed. London: Chapman and Hall, 1990:325 -364.

20. Bigger Jr JT: Relation between left ventricular dysfunction and ventricular a r-rhythmias after myocardial infarction. Am J Cardiol 1986;57:8B-14B.

21. Maggioni AP, Zuanetti G, Franzosi MG, Rovelli F, Santoro E, Staszewsky L,Tavazzi L, Tognoni G, on behalf of GISSI -2 investigators: Prevalence and progno s-tic significance of ventricular arrhythmias after acute myocardial infarction in thefibrinolytic era. GISSI -2 results. Circulation 1993;87:312-322.

CHAPTER 2

68

22. Bigger Jr JT, Fleiss JL, Kleiger R, Miller JP, Rolnitzky LM, and the MulticenterPost-Infarction Research Group: The relationships among ventricular arrhythmias,left ventricular dysfunction, and mortality in the 2 years after myocardial infarction.Circulation 1984;69:250-258.

23. Mukharji J, Rude RE, Poole WK, Gustafson N, Thomas LJ, Strauss HW, Jaffe AS,Muller JE, Roberts R, Raabe DS, Croft CH, Passamani E, Braunwald E, WillersonJT, and the MILIS study group: Risk factors for sudden death after acute myocardialinfarction: two -year follow-up. Am J Cardiol 1984;54:31-36.

24. Allessie MA, Bonke FM, Schopman FJG: Circus movement in rabbit atrial muscleas a mechanism of tachycardia: II. The role of nonuniform recovery of excitabilityin the occurrence of unidirectional block, as studied with multiple microelectrodes.Circ Res 1976;39:168-177.

25. Kuo CS, Atarashi H, Reddy CP, Surawicz B: Dispersion of ventricular repolariz a-tion and arrhythmia: study of two consecutive ventricular premature complexes.Circulation 1985;72:370-376.

26. Saffitz JE, Corr PB, Sobel BE: Arrythmogenesis and ventricular dysfunction aftermyocardial infarction. Is anomalous cellular coupling the elusive link? Circulation1993;87:1742-1745.

27. van Gilst WH, Kingma JH: Captopril during thrombolysis in myoca rdial infarction:a 3 months follow -up. J Am Coll Cardiol 1993;21:224A. (abstract)

28. Ray SG, Pye M, Oldroyd KG, Christie J, Connelly DT, Northridge DB, Ford I,Morton JJ, Dargie HJ, Cobbe SM: Early treatment with captopril after acute my o-cardial infarction. Br Heart J 1993;69:215-222.

29. Pipilis A, Flather M, Collins R, Hargreaves A, Kolettis T, Boon N, Foster C, A p-pleby P, Sleight P: Effects on ventricular arrhythmias of oral captopril and of oralmononitrate started early in acute myocardial infarction: results of a randomizedplacebo controlled trial. Br Heart J 1993;69:161-165.

30. Bussmann WD, Micke G, Hildenbrand R, Klepzig H Jr: Captopril bei akutemHerzinfarkt: Einfluss auf Infarctgrösse und Rhythmusstörungen. Dtsch MedWochenschr 1992;117:651 -657.

31. Fonarow GC, Chelimsky -Fallick C, Stevenson LW, Luu M, Hamilton MA,Moriguchi JD, Tillisch JH, Walden JA, Albanese E: Effect of direct vasodilatationwith hydralazine versus angiotensin -converting enzyme inhibition with captopril onmortality in advanced heart failure: the Hy -C trial. J Am Coll Cardiol1992;19:842-850.

32. Fletcher RD, Cintron GB, Johnson G, Orndorff J, Carson P, Cohn JN: Enalapril d e-creases prevalence of ventricular tachycardia in patients with chronic congestiveheart failure. The V -HeFT II VA Cooperative Studies Group. Circulation1993;87(6 Suppl):VI49 -VI55.

33. Cleland JG, Dargie HJ, Hodsman GP, Ball SG, Robertson JI, Morton JJ, East BW,Robertson I, Murray GD, Gillen G: Captopril in heart failure. A double blind co n-trolled trial. Br Heart J 1984;52:530-535.

34. Echt DS, Liebson PR, Mitchell LB, Peters RW, Obias -Manno D, Barker AH, Aren s-berg D, Baker A, Friedman L, Greene L, Huther ML, and the CAST investigators:

DILATATION AND VENTRICULAR ARRHYTHMIAS

69

Mortality and morbidity in patients receiving encainide, flecainide or placebo. TheCardiac Arrhythmia Suppression Trial. N Engl J Med 1991;324:781 -788.

35. Pfeffer MA, Brauwnwald E, Moyé LA, Basta L, Brown EJ, Cuddy TE, Davis BR,Geltman EM, Goldman S, Flaker GC, Klein M, Lamas GA, Packer M, Rouleau J,Rouleau JL, Rutherford J, Wertheimer JH, Hawkins CM, on behalf of the SAVE i n-vestigators: Effect of captopril on mortality and morbidity in patients with left ve n-tricular dysfunction after myocardial infarction. Results of the Survival and Ve n-tricular Enlargement Trial. N Engl J Med 1992;327:669 -677.

CHAPTER 3

Association of left ventricular remodeling andnonuniform electrical recovery expressed by nondipolar

QRST integral map patterns in survivors of a firstanterior myocardial infarction

Jan-Henk E. Dambrink,1 MD; Arne Sippens Groenewegen,2 MD;Wiek H. van Gilst,3 PhD; Kathinka H. Peels,4 MD; Cornelis A. Grimbergen,5

PhD; and J. Herre Kingma,1,3 MD, for the CATS investigators6

1Department of Cardiology, St Antonius Hospital, Nieuwegein2Department of Cardiology, Heart-Lung Institute, University Hospital Utrecht,Utrecht

3Department of Clinical Pharmacology, University of Groningen, Groningen4Department of Cardiology, Catharina Hospital, Eindhoven5Laboratory of Medical Physics, Faculty of Medicine, University ofAmsterdam, Amsterdam

6see Appendix

Published in:Eur Heart J 1993;14(suppl):24 (abstract)Circulation 1995;92:300-310

REMODELING AND NONUNIFORM RECOVERY

71

Abstract

Background. Progressive left ventricular dilatation after myocardial infarctionis associated with a high mortality rate, the majority of which is arrhythmogenicin origin. The underlying mechanism of this relation remains unknown. It hasbeen suggested, however, that left ventricular dilatation is accompanied bychanges in repolarization characteristics which may facilitate the occurrence oflife-threatening ventricular arrhythmias.Methods and results. We examined 62-lead body surface QRST integral mapsduring sinus rhythm in 78 patients, 349 ± 141 days after thrombolysis for a firstanterior myocardial infarction. Visual map analysis was directed at discriminat-ing dipolar (uniform repolarization) from nondipolar patterns (nonuniform re-polarization). In addition, the nondipolar content of each map was assessedquantitatively using eigenvector analysis. Nondipolar map patterns were presentin almost one third of the patients (32%). Left ventricular end-systolic and end-diastolic volumes were assessed echocardiographically before discharge andafter three and 12 months using the modified biplane Simpson’s rule. Increase inleft ventricular end-systolic volume one year after myocardial infarction wasmore pronounced in patients with nondipolar QRST integral map patterns (15 ±14 versus 4 ± 8 ml/m2, p=0.017). In patients with an increase in end-systolicvolume of more than 16 ml/m2 (upper quartile), the prevalence of nondipolarmaps was 89% compared to 29% in patients with dilatation of less than 16ml/m2. In addition, the nondipolar content of maps in patients in the upperquartile was significantly increased compared to the lower quartiles (49 ± 14versus 37 ± 12%, p=0.013). A logistic regression analysis revealed that an end-systolic volume of more than 42 ml/m2 after one year contributed independentlyto the appearance of nondipolar maps. Patients with high-grade ventricular ar-rhythmias showed a higher nondipolar content (49 ± 17% versus 39 ± 10%,p=0.013). QTc dispersion did not discriminate between patients with and with-out high-grade ventricular arrhythmias. Finally, the association between leftventricular remodeling and nondipolar map patterns was confirmed prospec-tively in an additional group of 15 patients.Conclusions. Nondipolar map patterns are present in 32% of cases afterthrombolysis for a first anterior myocardial infarction and are associated withmore left ventricular dilatation. These data support the hypothesis that left ven-tricular dilatation after myocardial infarction leads to changes in repolarizationcharacteristics which may facilitate the occurrence of life-threatening ventricu-lar arrhythmias.

CHAPTER 3

72

Introduction

Left ventricular remodeling after myocardial infarction is characterized byexpansion of the infarcted area and dilatation and hypertrophy of the noninfarc-ted myocardium.1 This process is accompanied by an overall increase in leftventricular volume. An increase in left ventricular end-systolic volume aftermyocardial infarction has been identified as an important risk factor for suddencardiac death.2 At present, the exact mechanism of this association is unclear. Ithas been reported that increased loading conditions and dilatation of the heart isparalleled by changes in repolarization characteristics3-6 and the occurrence ofearly afterdepolarizations.4,6 These altered electrophysiologic conditions maypredispose to the occurrence of ventricular arrhythmias.7-9

Local repolarization properties during sinus rhythm can be assessed nonin-vasively using multiple-lead electrocardiography (body surface mapping). TheQRST deflection area or isointegral has been identified both in theory10-12 and inanimal experiments13 as a marker of changes in local repolarization properties,irrespective of the site of origin of ventricular activation.14 A high incidence ofnondipolar QRST integral map patterns has been reported in patients withdocumented sustained ventricular tachycardia or ventricular fibrillation.15-18 Incontrast, normal subjects usually demonstrate a dipolar QRST integral map pat-tern.19,20 It has been suggested that nondipolar QRST integral map patterns re-flect nonuniformity of local recovery properties and can be used to identify pa-tients at risk of developing life-threatening ventricular arrhythmias.16,21,22

In this study, we hypothesize that left ventricular remodeling after myocardialinfarction leads to changes in local repolarization properties, which can be de-tected noninvasively as nondipolar QRST integral map patterns and may ex-press a potentially arrhythmogenic state.

Methods

Patients. This study was part of the Captopril And Thrombolysis Study (CATS),in which the effect of captopril treatment, started during thrombolysis, wasevaluated in patients with a first anterior myocardial infarction.23 Main end-points included left ventricular remodeling, neurohumoral activation and ven-tricular arrhythmias. In CATS, 298 patients were included in 12 hospitals in TheNetherlands. Selection criteria included a typical history of chest pain consistentwith acute myocardial infarction, onset of symptoms no longer than 6 hours be-fore admission, and ECG criteria for acute anterior myocardial infarction includ-ing at least 1 mm ST segment elevation in leads I and aVL and/or 2 mm ST

REMODELING AND NONUNIFORM RECOVERY

73

segment elevation in two or more precordial leads of the 12-lead electrocardio-gram, consistent with anterolateral, anteroseptal and/or anterior myocardial in-farction. Patients had to be eligible for thrombolytic therapy. Exclusion criteriaincluded presence of a prior myocardial infarction, left bundle branch block andsevere heart failure (Killip class III or IV). In 78 of these 298 patients, 62-leadbody surface mapping was performed during sinus rhythm. Baseline character-istics of all CATS patients and the patients who underwent a body surfacemapping recording are listed in Table 3.1. The prevalence of nondipolar mappatterns in the chronic phase of myocardial infarction was determined in all 78patients who underwent body surface mapping. The relation between nondipo-larity of QRST integral maps and remodeling was analyzed in a subset of 50patients in whom left ventricular volume assessment at one year was available.In addition, the association between nondipolarity and the occurrence of ven-tricular arrhythmias was investigated in 50 patients with Holter monitoring per-formed at one year.

Body surface mapping. All patients were without angina pectoris and in sinusrhythm at the time of investigation. Patients with right bundle branch block orsecond or third degree AV-block were excluded. None of the patients had leftbundle branch block. The time interval between body surface mapping andmyocardial infarction varied between 5 and 28 months (349 ± 141 days). Bodysurface mapping was performed using a portable microcomputer based mapping

Table 3.1. Characteristics of all CATS patients and those participating in the bodysurface mapping substudy at hospital admission

All CATS patients Body surface mapping

N 298 78MaleAge (years)α-HBDH Q72 (U/l)LVEDVI (ml/m2)LVESVI (ml/m2)LVEF (%)

224/298 (75%)59 ± 10

1277 ± 100756 ± 1325 ± 1055 ± 10

65/78 (83%)58 ± 8

1323 ± 84353 ± 1225 ± 953 ± 9

WMSI 1.91 ± 0.36 1.93 ± 0.37

α-HBDH Q72 indicates cumulative alpha-hydroxybutyrate dehydrogenase over the first72 hours after myocardial infarction; CATS, Captopril And Thrombolysis Study;LVEDVI, left ventricular end-diastolic volume indexed for body surface area; LVEF, leftventricular ejection fraction; LVESVI, left ventricular end-systolic volume indexed forbody surface area; WMSI, wall motion score index.

CHAPTER 3

74

system that includes a 62-lead electrode array (Figure 3.1). This array also in-cludes the six standard precordial leads (V1 - V6). The procedure of recording ofECG signals has been described before.24,25 In summary, 62 unipolar lead re-cordings were made with Wilson’s central terminal as reference. Single beatECG tracings were stored on floppy disk after A/D conversion with a samplingfrequency of 500 Hz. The noise of this system was estimated at 0.03 mV. Thestored data were then transferred to an Amiga 500 microcomputer(Commodore-Amiga Inc.) for further processing. Offset differences and/or lin-ear baseline drifting were corrected by a computer algorithm. ECG signals ofunacceptable quality were replaced by values calculated from surroundingleads. The average number of leads that was rejected per measurement was 0.5± 0.7. Isopotential maps were generated at a 2 ms interval during the QRSTcomplex. The onset of the QRS complex was defined as the time instant whereearliest activation exceeded 0.05 mV; offset of the T-wave was set at the pointin time where the maximum positive or negative voltage dropped below 0.05mV. U-waves were not included in the QRST interval. An integral map of thecomplete QRST interval was thereafter constructed19 and a hardcopy of this mapwas produced. QRST integral maps were visually evaluated by two observersunaware of left ventricular dimensions. Maps were defined nondipolar if 3 ormore extrema were present. An extreme was considered present if an area ofequal polarity included recordings of at least two unipolar leads. Maps with so-called pseudopods16 without additional extrema were considered dipolar. If dif-ferences existed between evaluations of both observers, maps were reevaluatedand classified after discussion until consent was reached.

In addition, a supportive quantitative analysis was carried out. Singular valuedecomposition was applied to the QRST intervals of all 78 patients, using Mat-lab software (The MathWorks Inc.) installed on a Sun Sparc Station 2 computer(Sun Microsystems Inc.). The first 12 eigenvectors of the entire set were de-rived, and the QRST integral map of each patient was expressed in terms ofthese eigenvectors as described by Lux et al.26 These authors demonstrated thatby using the first 12 eigenvectors of the complete dataset, each individual mapcould be reconstructed without any significant loss of information, yielding a16: 1 data reduction. In addition, the first three eigenvectors of their set revealeddipolar patterns, whereas eigenvectors 4 - 12 showed nondipolar patterns.Therefore, the nondipolar content of each map was defined as the contributionof eigenvectors 4 - 12 relative to the contribution of eigenvectors 1 - 12. In thefirst 20 patients, two consecutive measurements were carried out to assess thereproducibility of this method.

REMODELING AND NONUNIFORM RECOVERY

75

Echocardiography. According to the CATS protocol, 2D-echocardiography wasperformed within 24 hours after admission, after three days, before dischargeand at three and 12 months after myocardial infarction. Left ventricular end-systolic and end-diastolic volumes were calculated from a two- and four-chamber view using the modified biplane Simpson’s rule.27 The ejection frac-tion was calculated from these volume measurements. Measurements weremade off-line from end-diastolic and end-systolic stillframes using a DataviewMicrosonics cardiac analysis system (Nova Microsonics). Left ventricular vol-umes were indexed for body surface area. Furthermore, regional wall motionabnormalities were evaluated using the wall motion score recommended by theAmerican Society of Echocardiography.27 In this scoring system the left ventri-cle is divided into 16 segments, scoring each segment as 1 for normokinesia, 2for hypokinesia, 3 for akinesia, 4 for dyskinesia and 5 for an aneurysmal seg-ment. A Wall Motion Score Index (WMSI) was computed as the sum of scoresof all segments divided by the number of segments evaluated. Twelve evaluablesegments were considered a minimum to reliably assess WMSI.

CHAPTER 3

76

Holter recording. The occurrence of ventricular arrhythmias was assessed bytwo-channel 24-hour ambulatory Holter monitoring (Reynolds Medical Trackerrecorder) 12 months after myocardial infarction. Analysis was performed withhelp of a Reynolds Medical Pathfinder 3 analysis system. Tapes were analyzedfor the presence of uniform and multiform premature ventricular beats, pairs ofpremature ventricular beats and ventricular tachycardia. Pairs were defined as arepetition of two ventricular beats with a maximum interval of 0.6 s. Ventriculartachycardia was defined as a repetition of three or more ventricular beats with arate exceeding 100 beats per minute. High-grade ventricular arrhythmias weredefined as Lown classes 4A and B (paired ventricular premature beats and ven-tricular tachycardia) since patients with repetitive forms of ventricular premature

Figure 3.1. Diagram showing the location of the 62 electrodes on the thorax. Seven elec-trode columns cover the front (left panel) and another 7 columns cover the back (rightpanel) of the thorax which results in 41 anterior and 21 posterior electrodes. The minimalvertical interelectrode distance is 3.3 cm; the horizontal distance between columns isadapted to the size of the thorax of the individual patient. The density of electrodes ismaximal in the precordial region. Leads V1 to V6 are indicated by open circles. The selec-tion of lead positions in the array has previously been described.25

REMODELING AND NONUNIFORM RECOVERY

77

beats are especially at high risk of sudden cardiac death.28 Life-threatening ven-tricular arrhythmias, as mentioned in the text, were defined as primary ventricu-lar fibrillation, ventricular tachycardia degenerating into ventricular fibrillationor ventricular tachycardia leading to hemodynamic collapse.

Infarct size. Enzymatic infarct size was determined by analysis of the cumula-tive alpha-hydroxybutyrate dehydrogenase (α-HBDH) washout curve, calcu-lated from α-HBDH samples twice daily during the first five days, as describedby van der Laarse et al.29

Twelve-lead electrocardiogram. The presence of a Q-wave myocardial infarc-tion was investigated by analysis of the 12-lead ECG. QT intervals were meas-ured from the onset of the QRS to the end of the T-wave, defined as a return tothe T-P baseline. When U-waves were present the QT interval was measured tothe nadir of the curve between the T- and U-waves. When the end of the T-wavecould not be reliably identified, that lead was not included in subsequent analy-sis. QT- and QTc-dispersion was determined in all patients in the body surfacemapping substudy, and was defined as the difference between the maximumand minimum QT interval measured on the 12-lead electrocardiogram. Bazetts’formula was used to correct for heart rate: QTc = QT/√RR (RR interval in sec-onds).

Coronary angiography. Coronary angiography was performed when considerednecessary by the individual investigator. All available angiograms were studiedfor the presence or absence of an occluded infarct related artery. The severity ofcoronary artery disease was expressed as one-, two- or three-vessel disease.

Statistical analysis. Data are presented as mean ± standard deviation. Differ-ences were evaluated using the paired or unpaired Student’s t-test. Logistic re-gression analysis was used to identify variables independently predicting theoccurrence of nondipolar map patterns and high-grade ventricular arrhythmias,respectively.30 A two-sided p-value of less than 0.05 was considered statisticallysignificant. Calculations were made using an IBM PS/2 personal computer andSPSS/PC+ v 4.0 software.

CHAPTER 3

78

Results

Body surface mapping

Visual analysis revealed nondipolar maps in 25 of 78 patients (32%) and di-polar maps in the remaining 53 of 78 patients (68%). Examples of a typical di-polar and nondipolar QRST integral map pattern are given in Figure 3.2. Inpanel A, a dipolar QRST integral map of a patient with a limited myocardial in-farction and no left ventricular dilatation is displayed. Panel B features a nondi-polar QRST integral map of a patient with a large myocardial infarction and anearly two-fould increase in end-systolic volume after one year. Body surfacemaps were recorded at 200 and 221 days after myocardial infarction, respec-tively.

Quantitative assessment of nondipolarity was performed using singular valuedecomposition. The first 12 eigenvectors of the complete dataset and their sin-gular values are represented in Figure 3.3. These 12 eigenvectors include theQRST complexes of all 78 patients. This computed set of 12 eigenvectors wasused to calculate the nondipolar content of the QRST interval of each patient asthe contribution of eigenvectors 4 - 12 relative to eigenvectors 1 - 12. In the first20 patients, the nondipolar content was assessed in two consecutive measure-ments. It was found, that nondipolarity between measurements correlated well,r=0.89, p < 0.001. The mean difference between measurements was 1 ± 7%(range 0-12%).

The nondipolar content of the maps which were identified as dipolar was 36 ±10% and of maps considered nondipolar 52 ± 13%. The difference betweenthese groups was statistically significant (p < 0.001). In contrast, QTc-dispersionon the standard 12-lead ECG was not different between patients with dipolarand nondipolar maps (102 ± 46 versus 97 ± 36 ms), and there was no correla-tion between QTc-dispersion on the 12-lead ECG and nondipolar content(r=0.06, p=0.632). Additional prospective data are presented in the Appendix ofthis chapter.

REMODELING AND NONUNIFORM RECOVERY

79

CHAPTER 3

80

Echocardiography

Echocardiographic data of patients with and without nondipolar maps aresummarized in Table 3.2. Left ventricular dimensions were significantly differ-ent in patients with- and without nondipolar maps. End-systolic volume indexedfor body surface area (LVESVI) was already higher in the nondipolar group atday 1. This difference persisted until month 3 and 12. The absolute increase inend-systolic volume was more pronounced in patients with nondipolar maps.End-diastolic volume index (LVEDVI) was also larger in the nondipolar groupfrom day 1 on, and here also more dilatation appeared to be present. However,the difference in increase in end-diastolic volume between both groups was notstatistically significant.

←Figure 3.2. Panel A. Dipolar QRST integral map of a patient without left ventriculardilatation. The interval selected for computing the QRST integral is indicated by the verti-cal bars and the shaded area in lead V1 below the map. The position of lead V1 is also indi-cated on the map. Maximal and minimal QRST integral values are indicated by plus- andminus signs, respectively, and their values are given in mV.ms below the map. Positiveisointegral lines in the shaded area are indicated by solid lines with an interval of 10mV.ms, negative isointegral lines are represented by dashed lines with an interval of 4mV.ms and the single dotted line indicates the zero level. The position of the sternum andspinal column are indicated at the left and right top sides of the map, respectively. One maynote a clear dipolar pattern featuring a minimum at the upper anterior thorax and a maxi-mum at the left axilla. This patient had an enzymatic infarct size of α-HBDH Q72 639 U/l,a wall motion score index of 1.63, and an ejection fraction of 64%. The nondipolar contentwas 40%; end-systolic volume decreased from 17 ml/m2 upon admission to 13 ml/m2 12months after myocardial infarction. Panel B. Nondipolar QRST integral map of a patientwith left ventricular dilatation. Intervals between positive and negative isointegral lines areset at 6 mV.ms and 4.5 mV.ms, respectively. A total of 4 extremes can be observed. The twoextremes indicated by the plus and minus signs represent maximal and minimal QRST inte-gral values. A second negative extreme is positioned anteriorly at the right lower thoraxand is surrounded by a positive area. Another positive extreme is present at the left lowerback. The positive area located at the right scapula represents the QRST integral of onlyone scalar ECG lead and was therefore not considered as a separate extreme. One may notethat a saddle phenomenon, first described by Taccardi et al,41 is formed by the two positiveextremes and the negative extreme in between. This patient had an enzymatic infarct size ofa-HBDH Q72 3593 U/l, a wall motion score index of 2.44 and an ejection fraction of 39%.The nondipolar content was 73%; end-systolic volume increased from 33 ml/m2 upon ad-mission to 61 ml/m2 after 12 months.

REMODELING AND NONUNIFORM RECOVERY

81

A lower ejection fraction and higher wall motion score index were found inpatients with nondipolar map patterns. Ejection fraction remained stable in thefirst 12 months after myocardial infarction. The wall motion score index showeda small improvement during the first year of follow up in patients with dipolarmaps, whereas this index remained unchanged in those with nondipolar mappatterns.

In Figure 3.4, the percentage of patients with nondipolar maps is given forthe four quartiles of left ventricular dilatation one year after myocardial infarc-tion (see also legend of Figure 3.4). Especially in the upper quartile (increase inend-systolic volume of more than 16 ml/m2), the prevalence of nondipolar maps

Figure 3.3. The first 12 eigenvectors of the complete dataset are displayed. Each eigenvec-tor is represented in the body surface map format (see legends of Figure 3.2 for further de-tails). Rank number and singular value of each eigenvector are indicated below the map.The singular value is equivalent to the relative contribution of each eigenvector to thecomplete data set. Eigenvectors are sorted by singular value in descending order. The firstthree eigenvectors demonstrate a dipolar pattern; eigenvectors 4 to 12 show nondipolarpatterns. The map patterns displayed are comparable to those described by Lux et al..26 Therelatively high importance of the eigenvector ranked as number 1 is a characteristic featurein patients with anterior wall infarction.26

CHAPTER 3

82

was high (8/9 patients, 89%), paralleled by a higher nondipolar content in thesepatients (37 ± 12% in the lower three quartiles versus 49 ± 14% in the upperquartile, p=0.013). In contrast, there was no difference in QTc-dispersion on the12-lead ECG between these groups (90 ± 41 ms versus 105 ± 37 ms).

In Table 3.3, other variables besides left ventricular dilatation which mightexplain the presence of a nondipolar map pattern are considered. Patients withnondipolar map patterns showed a larger enzymatic infarct size. This was notparalleled by more Q-wave infarctions or a higher percentage of patients with anoccluded infarct related artery. However, multi-vessel disease was more promi-nent in patients with nondipolar maps. The number of interventions, the timeinterval between infarction and body surface mapping, heart rate and medica-tion were comparable between groups. A logistic regression analysis was per-formed to assess whether left ventricular remodeling contributed independentlyto the presence of nondipolar maps. End-systolic volume at one year was se-lected as the parameter of remodeling. First, enzymatic infarct size (α-HBDHQ72 of more than 1089 U/l, median value in this subset), was entered into themodel. After this, four levels of end-systolic volume (quartiles) were added. Anend-systolic volume of more than 42 ml/m2 (upper quartile) was found to con-tribute independently to the appearance of nondipolar maps (Table 3.4). Al-though patients with an end-systolic volume of more than 42 ml/m2 in the sub-group with an infarct size of more than 1089 U/l were randomized to captoprilmore frequently than patients with an end-systolic volume of less than 42 ml/m2

(55% versus 33%), patients with a large end-systolic volume more frequentlyneeded open captopril treatment (36% versus 0%), showed more Q-wave in-farctions (91% versus 67%), had more multi-vessel coronary artery disease(50% versus 38%) and showed an occluded infarct-related artery in a higherproportion of patients (50 versus 25%). None of these differences were statisti-cally significant, possibly due to the small number of patients in these sub-groups. Additional prospective data are presented in the Appendix of this chap-ter.

Incidence of ventricular arrhythmias detected during Holter monitoring afteroneyear

Holter monitoring after one year showed that patients with nondipolar mapsdemonstrated ventricular arrhythmias more frequently (Table 3.5). However,these differences were not statistically significant. The overall incidence of ven-tricular tachycardia was low. Only one patient had nonsustained ventriculartachycardia, and this patient showed a dipolar map pattern. In patients with left

REMODELING AND NONUNIFORM RECOVERY

83

ventricular dilatation of more than 16 ml/m2, ventricular arrhythmias were rela-tively frequent. All 4 of 6 patients with high-grade ventricular arrhythmias(Lown 4A and B) in this subgroup had nondipolar map patterns. The nondipolarcontent of maps in 14 patients with high-grade ventricular arrhythmias was 49 ±10% versus 39 ± 17% in 41 patients without these arrhythmias (p=0.013). Therewas no significant difference in QTc-dispersion on the 12-lead ECG betweenpatients with and without high-grade ventricular arrhythmias (116 ± 44 ms ver-sus 101 ± 38 ms).

In Table 3.6, a number of possible determinants of high-grade ventricular ar-rhythmias are listed. A logistic regression analysis (forward stepwise model)

0

20

40

60

80

100

*

8/92/93/93/9

IVIIIIII

(%)

Nondipolar maps (%)

(% ± SD)Nondipolar content

Figure 3.4. The percentage of patients with nondipolar maps are given for the 4 quartilesof left ventricular dilatation, defined as increase in left ventricular end-systolic volume in-dexed for body surface area (∆LVESVI) in the first year after myocardial infarction.Quartiles of left ventricular dilatation were the following: I. ∆LVESVI up to 1 ml/m2; II.∆LVESVI 1 to 8 ml/m2; III. ∆LVESVI 8 to 16 ml/m2; and IV. ∆LVESVI more than 16 ml/m2.Especially among patients in the upper quartile (IV), the prevalence of nondipolar mapswas high (8 out of 9 patients, 89%). This finding was supported by an increase in nondipo-lar content in QRST intervals of patients in the upper quartile.

CHAPTER 3

84

showed that change in end-systolic volume after one year was the only inde-pendent predictor of these arrhythmias.

Medication and nondipolar map patterns

REMODELING AND NONUNIFORM RECOVERY

85

There were no significant differences in concomitant medication between pa-tients with dipolar and nondipolar map patterns (Table 3.7). However, patientswith dipolar map patterns tended to be randomized to captopril more frequently,and used beta-blockers in a high percentage of patients. Apparently, patients

Table 3.2 . Echocardiographic data of patients in the body surface mapping substudy

Nondipolar QRST Dipolar QRST p-valueintegral (N) integral (N)

LVEDVI (ml/m 2) entry day 3 predischarge 3 months 12 months

59 ± 12 (20)61 ± 11 (20)69 ± 19 (22)72 ± 19 (15)77 ± 23 (18)

48 ± 10 (30)52 ± 12 (35)54 ± 11 (33)58 ± 15 (34)58 ± 16 (32)

0.0030.0090.0020.0090.002

LVESVI (ml/m 2) entry day 3 predischarge 3 months 12 months

31 ± 8 (20)31 ± 8 (20)36 ± 13 (22)36 ± 15 (15)44 ± 19 (18)

21 ± 7 (30)22 ± 8 (35)23 ± 8 (33)25 ± 11 (34)26 ± 12 (32)

< 0.001< 0.001< 0.001

0.0050.001

LVEF (%) entry day 3 predischarge 3 months 12 months

48 ± 8 (20)48 ± 9 (20)49 ± 7 (22)51 ± 11 (15)45 ± 13 (18)

57 ± 8 (30)57 ± 9 (35)58 ± 8 (33)58 ± 9 (34)56 ± 10 (32)

< 0.001 0.001

< 0.001 0.026 0.004

WMSI entry day 3 predischarge 3 months 12 months

2.11 ± 0.29 (20)2.12 ± 0.33 (20)2.08 ± 0.38 (22)1.99 ± 0.37 (15)2.08 ± 0.49 (18)

1.80 ± 0.37 (30)1.81 ± 0.32 (35)1.73 ± 0.39 (33)1.74 ± 0.37 (34)1.66 ± 0.39 (32)

0.0030.0010.0020.0330.004

Change in firstyear LVEDVI LVESVI LVEF

18 ± 16 (16)15 ± 14 (16)-5 ± 12 (16)

8 ± 14 (20)4 ± 8 (20)

-1 ± 10 (20)

0.0590.0170.355

WMSI 0.00 ± 0.35 (16) -0.15 ± 0.36 (20) 0.215

Abbreviations see Table 3.1.

CHAPTER 3

86

with nondipolar map patterns more frequently needed open captopril treatmentfor congestive heart failure.

Discussion

Left ventricular dysfunction is an important determinant of life-threateningventricular arrhythmias after myocardial infarction.2,31 In recent years, experi-mental evidence has become available indicating that left ventricular dilatation

Table 3.3. Other characteristics of patients with and without nondipolar map patt erns

Nondipolar (N) Dipolar (N) p-value

Demographics Age (years) MaleMyocardial injury α-HBDH Q72 (U/l) Q-wave infarctionExtent of CAD ≥ 2 vessel disease LAD occluded Time to CAG (days)*

Interventions PTCA CABGBody SurfaceMapping Days after infarction Heart rate (beats/min)Medication Beta-blockers Captopril randomized open label

58 ± 8(25)92% (23/25)

1818 ± 881 (23)76% (19/25)

50% ( 7/14)14% ( 2/14)

67 ± 127 (14)

12% (3/25) 4% (1/25)

357 ± 149 (25) 67 ± 11 (24)

33% ( 8/24)

40% (10/25)27% ( 6/22)

58 ± 9 (53)79% (42/53)

1087 ± 721 (48)66% (35/53)

21% ( 8/39)18% ( 7/39)45 ± 73 (39)

20% (11/53)5% ( 3/53)

345 ± 138 (53) 69 ± 11 (50)

43% (16/37)

52% (28/53)10% ( 4/38)

0.9430.205

< 0.0010.531

0.0460.9190.545

0.5291.000

0.7210.555

0.613

0.4150.149

Digoxin 9% ( 2/22) 7% ( 3/38) 1.000

CABG indicates coronary artery bypass grafting; CAD, coronary artery disease; LAD, LeftAnterior Descending artery; PTCA, percutaneous transluminal coronary angioplasty.Other abbreviations see Table 3.1. *Coronary angiograms were made 31 ± 68 days (range 0- 408) after myocardial infarction.

REMODELING AND NONUNIFORM RECOVERY

87

is associated with increased dispersion in repolarization.3-6 These studies suggestthat left ventricular remodeling after myocardial infarction, a process which ischaracterized by left ventricular dilatation, may also lead to changes in repolari-zation characteristics. Increased dispersion in refractoriness in dilated ventriclesmay lower the threshold for the occurrence of life-threatening ventricular ar-rhythmias.32,33

In the present study, body surface mapping was used to detect changes in re-polarization characteristics accompanying left ventricular dilatation after myo-cardial infarction. Increased dispersion in repolarization, represented by nondi-polar map patterns, appeared to be more frequent in patients with left ventriculardilatation, especially when dilatation exceeded 16 ml/m2 (89% prevalence ofnondipolar maps). This was supported by a higher nondipolar content of QRSTintegral map patterns in these patients. In addition, when corrected for infarctsize, end-systolic volume after one year still contributed independently to theappearance of nondipolar QRST integral map patterns. The present data indicatethat left ventricular remodeling after myocardial infarction is accompanied byaltered repolarization characteristics, which may facilitate the occurrence of life-threatening ventricular arrhythmias.

Changes in left ventricular dimensions and the occurrence of ventriculararrhythmias

The prognostic significance of left ventricular dimensions was demonstratedby White et al.2 who found that end-systolic volume was the most powerfulpredictor of death after myocardial infarction. Since death was found to be sud-den in 70% of cases in this study, left ventricular dilatation was believed to beassociated with an increased risk for fatal ventricular arrhythmias. In two recentexperimental studies,34,35 the relation between left ventricular volume changesand the occurrence of ventricular arrhythmias was investigated. Hansen et al.35

observed in their isolated canine heart model that a sudden increase in left ven-

Table 3.4. Results of logistic regression analysis

B SE p-value OR (95%-CI)

α-HBDH Q72 > 1089 U/l 0.8649 0.8232 0.2934 2.37 (0.47-11.8)LVESVI > 42 ml/m2 2.1972 0.9930 0.0269 9.00 (1.30-62.8)Constant -1.5580 0.5501

B indicates regression coefficient; CI, confidence intervals; OR, odds ratio; SE, standarderror. Other abbreviations see Table 3.1.

CHAPTER 3

88

tricular volume applied during diastole induced frequent ventricular prematurebeats and occasional nonsustained ventricular tachycardia. In a similar study,Franz et al.34 investigated the arrhythmogenic effect of slow versus rapid in-crease in left ventricular volumes in the isolated Langendorff perfused rabbitheart. Both forms of volume changes induced an increase in ventricular ar-rhythmias, but especially rapid volume pulses caused frequent ectopic ventricu-lar excitation.

These studies confirm that an increase in left ventricular volume is accompa-nied by frequent ventricular arrhythmias. In the present study, high-grade ven-tricular arrhythmias were a relatively rare observation (Table 3.5). However,paired ventricular premature beats were especially frequent in patients withdilatation more than 16 ml/m2 in the presence of nondipolar map patterns.

Underlying mechanisms

Changes in repolarization characteristics. Changes in repolarization charac-teristics accompanying increased loading conditions or increase of left ventricu-lar volume have been described both in animal experiments3 and in clinicalstudies.4-6 Lerman et al. demonstrated in dogs that increasing the preload of theheart significantly shortens the refractory period.3 Obstruction of the right ven-

Table 3.5. Ventricular arrhythmias detected during Holter monitoring in patients withand without nondipolar maps or dilatation after 12 months

Nondipolar Dipolar p- value Dilatation NoDilatation

p-value

(N) (N) (N) (N)

VPBs 33% 21% 0.510 66% 15% 0.028 >10/hour ( 6/18) ( 8/37) (4/6) ( 3/20)Bigemini of VPBs

11%(2/18)

10%(4/37)

1.000 33%(2/6)

5%(1/20)

0.123

Multiform VPBs

77%(14/18)

62%(23/37)

0.394 100%(6/6)

60%(12/20)

0.132

Paired VPBs

33%(6/18)

18%(7/37)

0.314 66%(4/6)

0%(0/20)

0.001

VT 0% 2% 1.000 0% 0% - **

(0/18) (1/37) (0/6) (0/20)

VPBs indicates ventricular premature beats; VT, ventricular tachycardia. *Dilatation wasdefined as an increase in end-systolic volume after 12 months of more than 16 ml/m 2. **pis not defined.

REMODELING AND NONUNIFORM RECOVERY

89

tricular outflow tract during balloon valvuloplasty4 and performance of the Val-salva maneuver6 have also been found to reduce the refractory period. In pa-tients being weaned off from extracorporal life-support after bypass surgery,Taggart et al.5 observed a shortening of the refractory period when the left ven-tricle was filled to normal dimensions. Since these changes in refractory periodwere not uniformly distributed over the ventricle, dispersion in refractorinesswas considered to be increased.

No long-term studies are available showing the relation between left ventricu-lar dilatation and changes in repolarization characteristics. All of the previousstudies describe an experimental situation in which left ventricular volume wasincreased for a short period of time. The present study describes repolarizationcharacteristics after chronic left ventricular dilatation.

Early afterdepolarizations. Early afterdepolarizations (EADs) were observedduring increased loading conditions in two of the previously mentioned clinicalstudies.4,6 This phenomenon was detected during balloon valvuloplasty for pul-monary valve stenosis4 and during a Valsalva maneuver carried out during rou-tine coronary angiography6 both by measurement of monophasic action poten-tials. Early afterdepolarizations are oscillations in the membrane potential thatoccur during the plateau of the repolarization phase which may induce single ormultiple action potentials when they reach threshold level9 (triggered activity).

In the above mentioned studies, EADs were found in a minority of cases. Theexact role of EADs in the cause of ventricular arrhythmias in patients with leftventricular dysfunction remains unclear.36

Influence of thrombolytic therapy. Arrhythmogenic vulnerability is decreasedafter thrombolytic therapy.37 Previous studies have shown that successfulthrombolysis leads to a reduced incidence of late potentials, a marker of thepresence of an arrhythmogenic substrate that may cause reentrant ventricular

Table 3.6. Possible determinants of high-grade ventricular arrhythmias

high-grade No high-grade p-valuearrhythmias N arrhythmias N

α-HBDH Q72 (U/l) 1831 ± 868 11 1071 ± 752 40 0.006LVESVI (ml/m2)LVEF (%)∆LVESVI (ml/m2)nondipolar

45 ± 1646 ± 1328 ± 10

884

29 ± 1556 ± 115 ± 10

292922

0.0150.041

< 0.001

content (%) 49 ± 17 13 39 ± 10 42 0.013

∆LVESVI indicates change in LVESVI after one year. Other abbreviations see Table 3.1.

CHAPTER 3

90

tachyarrhythmias.38 In addition, it is known that a patent infarct-related artery isassociated with less left ventricular dilatation during follow up.39 In the presentstudy, left ventricular dilatation is associated with increased dispersion in refrac-toriness. Therefore, a relation between patency of the infarct related artery andthe presence of nondipolar map patterns may be expected. The finding thatnondipolar map patterns were relatively more frequent in patients with an oc-cluded infarct-related artery is consistent with this hypothesis. However, a sig-nificant difference in nondipolarity could not be detected, possibly due to asmall patient cohort.

If this trend could be confirmed, it would imply that thrombolytic therapy notonly reduces the risk of the development of an anatomical substrate after myo-cardial infarction, but also is capable of reducing dispersion in refractoriness, another important arrhythmogenic factor.

Noninvasive assessment of repolarization characteristics by means of bodysurface mapping

Wilson et al.10 suggested that the deflection area of the QRST complex(QRST integral or ventricular gradient) was determined by differences in dura-tion of repolarization. Subsequent theoretical models confirmed this relation.11,12

Abildskov et al. provided direct evidence for this relationship in an experimentalstudy in dogs.13 Changes in refractory period induced by local warming wereclosely associated with similar changes in QRST integrals measured at the sameepicardial site. This association was found to be independent of the site of pac-ing (five sites). The authors concluded that QRST integrals represent primaryrepolarization characteristics, which are independent of the site of ventricular

Table 3.7. Medication at the time of body surface mapping

Nondipolar Dipolar p-value

Captoprilrandomizedopen label

Beta-blockerNitratesCalcium antagonistsDigoxin

40% (10/25)24% ( 6/25)32% ( 8/25)28% ( 7/25)20% ( 5/25) 8% ( 2/25)

52% (28/53)11% ( 6/53)42% (22/53)11% ( 6/53)17% ( 9/53) 8% ( 4/53)

0.4150.1841.0000.1010.7591.000

Anti-arrhythmic agents* 8% ( 2/25) 6% ( 3/53) 0.642* amiodarone or sotalol.

REMODELING AND NONUNIFORM RECOVERY

91

activation, as opposed to secondary repolarization characteristics which are de-pendent on the activation site as well as on a disturbed activation sequence dueto bundle branch block or prior infarction.

In man, QRST integrals have been studied extensively using body surfacemapping.15-20 It has been demonstrated in normal subjects that QRST integralmaps obtained during sinus rhythm demonstrate a homogeneous dipolar distri-bution over the thorax.19, 20 However, maps of patients with documented ven-tricular arrhythmias show inhomogeneous nondipolar patterns in a majority ofcases (63-71%).15-18,21 This association suggests that there is an increased vul-nerability for ventricular arrhythmias if QRST integrals are inhomogeneouslydistributed over the body surface. The underlying mechanism has not beenclarified in man. However, based on the previously mentioned animal experi-ments, it may be deduced that regional differences in QRST integrals on thebody surface reflect local, potentially arrhythmogenic, differences in repolariza-tion properties in the heart.

In our study, nondipolar maps were found in 32% of all patients. A similarfraction has been found in other postinfarction studies,16 despite the applicationof different criteria for the classification of QRST integral maps into dipolar andnondipolar patterns. In contrast to previously mentioned studies, nondipolarityof QRST integral maps did not provide an index of serious ventricular arrhyth-mias in the present study, since no patient died suddenly or suffered a clinicallyrelevant ventricular arrhythmia.

Nondipolar QRST integral map patterns and left ventricular dilatation

We hypothesized in this study that left ventricular dilatation leads to alteredrepolarization properties which can be detected by means of body surfaceQRST integrals. The incidence of nondipolar QRST integral map patterns washigh in patients with left ventricular dilatation: 8 out of 9 patients with an in-crease in LVESVI of more than 16 ml/m2 (upper quartile) had nondipolar pat-terns. This relation was further confirmed by the finding of a significant increasein nondipolar content in patients with left ventricular dilatation of more than 16ml/m2 (Figure 3.4). These data indicate that remodeling is frequently accompa-nied by changes in repolarization properties which represent an underlyingelectrophysiologic condition that may facilitate the occurrence of ventricular ar-rhythmias. It should be noted that these changes could not be detected usingQTc-dispersion on the standard 12-lead electrocardiogram.

Until now, studies investigating the association between left ventricular di-mensions and dispersion in repolarization were carried out in noninfarcted ven-tricles. Therefore, when studying the effect of remodeling on repolarization

CHAPTER 3

92

characteristics, infarct size should be taken into account. In this study it isshown that end-systolic volume after one year significantly contributes to theappearance of nondipolar map patterns, irrespective of enzymatically deter-mined infarct size.

Nondipolar content of QRST integral map patterns and ventricular arrhythmias

In this study, the nondipolar content of QRST integral maps was significantlyhigher in patients with high-grade ventricular arrhythmias during Holter moni-toring. However, logistic regression analysis revealed that the change in end-systolic volume after one year was the only independent predictor of these ar-rhythmias. The nondipolar content of the QRST integrals did not contribute in-dependently. This is probably due to the fact that in this study left ventriculardilatation and nondipolarity of QRST integral maps were closely related. Thefinding that left ventricular dilatation is a stronger predictor of ventricular ar-rhythmias than nondipolarity of QRST integral maps may indicate that, besidesdispersion in refractoriness, other mechanisms such as slow conduction,40 myo-cardial stretch,35 or early afterdepolarizations6 are responsible for the arrhyth-mogenic effects of left ventricular dilatation.

Limitations of the study

Body surface mapping was performed within a time window of 5 to 28months (‘cross-sectional’) after myocardial infarction had occurred and was cor-related with echocardiograms obtained at day 1, day 3, before discharge andafter three and 12 months. QRST integral map patterns obtained at the samepoint in time when echocardiographic recordings were made might have showndifferent results. In addition, if the development of dispersion in repolarization isrelated to ventricular remodeling, a time-dependent effect on the developmentof nondipolar QRST integral maps may be expected. In the present study, therewas no difference in the prevalence of nondipolar maps before or after 12months of follow up (Table 3.3). Furthermore, there was no relation betweenduration of follow up before body surface mapping and nondipolar content ofthe map. Therefore, a time dependent effect could not be demonstrated in thisstudy. However, in the Appendix of this chapter 15 new patients who under-went serial body surface mapping and echocardiography after myocardial in-farction are described. Data from these patients show that changes in nondipo-larity do occur during follow up after myocardial infarction. In this study it isdemonstrated that a change from a dipolar pattern to a nondipolar pattern isparalleled by an increase in end-systolic volume, and a conversion from a non-

REMODELING AND NONUNIFORM RECOVERY

93

dipolar pattern to a dipolar pattern is accompanied by a decrease in end-systolicvolume. Thus, despite the occurrence of time dependent changes in map pat-terns, the relation of nondipolarity in map pattern and increased end-systolicvolume remains intact.

In the retrospective study there was no direct correlation between the devel-opment of life-threatening ventricular arrhythmias and the presence or absenceof nondipolar map patterns. This finding is not surprising, since in the totalCATS population only 6/298 (2%) patients died suddenly during the first yearafter myocardial infarction. In addition, this study investigated patients surviv-ing for at least five months after myocardial infarction, who were well enough tovisit the out-patient clinic to undergo body surface mapping. Therefore, a groupof relatively well patients was selected. This may also explain why no seriousventricular arrhythmias were documented and the relation with nondipolar mappatterns could not be established. Still, the nondipolar content of maps in pa-tients with high-grade ventricular arrhythmias during Holter monitoring wassignificantly higher than in patients without these arrhythmias, underlining theassociation between ventricular arrhythmias and dispersion in refractoriness.

Even though the present study demonstrates the relation between left ven-tricular remodeling and nondipolarity of QRST integral maps, these data do notallow prediction of repolarization characteristics on the basis of an enlargedend-systolic volume. It does imply that, when assessing risk in postinfarctionpatients, an enlarged end-systolic volume should be considered a potent riskfactor, with increased dispersion in refractoriness as a possible underlying ar-rhythmogenic mechanism.

Conclusions

The present study describes a correlation between left ventricular remodelingand nondipolar QRST integral map patterns after myocardial infarction. Thesefindings do not allow inference about causality or prognosis. However, they doprovide a step towards the understanding of the relation between left ventriculardilatation and life-threatening ventricular arrhythmias.

The present study underlines that left ventricular remodeling may prove, bymeans of altered repolarization properties, an important risk factor for the occur-rence of life-threatening ventricular arrhythmias after myocardial infarction.

CHAPTER 3

94

Appendix

To investigate the presence or absence of time dependent changes in nondi-polarity of QRST integral maps after acute myocardial infarction, 15 other pa-tients underwent serial body surface mapping 8 ± 5 and 640 ± 223 days aftermyocardial infarction. Left ventricular dimensions were assessed by echocardi-ography before hospital discharge and 12 months after myocardial infarction.Measurements were performed as described in the Methods section of thischapter. Infarct location was assessed by standard 12-lead electrocardiographiccriteria. Of all patients, 6 had anterior, 8 inferior and 1 lateral myocardial infarc-tion. Only one patient had a previous myocardial infarction. All patients re-ceived thrombolytic therapy after hospital admission. Coronary angiographywas performed before hospital discharge. In 12/15 patients (80%), the infarctrelated artery was patent.

In Table 3.8 it is shown that there was no significant increase in nondiplarcontent during follow up. End-systolic and end-diastolic volume increased sig-nificantly, but the ejection fraction remained relatively unchanged. At baseline,4 of 15 patients (27%) had nondipolar maps. Of these patients, two also had anondipolar map at follow up, and two featured a change of a nondipolar into adipolar map. Two other patients had dipolar maps at baseline, and developednondipolar maps at follow up. The two patients who demonstrated conversionfrom dipolar to nondipolar maps showed an increase in end-systolic volume of7 and 5 ml/m2 resulting in LVESVI of 39 and 30 ml/m2, respectively. The non-dipolar content increased with 26 and 3% to 58 and 37%, respectively. The twopatients of whom maps changed from nondipolar to dipolar showed a decreasein end-systolic volume of 5 and 2 ml/m2, resulting in LVESVI of 18 and 26ml/m2, respectively. The nondipolar content of the maps in these patients de-

Table 3.8. Data on 15 postinfarction patients followed prospectively

Baseline Follow-up Change p-value

Nondipolar content (%) 34 ± 13 37 ± 10 3 ± 11 0.338LVESVI (ml/m2)LVEDVI(ml/m2)

21 ± 643 ± 10

25 ± 749 ± 11

4 ± 7 6 ± 10

0.0380.042

LVEF (%) 51 ± 6 49 ± 8 2 ± 7 0.379

Abbreviations see Table 3.1.

REMODELING AND NONUNIFORM RECOVERY

95

creased with 14 and 9% to 43 and 35%, respectively. At baseline, there were nosignificant differences in infarct size, end-systolic volume, end-diastolic volumeand ejection fraction between patients with dipolar or nondipolar map patterns.However, patients who showed nondipolar map patterns at follow up had a sig-nificantly larger end-systolic volume after one year (33 ± 5 vs 22 ± 5 ml/m2,p=0.004). The increase in end-systolic volume was also more pronounced inpatients with nondipolar maps, although this difference was not significant (9 ±5 vs 2 ± 6, p=0.074). Multiple regression analysis, using nondipolar content atfollow-up as a continuous dependent variable, revealed that the nondipolarcontent at baseline and end-systolic volume after one year were the only inde-pendent predictors of the nondipolar content at follow-up, explaining almost60% of the variation of this variable (R2=0.59).

We conclude that the data from this small set of prospectively evaluated pa-tients supports the hypothesis that left ventricular remodeling and nondipolarcontent as a measure of dispersion in refractoriness are clearly related.

References

1. Pfeffer MA, Brauwnwald E: Ventricular remodeling after myocardial infarction.Experimental observations and clinical implications. Circulation 1990;81:1161-1172.

2. White HD, Norris RM, Brown MA, Brandt PWT, Whitlock RML, Wildt CJ: Leftventricular end-systolic volume as the major determinant of survival after recoveryfrom myocardial infarction. Circulation 1987;76:44-51.

3. Lerman BB, Burkhoff D, Yue DT, Franz MR, Sagawa K: Mechanoelectrical fee d-back: independent role of preload and contractility in modulation of canine ve n-tricular excitability. J Clin Invest 1985;76:1843-1850.

4. Levine JH, Guarnieri T, Kadish AH, White RI, Calkins H, Kan JS: Changes in my o-cardial repolarization in patients undergoing balloon valvuloplasty for congenitalpulmonary stenosis: evidence for contraction-excitation feedback in humans. Circu-lation 1988;77:70-77.

5. Taggart P, Sutton PM, Treasure T, Lab M, O'Brien W, Runnalls M, Swanton RH,Emanuel RW: Monophasic action potentials at discontinuation of cardiopulmonarybypass: evidence for contraction-excitation feedback in man. Circulation1988;77:1266-1275.

6. Taggart P, Sutton P, John R, Lab M, Swanton H: Monophasic action potential r e-cordings during acute changes in ventricular loading induced by the Valsalva m a-neuver. Br Heart J 1992;67:221-229.

7. Han J, Moe GK: Nonuniform recovery of excitability in ventricular muscle. CircRes 1964;14:44-60.

8. Allessie MA, Bonke FM, Schopman FJG: Circus movement in rabbit atrial muscleas a mechanism of tachycardia: II. The role of nonuniform recovery of excitability

CHAPTER 3

96

in the occurrence of unidirectional block, as studied with multiple microelectrodes.Circ Res 1976;39:168-177.

9. Binah O, Rosen MR: Mechanisms of ventricular arrhythmias. Circulation1992;85(1 Suppl):I25-I31.

10. Wilson FN, Macleod AG, Barker PS, Johnston FD: The determination and the si g-nificance of the areas of the ventricular deflections of the electrocardiogram. AmHeart J 1934;10:46-61.

11. Burger HC: A theoretical elucidation of the notion “ventricular gradient”. Am HeartJ 1957;53:240-246.

12. Plonsey R: A contemporary view of the ventricular gradient of Wilson. J Electro-cardiol 1979;12:337-341.

13. Abildskov JA, Evans AK, Lux RL, Burgess MJ: Ventricular recovery properties andQRST deflection area in cardiac electrograms. Am J Physiol 1980;239:H227-H231.

14. Lux RL, Urie PM, Burgess MJ, Abildskov JA: Variability of the body surface distr i-butions of QRS, ST-T and QRST deflection areas with varied activation sequence indogs. Cardiovasc Res 1980;14:607-612.

15. SippensGroenewegen A, Muilwijk SLC, Kingma JH, van Hemel NM, Hauer RNW,Janse MJ, Dunning AJ: Incidence of nonuniform ventricular recovery in patientswith postinfarction ventricular arrhythmias: significance of infarct location andtype of spontaneously o ccurring arrhythmia. Eur Heart J 1991;12(II-IV):109.

16. Gardner MJ, Montague TJ, Armstrong CS, Horacek BM, Smith ER: Vulnerability toventricular arrhythmia: assessment by mapping of body surface potential. Circula-tion 1986;73:684-692.

17. Mitchell LB, Hubley-Kozey CL, Smith ER, Wyse DG, Duff HJ, Gillis AM, HoracekBM: Electrocardiographic body surface mapping in patients with ventricular tach y-cardia. Assessment of utility in the identification of effective pharmacological the r-apy. Circulation 1992;86:383-393.

18. Tsunakawa H, Nishiyama G, Kusahana Y, Harumi K: Identification of susceptibilityto ventricular tachycardia after myocardial infarction by nondipolarity of QRSTarea maps. J Am Coll Cardiol 1989;14:1530-1536.

19. Montague TJ, Smith ER, Cameron DA, Rauraharju PM, Klassen GA, FelmingtonCS, Horacek BM: Isointegral analysis of body surface maps: surface distributionand temporal variability in normal subjects. Circulation 1981;63:1166-1172.

20. Abildskov JA, Green LS, Lux RL : Detection of disparate ventricular repolarisationby means of the body surface electrocardiogram, in Grune, Stratton (eds): Cardiacelectrophysiology and arrhythmias. Orlando, Fla,USA, 1985, pp. 495-499.

21. Yasumura S, Kubota I, Ikeda K, Tsuiki K, Yasui S: Using body surface mapping todetect vulnerability to ventricular arrhythmias in patients with coronary artery di s-ease. J Electrocardiol 1987;20:114-120.

22. Abildskov JA, Green LS: The recognition of arrhythmia vulnerability by body su r-face electrocardiographic mapping. Circulation 1987;75(4 Pt 2):III79-III85.

23. Kingma JH, van Gilst WH, Peels KH, Dambrink J-HE, Verheugt FWA, WielengaRP: Acute intervention with captopril during thrombolysis in patients with first a n-terior myocardial infarction. Eur Heart J 1994;15:898-907.

REMODELING AND NONUNIFORM RECOVERY

97

24. Reek EJ, Grimbergen CA, van Oosterom A: A low cost 64 channel microcomputerbased data acquisition system for bedside registration of body surface maps, ind’Alche P (ed): Advances in Electrocardiology. Caen, France, Centre de Public a-tions de l’Universite de Caen, 1985, pp. 112-114.

25. SippensGroenewegen A, Spekhorst H, van Hemel NM, Kingma JH, Hauer RNW,Janse MJ, Dunning AJ: Body surface mapping of ectopic left and right ventricularactivation. QRS spectrum in patients without structural heart disease. Circulation1990;82:879-896.

26. Lux RL, Evans AK, Burgess MJ, Wyatt RF, Abildskov JA: Redundancy reductionfor improved display and analysis of body surface potential maps. Circ Res1981;49:186-196.

27. Schiller NB, Shah PM, Crawford M, DeMaria A, Devereux R, Feigenbaum H, Gu t-gesell H, Reichek N, Sahn D, Schnittger I, Silverman NH, Tajik AJ: Recommend a-tions for quantitation of the left ventricle by two-dimensional echocardiography.American Society of Echocardiography Committee on Standards, Subcommittee onQuantitation of Two-Dimensional Echocardiograms. J Am Soc Echocardiogr1989;2:358-367.

28. Bigger JT Jr, Weld FM, Rolnitzky LM: Which postinfarction ventricular arrhyt h-mias should be treated? Am Heart J 1982;103(4 Pt 2):660-666.

29. van der Laarse A, Kerkhof PL, Vermeer F, Serruys PW, Hermens WT, VerheugtFW, Bar FW, Krauss XH, van der Wall EE, Simoons ML: Relation between infarctsize and left ventricular performance assessed in patients with first acute myocardialinfarction randomized to intracoronary thrombolytic therapy or to conventionaltreatment. Am J Cardiol 1988;61:1-7.

30. Altman DG: Relation between several variables, in Practical statistics for medicalresearch, 1st ed. London, Chapman and Hall, 1990, pp. 325-364.

31. Multicentre Postinfarction Research Group: Risk stratification and survival aftermyocardial infarction. N Engl J Med 1983;309:331-336.

32. Kuo CS, Atarashi H, Reddy CP, Surawicz B: Dispersion of ventricular repolariz a-tion and arrhythmia: study of two consecutive ventricular premature complexes.Circulation 1985;72:370-376.

33. Kuo CS, Reddy CP, Munakata K, Surawicz B: Mechanism of ventricular arrhyt h-mias caused by increased dispersion of repolarization. Eur Heart J 1985;6 SupplD:63-70.

34. Franz MR, Cima R, Wang D, Profitt D, Kurtz R: Electrophysiological effects ofmyocardial stretch and mechanical determinants of stretch-activated arrhythmias.Circulation 1992;86: 968-978.

35. Hansen DE, Craig CS, Hondeghem LM: Stretch-induced arrhythmias in the isolatedcanine ventricle. Circulation 1990;81:1094-1105.

36. Taggart P, Sutton P, Lab M: Interaction between ventricular loading and repolaris a-tion: relevance to arrhythmogenesis. Br Heart J 1992;67:213-215.

37. Kersschot IE, Brugada P, Ramento l M, Zehender M, Waldecker B, Stevenson WG,Geibel A, De Zwaan C, Wellens HJ: Effects of early reperfusion in acute myocardialinfarction on arrhythmias induced by programmed stimulation: a prospective, ra n-domized study. J Am Coll Cardiol 1986;7:1234-1242.

CHAPTER 3

98

38. Gang ES, Lew AS, Hong M, Wang FZ, Siebert CA, Peter T: Decreased incidence ofventricular late potentials after successful thrombolytic therapy for acute myoca r-dial infarction. N Engl J Med 1989;321:712-716.

39. Jeremy RW, Hackworthy RA, Bautovich G , Hutton BF, Harris PJ: Infarct arteryperfusion and changes in left ventricular volume in the month after acute myoca r-dial infarction. J Am Coll Cardiol 1987;9:989-995.

40. Dambrink J-HE, Tobé TJM, Peels CH, van Gilst WH, Kingma JH, on behalf of theCATS investigators group: Relation between left ventricular dilatation and QRS d u-ration after a first anterior myocardial infarction. Eur Heart J 1994;15S:494.

41. Taccardi B: Distribution of heart potentials on the thoracic surface of normal humansubjects. Circ Res 1963;12:341-352.

CHAPTER 4

Does left ventricular dilatation contribute to an increasedfiltered QRS duration after myocardial infarction?

J-H.E. Dambrink,1 MD; T.J.M. Tobé,2 MD; N.M. van Hemel,1 MD; W.H. vanGilst,3 PhD; and J.H. Kingma,1,3 MD for the CATS Investigators4

1Department of Cardiology, St Antonius Hospital, Nieuwegein2Department of Internal Medicine, Sophia Ziekenhuis, Zwolle3Department of Clinical Pharmacology, University of Groningen, Groningen4see Appendix

Published in:Eur Heart J 1994;15(suppl):49 (abstract)Submitted

DILATATION AND QRS DURATION

99

Introduction

Left ventricular dilatation after myocardial infarction is associated with a highmortality rate, an important part of which is arrhythmogenic in origin.1 Untilnow, the mechanism of this association is not fully understood. Myocardial cellslippage may lead to slow, heterogeneous conduction,2 which could facilitatethe occurrence of ventricular arrhythmias based on reentry. Slow conductionmay be detected as a prolongation of the filtered QRS (fQRS) duration on thesignal averaged electrocardiogram.3

We hypothesized that after myocardial infarction, left ventricular dilatationresults in a prolonged QRS duration, which could explain its association withthe occurrence of life-threatening ventricular arrhythmias.

Methods

All patients in this study were participating in the Captopril And Thromboly-sis Study (CATS), the methods of which have been described before.4 In brief,this study investigated the effect of captopril in patients with a first, anteriorwall, myocardial infarction. All patients received thrombolytic therapy. Serial2D-echocardiography, including off-line left ventricular volume measurementswere part of the study protocol. Patients with left- or right bundle branch blockwere excluded. Signal-averaged electrocardiography was performed in thechronic phase of myocardial infarction after a mean of 8 months. An Arrhyth-mia Research Technology (ART) EPX 1200 electrocardiograph was used in allpatients. The Frank X,Y,Z leads were used, and the signal of each lead was fil-tered with a high-pass bidirectional filter. The cut-off frequency set at 40 Hz.

Table 4.1. Patient characteristics

N 64 fQRS (ms) 94 ± 11Age (years)Male (%)LVEF (%)Killip class I (%)

59 ± 978

56 ± 983

average beatsnoise (mV)captopril (%)beta-blocker (%)

93 ± 410.5 ± 0.1

4416

Killip class II (%) 17 digoxin (%) 3

fQRS indicates filtered QRS duration; LVEF, left ventricular ejection fraction. Clinicaldata and medication were assessed at randomization.

CHAPTER 4

100

Filtered QRS duration was selected as the parameter of slow conduction, sincethis variable is most predictive of risk for ventricular arrhythmias.3 Patients withserial echocardiographic measurements at day 1 and month 3, as well as an en-zymatic estimate of infarct size (cumulative α-HBDH) and signal-averagedelectrocardiography were included in this substudy (N=64). Student’s t-test wasused for continuous, and Chi-square for discrete data. A p-value of < 0.05 wasconsidered statistically significant. Data are presented as mean ± standard de-viation unless stated otherwise.

Table 4.2. Possible determinants of QRS duration

fQRS < 100 ms fQRS > 100 ms p-value

NDemographics Age (years) Male (%)Myocardial injury α-HBDH Q72 (U/l) Wall motion score indexLeft ventriculardimensions End-systolic volume (ml/m2)

within 24 hoursat 3 monthsincrease

End-diastolic volume(ml/m2)

within 24 hoursat 3 monthsincrease

Ejection fraction (%)within 24 hours

Medication Captopril (%) Beta-blocker (%) Digoxin (%)

52

58 ± 979

1208 ± 8801.89 ± 0.35

23.3 ± 7.726.2 ± 11.73.0 ± 9.1

52.4 ± 10.558.0 ± 14.35.6 ± 13.9

56 ± 9

42154

12

62 ± 775

1159 ± 10612.03 ± 0.33

27.8 ± 10.333.5 ± 19.05.7 ±12.8

59.9 ± 14.367.8 ± 25.57.9 ± 16.2

55 ± 9

50170

0.1390.715

0.8670.213

0.0900.0940.389

0.0410.0760.622

0.620

0.8720.7411.000

α-HBDH Q72 indicates cumulative alpha-hydroxybutyrate dehydrogenase over the first72 hours.

DILATATION AND QRS DURATION

101

Results

In Table 4.1, the baseline characteristics of the patients studied are given. Pa-tient details were similar to those not included in this substudy. A relatively highejection fraction and a low incidence of heart failure is evident in these patients,who all received thrombolytic therapy. A high percentage of patients used anACE inhibitor as part of the study protocol. Two patients had ventricular ar-rhythmias requiring anti-arrhythmic treatment after the acute phase of myocar-dial infarction. One patient had ventricular tachycardia at day 12 and wastreated successfully with mexiletine (fQRS 88 ms). Another patient had ven-tricular fibrillation at day 11 and was cardioverted to sinus rhythm (fQRS 103ms). No other patients had late arrhythmic events. Signal-averaged electrocar-diography was performed 248 ± 110 days after myocardial infarction. Patientswere divided into those with a fQRS duration of more than 100 ms (80th per-centile) and those with a shorter fQRS duration (Table 4.2). End-diastolic vol-ume assessed within 24 hours was significantly larger in patients with a pro-longed fQRS. There was also a trend towards a higher end-systolic volumewithin 24 hours, and larger end-systolic and end-diastolic volume after 3months of follow up in this group. There were no significant differences indemographics, infarct size or medication. To assess the contribution of end-systolic and end-diastolic volume to fQRS duration, a multiple regressionanalysis was carried out using fQRS as a continuous dependent variable (Table4.3). First, the most powerful determinants of fQRS, not including volumemeasures, were entered into the model (entry criterion: p < 0.25). After this,end-diastolic volume within 24 hours was added. This variable contributed in-dependently. No other variable, including changes of end-systolic or end-diastolic volume, could improve this predictive model of fQRS. Finally, in Fig-ure 4.1 the fQRS duration in four groups, separated by quartiles of early end-diastolic volume, is depicted. It is shown that patients with a large early end-diastolic volume have a significantly longer fQRS duration at follow-up com-pared to those with smaller volumes.

B SE t p-value

Age 0.1165 0.1443 0.807 0.4229Wall motion score index -0.5292 3.8602 -0.137 0.8914End-diastolic volume

within 24 hours (ml/m2) 0.3668 0.1144 3.205 0.0022Constant 68.2169 11.0817

B indicates regression coefficient; SE, standard error. R 2 = 0.16, F-value = 3.917, p =0.0127 .

CHAPTER 4

102

Discussion

This study demonstrates the importance of early (within 24 hours) dilatationfor the duration of the filtered QRS complex late after myocardial infarction. Afurther increase in left ventricular volume during the first few months did notlead to a longer fQRS. In addition, other parameters of infarct size, such asejection fraction, wall motion score and enzymatic infarct size did not add to aprolonged fQRS. Other studies investigating the association between slow con-duction, quantified by filtered QRS duration or the presence of late potentials,and left ventricular volume have also demonstrated a relation between these twofactors. Zaman et al.5 assessed the presence of late potentials in relation to left

90

100

110 *fQRS ± SEM (ms)

Q I Q II Q III Q IV

Figure 4.1. Filtered QRS duration in four groups, separated by quartiles of end-diastolicvolume, are depicted. End-diastolic volume was assessed within 24 hours after admissionfor myocardial infarction. The quartiles of end-diastolic volume were separated as follows:bottom quartile (Q I), up to 46.5 ml/m2; lower middle quartile (Q II), 46.5 up to 54.4ml/m2; upper middle quartile (Q III), 54.4 up to 59.6 ml/m2; and top quartile (Q IV), morethan 59.6 ml/m2. The difference in fQRS between the top quartile compared to the threelower quartiles was statistically significant. fQRS indicates filtered QRS duration; SEM,standard error of the mean; Q, quartile. * p=0.032.

DILATATION AND QRS DURATION

103

ventricular volume up to 42 days in 52 patients with anterior myocardial infarc-tion. They found that the presence of late potentials was related to left ventricu-lar dilatation up to seven days after myocardial infarction. However, at 42 daysthis relation was lost. In contrast, Hochman et al.6 showed a positive correlationbetween left ventricular volume and fQRS duration up to six months after myo-cardial infarction. Data on left ventricular dilatation were not provided. In thepresent study, both early and late assessments of left ventricular volume ap-peared to correlate with fQRS. However, when contributions of ventricular vol-ume were broken down into early and late components of dilatation, only earlydilatation, expressed as an increased end-diastolic volume within 24 hours, ap-peared to determine fQRS duration.

This finding may be explained by the fact that, before completion of scar tis-sue formation, the infarcted area is very vulnerable to straining forces7 whichmay cause cellular uncoupling.2 This could in turn result in delayed conductionas the electrophysiological expression of this process. Dilatation of the nonin-farcted area, mostly occurring after completion of scar tissue formation, is lesscontributing to conduction delay according to the present data. These findingsconfirm that left ventricular dilatation is a significant determinant of filteredQRS duration. This effect is an early one, since end-diastolic volume within 24hours after hospital admission, and not the subsequent increase during the firstthree months after myocardial infarction predicted filtered QRS duration. Theassociation between left ventricular dilatation and slow conduction may con-tribute to the understanding of increased arrhyhmogenesis in patients with leftventricular dilatation.

References

1. White HD, Norris RM, Brown MA, Brandt PWT, Whitlock RML, Wildt CJ. Leftventricular end-systolic volume as the major determinant of survival after recoveryfrom myocardial i nfarction. Circulation 1987;76:44-51.

2. Saffitz JE, Corr PB, Sobel BE. Arrhythmogenesis and ventricular dysfunction aftermyocardial infarction. Is anomalous cellular coupling the elusive link? Circulation1993;87:1742-1745.

3. Lander P, Berbari EJ, Rajagopalan CV, Vatterott P, Lazzara R. Critical analysis ofthe signal-averaged electrocardiogram. Improved identification of late potentials.Circulation 1993;87:105-117.

4. Kingma JH, van Gilst WH, Peels KH, Dambrink J-HE, Verheugt FWA, WielengaRP for the CATS investigators. Acute intervention with Captopril during thro m-

CHAPTER 4

104

bolysis in patients with first anterior myocardial infarction. Eur Heart J1994;15:898-907.

5. Zaman AG, Morris JL, Smyllie JH, Cowan JC. Late potentials and ventricular e n-largement after myocardial infarction. A new role for high-resolution electroca r-diography? Circulation 1993;88:905-914.

6. Hochman JS, Morgan CD, Steinberg JS, Burns JR, Freeman MR, Dorian P, Ar m-strong PW, for LATE Ancillary Investigators Toronto and New York. Filtered QRSduration and left ventricular volume post MI are positively related; results of a ra n-domised controlled trial of late thrombolysis. Circulation 1993;88:I-312.

7. Brauwnwald E, Pfeffer MA. Ventricular enlargement and remodelling followingacute myocardial infarction: mechanisms and management. Am J Cardiol1991;68:1D-6D.

CHAPTER 5

Norepinephrine levels early after thrombolytic therapy:association with angiographic findings and ventricular

arrhythmias

J-H.E. Dambrink,1 MD; W.H. van Gilst,2 PhD; H.W.M. Plokker,1 MD;A.C. Bredero,3MD; K.I. Lie,4 MD; and J. H. Kingma,1,2 MD

for the CATS investigators5

1Department of Cardiology, St Antonius Hospital, Nieuwegein2Department of Clinical Pharmacology, University of Groningen, Groningen3Department of Cardiology, Diakonessenhuis, Utrecht4Department of Cardiology, Groningen University Hospital, Groningen5see Appendix

Submitted

NOREPINEPHRINE AND ARRHYTHMIAS

107

Abstract

Background. Successful thrombolytic therapy for acute myocardial infarction re-sults in increased electrophysiological stability. It is not clear to what extentmodulation of sympathetic activity contributes to this effect.Methods. 298 patients with a first anterior myocardial infarction were random-ized to captopril or placebo in a multi-center clinical trial. All patients weretreated with streptokinase before randomization. Norepinephrine levels weremeasured up to 96 hours after the onset of thrombolytic therapy. Patency of theinfarct-related artery was assessed by coronary angiography. Full data includingcumulative norepinephrine levels and coronary angiography were available in125 patients. Presence of ventricular arrhythmias was assessed by Holter moni-toring within 24 hours, before discharge and three and 12 months after myocar-dial infarction.Results. Norepinephrine levels (± SEM) were significantly reduced in patientswith a patent left anterior descending artery (723 ± 38 versus 976 ± 103 pg/ml,p=0.017); this effect was independent of age, sex, left ventricular function andheart failure class. Complex ventricular arrhythmias were more frequent in pa-tients with increased norepinephrine levels within 24 hours, but not at three and12 months after myocardial infarction. During follow up, two patients died sud-denly, three had ventricular fibrillation, and three had ventricular tachycardia re-quiring anti-arrhythmic treatment. Five of these patients (63%) had an occludedinfarct-related artery, compared to 18% of those without arrhythmic events(p=0.011). Norepnephrine levels were not different in patients with or without ar-rhythmic events 687 ± 215 versus 781 ± 433 pg/ml). An occluded left anterior de-scending artery and a wall motion score > 2 proved the only independent predic-tors of life-threatening ventricular arrhythmias.Conclusion. Sympathetic activity is reduced after successful thrombolysis. Since areduced sympathetic activity per se did not result in a lower incidence of ven-tricular arrhythmias, the increased electrophysiological stability after successfulreperfusion can not be explained by a reduction of sympathetic activity alone.

Introduction

After myocardial infarction, increased sympathetic activity is an important riskfactor for the occurrence of ventricular arrhythmias. In animal experiments it hasbeen shown that depletion of catecholamine stores may partially or even totallyabolish ventricular arrhythmias after myocardial infarction.1, 2 In the clinical set-ting, patients with ventricular arrhythmias after myocardial infarction are char-acterized by high norepinephrine levels.3 In addition, norepinephrine levels haveshown to be of prognostic importance after myocardial infarction4, 5 and in sub-

CHAPTER 5

108

sequent heart failure.6 The effect of successful thrombolytic therapy on sympa-thetic activity is not well known. Its beneficial effect on infarct size,7 left ven-tricular function8 and remodeling9 may lead to a blunted neurohumoral response.Vice versa, a reduction of neurohumoral activation after thrombolytic therapymay favorably alter the remodeling process and limit the degree of left ventricu-lar dysfunction.

In this study, serial measurements of norepinephrine levels were performed inpatients with and without successful thrombolytic therapy after a first anteriormyocardial infarction. The aim of this study was 1) to assess the effect of suc-cessful thrombolytic therapy on norepinephrine values, and its relation to otherdeterminants of sympathetic activity, and 2) to investigate the importance ofsympathetic activity for the occurrence of ventricular arrhythmias after throm-bolytic therapy.

Methods

Patients. This study was part of the Captopril And Thrombolysis Study(CATS), in which the effect of captopril treatment started during thrombolysiswas evaluated in patients with a first anterior myocardial infarction. The methodsof this study are described elsewhere in detail.10 In brief, 298 patients were in-cluded in 12 hospitals in The Netherlands. Selection criteria included a typicalhistory of chest pain consistent with myocardial infarction with onset of symp-toms no longer than 6 hours before admission, and electrocardiographic criteriafor acute anterior myocardial infarction. Exclusion criteria included presence ofleft bundle branch block and severe heart failure (Killip class III or IV). Informedconsent was obtained by witnessed oral consent, later confirmed by written in-formed consent following the acute phase of myocardial infarction.

Coronary angiography. Coronary angiography was performed when clinicallyindicated, as determined by the individual investigator. The first available angio-gram in each patient was studied for the presence or absence of an occluded in-farct-related artery (TIMI flow grade 0), which was the left anterior descendingartery in this study. The severity of coronary artery disease was expressed as one-, two- or three- vessel disease. A 50% or more reduction of vessel diameter wasconsidered significant.

Infarct size. Enzymatic infarct size was estimated by cumulative alpha-hydroxybutyrate dehydrogenase values over the first 72 hours (α-HBDH Q72)after myocardial infarction as described by van der Laarse et al.7 This method isnot influenced by the presence or absence of reperfusion.

NOREPINEPHRINE AND ARRHYTHMIAS

109

Echocardiography. Regional wall motion abnormalities were evaluated usingthe wall motion score recommended by the American Society of Echocardiogra-phy.11 In this scoring system the left ventricle is divided into 16 segments, scor-ing each segment as 1 for normokinesia, 2 for hypokinesia, 3 for akinesia, 4 fordyskinesia and 5 for an aneurysmal segment. A Wall Motion Score Index(WMSI) was computed as the sum of scores of all segments divided by the num-ber of segments evaluated.

Ventricular arrhythmias. The occurrence of ventricular arrhythmias was as-sessed by two-channel 24-hour ambulatory Holter monitoring (Reynolds MedicalTracker recorder) at entry, before hospital discharge and three and 12 monthsafter myocardial infarction. Analysis was performed with help of a ReynoldsMedical Pathfinder 3 analysis system. Complex ventricular arrhythmias were de-fined as 10 or more premature ventricular beats per hour and/or pairs and/orventricular tachycardia.12 Arrhythmic events included sudden cardiac death,

Table 5.1. Characteristics of patients with an occluded or patent infarct-related artery

LAD occluded LAD patent p-value

N 26 99Demographics Age (years) Male (%)Myocardial injury α-HBDH Q72 (U/l) WMSIExtent of CAD ≥ 2-vessel disease (%)Functional class Heart failure (Killip)Haemodynamics Heart rate (beats/min) Syst. blood pressure (mmHg) Rate-pressure productMedication Captopril (%) Beta-blocker (%)

58 ± 981

1612 ± 10711.96 ± 0.51

39

1.27 ± 0.45

87 ± 18131 ± 20

11425 ± 3295

3915

58 ± 988

1169 ± 11381.82 ± 0.46

36

1.17 ± 0.38

80 ± 13123 ± 19

9863 ± 2203

536

0.9840.535

0.0940.159

0.975

0.276

0.0510.0670.033

0.2910.256

Digoxin (%) 15 0 0.002

α-HBDH Q72 indicates cumulative alpha-hydroxybutyrate dehydrogenase over the first 72 hours aftermyocardial infarction; CAD, coronary artery disease; LAD, Left anterior descending artery; WMSI, wallmotion score index.

CHAPTER 5

110

ventricular fibrillation, or ventricular tachycardia requiring anti-arrhythmic ther-apy, including DC cardioversion. Sudden death was defined as death within 1hour of symptoms, but also included unwitnessed death in patients who werepreviously stable.

Norepinephrine levels. Blood samples were collected for assessment of nor-epinephrine levels at 0, 12, 24, 48, 72 and 96 hours after the start of study medi-cation, which was at the completion of streptokinase infusion. Norepinephrinewas measured using a sensitive assay with electrochemical detection. Cumulativenorepinephrine values were calculated following the trapezium rule:13 for eachtime interval between measurements, the area under the curve was approximatedby the product of the time interval and the mean of the norepinephrine values atthe beginning and the end of this interval. These products obtained from all in-tervals were then added up and divided by the total time interval (usually 96hours), thus producing a cumulative value expressed in pg/ml. Missing valueswere replaced by values interpolated from adjacent measurements. Patients withtwo or more adjacent missing values were not included in the analysis.

Statistical analysis. Results are presented as means with standard deviation,except when stated otherwise. Differences between groups were examined usingthe Student’s t-test. The Mann-Whitney U-test was used if data were not distrib-uted normally. The Chi-square test was used for discrete data. Logistic regressionanalysis was used to identify independent relations between baseline characteris-tics and sympathetic activity, quantified by cumulative norepinephrine values.Calculations were made using SPSS/PC+ software.

Results

During the follow-up period, coronary angiography was performed in 163 pa-tients after a median of 7 days (range 0 - 408 days, 75% within 24 days). In 125of these patients, cumulative norepinephrine levels were available (42% of thetotal CATS population). In these patients, the relationship between norepineph-rine levels and patency status of the infarct-related artery was investigated. Com-pared to the total CATS population, the selected patients were younger (58 ± 9 vs60±10 years) and more frequently of male gender (86% vs 67%). Enzymatic in-farct size, left ventricular function and use of medication were comparable be-tween selected and non-selected patients.

NOREPINEPHRINE AND ARRHYTHMIAS

111

Successful reperfusion and norepinephrine levels

Figure 5.1 shows norepinephrine levels in patients with- and without success-ful reperfusion up to 96 hours after thrombolytic therapy. Of all 125 patients, 26(21%) had an occluded infarct-related artery on their first coronary angiogram.

It is shown that norepinephrine levels during the first 96 hours are higher inpatients with an occluded infarct-related artery, although differences between thetwo groups are not significantly different at all points in time. In the same figure,cumulative norepinephrine levels over 96 hours are depicted. Patients with anoccluded left anterior descending artery clearly had higher cumulative norepi-nephrine values. In Table 5.1, other characteristics of patients with or without an

400

600

800

1000

1200

1400

1600

0 50 100Hours

Cumulativeover 96 hours

Norepinephrine± SEM (pg/ml)

LAD patent

LAD occluded*

***

*

Figure 5.1. Serial and cumulative norepinephrine levels during the first 96 hours afterthe onset of thrombolytic therapy were significantly reduced in patients with a patentinfarct-related artery. LAD indicates left anterior descending artery. * p < 0.05, ** p <0.01

CHAPTER 5

112

occluded infarct-related artery are listed. Patients with an occluded left anteriordescending artery showed a trend towards a larger enzymatic infarct size andworse left ventricular function as assessed by the echocardiographic wall motionscore. In addition, these patients showed a higher heart rate, blood pressure andrate-pressure product. Finally, all patients using digoxin in this study populationwere found in the group with an occluded left anterior descending artery.

Other determinants of increased norepinephrine levels

To assess whether the reduction in norepinephrine levels in patients with suc-cessful reperfusion could be attributed to reperfusion itself, or to other associatedfactors, a number of possible determinants of increased norepinephrine levelswere investigated (Table 5.2). On univariate analysis, patency of the infarct-

Table 5.2. Characteristics of patients with high or low cumulative norepinephrinelevels

Norepinephrine Norepinephrine p-value above median below median

N 62 63Demographics Age (years) Male (%)Myocardial injury α-HBDH Q72 (U/l) WMSIExtent of CAD ≥ 2-vessel disease (%) LAD occluded (%)Functional class Heart failure (Killip)Haemodynamics Heart rate (beats/min) Syst. blood pressure (mmHg) Rate-pressure productMedication Captopril (%) Beta-blocker (%)

59 ± 890

1370 ± 11921.91 ± 0.45

4031

1.23 ± 0.42

83 ± 16127 ± 20

10584 ± 2617

5512

57 ± 983

1139 ± 10661.78 ± 0.44

3511

1.16 ± 0.37

80 ± 13122 ± 19

9828 ± 2439

455

0.4610.897

0.2790.129

0.6250.014

0.367

0.2390.1740.110

0.3260.310

Digoxin (%) 6 0 0.123

Abbreviations as in Table 5.1

NOREPINEPHRINE AND ARRHYTHMIAS

113

related artery was the only variable that was significantly associated with highcumulative norepinephrine levels, defined as values above the median of thestudy population. There was a trend towards a higher echocardiographic wallmotion score in patients with high cumulative norepinephrine levels, reflecting aslightly more pronounced left ventricular dysfunction in this group. In addition,heart rate and blood pressure tended to be higher in patients with high norepi-nephrine levels. A logistic regression analysis was carried out to determinewhether the relationship between patency of the infarct-related artery and norepi-nephrine levels was independent of other known determinants of sympathetic ac-tivity (Table 5.3). After correction for age, sex, functional class and left ven-tricular function, an occluded infarct-related artery still appeared to be an im-portant determinant of norepinephrine levels (odds ratio 3.47).

Association with ventricular arrhythmias during follow up

In order to evaluate early and late ventricular arrhythmias, Holter monitoringwas performed during the first 24 hours, before hospital discharge, and at threeand 12 months after myocardial infarction. In addition, data on life-threateningarrhythmic events were collected. In Figure 5.2 the incidence of ventricular ar-rhythmias during Holter monitoring in patients with high and low norepinephrinelevels is depicted. During the first 24 hours after admission complex ventriculararrhythmias were more frequent in patients with cumulative norepinephrine val-ues above the median. However, later Holter recordings up to one year aftermyocardial infarction did not show a higher incidence of late ventricular ar-rhythmias in patients with increased norepinephrine levels.

Table 5.3. Results of logistic regression analysis (explanation see text)

B SE p-value OR (95%-CI)

Age > 60 0.2309 0.3984 0.5623 1.26 (0.58 - 2.75)Male genderKillip-class ≥ 2WMSI > 2LAD occluded

0.83880.13250.47471.2436

0.60620.49660.39760.5098

0.16650.78960.23250.0147

2.31 (0.70 - 7.59)1.14 (0.43 - 3.02)1.61 (0.74 - 3.50)3.47 (1.28 - 9.42)

Constant - 1.3935 0.8254B indicates regression coefficient; CI, confidence intervals; OR, odds ratio; SE, standard error. Otherabbreviations as in Table 5.1.

CHAPTER 5

114

Of all 125 patients, eight had late arrhythmic events during the first year offollow up. Three patients had ventricular tachycardia, another three patients hadventricular fibrillation and two patients died suddenly. Details of these patientsare given in Table 5.4. Patients with ventricular arrhythmias during follow upwere all male, of higher age than those without arrhythmias, and characterized bya larger infarct size, worse left ventricular function and, frequently, an occludedinfarct-related artery (Table 5.5). A stepwise logistic regression model showedthat a wall motion score > 2 and and an occluded infarct-related artery were in-dependent predictors of arrhythmic events during the first year of follow up. High

0

20

40

60

80

100

������������������������

������

����������

����������

Entry Predischarge 3 months 12 months

Norepinephrine < medianNorepinephrine > median

����

(%)

*

Figure 5.2. Complex ventricular arrhythmias during Holter moni-toring in patients with cumulative norepinephrine levels ofmore or less than the median value. Patients with high nor-epinephrine levels had a higher incidence of complex ven-tricular arrhythmias within 24 hours after the onset ofthrombolytic therapy. However, at discharge and 3 and 12months after myocardial infarction, this difference was nolonger present.* p=0.015.

NOREPINEPHRINE AND ARRHYTHMIAS

115

cumulative norepinephrine levels did not contribute independently to the risk ofventricular arrhythmias.

Discussion

These results show that after successful thrombolytic therapy, sympathetic ac-tivity, quantified by cumulative norepinephrine levels, is reduced and that thisbeneficial effect is independent of age, sex, left ventricular function and func-tional class. Although complex ventricular arrhythmias within 24 hours aftermyocardial infarction were more frequent in patients with increased sympatheticactivity, late arrhythmic events during follow up were not predicted by high cu-mulative norepinephrine levels. Left ventricular function and patency status ofthe infarct-related artery proved the only two independent determinants of theseevents on multivariate analysis.

Sympathetic activity after successful thrombolytic therapy

To our knowledge, the effect of successful thrombolytic therapy on norepi-nephrine levels has not been investigated systematically. Sigurtsson et al.14 re-cently reported on neurohumoral activation after myocardial infarction in aCONSENSUS II substudy. They found no difference in norepinephrine levelswithin 24 hours between patients with and without thrombolytic therapy. How-

Table 5.4. Ventricular arrhythmias during the first year of follow up

Pat Age Arrhythmia Days α-HBDH WMSI Nor 96h LAD (y) Q72 (U/l) (pg/ml)

1 54 VT, VF 44 1357 1.25 807 Occl234567

737055696768

SCDSCDVTVFVFl

nsVT

71711914814

579543

3445-

21291611

2.252.062.082.752.502.71

594696738638999777

PatPat

OcclOcclOcclPat

8 55 VT 11 1428 1.50 251 OcclDays indicates days after myocardial infarction; LAD, left anterior descending artery; Nor 96h, cumula-tive norepinephrine levels over 96 hours; nsVT, nonsustained VT; Occl, occluded; Pat, patient; SCD,sudden cardiac death; VF, ventricular fibrillation; VT, ventricular tachycardia; WMSI, wall motionscore index.Other abbrevations as in Table 5.1.

CHAPTER 5

116

ever, the numbers of patients investigated were small, and results of coronary an-giography were not given. Some information is available regarding the effect ofthrombolysis on autonomic activity as assessed by heart rate variability. Casoloet al.15, 16 found a beneficial effect of rtPA treatment on heart rate variability as-sessed two and three months after myocardial infarction. Pedretti et al.17 de-scribed a beneficial effect of thrombolytic therapy on heart rate variability in pa-tients with anterior infarcts, assessed 21 ± 6 days after hospital admission. Theseinvestigators also demonstrated that this effect was independent of the residualleft ventricular function. However, in none of these mentioned studies was coro-nary angiography performed to evaluate the result of thrombolytic therapy. Ourstudy clearly demonstrates a reduction of serial and cumulative norepinephrinelevels after successful thrombolytic therapy. Similar to the study of Pedretti etal.,17 this effect was not influenced by the degree of left ventricular dysfunction,quantified by a wall motion score.

Ventricular arrhythmias and neurohumoral activation after thrombolytictherapy

Several large multi-center trials have reported a reduced incidence of ven-tricular arrhythmias, especially in-hospital ventricular fibrillation, after throm-bolytic therapy.18-20 In addition, ventricular arrhythmias are less frequently in-

Table 5.5. Characteristics of patients with and without arrhythmic events

Arrhythmic events No events p-valueN Age (years) Male (%) α-HBDH Q72 (U/l) WMSI ≥ 2-vessel disease (%) LAD occluded (%) Heart failure (Killip)Cumulative norepinephrine(pg/ml)Captopril (%)Beta-blocker (%)

864 ± 8

1001585 ± 9942.14 ± 0.54

3863

1.38 ± 0.52687 ± 215

5013

11758 ± 9

861237 ± 11431.83 ± 0.44

3718

1.18 ± 0.39781 ± 433

508

0.0520.5310.4340.0580.7370.0110.1820.545

0.7320.850

Digoxin (%) 0 3 0.612

Abbrevations as in Table 5.1.

NOREPINEPHRINE AND ARRHYTHMIAS

117

ducible after successful thrombolysis.21 In concordance with these findings, pa-tients with ventricular arrhythmias in the present study frequently (63%) showedan occluded infarct-related artery. The underlying mechanism of this relation hasnot been fully elucidated. Next to a reduced infarct size and better preservation ofleft ventricular function, a reduced incidence of late potentials as marker of ananatomical substrate has been suggested as a possible explanation.22-24 As men-tioned before, a beneficial effect on heart rate variability after thrombolytic ther-apy has been described.17, 25 In addition, the present study clearly showed a re-duction of norepinephrine levels in patients with successful thrombolysis, andventricular arrhythmias were less frequent during follow up in patients with apatent infarct-related artery. However, cumulative norepinephrine levels were notincreased in patients with ventricular arrhythmias during the follow-up period.Moreover, in contrast to the high occurrence of ventricular arrhythmias in theacute phase of myocardial infarction, later Holter recordings up to one year aftermyocardial infarction did not show more ventricular arrhythmias in patients withincreased norepinephrine values.

These data indicate that the beneficial effect of successful thrombolysis on theoccurrence of ventricular arrhythmias can not be explained by a reduction of nor-epinephrine levels alone.

Conclusions

Sympathetic activity as assessed by serial and cumulative norepinephrine val-ues is reduced by successful reperfusion, and this effect is independent of age,sex, left ventricular function and heart failure class. However, the importance ofsympathetic activation as an arrhythmogenic factor appears to be short lived afterthrombolytic therapy. The increased electrophysiological stability after success-ful reperfusion can not be explained by its beneficial effect on sympathetic activ-ity alone.

CHAPTER 5

118

References

1. Schaal SF, Wallace AG, Sealy WC: Protective influence of cardiac denervation against ar-rhythmias of myocardial infarction. Cardiovasc Res 1969;3:241-244.

2. Ebert PA, Vanderbeek RB, Allgood RJ, Sabiston Jr DC: Effect of chronic cardiac denervationon arrhythmias after coronary artery ligation. Cardiovasc Res 1970;4:141-147.

3. McAlpine HM, Morton JJ, Leckie B, Rumley A, Gillen G, Dargie HJ: Neuroendocrine acti-vation after acute myocardial infarction. Br Heart J 1988;60:117-124.

4. Karlsberg RP, Cryer PE, Roberts R: Serial plasma catecholamine response early in the courseof clinical acute myocardial infarction: relationship to infarct extent and mortality. Am Heart J1981;102:24-29.

5. Benedict CR, Grahame-Smith DG: Plasma adrenaline and noradrenaline concentrations anddopamine-beta-hydroxylase activity in myocardial infarction with and without cardiogenicshock. Br Heart J 1979;42:214-220.

6. Cohn JN, Levine TB, Olivari MT et al: Plasma norepinephrine as a guide to prognosis in pa-tients with chronic congestive heart failure. N Engl J Med 1984;311:819-823.

7. van der Laarse A, Kerkhof PL, Vermeer F et al: Relation between infarct size and left ven-tricular performance assessed in patients with first acute myocardial infarction randomised tointracoronary thrombolytic therapy or to conventional treatment. Am J Cardiol 1988;61:1-7.

8. Smalling RW, Fuentes F, Matthews MW et al: Sustained improvement in left ventricularfunction and mortality by intracoronary streptokinase administration during evolving myocar-dial infarction. Circulation 1983;68:131-138.

9. Marino P, Zanolla L, Zardini P, on behalf of GISSI: Effect of streptokinase on left ventricularremodelling and function after myocardial infarction: The GISSI (Gruppo Italiano per lo stu-dio della Streptochinaisi nell'Infarto Miocardico) Trial. J Am Coll Cardiol 1989;14:1149-1158.

10. Kingma JH, van Gilst WH, Peels KH, Dambrink J-HE, Verheugt FWA, Wielenga RP, for theCATS investigators: Acute intervention with captopril during thrombolysis in patients withfirst anterior myocardial infarction. Eur Heart J 1994;15:898-907.

11. Schiller NB, Shah PM, Crawford M et al: Recommendations for quantitation of the left ven-tricle by two-dimensional echocardiography. American Society of Echocardiography Com-mittee on Standards, Subcommittee on Quantitation of Two-Dimensional Echocardiograms. JAm Soc Echocardiogr 1989;2:358-367.

12. Maggioni AP, Zuanetti G, Franzosi MG et al on behalf of GISSI-2 investigators: Prevalenceand prognostic significance of ventricular arrhythmias after acute myocardial infarction in thefibrinolytic era. GISSI-2 results. Circulation 1993;87:312-322.

13. Altman DG: Relation between several variables. In: Practical statistics for medical research.1st ed. London: Chapman and Hall, 1990:325-364.

14. Sigurdsson A, Held P, Swedberg K: Short- and long-term neurohormonal activation followingacute myocardial infarction. Am Heart J 1993;126:1068-1076.

15. Ellis SG, Henschke CI, Sandor T, Wynne J, Brauwnwald E, Kloner RA: Time course of func-tional and biochemical recovery of myocardium salvaged by reperfusion. J Am Coll Cardiol1983;1:1047-1055.

16. Pfeffer MA, Brauwnwald E: Ventricular remodelling after myocardial infarction. Experimen-tal observations and clinical implications. Circulation 1990;81:1161-1172.

17. Pedretti RF, Colombo E, Sarzi Braga S, Caru B: Effect of thrombolysis on heart rate variabil-ity and life-threatening ventricular arrhythmias in survivors of acute myocardial infarction. JAm Coll Cardiol 1994;23:19-26.

NOREPINEPHRINE AND ARRHYTHMIAS

119

18. Gruppo per lo studio della streptochinasi nell'infarto miocardico (GISSI): Effectiveness of in-travenous thrombolytic treatment in acute myocardial infarction. Lancet 1986;i:397-402.

19. ISIS-2 (Second International Study of Infarct Survival Group) Collaborative Group: Ran-domised trial of intravenous streptokinase, oral aspirin, both, or neither among 17,187 casesof suspected acute myocardial infarction: ISIS-2. Lancet 1988;ii:349-360.

20. Wilcox RG, von der Lippe G, Olsson CG, Jensen G, Skene AM, Hampton JR: Trial of tissueplasminogen activator for mortality reduction in acute myocardial infarction. Anglo-Scandinavian Study of Early Thrombolysis (ASSET). Lancet 1988;2:525-530.

21. Kersschot IE, Brugada P, Ramentol M et al: Effects of early reperfusion in acute myocardialinfarction on arrhythmias induced by programmed stimulation: a prospective, randomisedstudy. J Am Coll Cardiol 1986;7:1234-1242.

22. Pedretti R, Laporta A, Etro MD et al: Influence of thrombolysis on signal-averaged electro-cardiogram and late arrhythmic events after acute myocardial infarction. Am J Cardiol1992;69:866-872.

23. Vatterott PJ, Hammill SC, Bailey KR, Wiltgen CM, Gersh BJ: Late potentials on signal-averaged electrocardiograms and patency of the infarct related artery in survivors of acutemyocardial infarction. J Am Coll Cardiol 1991;17:330-337.

24. Gang ES, Lew AS, Hong M, Wang FZ, Siebert CA, Peter T: Decreased incidence of ven-tricular late potentials after successful thrombolytic therapy for acute myocardial infarction. NEngl J Med 1989;321:712-716.

25. Casolo GC, Stroder P, Signorini C et al: Heart rate variability during the acute phase of myo-cardial infarction. Circulation 1992;85:2073-2079.

CHAPTER 6

Association between reduced heart rate variability andleft

ventricular dilatation in patients with a first anteriormyocardial infarction

Jan-Henk E. Dambrink,1 MD; Ype S. Tuininga,2 MD; Wiek H. van Gilst,3 PhD;Kathinka H. Peels,4 MD; K.I. Lie,2 MD; and J. Herre Kingma,1,3 MD

for the CATS investigators5

1Department of Cardiology, St Antonius Hospital, Nieuwegein2Department of Cardiology, Groningen University Hospital, Groningen3Department of Clinical Pharmacology, University of Groningen, Groningen4Department of Cardiology, Catharina Hospital, Eindhoven5see Appendix

Published in:Circulation 1993;88:I-107 (abstract)British Heart J 1994;72:514-520

HEART RATE VARIABILITY AND DILATATION

123

Abstract

Background. Reduced heart rate variability has been identified as animportant prognostic factor after myocardial infarction. This parameter isthought to reflect an imbalance between sympathetic and parasympatheticactivity, which may lead to unfavorable loading conditions, thus promoting leftventricular dilatation.Methods. Before discharge, 24-hour Holter monitoring was performed toevaluate heart rate variability as a possible risk factor for left ventricular dila-tation after myocardial infarction. Patients were divided in a group with re-duced (index of ≤ 25) and a group with normal heart rate variability (index of >25). Left ventricular volumes were assessed by echocardiography before dis-charge and three and 12 months after myocardial infarction. Extent of myocar-dial injury, severity of coronary artery disease, functional class, hemodynamicsand medication were also considered as possible determinants of left ventriculardilatation. All patients participated in a multi-center clinical trial, in which 298patients were randomized to captopril or placebo after a first anterior myocar-dial infarction. All patients were treated with streptokinase before randomiza-tion. In the present substudy, full data including heart rate variability and echo-cardiographic measurements were available in 80 patients.Results. Before discharge, end-systolic and end-diastolic volume was not differ-ent in patients with or without a reduced heart rate variability. However, after12 months, end-systolic volume (mean ± SD) had increased in patients with areduced heart rate variability with 6 ± 14 ml/m2 (p=0.043), and end-diastolicvolume had increased with 8 ± 17 ml/m2 (p=0.024). Left ventricular volumeswere unchanged in patients with a normal heart rate variability. In addition,patients with left ventricular dilatation were characterized by a larger enzymaticinfarct size, higher heart rates and rate-pressure products. A reduced heart ratevariability index before discharge proved an independent risk factor for the oc-currence of left ventricular dilatation during follow up. Measurement of heartrate variability after three months had no predictive value for this event.Conclusion. Assessment of the heart rate variability index before discharge,but not at three months, yields important additional information for identifyingpatients at risk for left ventricular dilatation.

Introduction

Reduced heart rate variability after acute myocardial infarction is an impor-tant risk factor for mortality1-3 and the occurrence of life-threatening ventriculararrhythmias4-6 after hospital discharge. Changes in heart rate variability are

CHAPTER 6

124

thought to reflect an imbalance between sympathetic and parasympathetic ac-tivity.7-9 After myocardial infarction, a relative increase in sympathetic activitymay result in a higher wall stress by elevating loading conditions. An increase inwall stress may enhance dilatation of the left ventricle,10,11 and by this mecha-nism increased sympathetic activity may form an important causative factor inthe process of left ventricular remodeling.12 Since persisting sympathetic activityafter myocardial infarction is usually paralleled by activation of the renin-angiotensin system, wall stress may increase even more, and thus activation ofboth systems may contribute to progressive dilatation of the ventricle.

In this study, we investigated the association between heart rate variability as-sessed before hospital discharge and at three months and left ventricular dilata-tion during one year of follow up after a first anterior myocardial infarction.Since heart rate variability can be assessed reliably and reproducibly,13,14 thismay provide important additional information for identifying patients at risk af-ter myocardial infarction.

Methods

Patients. This study was part of the Captopril And Thrombolysis Study(CATS), in which the effect of captopril treatment, started during thrombolysis,was evaluated in patients with a first anterior myocardial infarction.15 Informedconsent was obtained by witnessed oral consent, later confirmed by written in-formed consent following the acute phase of myocardial infarction. Main end-points included left ventricular remodeling, neurohumoral activation and ven-tricular arrhythmias. In the CATS study, 298 patients were included in 12 hospi-tals in The Netherlands. The study was approved by the Institutional ReviewBoard of all participating hospitals. Selection criteria included a typical historyof chest pain consistent with myocardial infarction with onset of symptoms nolonger than 6 hours before admission, and electrocardiographic criteria for acuteanterior myocardial infarction including at least 1 mm ST segment elevation inleads I and aVL and/or 2 mm ST segment elevation in two or more precordialleads of the 12-lead electrocardiogram, consistent with anterolateral, anterosep-tal and/or anterior wall infarction. Patients had to be eligible for thrombolytictherapy. Exclusion criteria included presence of a prior myocardial infarction,left bundle branch block and severe heart failure (Killip class III or IV). Twenty-four-hour Holter monitoring before discharge and at month 3 was part of theCATS study protocol.

Heart rate variability assessment. Heart rate variability was assessed at dis-charge and after three months using an automated procedure described by

HEART RATE VARIABILITY AND DILATATION

125

Malik et al..16 The width of the frequency distribution curve of all selected RRintervals was used as an index for heart rate variability. This method is operatorindependent and has been validated previously.16 In addition, patients were di-chotomized into a group with a heart rate variability index less than or equal to25 and a group with a heart rate variability index more than 25. It has beendemonstrated that patients with a heart rate variability index below 25 have anincreased risk for serious arrhythmic events, suggesting a clinically relevantchange in autonomic balance.4 The two groups thus dichotomized were used toevaluate the association between heart rate variability and left ventricular dilata-tion during follow up. Tracker Holter recording equipment and a Reynoldspathfinder 3 analysis system was used for heart rate variability assessment. Thistype of analysis system has been validated for heart rate variability measure-ments.17 The Reynolds analysis system identifies the different shapes of aberrantbeats. The triggering level for this identification can be adjusted by the operator.Aberrant-normal and normal-aberrant RR intervals were excluded from analysis;only normal-normal RR intervals were used.

Echocardiography. Echocardiograms were made before discharge and atthree and 12 months after myocardial infarction. Left ventricular end-systolicand end-diastolic volumes were calculated from a two- and four-chamber viewusing the modified biplane Simpson’s rule.18 From these volume measurementsthe ejection fraction was calculated. Measurements were made off-line fromend-diastolic and end-systolic stillframes using a Dataview Microsonics cardiacanalysis system (Nova Microsonics). Left ventricular volumes were indexed forbody surface area. Left ventricular dilatation was defined as the increase in end-systolic volume indexed for body surface area between hospital discharge andone year after myocardial infarction. Furthermore, regional wall motion abnor-malities were evaluated using the wall motion score recommended by theAmerican Society of Echocardiography.18 In this scoring system the left ventri-cle is divided into 16 segments, scoring each segment as 1 for normokinesia, 2for hypokinesia, 3 for akinesia, 4 for dyskinesia and 5 for an aneurysmal seg-ment. A wall motion score index was computed as the sum of scores of all seg-ments divided by the number of segments evaluated. Twelve evaluable seg-ments were considered a minimum to reliably assess the wall motion score.

Infarct size. Enzymatic infarct size was estimated by cumulative alpha-hydroxybutyrate dehydrogenase values over the first 72 hours (α-HBDH Q72)after myocardial infarction as described by van der Laarse et al..19 This methodis not influenced by the presence or absence of reperfusion.

Statistical analysis. Results are presented as means ± standard deviation. Dif-ferences between groups were examined using the Student’s t-test. Logistic re-gression analysis was applied to identify independent relations between baseline

CHAPTER 6

126

characteristics and left ventricular dilatation.20 Calculations were made usingSPSS/PC+ software.

Results

Heart rate variability

Holter tapes of 199 CATS patients (199/298, 66%) were available for heartrate variability assessment at discharge. Analysis was not possible in 24 casesdue to speed errors (3 tapes) or incompatibility of Holter tapes and the analysissystem used (21 tapes). Therefore, data on heart rate variability were availablein 175 out of 298 (58%) cases. Baseline characteristics of all CATS patients andpatients that were part of the heart rate variability study are listed in Table 6.1.

There were no significant differences in age, gender, infarct size and parame-ters of left ventricular dysfunction between both groups. The mean recordingtime was 22.8 ± 3.2 hours (range 4.4 to 26, 90% of recordings longer than 21hours), and 92,849 ± 22,538 RR intervals were analyzed (range 14,426 to145,033 intervals). Before discharge, 74 patients (42%) had a reduced heart ratevariability (index of less or equal to 25) and 101 (58%) had a normal heart ratevariability (index of more than 25). The heart rate variability index was 19.09 ±

Table 6.1. Patient characteristics upon hospital admission

Total CATS population Cohort with heart ratevariability assessment

N 298 175Male (%)Age (years)α-HBDH Q72 (U/l)LVEDVI (ml/m2)LVESVI (ml/m2)LVEF (%)

7559 ± 10

1277 ± 100756 ± 1325 ± 1055 ± 10

7959 ± 10

1323 ± 84354 ± 1224 ± 957 ± 9

WMSI 1.91 ± 0.36 1.88 ± 0.37

α-HBDH Q72 indicates cumulative α-hydroxybutyrate dehydrogenase over the first 72hours after myocardial infarction (enzymatic infarct size); LVEDVI, left ventricular end-diastolic volume indexed for body surface area, LVEF, left ventricular ejection fraction;LVESVI, left ventricular end-systolic volume indexed for body surface area; WMSI, wallmotion score index.

HEART RATE VARIABILITY AND DILATATION

127

4.06 and 35.43 ± 8.40 in both groups, respectively. A second heart rate vari-ability assessment was available in 120/175 patients (69%) after three months.

Serial measurements in this subgroup revealed that after three months, heartrate variability had increased both in patients with (index of less or equal to 25)and patients without (index of more than 25) a reduced heart rate variability atdischarge. However, the increase in heart rate variability was more pronouncedin patients with an index of less or equal to 25 (increase 11.16 ± 7.37 versus5.97 ± 12.14, p=0.004). After three months, 31/120 (25%) patients who had anindex of less or equal to 25 at discharge showed an improvement to an index ofmore than 25 and 5/120 (4%) patients with an index of more than 25 at dis-charge had a worsening of their index to of less or equal to 25. All other patientsremained in the same category as they were at discharge (70/120 (58%) with anindex of more than 25 and 14/120 (11%) with an index of less or equal to 25).

Echocardiography

LVESVI (ml/m2)

p = 0.043

p = 0.008p = 0.05

0

10

20

30

40

50

60

70

Predischarge 12 months Increase

HRVI 25≤

HRVI 25>

F i g u r e 6 . 1 .F i g u r e 6 . 1 . A small but not significant difference in left ventricular end-systolic volume indexed for body sur-

face area (LVESVI) was present before discharge in patients with (HRVI ≤ 25) and without reduced heart rate

variability (HRVI > 25). After one year, LVESVI had increased in those with HRVI ≤ 25, but in patients withHRVI > 25, LVESVI had remained unchanged. HRVI indicates heart rate variability index; LVESVI, left ven-tricular end-systolic volume.

CHAPTER 6

128

Before discharge, an echocardiogram was available for all patients (175/175).Assessment of left ventricular volumes was possible in 125/175 cases (71%).Serial measurements before discharge and after 12 months were available for80/175 patients (45%). Echocardiographic follow-up of the two groups with andwithout a reduced heart rate variability at discharge is summarized in Table 6.2.

Before discharge, there was a slight but not significant difference in end-systolic volume between both groups. End-diastolic volume was also compara-ble between groups, but wall motion abnormalities were more pronounced inpatients with reduced heart rate variability. Ejection fraction was lower beforedischarge in the group with an index of less or equal to 25. After three months,both end-systolic and end-diastolic volume had increased in patients with a re-duced heart rate variability, whereas left ventricular dimensions decreased in

Table 6.2. Summary of all echo data

HRVI ≤ 25 N HRVI > 25 N p-value

Diastolic volume (ml/m 2) predischarge 59 ± 16 51 59 ± 13 74 0.916 3 months 12 months

64 ± 1965 ± 19

4436

58 ± 1358 ± 18

6368

0.0650.079

Systolic volume (ml/m 2) predischarge 3 months 12 months

29 ± 1332 ± 1433 ± 14

514436

25 ± 1124 ± 1025 ± 13

746368

0.0500.0030.008

Ejection fraction (%) predischarge 3 months 12 months

52 ± 1151 ± 1051 ± 10

514436

58 ± 959 ± 959 ± 10

746368

< 0.001< 0.001< 0.001

Wall motion score i ndex predischarge 3 months 12 months

1.97 ± 0.381.91 ± 0.451.97 ± 0.46

514436

1.69 ± 0.431.61 ± 0.391.58 ± 0.39

746368

< 0.001< 0.001< 0.001

Change in first year diastolic volume systolic volume ejection fraction

8 ±176 ±14-4 ±10

282828

-1 ± 120 ± 80 ± 9

525252

0.0240.0430.162

wall motion score 0.07 ± 0.31 28 -0.08± 0.24 52 0.018

HRVI indicates heart rate variability index.

HEART RATE VARIABILITY AND DILATATION

129

patients with an index of more than 25. Between three and 12 months, a smallincrease in left ventricular volumes was observed in both groups. The total in-crease in systolic and diastolic volume after one year was more pronounced inpatients with a reduced heart rate variability.

Changes in end-systolic volume in both groups after one year of follow up isshown in Figure 6.1. Ejection fraction had remained relatively stable in bothgroups, but the wall motion score index showed a slight improvement in pa-tients with an index of more than 25, while in patients with a reduced heart ratevariability a worsening of this index was observed. Figure 6.2 features the per-centage of patients with reduced heart rate variability at discharge in five sub-groups of left ventricular dilatation. The percentage of patients with a reducedheart rate variability was higher in the subgroups with 5-10, 10-15 and morethan 15 ml/m2 increase in end-systolic volume after one year in comparison topatients with no dilatation.

Increase in LVESVI (ml/m2)

0

20

40

60

80

100 (%)

≤ >1510-155-100-50

F i g u r e 6 . 2 .F i g u r e 6 . 2 . The percentage of patients with a reduced heart rate variability (index ≤ 25) is given for fivecategories of left ventricular dilatation. More dilatation, quantified as increase in end-systolic volume after 12months, is accompanied by a higher frequency of patients with a reduced heart rate variability.

CHAPTER 6

130

In Table 6.3, univariate analysis of possible predictive factors for the devel-opment of left ventricular dilatation is shown. Besides a reduced heart rate vari-ability, only an enzymatic infarct size of more than 1000 U/l significantly in-

Table 6.3. Predischarge characteristics of patients with and without left ventricular dilatation > 5 ml/m2 after 12 months

Dilatation N No dilatation N p-value

Demographics Age (years) Male (%)Myocardial injury α-HBDH Q72 (U/l) Ejection fraction (%) WMSIExtent of CAD ≥ 2-vessel disease (%) LAD occluded (%)Functional class Heart failure (Killip) Angina pectoris (NYHA)LV remodeling LVESVI (ml/m2) LVEDVI (ml/m2)Hemodynamics Heart rate (beats/min) Blood pressure (mmHg) systolic diastolic Rate-pressure productNeurohumoralactivation Heart rate variabilityMedication Beta-blockers (%) ACE inhibitors (%) Diuretics (%)

56.5 ± 8.475

1552 ± 97455 ± 10

1.92 ± 0.42

3628

1.4 ± 0.61.2 ± 0.5

28 ± 1258 ± 15

76 ± 8

117 ± 1772 ± 11

9278 ± 1756

24 ± 7

285520

4040

354040

2525

3939

4040

24

393924

24

404040

57.8 ± 10.484

1072 ± 76656 ± 10

1.78 ± 0.45

2916

1.3 ± 0.51.4 ± 0.6

26 ± 1261 ± 16

69 ± 8

120 ± 1775 ± 10

8321 ± 1630

32 ± 10

334817

7979

717979

4949

7979

7979

56

757552

56

797979

0.4950.385

0.0070.5640.090

0.2820.381

0.7100.210

0.3610.300

0.002

0.3650.1210.023

0.001

0.6940.6050.822

Digoxin (%) 8 40 3 79 0.428

CAD indicates coronary artery disease; LAD, left anterior descending artery. Otherabbreviations see Table 6.1. Coronary angiograms were made 31 ± 68 days after myocardialinfarction.

HEART RATE VARIABILITY AND DILATATION

131

creased the risk for dilatation during the first year of follow up. Age of the pa-tient, left ventricular volumes, ejection fraction and wall motion score index atdischarge had no significant predictive value for the occurrence of left ventricu-lar dilatation. Logistic regression analysis was applied to investigate whetherheart rate variability provided independent prognostic information for the de-velopment of dilatation (Table 6.4). A stepwise regression model (entry crite-rion: p < 0.15) revealed that a heart rate variability index of less or equal to 25was a stronger predictor for the occurrence of left ventricular dilatation of morethan 5 ml/m2 than an enzymatic infarct size of more than 1000 U/l.

Heart rate variability in patients with and without dilatation after three months

Before discharge, there was a significant difference in heart rate variabilityindex between patients with and without left ventricular dilatation more than 5ml/m2 after one year (24.24 ± 7.34 versus 32.33 ± 9.91), p < 0.001). After threemonths, heart rate variability had improved in patients with and without dilata-tion. However, this improvement was more pronounced in patients with dilata-tion, and after three months heart rate variability was no longer different be-tween these groups (36.30 ± 12.33 versus 38.91 ± 10.09, p=0.371).

Ventricular arrhythmias during follow up

In Table 6.5, the six patients that showed ventricular arrhythmias or died sud-denly in the first year after myocardial infarction are listed. Before discharge,heart rate variability was reduced in 3/6 (50%) of these patients (HRVI ≤ 25)compared to 74/175 (42%) in all patients. It is interesting to note that two pa-tients (cases 2 and 3) already had a large end-systolic volume at discharge(above the 75th percentile), and another three patients (case 1, 5 and 6) showedan increase in end-systolic volume after three months above the 75th percentileof the study population. One patient (case 4) had a normal end-systolic volume;

Table 6.4 . Independent predictors of left ventricular dilatation

B SE p-value OR (95%-CI)

Constant -1.9710 0.5248 0.0002HRVI ≤ 25 1.5934 0.5660 0.0049 4.92 (1.62 - 14.92)Infarct size > 1000 U/l 1.7524 0.5775 0.1856 2.15 (0.69 - 6.66)

B indicates regression coefficient; CI, confidence intervals; HRVI, heart rate variabilityindex; OR, odds ratio; SE, standard error.

CHAPTER 6

132

follow-up echocardiography was not available in this case. These data show thatearly or late left ventricular dilatation dilatation occurred in 5/6 (83%) of the pa-tients with arrhythmic events.

Selection of the optimal value of the heart rate variability index discriminatingbetween dilatation and no dilatation

In Figure 6.3, the Receiver Operator Characteristics (ROC) curve is plottedfor the heart rate variability index as a predictor of left ventricular dilatation. Inthis study, the cut-off point between normal and abnormal heart rate variabilitywas selected at a heart rate variability index of 25 on the basis of data in the lit-erature4. This value has a sensitivity of 62% and a specificity of 76% in detect-ing left ventricular dilatation. In this figure it can be seen, that an index of 30would yield a higher sensitivity of 83%, although this is accompanied by alower specificity (60%).

The positive predictive value of a heart rate variability index ≤ 25 for the oc-currence of left ventricular dilatation was 54% (15/28), and the negative predic-tive value was 83% (43/52).

Table 6.5 . Data from the six patients that had ventricular arrhythmias or diedsuddenly in the first year after myocardial infarction

LVESVI LVESVI atCase Event Days after HRVI at discharge 3 months

AMI (ml/m2) (ml/m2)

1 VT* 12 26 19 332 VF* 14 17 37 n.a.3 VT 19 12 43 n.a.4 SCD 171 17 23 n.a.5 SCD 171 53 19 376 SCD 277 34 14 23mean - 110 ± 112 26 ± 15 26 ± 12 31 ± 7all pts - - 29 ± 11 27 ± 12 29 ± 15

AMI indicates acute myocardial infarction; HRVI, heart rate variability index; LVESVI,left ventricular end-systolic volume indexed for body surface area; VF, ventricularfibrillation; VT, ventricular tachycardia; SCD, sudden cardiac death; n.a., not available;pts, patients. *After HRVI assessment but in hospital.

HEART RATE VARIABILITY AND DILATATION

133

Discussion

This study demonstrates the predictive value of a reduced heart rate variabil-ity before discharge for the occurrence of left ventricular dilatation during fol-low up after a first anterior myocardial infarction. Assessment of the heart ratevariability index before discharge yields important additional information notobtained by considering infarct size or left ventricular dysfunction alone. A de-crease in heart rate variability at discharge may reflect persistent sympatheticactivation, which can be harmful in the early phase of remodeling when scar tis-sue formation is not yet fully completed.21 This is the first study linking persist-ing sympathetic activation after myocardial infarction to the occurrence of leftventricular dilatation during follow up.

Heart rate variability as an indicator for neurohumoral activation

Sensitivity (%)

Specificity (%)0 20 40 60 80 100

0

20

40

60

80

10045

40

3530

25

20

15

Heart rate variabilityROC curve

Figure 6.3. Receiver Operator Characteristics (ROC) curve of the heart rate variability index as a predictorof left ventricular dilatation. For explanation see text.

CHAPTER 6

134

A reduced heart rate variability is thought to reflect an imbalance betweensympathetic and parasympathetic activity.7-9 Direct vagal nerve stimulation22

and the infusion of atropine8,23 or isoproterenol23 cause reproducible shifts in thepower spectrum of heart rate variability in normal subjects. In patients with heartfailure, it has been demonstrated that a reduced heart rate variability is directlyrelated to plasma norepinephrine levels and muscle sympathetic nerve activity.7

These studies indicate that changes in heart rate variability reflect changes inautonomic balance. No studies are available that have demonstrated a relationbetween a reduced heart rate variability and activation of the renin-angiotensinsystem. However, especially in patients with considerable left ventricular dys-function, persisting sympathetic activity after myocardial infarction is paralleledby activation of the renin-angiotensin system.24 Therefore, heart rate variabilityafter myocardial infarction may well reflect the degree of general neurohumoralactivation after myocardial infarction.

Infarct size and left ventricular dilatation

For dilatation to occur, a certain degree of myocardial injury is necessary. Inrat studies25 it has been demonstrated that if myocardial infarction is limited toless than 40% of the free wall, the loss of contractile tissue is compensated byphysiologic hypertrophy of the noninfarcted area, and the resulting dilatation islimited. However, when infarct size exceeds 40% of the free wall, compensatorymechanisms fail, hypertrophy becomes pathological, and progressive dilatationoccurs. This association between infarct size and left ventricular dilatation hasbeen confirmed in man26 and has also been reproduced in the present study.After the acute phase of myocardial infarction, the process of infarct healing be-comes an important determinant of left ventricular dilatation. This process canlast for weeks after myocardial infarction.21 During this period, the infarctedarea is highly vulnerable to an increase in wall stress.27

Neurohumoral activation and left ventricular dilatation

Increased loading conditions, leading to higher levels of wall stress, havebeen shown to promote left ventricular dilatation in man.10 Increased sympa-thetic activity, possibly paralleled by activation of the renin-angiotensin system,will increase pre- and afterload due to vasoconstriction and fluid retention lead-ing to higher wall stress levels. Therefore, neurohumoral activation early aftermyocardial infarction is liable to promote left ventricular dilatation by elevatingwall stress in a vulnerable phase of left ventricular remodeling. This is supportedby the finding that in this study, heart rate variability was reduced in patients

HEART RATE VARIABILITY AND DILATATION

135

with left ventricular dilatation at discharge. However, after three months no dif-ference in heart rate variability could be detected between patients with andwithout left ventricular dilatation, suggesting that a reduced heart rate variabilityis only associated with left ventricular dilatation in the early phase after myo-cardial infarction.

Other determinants of left ventricular dilatation

A number of other possible determinants of left ventricular dilatation wereconsidered in this study (Table 6.3). Occlusion of the infarct-related artery isknown to promote left ventricular dilatation.28 In the present study, the left ante-rior descending artery was occluded in 28% of patients with dilatation versus16% of patients without dilatation. However, this difference was not statisticallysignificant. It should be noted that coronary angiography was not part of thestudy protocol, and performed at a wide range of time intervals, usually becauseof recurrent angina pectoris. Coronary angioplasty and other procedures mayhave troubled the relation between patency of the infarct-related artery and leftventricular dilatation.

The rate-pressure product before discharge was higher in patients with leftventricular dilatation, and was primarily determined by a higher mean heart rate.This finding confirms the assumption that a higher workload promotes left ven-tricular dilatation. However, this parameter did not independently predict theoccurrence of dilatation (Table 6.4). Finally, there was no significant effect ofmedication, including study medication, on the occurrence of left ventriculardilatation in this subgroup of patients. Other studies have demonstrated thebeneficial effect of ACE inhibition on left ventricular remodeling.29-31 This lackof effect may be due to the limited number of patients studied.

Implications of the study

Identifying patients at risk for left ventricular dilatation after myocardial in-farction is a difficult but important task. The importance of infarct size as a de-terminant of dilatation is confirmed by the present study. Left ventricular end-systolic- and end-diastolic volume, ejection fraction and wall motion score haveno additive value for the prediction of dilatation one year after myocardial in-farction. An occluded infarct-related vessel is an independent risk factor for theoccurrence left ventricular dilatation28,32,33 and, when available, should be takeninto account when assessing the risk for ventricular dilatation after myocardialinfarction. A reduced heart rate variability before discharge as an indicator ofpersisting neurohumoral activation proves to be a strong predictor of left ven-

CHAPTER 6

136

tricular dilatation. Selection of a heart rate variability index of 30 as the cut-offpoint between ‘reduced’ and ‘normal’ heart rate variability would even improvethe sensitivity of this method. However, the predictive value of heart rate vari-ability for the occurrence of dilatation is limited to the late in-hospital period,possibly reflecting the importance of persisting sympathetic activity in the early(vulnerable) phase of remodeling. Assessment of heart rate variability beforedischarge can be helpful in identifying patients at risk for dilatation after myo-cardial infarction at relatively little cost in time and manpower.

Limitations

In this study, a selected patient population was investigated: only patientswith a first anterior myocardial infarction treated with streptokinase were in-cluded. Therefore, results of this study should be extrapolated to other patientgroups with caution. However, both patients with very small and very large in-farcts were part of the study, and therefore the complete spectrum of left ven-tricular remodeling was investigated.

Conclusions

Assessment of the heart rate variability index before discharge, but not threemonths later, yields important additional information for identifying patients atrisk for left ventricular dilatation after myocardial infarction. This informationcan be obtained at relatively little cost in time and manpower. The therapeuticstrategy for patients with a reduced heart rate variability remains to be estab-lished.

References

1. Klieg RE, Miller JP, Bigger JT, Moss AJ: Decreased heart rate variability and its a s-sociation with increased mortality after acute myocardial infarction. Am J Cardiol1987;59:256-262.

2. Casolo GC, Stroder P, Signorini C, et al: Heart rate variability during the acute phaseof myocardial infar ction. Circulation 1992;85:2073 -2079.

3. Bigger JT, Fleiss JL, Steinman RC, Rolnitzky LM, Kleiger RE, Rottman JN: Correl a-tions among time and frequency domain measures of heart period variability twoweeks after acute myocardial I nfarction. Am J Cardiol 1992;69:891-898.

4. Cripps TR, Malik M, Farrell TM, Camm AJ: Prognostic value of reduced heart ratevariability after myocardial infarction: clinical evaluation of a new analysis method.Br Heart J 1991;65:14-19.

HEART RATE VARIABILITY AND DILATATION

137

5. Farrell TG, Bashir Y, Cripps T, et al: Risk stratification for arrhythmic events inpostinfarction patients based on heart rate variability, ambulatory electroca r-diographic variables and the signal -averaged electrocardiogram. J Am Coll Cardiol1991;18:687-697.

6. Martin GJ, Magid NM, Myers G, et al: Heart rate variability and sudden death seco n-dary to coronary artery disease during ambulatory electrocardiographic monitoring.Am J Cardiol 1987;60:86-89.

7. Kienzle MG, Ferguson DW, Birkett CL, Myers GA, Berg WJ, Mariano DJ: Clinical,hemodynamic and sympathetic neural correlates of heart rate variability in congestiveheart failure. Am J Cardiol 1992;69:761-767.

8. Hayano J, Sakakibara Y, Yamada A, et al: Accuracy of assessment of cardiac vagaltone by heart rate variability in normal subjects. Am J Cardiol 1991;67:199-204.

9. Saul JP, Arai Y, Berger RD, Lill y LS, Colucci WS, Cohen RJ: Assessment of aut o-nomic regulation in chronic congestive heart failure by heart rate spectral analysis.Am J Cardiol 1988;61:1292 -1299.

10. Piérard LA, Albert A, Gilis F, Sprynger M, Carlier J, Kulbertus HE: Hemodynamicprofile of patients with acute myocardial infarction at risk of infarct expansion. Am JCardiol 1987;60:5-9.

11. Jugdutt BI, Michorowski BL, Kappagoda CT: Exercise training after anterior Q wavemyocardial infarction: Importance of regional left ventricular function and topogr a-phy. J Am Coll Cardiol 1988;12:362-372.

12. Pfeffer MA, Brauwnwald E: Ventricular remodelling after myocardial infarction.Experimental observ ations and clinical implications. Circulation 1990;81:1161 -1172.

13. Bigger JT Jr, Fleiss JL, Rolnitzky LM, Steinman RC: Stability over time of heart p e-riod variability in patients with previous myocardial infarction and ventricular a r-rhythmias. The CAPS and ESVEM investigators. Am J Cardiol 1992;69:718-723.

14. Van Hoogenhuyze D, Weinstein N, Marti n GJ, et al: Reproducibility and relation tomean heart rate of heart rate variability in normal subjects and in patients with co n-gestive heart failure secondary to coronary artery disease. Am J Cardiol1991;68:1668-1676.

15. van Gilst WH, Kingma JH, for the CATS investigators group: Early intervention withangiotensin-converting enzyme inhibitors during thrombolytic therapy in acute my o-cardial infarction: Rationale and design of Captopril and Thrombolysis Study. Am JCardiol 1991;68:111D -115D.

16. Malik M, Farrell T, Cripps T, Camm J: Heart rate variability in relation to prognosisafter myocardial infarction: selection of optimal processing techniques. Eur Heart J1989;10:1060-1074.

17. Mølgaard H, Sørensen KE, Bjerregaard P: Circadian variation and influence of riskfactors on heart rate variability in healthy Subjects. Am J Cardiol 1991;68:777-784.

18. Schiller NB, Shah PM, Crawford M, et al: Recommendations for quantitation of theleft ventricle by two -dimensional echocardiography. American Society o f Echocardi-ography Committee on Standards, Subcommittee on Quantitation ofTwo-Dimensional Echocardiograms. J Am Soc Echocardiogr 1989;2:358-367.

19. van der Laarse A, Kerkhof PL, Vermeer F, et al: Relation between infarct size and leftventricular performance assessed in patients with first acute myocardial infarction

CHAPTER 6

138

randomised to intracoronary thrombolytic therapy or to conventional treatment. Am JCardiol 1988;61:1-7.

20. Altman DG: Relation between several variables. In: Practical statistics for medi cal re-search. 1st ed. Lo ndon: Chapman and Hall, 1990:325 -364.

21. Weisman HF, Healy B: Myocardial infarct expansion, infarct extension and reinfar c-tion: Pathophys iologic concepts. Prog Cardiovasc Dis 1987;307:73-110.

22. Kamath MV, Upton AR, Talalla A, Fallen EL: Effect of vagal nerve electrostimul a-tion on the power spectrum of heart rate variability in man. PACE 1992;15:235-243.

23. Binkley PF, Nunziata E, Haas GJ, Nelson SD, Cody RJ: Parasympathetic withdrawalis an integral component of autonomic imbalance in congestive heart failure: demo n-stration in human subjects and verification in a paced canine model of ventricularfailure. J Am Coll Cardiol 1991;18:464-472.

24. McAlpine HM, Morton JJ, Leckie B, Rumley A, Gillen G, Dargie HJ: Neuroendocrineactivation after acute myocardial infarction. Br Heart J 1988;60:117-124.

25. Anversa P, Olivetti G, Capasso JM: Cellular basis of ventricular remodelling aftermyocardial infar ction. Am J Cardiol 1991;68:7D-16D.

26. Gaudron P, Eilles C, Ertl G, Kochsiek K: Early remodelling of the left ventricle in p a-tients with myoca rdial infarction. Eur Heart J 1990;11(Suppl. B):139 -146.

27. Weisman HF, Bush DE, Mannisi JA, Weisfeldt ML, Healy B: Cellular mechanisms ofmyocardial infarct expansion. Circulation 1985;57:186-201.

28. Jeremy RW, Hackworthy RA, Bautovich G, Hutton BF, Harris PJ: Infarct artery pe r-fusion and changes in left ventricular volume in the month after acute myocardial i n-farction. J Am Coll Cardiol 1987;9:989-995.

29. Pfeffer MA, Lamas GA, Vaughan DE, Parisi AF, Brauwnwald E: Effect of captoprilon progressive ventricular dilatation after anterior myocardial infarction. N Engl JMed 1988;319:80-86.

30. Nabel EG, Topol EJ, Galeana A, et al: A randomised, placebo -controlled trial ofcombined early intravenous captopril and recombinant tissue -type plasminogen act i-vator therapy in acute my ocardial infarction. J Am Coll Cardiol 1991;17:467-473.

31. Sharpe N, Smith H, Murphy J, Greaves S, Hart H, Gamble G: Early prevention of leftventricular dysfunction after myocardial infarction with angi o-tensin-converting-enzyme inhibition. Lancet 1991;337:872 -876.

32. White HD, Norris RM, Brown MA, Brandt PWT, Whitlock RML, Wildt CJ: Left ve n-tricular end-systolic volume as the major determinant of survival after recovery frommyocardial infarction. Circulation 1987;76:44-51.

33. Gibson RS, Bishop HL, Stamm RB, Crampton RS, Beller GA, Martin RP: Value ofearly two dimensional echocardiography in patients with acute myocardial infarction.Am J Cardiol 1982;49:1110 -1119.

CHAPTER 7

Acute intervention with captopril during thrombolysis inpatients with first anterior myocardial infarction

Results from the Captopril And Thrombolysis Study (CATS)

J. Herre Kingma,1,2 MD; Wiek H. van Gilst,2 PhD; Kathinka H. Peels,3 MD;Jan-Henk E. Dambrink,1 MD; Freek W.A. Verheugt,5 MD;

Robert P. Wielenga,4 MD; for the CATS investigators6

1Department of Cardiology, St Antonius Hospital, Nieuwegein2Department of Clinical Pharmacology, University of Groningen, Groningen3Department of Cardiology, Catharina Hospital, Eindhoven4Department of Cardiology, Ignatius Hospital, Breda5Department of Cardiology, Free University Hospital, Amsterdam6see Appendix

Published in:Eur Heart J 1994;15:898-907

THREE-MONTH CATS RESULTS

141

Abstract

Background. Left ventricular dysfunction and prognosis after myocardial in-farction can be improved by angiotensin-converting enzyme inhibition startedafter the ischemic phase. Experimental evidence suggests that intervention dur-ing thrombolysis may lead to even further benefit.Methods. In a randomized, double-blind placebo controlled trial, 298 patientswith a first anterior myocardial infarction, eligible for thrombolytic therapywere treated with 6.25 mg captopril or placebo, started immediately uponstreptokinase infusion and titrated to 25 mg TID. Effects of captopril by an in-tention to treat analysis on left ventricular volumes, ventricular arrhythmias,neurohumoral activation and enzymatic infarct size were measured.Results. During dose titration, mean blood pressure and heart rate were notdifferent in both groups. However, first dose hypotension was reported in 18patients on placebo and 31 patients on captopril (p < 0.05). At discharge 80%of patients were on study medication. Left ventricular volumes were significantlyincreased in both groups at three months. Left ventricular volumes in the capto-pril group tended to be lower, but differences were not statistically significant.Incidence of accelerated idioventricular rhythm and nonsustained ventriculartachycardia in captopril patients was lower (p < 0.05), paralleled by transientlylower norepinephrine levels (p < 0.05) upon thrombolysis. Enzymatic infarctsize showed to be smaller in captopril patients, especially in larger infarcts (p <0.05). A 34% (95% confidence interval; 0-56%) lower incidence of heart failureduring three months of follow up was reported in the captopril group.Conclusion. Captopril is well tolerated, although first dose hypotension wasmore common in patients on captopril. In agreement with experimental studiescaptopril reduces repetitive ventricular arrhythmias and catecholamine levels inthe acute thrombolytic phase of myocardial infarction. Although left ventricularvolumes were not significantly smaller in captopril patients, in the chronicphase, these patients showed a reduced incidence of heart failure.

Introduction

Over the last decade the use of thrombolytic therapy has drastically changedthe approach to acute myocardial infarction. With intravenous thrombolyticdrugs, recanalization is achieved in 68 - 76% of the patients.1 Early restorationof flow to the jeopardized myocardium greatly contributes to maintenance offunction2 and improvement of survival3,4 irrespective of the type of thrombolyticagent used.1

Although not proven to be of clinical importance, experimental observationsindicate that early and acute re-establishment of oxygen supply to a previously

CHAPTER 7

142

ischemic area, may result in a paradoxical increase of myocardial injury knownas ‘reperfusion injury’.5 There is experimental evidence that the use of the an-giotensin-converting enzyme (ACE) inhibitor captopril before or during the timeof reperfusion after ischemia, can prevent or limit the occurrence of reperfusionarrhythmias and myocardial injury both in vitro6 and in vivo.7 Use of ACE in-hibitors in patients with a reduced left ventricular ejection fraction following theacute phase of myocardial infarction can improve ventricular function by reduc-ing diastolic and systolic ventricular expansion.8,9 Recently, it was demonstratedthat long term use of captopril in patients with asymptomatic left ventriculardysfunction after myocardial infarction could reduce mortality, the incidence ofheart failure and the rate of reinfarction.10 When captopril was given within 24hours after the onset of symptoms a favorable effect on left ventricular remodel-ing was demonstrated.11 However, treatment with intravenous enalapril within24 hours after the onset of symptoms of myocardial infarction did not reveal aneffect on six months survival.12 A modest but significant effect on mortality ofcaptopril13 and lisinopril14 respectively was observed when treatment is startedwithin the first 24 hours after myocardial infarction. The reduction of mortalitybecame even more apparent in patients with clinical evidence of transient or on-going heart failure when oral ramipril was administered between the second andninth day after myocardial infarction.15

On the basis of the aforementioned experimental evidence, we hypothesizedthat immediate concomitant administration of captopril would enhance the ef-fects of thrombolysis in the acute phase of myocardial infarction and that itscontinued use would result in preservation of left ventricular function in thechronic phase. To test this hypothesis we designed the Captopril And Throm-bolysis Study (CATS). The primary objectives included attenuation of left ven-tricular volume expansion, a reduction of ventricular arrhythmias and a decreasein norepinephrine release following thrombolysis. An important aim was the as-sessment of hemodynamic safety and tolerance of oral captopril administeredimmediately following thrombolytic therapy with streptokinase in patients with afirst anterior myocardial infarction.

Methods

Organization. CATS was a double-blind, randomized, placebo-controlledstudy in patients with a first anterior wall myocardial infarction treated withthrombolytic therapy with intravenous streptokinase. A total of 298 patientswere enrolled in 12 hospitals in the Netherlands and assigned to either placeboor captopril. The coordination center was the St Antonius Hospital, Nieuwegein,

THREE-MONTH CATS RESULTS

143

The Netherlands. The study was controlled and supervised by two study-directors with support of a steering committee including the principal investiga-tors from all participating hospitals and representatives from the sponsor. Allmeasurements of the primary objectives and enzymatic infarct size were per-formed in central core laboratories and data were reviewed and stored in thedatabase in the coordination center. Data on safety including serious adverseclinical events were reviewed by the study directors and filed in the database ofthe sponsor. An independent Data Quality Committee reviewed the data onclinical endpoints. The progress and conduct of the study with emphasis on thesafety of patients, was supervised by an international Policy Advisory Board. Atregular intervals the Board was provided by the sponsor with blinded data onserious adverse clinical events represented per treatment group. If there wouldbe a strong indication that adverse clinical events would occur more frequentlyin one of the treatment groups, the Board had access to the randomization codeand could decide to terminate the study prematurely. Data analysis in the coor-dination center and by the sponsor at the end of the study was supervised by amember of the Policy Advisory Board (J.G.P.T.). The study was approved bythe Institutional Review Board of all participating hospitals.Objectives of the study. Primary objectives were the assessment of the effect ofcaptopril relative to placebo:

- on the preservation of left ventricular volume as measured by serial 2D-echocardiography at three months after myocardial infarction;

- on the incidence of episodes of nonsustained ventricular tachycardia (VT)and accelerated idioventricular rhythm (AIVR) assessed by serial ambula-tory ECG monitoring during the first 12 hours after admission;

- on neurohumoral activation as determined by serial assay of norepinephrinelevels and plasma renin activity in the acute phase.

Secondary objectives were the assessment of the effect of captopril, relative toplacebo:

- on infarct size, calculated from α-hydroxybutyrate dehydrogenase (α-HBDH) determinations;

- on radionuclide left ventricular ejection fraction, obtained at rest threemonths after myocardial infarction;

- on clinical event rate during follow up, including mortality, reinfarction,development of heart failure, unstable angina, need for percutaneoustransluminal coronary angioplasty (PTCA) and/or coronary artery bypassgrafting (CABG).

CHAPTER 7

144

Eligibility of patients. Patients were considered eligible for enrollment, if afirst anterior wall myocardial infarction within 6 hours after the onset of symp-toms was present, and if they were treated with thrombolytic therapy (1.5 mil-lion international units intravenous streptokinase administered in 30 minutes).Informed consent was obtained by witnessed oral consent, later confirmed bywritten consent following the acute phase of myocardial infarction. Diagnosiswas based on the presence of characteristic symptoms of acute anterior myo-cardial infarction, with at least 1 mm ST-segment elevation in lead I and aVLand/or 2 mm ST elevation in two precordial leads of the 12-lead electrocardio-gram compatible with infarction of the anterior wall and adjacent areas, includ-ing septal and lateral portions of the left ventricle. Patients were excluded ifthere was known intolerance to ACE inhibitors, renal insufficiency, systolicblood pressure over 200 mmHg or below 100 mmHg and diastolic blood pres-sure over 120 mmHg or below 55 mmHg. Additional exclusion criteria were se-vere valvular heart disease, arrhythmias requiring anti-arrhythmic therapy, seri-ous systemic or metabolic disease except diabetes mellitus, AV conduction dis-turbances, (PR interval ≥ 0.24 s), left bundle branch block, a history of transientischemic attacks and a cerebrovascular accident within six weeks.

Treatment protocol. Immediately after admission to the coronary care unit,1,500,000 international units of streptokinase was administered in 30 minutes,by continuous intravenous infusion. Nitrates were withheld during this phaseand were only allowed when specifically indicated, such as for elevated systolicblood pressure or severe angina pectoris. Double-blind medication was initiatedimmediately upon completion of the streptokinase infusion, provided the sys-tolic blood pressure was stable and above or equal to 100 mmHg. If systolicblood pressure was below 100 mmHg, the initiation of double-blind medicationcould be postponed for a maximum of 1 hour following start of the streptok-inase infusion. Initially, an oral dose of 6.25 mg was given, repeated after 4 and8 hours and at 16 and 24 hours doses of 12.5 mg and 25 mg respectively wereadministered. Dose titration was continued provided the systolic blood pressure,measured immediately prior to the next scheduled dose of study medication wasabove or equal to 95 mmHg. If the systolic blood pressure was below 95mmHg, the study medication was withheld until the next dosing time. The targetmaintenance dose of the double-blind study medication was 25 mg TID, admin-istered from day 2 until three months after myocardial infarction. During thethree months of double-blind therapy, concomitant therapy with calcium an-tagonists, beta-blockers or nitrates was instituted only for specific indications,e.g. angina and hypertension. However the protocol did not prohibit the use ofbeta-blockers for secondary prevention. Also the use of aspirin was at the dis-

THREE-MONTH CATS RESULTS

145

cretion of the local investigator. The recommended dose of aspirin was 80 mg tominimize a possible interaction with captopril. Blood pressure and heart ratewere monitored during the initial phase with an automatic cuff blood pressuremeasurement device (Dynamap, Criticon, Germany) with digital print-out everythree minutes during the first hour.

Echocardiography. For echocardiographic measurement of left ventricularend-diastolic and end-systolic volumes, apical four- and two-chamber viewswere obtained and recordings made with respiration held at end-expiration orpartial inspiration. Patient angulations, transducer position and respiratory phasewere recorded. End-diastolic and end-systolic frames were outlined from bothapical views using the Dataview Microsonics cardiac analysis system (NovaMicrosonics). Left ventricular volumes were calculated by the biplane(modified) Simpson’s rule method using both views. The mean of two meas-urements on consecutive cycles was taken for each examination. Left ventricu-lar end-diastolic volume index and left ventricular end-systolic volume index,were derived from body surface area, which was estimated at each time-point.Initial echocardiographic determination of left ventricular volume was donewhen possible, within 12-24 hours after admission and at least 4 hours after thelast dose of double blind study medication. All subsequent echocardiographicdeterminations obtained at day 3, prior to discharge and at three months weremade not earlier than 8 hours post-dose of double blind study medication. Toassure that return to the same echocardiographic view, a method of locating thetransducer in reference to bony landmarks was applied. Lateral rotation and ele-vation of the upper body were noted and maintained at subsequent studies.

Long-term ambulatory ECG. The presence of ventricular arrhythmias was as-sessed by two-channel 24-hour ambulatory ECG, (Reynolds Medical Trackerrecorder) at baseline for at least 12 hours, predischarge and after three monthsof double blind study medication. Analysis was performed using the ReynoldsMedical Pathfinder 3. The number and duration of episodes of AIVR, pairs ofventricular premature beats and VT were determined by real time counting.Premature ventricular beats were disregarded in the analysis. AIVR was definedas a repetition of three or more monomorphic ventricular beats with a rate ofless than 100 beats per minute. Pairs of ventricular premature beats were de-fined as a repetition of two ventricular beats with a maximum interval of 0.6 s.VT was defined as a repetition of three or more ventricular beats with a rate ex-ceeding 100 beats per minute.

Neurohormones. Neuroendocrine determinations, including samples for de-termination of norepinephrine and plasma renin activity, were collected at thestart of the study medication at 1 and 12 hours post-dose and twice daily there-after from day 2 to 5. Epinephrine and norepinephrine were measured using a

CHAPTER 7

146

sensitive HPLC assay with electrochemical detection. Plasma renin activity wasmeasured by means of a radio-immuno assay (Dupont, Ria NEN kit No NEA-022, 026).

Enzymatic infarct size and radionuclide ejection fraction. Infarct size was es-timated from the cumulative release of α-HBDH activity per liter of plasmawithin the first 72 hours, calculated from serial plasma α-HBDH determinationsfrom blood samples taken twice daily during the first five days, as described byvan der Laarse et al..16 Left ventricular ejection fraction was determined at restby radionuclide scanning after three months of double-blind therapy. All de-terminations were made not earlier than 8 hours post-dose of study medication.

Clinical events. Clinical events including development or worsening of con-gestive heart failure, serious rhythm disturbances, angina, reinfarction, need forPTCA or CABG, cardiac morbidity and mortality were noted in the patient rec-ord forms and transcribed on adverse event forms in case of serious events.

The presence of congestive heart failure was based on clinical judgment ofthe principal investigators. Patients with congestive heart failure were subdi-vided in those without medication, those in whom digoxin and/or diuretics werestarted, those in whom open treatment with an ACE inhibitor was started, thosein whom hospitalization was prolonged or who had to be rehospitalized. All pa-tients on diuretics and/or digoxin, and those in whom open treatment with anACE inhibitor was started were considered to have congestive heart failure.

Sample size consideration. Based on the anticipated variances in left ventricu-lar volumes as measured by 2D-echocardiography, sample size was estimated at280 patients. It was assumed that technically adequate echocardiographic meas-urements of left ventricular volume would be available in 70-75% of random-ized patients. In addition, left ventricular volume measurements beyond the ini-tial 24-hour recording, would not be present in approximately 8-10% of the pa-tients, due to early mortality. Therefore, it was estimated that from about 65% ofthe projected 280 randomized patients (180 patients) left ventricular volumeswould be available for evaluation. The effective sample size (N=180) was an-ticipated to be sufficient to detect a mean difference in left ventricular end-diastolic volume index between the treatment groups of 8 ml/m2 at three monthswith a power exceeding 0.8. The standard deviation for left ventricular end-diastolic volume index was assumed to be 18 ml/m2. A mean difference of 5ml/m2 in left ventricular end-systolic volume index at three months would bedetectable with similar power and a standard deviation of 12 ml/m2.

THREE-MONTH CATS RESULTS

147

Statistics. If not otherwise indicated, continuous variables were comparedusing Student’s t-test and categoral variables using Chi-square test. Results wereconsidered statistically significant if the p-values were less than or equal to 0.05,using the two-sided level of significance. Data on left ventricular volumes andenzymatic infarct size are represented by mean values with 95% confidence in-tervals. Left ventricular volume change and neurohumoral data over time wereevaluated by analysis of variance for repeated measurements. The incidence ofclinical endpoints is represented by relative risk estimates with corresponding95% confidence intervals. According to intention-to-treat principle all outcomeanalyses were based on the treatment group to which the patients had been ran-domly assigned.

Table 7.1. Baseline characteristics and medication in the two treatment groups

Characteristics Placebo CaptoprilN=149 N=149

Age (years) 60 ± 9 59 ± 10Male (%) 80 70Clinical history (%) ischemic heart disease hypertension diabetes mellitus current smoker

8.016.19.457.7

9.427.59.467.1

Killip class (%) class I class II

7525

7624

Concomitant medication* (%) aspirin beta-blockers calcium antagonists diuretics nitrates

31.511.4

05.49.4

32.914.1

012.110.7

Hemodynamics*

blood pressure (mmHg) systolic diastolic

131 ± 1.5 79 ± 1.1

136 ± 2.1 82 ± 1.3

heart rate (beats/min) 75 ± 1.2 78 ± 1.6* as assessed at randomization.

0 2 4 6 8 10 1270

80

90

100

110

120

130

140

Placebo

Captopril

Hours

mmHg ± SEM

6.25 mg 6.25 mg 6.25 mg

Figure 7.1 . Mean (± SEM) systolic and diastolic blood pressure after repeated doses of captopril com-pared to placebo.

CHAPTER 7

148

Results

During the enrollment period of the CATS study, 298 patients were included,with 149 patients allocated to each treatment group. The time from onset ofsymptoms to the time of streptokinase infusion was 166 ± 70 minutes in the pla-cebo group and 163 ± 76 minutes in the captopril group. The time of the firstdose of study medication from onset of symptoms was 213 ± 76 minutes and212 ± 86 minutes respectively. A complete clinical follow-up over the three-month period was obtained in 282 patients (94.6%). Fifteen patients (5.0%) diedand one patient was lost during follow up. Twenty-four hours after randomiza-tion, the target dose of 25 mg TID was reached in 95% of patients. At discharge,80% of patients were still on study medication. At the end of the three-monthfollow-up period, 79% of patients were still on study medication (placebo 81%and captopril 77%). After the dose titration phase, the mean dose administeredat 24 hours was 24.3 mg (placebo 24.2 mg and captopril 24.5 mg). Complianceof patients with study medication based on pill count was 79.9%. The clinicalcharacteristics of the two groups were comparable at baseline (Table 7.1), al-though smoking and hypertension tended to be more frequent in the captoprilgroup.

Safety: blood pressure and heart rate during titration phase

Heart rate remained unchanged from baseline in both groups during the titra-tion phase. Mean systolic and diastolic blood pressure decreased significantlyafter the first dose in both groups. Mean systolic blood pressure decreased by8.7 ± 2.3 mmHg in the placebo group and 13.3 ± 2.2 mmHg in the captoprilgroup (both p < 0.05, Figure 7.1).No significant blood pressure changes wererecorded after the second and third doses. There were no distinguishable differ-ences in mean blood pressure between the two groups during dose titration.Acute hypotension after the first dose, defined as a drop in systolic blood pres-sure below 90 mmHg, measured by automatic cuff blood pressure equipment,was reported in 18 patients in the placebo group and in 31 patients in the capto-pril group (relative risk (RR)1.3, 95% confidence interval (CI) 1.02 - 1.77). Thetotal incidence of hypotension during the titration phase was 22 (14.8%) in theplacebo and 33 (22.3%) in the captopril group (RR 1.3, 95% CI 0.96 - 1.66).The reported incidence of hypotension during the three-month follow-up was18.1% in the placebo and 26.8% in the captopril group (RR 1.5, 95% CI 0.96-2.28).

Echocardiographic measurements

THREE-MONTH CATS RESULTS

149

Biplane projections for appropriate measurement of left ventricular volumeswere available in 66% of echocardiograms. Both left ventricular end-systolicand end-diastolic volume indexes at first measurement (i.e., within 24 hours af-ter randomization) were at the upper range of normal (Figure 7.2). The placebogroup showed a sustained increase in left ventricular end-diastolic volume of6.2 ml/m2 (95% CI 1.7 - 10.8) over the three-month period. In the captoprilgroup there was an early but transient decrease in both left ventricular end-diastolic volume index and left ventricular end-systolic volume index. Overall,left ventricular end-diastolic volume for the captopril group increased by 5.4ml/m2 (95% CI 0.6 - 10.2) over the three-month period while both the left ven-tricular end-diastolic and end-systolic volume indexes for the captopril grouptended to be lower than for the placebo group at all time intervals.

CHAPTER 7

150

22

24

26

28

30

32

Placebo

Captopril

< 24 hours Day 3 Day 10 Month 3

LVESVI (ml/m2)

52

54

56

58

60

62

64

66

Placebo

Captopril

< 24 hours Day 3 Day 10 Month 3

LVEDVI (ml/m2)

Figure 7.2. Left ventricular end-systolic index (LVESVI, panel A) and end-diastolic volumeindex (LVEDVI, panel B) of patients allocated to captopril or placebo during the first threemonths.

THREE-MONTH CATS RESULTS

151

This difference was most pronounced at day 3 for the left ventricular end-diastolic volume index, with a difference of 3.5 ml/m2 (95% CI 0.0 - 7.0). Forleft ventricular end-systolic volume index, this difference was most pronouncedprior to discharge, being 3.3 ml/m2 (95% CI 1.1 - 5.5). The overall difference(repeated measurements analysis) between the captopril and placebo group wasnot statistically significant.

Ventricular arrhythmias

Of all Holter tape recordings 67% were adequate for analysis. The incidenceof ventricular arrhythmias, in particular of AIVR and nonsustained VT, washighest at admission. All arrhythmias diminished prior to discharge but tendedto become more frequent at three months. In the captopril group the number ofpatients with pairs of ventricular premature beats, AIVR and nonsustained VTwere lower at all time intervals than in patients allocated to placebo (Table 7.2).During the acute thrombolytic phase nonsustained VT and AIVR in the capto-pril group were significantly less (by 22% and 25%, respectively, both p < 0.05)than in the placebo group (Table 7.2).

Neurohumoral activation

Plasma samples for neurohumoral measurements were available in 92% ofthe patients. Norepinephrine levels and plasma renin activity were both in-creased at randomization (Table 7.3). In contrast to the placebo group, norepi-nephrine in the captopril group was significantly lower 1 hour after the first dose

Table 7.2 . Percentage of patients with episodes of AIVR, pairs of VPBs andnonsustained VT

AIVR PAIRS nsVT

Pl C Pl C Pl C

Thrombolysis 44.5 33.8* 59.9 52.1 40.9 31.7*

Discharge 2.4 1.6 16.8 15.8 4.0 1.6Three months 4.1 3.4 22.1 18.5 3.3 2.5

AIVR indicates accelerated idioventricular rhythm; C, captopril; PAIRS, pairs of ventric u-lar premature beats; nsVT, nonsustained ventricular tachycardia; PL, placebo. Based on data from all p a-tients with Holter monitoring, including those with failed registrations. Values comparedto placebo * p < 0.05 (Fisher’s Exact test).

CHAPTER 7

152

compared to baseline. However, after 12 hours no significant differences in no-repinephrine levels between the two groups could be detected. During the firstfive days plasma renin activity showed an increased in the captopril group andremained within normal range in the placebo group (p < 0.05, Table 7.3). In thecaptopril group ACE activity was significantly lower compared to baseline fromthe first hour after start of the study.

Enzymatic infarct size and radionuclide ejection fraction

Table 7.3. Plasma renin activity, norepinephrine levels and ACE activity during thefirst five days after acute myocardial infarction

Baseline 1 hour 12 hours 5 days

PRA P 2.86 ± 0.25 1.87 ± 0.46 1.78 ± 0.56 3.65 ± 0.70(U/l.h) C 3.01 ± 0.37 3.29 ± 0.14 2.04 ± 0.25 12.67 ± 2.02#

Norepinephrine(pg/ml)

PC

1173 ± 731081 ± 60

1080 ± 78914 ± 46*

813 ± 51812 ± 53

732 ± 38723 ± 53

ACE activity P 20.7 ± 1.5 20.1 ± 1.4 20.3 ± 1.3 19.2 ± 1.3(nmol/ml/min) C 19.4 ± 1.2 17.3 ± 1.2* 14.4 ± 1.3§ 12.8 ± 1.4#

ACE indicates angiotensin converting enzyme; C, captopril; P, placebo; PRA, plasma reninactivity.Values are represented as mean ± SEM. Differences compared to baseline values: * p <0.05,§ p < 0.001, # p < 0.0001.

Table 7.4. Peak plasma enzyme activities of CK and α-HBDH and calculatedcumulative release of α-HBDH in the two groups of patients

Placebo Captopril Difference (95% CI)

Peak CK (U/l) 1907 ± 1635 1618 ± 1360 289 (-61 - 639)peak α-HBDH (U/l)α-HBDH Q72 (U/l) small infarcts 1)

876 ± 7201390 ± 1109 587 ± 314

699 ± 4541166 ± 886 491 ± 278

177 (36 - 318)*

224 (-22 - 470)96 (-7 - 199)

large infarcts 2) 2193 ± 1035 1873 ± 738 320 (6 - 634)*

1) mean values of the lower two quartiles (50%) of infarcts; 2) mean values of the uppertwo quartiles (50%) of infarcts; a-HBDH indicates α-hydroxybutyrate dehydrogenase;Q72 cumulative α-HBDH release in the first 72 hours; CI, confidence interval; CK, cr e-atine phosphokinase. Values are represented as means ± SD. * p < 0.05.

THREE-MONTH CATS RESULTS

153

Serum creatine kinase and α-HBDH activity could be determined in 86.9% ofpatients. Peak creatine kinase, peak α-HBDH and cumulative α-HBDH releaseover 72 hours (α-HBDH Q72) were lower in the captopril patients than in theplacebo group (Table 7.4). This reduction in cardiac enzyme release indicatesan attenuation of myocardial injury, i.e., infarct size. However, this was not re-flected in a difference in nuclear left ventricular ejection fraction, measured atthree months which amounted to 43.9 ± 1.4% in the placebo group and 44.7 ±1.6% in the captopril group.

Clinical events

Incidence of clinical events are summarized in Table 7.5. Mortality rate waslow with no significant difference between the two treatment groups. However,there was a 34% (95% CI 0 - 56) lower reported incidence of heart failure in thecaptopril group compared to the placebo group. The incidence of coronary re-vascularization procedures was equal in both groups. Reinfarction occurred in14 patients, seven of which within 72 hours, with four cases PTCA related.

Table 7.5. Clinical endpoints. Figures represent numbers of patients

Placebo Captopril Relative riskN=149 N=149 (95% CI)

Mortality 6 9 1.50 (0.55 - 4.11)Heart failure hospitalization open label ACE inhibitor start diuretics/digitalis

42 7 2

21

28 2 1

22

0.66 (0.44 - 1.00)*

0.28 (0.07 - 1.21) 0.50 (0.05 - 5.18) 1.04 (0.60 - 1.81)

PTCA 291 272 0.92 (0.58 - 1.48)CABG 71 92 1.28 (0.49 - 3.34)Reinfarction overall < 72 hours

4

2

10

5

2.48 (0.83 - 7.43)

2.48 (0.52 - 11.92) > 72 hours 2 5 2.48 (0.52 - 11.92)

1One placebo patient had PTCA and CABG and is counted once under each procedure.2Three capt o-pril patients had PTCA and CABG and are counted once each under both procedures. * re-flecting p=0.05 (placebo vs captopril).

CHAPTER 7

154

Discussion

It has been suggested that the adjunctive use of ACE inhibitors during throm-bolytic therapy for myocardial infarction may improve clinical outcome.17 In thepresent study captopril was administered to patients with acute myocardial in-farction concomitantly with thrombolytic therapy. We demonstrate that at leastpart of the ACE inhibitor associated advantageous effects, observed in animalmodels in ischemia reperfusion i.e. a decrease in ventricular arrhythmias andnorepinephrine release, are present as well in patients with acute myocardial in-farction treated with thrombolytic therapy. However, remodeling indicated byan enlargement of left ventricular end-systolic or end-diastolic volumes was notprevented, although mean left ventricular volumes were smaller at all time inter-vals in the captopril group than in the placebo group.

Safety and tolerance

First dose hypotension and hypotension during the first 24 hours was morefrequent in the captopril group as compared to the placebo group, but in onlyfive patients did this necessitate discontinuation of treatment. Nabel et al.18 pub-lished data from 38 patients with acute myocardial infarction who underwentthrombolytic therapy with rtPA. Captopril was given intravenously in that study90 minutes after rtPA and no significant changes in blood pressure were ob-served. However, in the Captopril And Thrombolysis pilot Study,17 captoprilwhen administered intravenously concomitantly with streptokinase infusion,produced profound, although short lasting drops in systolic blood pressure.Transient hypotension may be observed during streptokinase infusion itself19

and is possibly related to transiently increased bradykinin levels.20 Use of intra-venous captopril, which acutely inhibits the breakdown of circulating bradyki-nin, apparently aggravates streptokinase induced hypotension. This unfavorableinteraction was avoided in the present study, since captopril was administeredorally, immediately upon completion of the streptokinase infusion and resultedin a more gradual ACE inhibition without interfering with the streptokinase. Theintravenous use of enalapril in acute myocardial infarction also induced unto-ward hypotensive effects, as observed in CONSENSUS II,12 although this wasunrelated to thrombolysis, since enalapril was administered approximately 18hours after admission and mainly in patients without concomitant thrombolytictherapy. One may speculate that the extent of suppression of ACE activity ob-tained after a relatively high dose of intravenous enalapril in CONSENSUS II12

may have been inappropriate, whereas in the CATS study only moderate dos-

THREE-MONTH CATS RESULTS

155

ages of captopril were used initially, resulting in a moderate but significant re-duction of plasma ACE activity.

Left ventricular volumes

Several reasons can be postulated for the absence of the anticipated reductionof left ventricular volumes with captopril compared to placebo. First, the con-secutive enrollment of patients in the CATS study yielded an overall equal dis-tribution of small, moderate and large infarcts with a limited proportion of pa-tients showing left ventricular expansion. The average extent of dilatation mayhave been too small for an effect to appear. The patients included in this studywere not selected on the basis of left ventricular dysfunction, and as such repre-sent the common population of patients eligible for thrombolysis treatment witha relatively good prognosis. Infarct-related vessel patency may be estimated tobe 68 - 76%,1 which is an important determinant of change in left ventricularvolume shortly after myocardial infarction.21 The second reason for the lack ofeffect may be the short observation period i.e. the short exposure to the risk as-sociated with left ventricular expansion in this three months study. Interestingly,despite the lack of effect on remodeling, the reported incidence of heart failureduring follow up was clearly reduced indicating that other mechanisms than re-modeling alone may have affected clinical outcome. A third reason could be theinsufficient sensitivity of quantitative echocardiography as a tool to reliably de-tect small changes in left ventricular volumes.

Norepinephrine, infarct size and ventricular arrhythmias

An important finding in animal models of ischemia followed by reperfusion,is the effect of ACE inhibition which blunts the norepinephrine release from themyocardium.22 Although this effect was modest in this group of patients, capto-pril appears to reduce the acute and transient increase of plasma norepinephrinelevels during the first hours following thrombolysis. In animal models, the effectof captopril by blunting the transient release of catecholamines upon reperfusionwas also associated with a reduction in myocardial damage.6 In the presentstudy the transiently lower norepinephrine plasma levels in the captopril groupwas paralleled by a similar trend towards a reduction of enzymatic infarct size.Peak α-HBDH levels and, more importantly, the calculated cumulative α-HBDHrelease over 72 hours was reduced in the captopril patients, especially in pa-tients with larger infarcts. The reduction in enzymatic infarct size was not asso-ciated with a significant improvement in left ventricular function, as discussed

CHAPTER 7

156

before. The clinical meaning of a lower incidence of nonsustained VT andAIVR in the captopril group is unclear. However, this finding is in concordancewith observations on ventricular arrhythmias and norepinephrine release in ani-mal ischemia-reperfusion models.6 AIVR has been identified as a clinicalmarker of reperfusion.23 Therefore, reduction of its incidence may reflect at-tenuation of reperfusion injury possibly mediated by a decrease of norepineph-rine levels. Although the observations of transiently reduced norepinephrinelevels, reperfusion arrhythmias and a smaller enzymatic infarct size are onlymodest, all endpoints in this study are in agreement with our experimental ob-servations.6,7

Clinical endpoints

No differences were observed in clinical outcome between the two groups ofpatients, except for the incidence of congestive heart failure. The extent of thisreduction is very comparable to the figures which were reported in SAVE,10

SOLVD,24 CONSENSUS II12 and AIRE.15 However, as mentioned before, thisobservation cannot be explained by reduction of volumes, i.e. attenuation of leftventricular remodeling. Alternatively, the reduced enzymatic infarct size in largeinfarcts may have enhanced the beneficial effects of thrombolysis itself. Rein-farction tends to be more frequent in the captopril group. Nevertheless, themean enzymatic infarct size was smaller in the captopril group, irrespective ofthe larger number of reinfarctions. In contrast, the SAVE study10 showed a re-duction of reinfarction rate, but this occurred during the chronic phase and onlybecame apparent after nine months of follow up.

Limitations of the study

Before any implications for clinical practice can be derived from this study,several limitations should be taken into account. First, the study population ishighly selected since only first anterior myocardial infarctions receiving strep-tokinase were randomized. This represents approximately 10-15% of the totalinfarct population. This selection may have affected the results on the bloodpressure responses, since inferior infarctions are more prone to hypotension.Secondly, the use of echocardiography for quantitative measurements of leftventricular volumes resulted in a loss of one third of the patients for evaluation.Thirdly, it is unknown at present whether captopril may alter the kinetics ofmyocardial enzymes. Although it has been shown that the use of cumulative α-HBDH as a measure of infarct size is independent of the occurrence of reperfu-sion, such an effect may have influenced the present results on infarct size on

THREE-MONTH CATS RESULTS

157

this study. Finally, the study was under powered to detect differences in leftventricular end-diastolic volume index less than 4 ml/m2. A study population ofat least 700 patients would have been needed to establish whether the presentlyfound differences are due to a true treatment effect.

Implications for clinical practice and future research

The design of the CATS study focused on intervention additional to throm-bolysis, as contrasted to most other studies, in which ACE inhibitors were stud-ied for secondary prevention after myocardial infarction. Several effects ob-served in this study are in favor of early use of captopril in acute myocardial in-farction, although one could argue that there appears to be a trend towards moreserious adverse clinical events in the captopril group, with respect to the com-bined incidence of reinfarction and death. However, the numbers in this studyare too small to allow for any conclusion. The outcome of the recently presentedISIS-413 and GISSI-314 studies indicate a modest favorable effect on mortalityduring the first weeks after infarction. Based on the outcome of these studies,there is no need to avoid early use of captopril for safety reasons. However, theresults from CONSENSUS II12 suggest that the early start of therapy with intra-venous enalapril does not actually improve survival. In CONSENSUS II12 ACEinhibition was used early, but not as an adjunct to thrombolysis. Furthermore,the drug was administered after a mean delay of 18 hours after admission. It isconceivable that the design of this study with a follow-up of only 180 days mayhave precluded the observation of a possible beneficial effect on mortality.

In conclusion, the CATS study provides important data on the effects of oralcaptopril administration of acute anterior myocardial infarction starting duringthrombolysis. Despite the outcome of the recently published large scale trialsseveral questions remain to be answered, as to whether earlier use of ACE in-hibitors, especially in association with thrombolysis, will improve the long termclinical outcome. Do pharmacological differences between the ACE inhibitorsused in these studies play a role? Which dose should be used and finally whichtype of patients will benefit most from ACE inhibition after myocardial infarc-tion?

CHAPTER 7

158

References

1. ISIS-3: a randomised comparison of streptokinase vs tissue plasminogen activatorvd anistreplase and of aspirin plus heparin vd aspirin alone among 41.299 cases ofsuspected acute myocardial infarction. Lancet 1992;339:753 -770.

2. Warren SE, Royal HD, Markis JE, Grossman W, McKay RG: Time course of leftventricular dilation after myocardial infarction: influence of infarct -related arteryand success of coronary thrombolysis. J Am Coll Cardiol 1988;11:12-19.

3. GISSI: Effectiveness of intravenous thrombolytic treatment in acute myocardial i n-farction. Lancet 1986;1:397-402.

4. ISIS-2: Randomised trial of intravenous streptokinase, ora l aspirin, both or neitheramong 17.187 cases of suspected acute myocardial infarction. Lancet1988;2:349-360.

5. Manning AS, Hearse DJ: Reperfusion -induced arrhythmias: mechanisms and pr e-vention. J Mol Cell Cardiol 1984;16:497-518.

6. van Gilst WH, de Graeff PA, Kingma JH, Wesseling H, de Langen CDJ: Captoprilreduces purine loss and reperfusion arrhythmias in the rat heart after coronary a r-tery occlusion . Eur J Pharmacol 1984;100:113 -117.

7. de Graeff PA, van Gilst WH, Kingma JH, de Langen CDJ, Wess eling H: Effects ofcaptopril in a closed -chest pig model against ischemia -reperfusion injury. NaunSchmied Arch Pharmacol 1986;280:181 -193.

8. Sharpe N, Smith H, Murphy J, Greaves S, Hart H, Gamble G: Early prevention ofleft ventricular dysfunction after myocardial infarction with angiotensin-converting-enzyme inhib ition. Lancet 1991;337:872-876.

9. Pfeffer MA, Lamas GA, Vaughan DE, Parisi AF, Braunwald E: Effect of captoprilon progressive ventricular dilatation after anterior myocardial infarction. N Engl JMed 1988;319:80-86.

10. Pfeffer MA, Braunwald E, Moyé LA, Basta L, Brown EJ, Cuddy TE et al: Effect ofcaptopril on mortality and morbidity in patients with left ventricular dysfunctionafter myocardial i nfarction. N Engl J Med 1992;327:669 -677.

11. Oldroyd KG, Pye MP, Ray SG: Effects of early captopril administration on infarctexpansion, left ventricular remodelling and exercise capacity after acute myocardialinfarction . Am J Cardiol 1991;68:713-718.

12. Swedberg K, Kjekshus J, Rasmussen K, R yden L, Wedel H, CONSENSUS II StudyGroup: Effects of early administration of enalapril on mortality in patients withacute myocardial i nfarction. N Engl J Med 1992;327:678 -684.

13. ISIS-4: A randomised factorial trial assessing early oral captopril, oral mononitrate,and intravenous magnesium sulphate in 58 050 patients with suspected acutemyocardial infarction. Lancet 1995;345:669-685.

14. GISSI-3: Effects of lisinopril and transdermal glyceryl trinitrate singly and togetheron 6-week mortality and ventricular function after acute myocardial infarction.Gruppo Italiano per lo Studio della Sopravvivenza nell'infarto Miocardico. Lancet1994;343:1115-1122.

THREE-MONTH CATS RESULTS

159

15. The Acute Infarction Ramipril Efficacy (AIRE) Study Investigators: Effect oframipril on mortality and morbidity of survivors of acute myocardial infarctionwith clinical evidence of heart failure. Lancet 1993;342:821-828.

16. van der Laarse A, Kerkhof PLM, Vermeer F, et al: Relation between infarct size andleft ventricular performance assessed in patients with first acute myocardial infar c-tion randomised to intracoronary thrombolytic therapy or to conventional trea t-ment. Am J Cardiol 1988;61:1-7.

17. Kingma JH, van Gilst WH, Peels CH, for the CATS Investigators Group: Angi o-tensin-converting enzyme inhibition during thrombolytic therapy in acute myoca r-dial infarction: the captopril and thrombolysis study. J Cardiovasc Pharmacol1992;19(Suppl. 4):S18 -24.

18. Nabel EG, Topol EJ, Galeana A, et al: A randomised placebo -controlled trial ofcombined early intravenous captopril and recombinant tissue -type plasminogen a c-tivator therapy in acute myocardial infarction. J Am Coll Cardiol 1991;17:467-473.

19. Lew AS, Laramee P, Cercek B, Shah PK, Ganz W: The hypotensive effect of intr a-venous streptokinase in patients with acute myocardial infarction. Circ1985;72:1321-1326.

20. Green J, Dupe RJ, Smith RAG, Harris GS, English PD: Comparison of the hypote n-sive effects of streptokinase -(human) plasmin activator complex and BRL 26921(p-anisoylated streptokinase -plasminogen activator complex) in the dog after highdose, bolus administr ation. Thromb Res 1984;36:29-36.

21. Jeremy RW, Hackworthy RA, Bautovich G, Hutton BF, Harris PJ: Infarct arteryperfusion and changes in left ventricular volume in the month after acute myoca r-dial infarction . J Am Coll Cardiol 1987;9:989-995.

22. Clough DP, Collis MG, Conway J, Hatton R, Keddie JR: Interaction of angi o-tensin-converting enzyme inhibitors with the function of the sympathetic nervoussystem. Am J Cardiol 1982;49:1410 -1414.

23. Gorgels AP, Vos MA, Letsch IS, et al: Usefulness of the accelerated idioventricularrhythm as a marker for myocardial necrosis and reperfusion during thrombolytictherapy in acute my ocardial infarction. Am J Cardiol 1988;61:231-235.

24. SOLVD Investigators: Effect of enalapril on survival in patients with reduced leftventricular ejection fractions and congestive heart failure. N Engl J Med1991;325:293-302.

CHAPTER 8

Which patient benefits from early angiotensin-convertingenzyme inhibition after myocardial infarction?

Results of one-year serial echocardiographical follow-up from theCaptopril And Thrombolysis Study (CATS)

Wiek H van Gilst,1,2 PhD; J Herre Kingma,1,3 MD; Kathinka H Peels,4 MD;Jan-Henk E Dambrink,3 MD; Martin St John-Sutton,5 MD

1 Department of Clinical Pharmacology, University of Groningen2 Thoraxcenter Groningen, Academic Hospital, Groningen3 Department of Cardiology, St. Antonius Hospital, Nieuwegein4 Department of Cardiology, Catharina Hospital, Eindhoven5 Cardiovascular Division, University of Pennsylvania, Philadelphia

Published in:Circulation 1995;92:I-24 (abstract)Submitted

ONE-YEAR CATS RESULTS

163

Abstract

Background. Remodeling of the heart starts in the acute phase of myocardialinfarction and is associated with adverse prognosis. Angiotensin-converting en-zyme inhibition, started in the subacute or chronic phase after myocardial in-farction improves prognosis. We investigated the effect of intervention withcaptopril within six hours on left ventricular volume and clinical symptoms ofheart failure in relation to actual infarct size.Methods and results. In CATS, 298 patients with a first anterior myocardialinfarction treated with IV streptokinase were randomized to either oralcaptopril (25 mg TID) or placebo. Left ventricular volume index was assessedby two dimensional echocardiography within 24 hours, at day 3, 10 and 90, andafter one year. A small but significant increase in left ventricular volume indexeswas observed after 12 months. Using a random coefficient model, no significanttreatment-effect on left ventricular volumes could be detected. In contrast, whensurvival models were used, the occurrence of left ventricular dilatation wassignificantly lower in captopril treated patients (p=0.018). In addition, theincidence of heart failure was lower in the captopril treated group (p < 0.03).This effect appeared early and was most obvious in patients with a medium sizeinfarct (p=0.04) and was not present in large infarcts.Conclusion. Very early treatment with captopril after myocardial infarctionsignificantly reduces the occurrence of early dilatation and the progression toheart failure. These data underscore the importance of early treatment. Fur-thermore, patients with intermediate infarct size benefit the most from thistreatment strategy.

Introduction

Left ventricular dilatation after myocardial infarction decreases prognosis.1,2

Initially, left ventricular dilatation is determined by infarct size and early infarctexpansion and in a later phase it is determined by increased wall stress and sub-sequent global ventricular enlargement.3 Therefore, the stage for left ventriculardilatation is already set in the acute phase of myocardial infarction. Conse-quently, infarct size limitation by rapid restoration of coronary blood flow to thearea at risk is considered the most effective strategy to prevent dilatation andsubsequent morbidity and mortality.4-7 Whether additional pharmacological in-tervention may further prevent the occurrence of dilatation and subsequentmorbidity following thrombolysis is not clear yet.

CHAPTER 8

164

Angiotensin-converting enzyme (ACE) inhibition, started within one to 14days after myocardial infarction, improves prognosis.8-12 It has also been dem-onstrated that long term use of captopril in patients with asymptomatic left ven-tricular dysfunction after myocardial infarction with or without thrombolysis re-duces mortality, the incidence of heart failure and the rate of reinfarction.8 Whencaptopril was given within 24 hours after the onset of symptoms a favorable ef-fect on left ventricular remodeling was demonstrated in a small selected groupof patients without thrombolysis.9 A modest but significant effect on mortality ofcaptopril10 and lisinopril11 was observed when treatment was started within thefirst 24 hours after myocardial infarction. The reduction of mortality becameeven more apparent in patients with clinical evidence of transient or ongoingheart failure if oral ramipril was administered between the second and ninth dayafter myocardial infarction.12

We hypothesized that immediate concomitant administration of captoprilwould further improve the beneficial effects of thrombolysis in the acute phaseof myocardial infarction and that its continued use would result in preservationof left ventricular structure and function in the chronic phase. To test this hy-pothesis we designed the Captopril And Thrombolysis Study (CATS). The ef-fect on ventricular arrhythmias, neurohormones and clinical events during thefirst three months of follow up for the two treatment groups is reported earlier.13

The aim of the present study was to evaluate the time course of ventriculardilatation as assessed by serial echocardiography during one year of follow upand its relation to 1) the effect of double-blind treatment with captopril, 2) pa-tient characteristics and infarct size at baseline, and 3) clinical outcome.

Methods

Organization. The study organisation has been reported earlier.13 In brief,CATS was a double-blind, randomized, placebo-controlled study in patients

Table 8.1. Reference values (ml/m2) used for the definition of the occurrence of leftventricular dilatation (see Methods)

Volume index Mean SD*

Left ventricular end-systolic 21.3 8.4Left ventricular end-diastolic 51.7 11.3

* SD indicates standard deviation .

ONE-YEAR CATS RESULTS

165

with a first anterior wall myocardial infarction treated with intravenousstreptokinase. Patients were enrolled in 12 hospitals in the Netherlands.Evaluation of all echocardiographic measurements and determination ofenzymatic infarct size was performed in central core laboratories. Anindependent Data Quality Committee reviewed the data on clinical endpoints.The progress and conduct of the study with emphasis on the safety of patients,were supervised by an international Policy Advisory Board. The study wasapproved by the Institutional Review Board of all participating hospitals (seeAppendix).

Objectives of the study. The primary and prespecified objective of the presentstudy was the assessment of the time-dependent effect of captopril relative toplacebo on the preservation of left ventricular volume as measured by serial 2D-echocardiography at 1, 3 and 10 days and at 3 and 12 months after myocardialinfarction. In addition, post hoc analyses (closely parallel to the primary objec-tive) of the relationship between volume changes, patient characteristics, infarctsize and/or clinical outcome were evaluated.

Eligibility of patients. Patients were considered eligible for enrollment if a firstanterior wall myocardial infarction was present, and if they were treated withthrombolytic therapy (1.5 million international units intravenous streptokinaseadministered in 30 minutes) within 6 hours after the onset of symptoms. In-formed consent was obtained by witnessed oral consent, later confirmed bywritten consent following the acute phase of myocardial infarction. Diagnosiswas based on the presence of characteristic symptoms of acute anterior myo-cardial infarction, with at least 1 mm ST-segment elevation in lead I and aVLand/or 2 mm ST elevation in two precordial leads of the 12-lead electrocardio-gram compatible with infarction of the anterior wall and adjacent areas, includ-ing septal and lateral portions of the left ventricle. Patients were excluded ifsystolic blood pressure was over 200 mmHg or below 100 mmHg, and if dia-stolic blood pressure was over 120 mmHg or below 55 mmHg. Additional ex-clusion criteria were reported earlier.13

Treatment protocol. Immediately after admission to the coronary care unit,1,500,000 international units of streptokinase was administered in 30 minutes,by continuous intravenous infusion. Double-blind medication was initiated im-mediately upon completion of the streptokinase infusion, provided the systolicblood pressure was stable and above or equal to 100 mmHg. If systolic bloodpressure was below 100 mmHg, the initiation of double-blind medication couldbe postponed for a maximum of one hour following the start of the streptok-inase infusion.

CHAPTER 8

166

Figure 8.1. Mean left ventricular end-systolic (panel A) and end-diastolic volume in-dex (panel B) during the one-year follow-up for the captopril treated group comparedto the placebo group.

0 90 180 270 360

22

24

26

28

30

32

34

Days after infarction

LVESVI (ml/m2)

Placebo

Captopril

0 90 180 270 360

52

54

56

58

60

62

64

66

Days after infarction

LVEDVI (ml/m2)

Placebo

Captopril

ONE-YEAR CATS RESULTS

167

Initially, an oral dose of 6.25 mg was given, repeated after 4 and 8 hours, and at16 and 24 hours doses of 12.5 mg and 25 mg, respectively, were administered.Dose titration was continued provided the systolic blood pressure measuredimmediately prior to the next scheduled dose of study medication, was above orequal to 95 mmHg. If the systolic blood pressure was below 95 mmHg, thestudy medication was withheld until the next dosing time. The target mainte-nance dose of the double-blind study medication was 25 mg TID, administeredfrom day 2 until 12 months after myocardial infarction. During the 12 months ofdouble-blind therapy, concomitant therapy with calcium antagonists, beta-blockers or nitrates was instituted only for specific indications, e.g. angina andhypertension. However, the protocol did not prohibit the use of beta-blockersfor secondary prevention. Also the use of aspirin was at the discretion of the lo-cal investigator with a recommended dose of 80 mg.

Table 8.2. Baseline characteristics of the two treatment groups

Characteristics Placebo CaptoprilN=149 N=149

Age (years) 60 ± 9 59 ± 10Male (%) 80 70Clinical history (%) ischemic heart disease hypertension diabetes mellitus current smoker

8.016.19.457.7

9.427.59.467.1

Killip class (%) class I class II

7525

7624

Concomitant medication* (%) beta-blockers calcium antagonists diuretics nitrates

16.114.810.710.7

22.212.812.89.4

Hemodynamics*

blood pressure (mmHg) systolic diastolic

125 ± 1977 ± 12

128 ± 21 77 ± 13

heart rate (beats/min) 81 ± 14 83 ± 16

* as assessed at admission.

CHAPTER 8

168

Echocardiography. For echocardiographic measurement of left ventricularend-diastolic and end-systolic volumes, apical four- and two-chamber viewswere obtained and recordings made with respiration held at end-expiration orpartial inspiration. Patient angulations, transducer position and respiratory phasewere recorded. End-diastolic and end-systolic frames were outlined from bothapical views using the Dataview Microsonics cardiac analysis system (NovaMicrosonics). Left ventricular volumes were calculated using the biplane(modified) Simpson’s rule. The mean of two measurements on consecutive cy-cles was taken for each examination. Left ventricular end-diastolic volume in-dex and left ventricular end-systolic volume index were derived from body sur-face area, which was estimated at each time-point. All initial echocardiographicdetermination of left ventricular volume were performed within 12-24 hours af-ter admission and at least 4 hours after the last dose of double-blind study medi-cation. All subsequent echocardiographic determinations obtained at day 3,prior to discharge and at three and 12 months were made no earlier than 8 hourspost-dose of double-blind study medication. To insure the return to the sameechocardiographic view, a method of locating the transducer in reference tobony landmarks was applied. Lateral rotation and elevation of the upper bodywere noted and maintained at subsequent studies.

Definition of left ventricular dilatation. Left ventricular volumes of all patientswith small enzymatic infarct size, i.e. a cumulative α-hydroxybutyrate dehydro-genase release (α-HBDH) less than 730 U/l, which represents the lower tertile ofall randomized infarct sizes, were pooled. This group of patients showed nochange between first and last echocardiographic evaluation. Mean left ventricu-lar end-diastolic and end-systolic volume index and corresponding standarddeviations of this reference group were used as the reference value for ourdefinition of left ventricular dilatation in the CATS study population (Table 8.1).Subsequently, all individual end-systolic and end-diastolic volume indexes ateach time point were compared with this reference value. A ventricle was con-sidered dilated if the following conditions were fulfilled:1) Left ventricular volume index (either end-systolic or end-diastolic) had to be

one standard deviation above the reference value. Thus, either left ventricularend-systolic volume index had to be above 29.7 ml/m2 or left ventricular end-diastolic volume index had to be above 63 ml/m2.

2) The sum of end-systolic and end-diastolic volume index had to be 1.5 stan-dard deviation above the sum of the respective reference values. Thus, thesum of left ventricular end-systolic and end-diastolic volume index should beabove 87.8 ml/m2.

ONE-YEAR CATS RESULTS

169

This dilatation criterion appeared to be robust since variation of the criterionbetween 0.5 and 1.5 standard deviation generated comparable results. The mo-ment of dilatation was assumed as halfway between the visit showing dilatationand the previous visit without dilatation.

Enzymatic infarct size. Infarct size was expressed as the cumulative release ofα-HBDH activity per liter of plasma within the first 72 hours. This was calcu-lated from serial plasma α-HBDH determinations from blood samples takentwice daily during the first five days, as described by van der Laarse et al.14

These investigators demonstrated that this method was independent of the resultof thrombolysis. Infarct size was considered small if enzyme release was lessthan 730 U/l, medium if enzyme release was between 730 and 1460 U/l andlarge if enzyme release was more than 1460 U/l. Cut-off points were derivedfrom tertiles of all randomized infarct sizes.

0 90 180 270 360

0,3

0,4

0,5

0,6

0,7

0,8

0,9

1,0

Captopril

Placebo

p=0.0182

Probability

Days after myocardial infarction

Figure 8.2. Survival curves representing the absence of dilatation during the one-yearfollow-up for the captopril treated group compared to the placebo group.

CHAPTER 8

170

Definition of heart failure. The presence of heart failure was based on clinicaljudgement of the principal investigators. All patients in whom diuretics and/ordigoxin was started, in whom open treatment with an ACE inhibitor was started,or in whom hospitalization was prolonged or who had to be rehospitalized dueto the presence of symptoms of heart failure, were considered to have heart fail-ure. Hospitalization was considered prolonged if the duration was above thenormal hospital stay for myocardial infarction which was on average nine days.Furthermore, the reason for this longer hospitalization was explicitly recorded inthe Case Record Form. All patients with prolonged hospitalization due to signsand symptoms of heart failure were evaluated by the Data Quality Committee.All other causes for prolonged hospitalization were excluded. An independentand blinded Data Quality Committee reviewed the data on clinical endpoints.

Statistics. One of the problems with standard multivariate approach to re-peated measures is that the data must be complete. Each subject must be meas-ured at every occasion to be included in the analysis. If there is a missing valuefor one of the repeated measures, then all the data for that subject are ignored.Therefore, a random coefficient model was used to analyse echocardiographicdata for the following reasons: 1) one or more echocardiograms are not evalu-able for a substantial group of patients due to a lack in echogeneity, 2) left ven-tricular end-systolic and end-diastolic volume indexes vary substantially be-

Table 8.3. Left ventricular dimensions for each treatment group and at each time-point during the one-year follow-up

Time Placebo Captopril

days ± SD LVESVI (ml/m2) LVESVI (ml/m2)0.3 ± 0.4 25.7 ± 9.5 25.7 ± 10.82.2 ± 0.48.7 ± 3.998.3 ± 16

369.1 ± 20.2

days ± SD0.3 ± 0.42.2 ± 0.48.7 ± 3.998.3 ± 16

26.2 ± 9.828.7 ± 11.330.0 ± 15.129.5 ± 14.8

LVEDVI (ml/m2)56.4 ± 13.357.2 ± 13.260.4 ± 14.462.7 ± 17.4

24.6 ± 9.126.6 ± 11.928.9 ± 15.730.0 ± 16.1

LVEDVI (ml/m2)56.1 ± 14.254.0 ± 11.859.1 ± 15.661.9 ± 19.4

369.1 ± 20.2 62.8 ± 17.6 61.4 ± 21.1

LVEDVI indicates left ventricular end-diastolic volume index; LVESVI, left ventricularend-systolic volume index; SD, standard deviation.

ONE-YEAR CATS RESULTS

171

tween patients at entry, and 3) echocardiograms are not obtained at exactly thesame points in time. The random coefficient model allows a comparison be-tween the two treatment groups using all echocardiographic data, accounting fordifferences in left ventricular volume index between patients at entry (randomintercept) and for differences in time in response to treatment within patients(random slope). P-values are approximated by a Chi-square distribution, basedon model differences in twice the negative log-likelihood.15 The occurrence ofdilatation according to our definition and the incidence of heart failure wasanalysed using survival techniques evaluated by the Log-Rank test. If not oth-erwise indicated, continuous variables other than left ventricular volumes werecompared using Student’s t-test and categorical variables using Fisher’s Exacttest (2-tail) or Chi-square test. Results were considered statistically significant ifthe p-values were less than 0.05, using the two-sided level of significance. Thestatistical software of ML3, version 2, was used to analyse the random coeffi-cient model. The remaining statistical analyses were performed with SAS, ver-sion 6.08. According to the intention-to-treat principle all outcome analyseswere based on the treatment group to which the patients had been randomly as-signed.

Results

During the enrollment period of CATS, 298 patients were included with 149patients allocated to each treatment group. The time from onset of symptoms tothe time of streptokinase infusion was 166 ± 70 minutes in the placebo groupand 163 ± 76 minutes in the captopril group. The time of the first dose of studymedication from onset of symptoms was 213 ± 76 minutes in the placebo groupand 212 ± 86 minutes in the captopril group, respectively. A complete clinicalfollow-up over the 12-month period was obtained in 245 patients (82.2%).Twenty three patients (8%) died (Table 8.7.) and 30 patients (13 patients in theplacebo group and 17 patients in the captopril group) were withdrawn from thestudy during the one-year follow-up. Twenty-four hours after randomization,the target dose of 25 mg TID was reached in 95% of patients (94% for the pla-cebo group and 95% for the captopril group). At discharge, 80% of all random-ized patients were still on study medication. At the end of the 12-month follow-up period, 89% of patients with complete clinical follow-up were still on studymedication (92% for the placebo group and 85% for the captopril group).

CHAPTER 8

172

0 90 180 270 360

20

25

30

35

40

45

Days after myocardial infarction

Large infarcts

Medium infarcts

Small infarcts

LVESVI (ml/m2)

0 90 180 270 360

45

50

55

60

65

70

75

80

Days after myocardial infarction

LVEDVI (ml/m2)

Large infarcts

Medium infarcts

Small infarcts

Figure 8.3. Filled dots represent mean (+ SEM) left ventricular end-systolic (panelA) and end-diastolic volume index (panel B) of the total study population as meas-ured during the one-year follow-up for small (bottom), medium (middle) and large(top) infarct sizes. Lines are calculated by the ML3 statistical model using the pa-rameters of Table 8.5.

ONE-YEAR CATS RESULTS

173

The clinical characteristics of the two groups were comparable at baseline(Table 8.2), although smoking and hypertension tended to be more frequent inthe captopril group.

Echocardiographic measurements of left ventricular volumes

Appropriate measurement of left ventricular volume index on at least onetime-point was available in 263 (88.3%) of the randomized patients. Character-istics of patients with available echocardiograms and calculated infarct sizes,which were used in the random coefficient model, in comparison to the remain-ing randomized population is presented in Table 8.4. Most important differencebetween the groups is the distribution of gender.

Table 8.4. Baseline characteristics of the patients used in the random coefficientmodel compared to the patients not used in the model

Characteristics Used in analysis Not used in analysisn = 263 n = 35

Age (years) 59 ± 9 62 ± 10 Male (%) 78 57*

Clinical history (%) ischemic heart disease hypertension diabetes mellitus current smoker

9.122.18.862.7

5.720.014.360.0

Killip class (%) class I class II

7723

6040

Concomitant medication (%) beta-blockers calcium antagonists diuretics nitrates

20.513.711

10.3

8.614.317.18.6

Blood pressure (mmHg) systolic diastolicHeart rate (beats/min)

127 ± 20 77 ± 13 82 ± 15

124 ± 18 77 ± 11 84 ± 13

Infarct size (α-HBDH Q72, U/l) 1121 ± 187 1176 ± 296

* p=0.012 Fischer’s Exact test (2-tailed)

CHAPTER 8

174

Left ventricular end-systolic and end-diastolic volume indexes during theone-year follow-up period for the two treatment groups are depicted in Figure8.1 and presented in Table 8.3. The total study population showed a small butsignificant increase in both left ventricular end-diastolic and end-systolic vol-ume index of 6.7 ml/m2 (95% CI 3.4 - 9.9) and 4.8 ml/m2 (95% CI 2.2 - 7.3)over the 12-month follow-up period. However, analysis with the random coef-ficient model revealed no significant effect of treatment with captopril onchanges in ventricular volume over the first 12 months after myocardial infarc-tion when compared to placebo. Factors which significantly influence left ven-tricular volume index are indicated in Table 8.5. As may be expected infarctsize and time were among the most powerful determinants of ventricular en-largement. The time factor is nonlinear and the largest changes occur in theacute phase. Additional factors associated with left ventricular volume indexesare different for systolic and diastolic volumes (Table 8.5).

Days after myocardial infarction0 90 180 270 360

0,0

0,2

0,4

0,6

0,8

1,0

Large

Medium

Small

Captopril

Placebo

Captopril

Captopril

Placebo

Placebo

p=0.27

p=0.04

p=0.74

Probability

Figure 8.4. Survival curves representing the absence of dilatation during the one-yearfollow-up for the captopril treated group compared to the placebo group and for dif-ferent infarct sizes.

ONE-YEAR CATS RESULTS

175

Occurrence of left ventricular dilatation

If, at any point in time, dilatation according to our definition occurred, thepatient remained dilated during the following echocardiographic evaluations.This enabled us to use survival techniques to evaluate treatment effects on theoccurrence of left ventricular dilatation. Figure 8.2 shows that a large percent-age of left ventricular dilatation already occurs in the in-hospital phase of myo-cardial infarction. Furthermore, during this period a significant effect of capto-pril treatment on the incidence of dilatation is already apparent, which is main-tained during the one-year follow-up.

Table 8.5. Factors significantly determining left ven tricular dimensions according tothe random coefficient model

LVESVI LVEDVI

Variable Coefficient p-value Coefficient p-value

Constant 18.4 <0.00001 47.0 <0.00001Time in days -0.19 <0.00001 0.18 <0.00001Infarct size medium large

3.09.2

0.02<0.00001

3.59.8

0.06<0.00001

T x I medium large

0.92.2

0.009<0.00001

1.02.9

0.03<0.00001

Male 3.2 0.013 5.0 0.003CHF 2.4 0.01 - NSCABG - NS -6.99 0.008Open label ACE 5.5 0.004 - NSUse of nitrates - NS 3.3 0.003Study medication -0.11 0.92 -0.44 0.76

ACE indicates angiotensin-converting enzyme; CABG, coronary artery bypass grafting;CHF, congestive heart failure; LVEDVI, left ventricular end-diastolic volume index;LVESVI, left ventricular end-systolic volume index; T x I, interaction term of time and i n-farct size.

CHAPTER 8

176

Enzymatic infarct size and echocardiography

Serum creatine kinase and α-hydroxybutyrate dehydrogenase activity weredetermined serially in 86.9% of patients. A wide range of infarct sizes was pres-ent in this population as can be seen in Table 8.6. Figure 8.3 presents thechange in left ventricular end-diastolic and end-systolic volumes as measuredover the 12 month follow-up period for small, intermediate and large infarcts.The change in left ventricular volume index for the different strata is predictedby the statistical model as indicated by the solid lines (Figure 8.3). The effect oftreatment with captopril in the three different strata on the occurrence of dilata-tion is indicated in Figure 8.4. The occurrence of dilatation increased with in-creasing infarct size. Captopril reduced the occurrence of dilatation in small in-farcts and significantly in medium infarcts. However, dilatation was not pre-vented by captopril in the large infarct group (Figure 8.4).

Days after myocardial infarction

0 90 180 270 360

0,65

0,70

0,75

0,80

0,85

0,90

0,95

1,00

Captopril

Placebo

p=0.037

Probability

Figure 8.5. Survival curves representing the absence of heart failure duringthe one-year follow-up for the captopril treated group compared to the placebogroup.

ONE-YEAR CATS RESULTS

177

Clinical events

Incidence of clinical events are summarized in Table 8.7. Mortality rate waslow (8%) with no significant difference between the two treatment groups. Theincidence of coronary revascularization procedures tended to be lower in thecaptopril group. Reinfarction occurred in 21 patients, seven of which within 72hours, in four cases related to PTCA.

Interestingly, the incidence of heart failure showed the same pattern as theoccurrence of ventricular dilatation during the one-year follow-up (Figure 8.5).Furthermore, a comparable, significant treatment effect was present (p=0.036).This treatment effect on progression to heart failure was, comparable to the oc-currence of dilatation, confined to the patients with medium size infarcts (Table8.6).

Table 8.6. Infarct size, occurrence of dilatation, left ventricular dimensions after oneyear and heart failure for patients with small medium and large infarcts inboth treatment groups

Placebo Captopril

Infarct sizeN

α-HBDH(SD)

N

Dilatation (%)LVESVILVEDVI

CHF(%)

α-HBDH(SD)

N

Dilatation (%)LVESVILVEDVI

CHF(%)

Small

87

414(199)

40

3020.7 ± 8.2

52.3 ± 13.5

25 383(211)

47

1919.5 ± 9.0

47.6 ± 11.9

21

Medium

88

1112(202)

43

6328.8 ± 14.961.8 ± 16.5

36 1087(213)

45

40@27.0 ± 9.9

58.8 ± 15.3

18$

Large 2515(1035)

8739.5 ± 14.1

45 2327(674)

8546.3 ± 15.7

43

88 47 74.5 ± 16.5 41 81.3 ± 20.4

CHF indicates congestive heart failure; LVEDVI, left ventricular end-diastolic volume i n-dex; LVESVI, left ventricular end-systolic volume index. Infarct size was considered smallif α-HBDH < 730 U/l, medium if α-HBDH was 730 - 1460 U/l, and large if α-HBDH >1460 U/l. @ indicates significant reduction when compared to placebo, Fisher’s Exact testp=0.036; $, borderline significant when co mpared to placebo, Fischer’s Exact test p=0.067.

CHAPTER 8

178

Discussion

The present data demonstrate that in our group of patients with a wide rangeof infarct sizes left ventricular end-systolic or end-diastolic volumes increasedonly slightly if all patients were taken together. Furthermore, after 12 months nosignificant effects of treatment on left ventricular volume indexes were foundbetween the total randomized groups. When patients were characterized in termsof the presence or absence of left ventricular dilatation, a significant effect oftreatment was observed. This effect on the incidence of dilatation coincided witha significant effect on the progression to overt heart failure. Treatment effectsappeared to be largest in the patients with intermediate infarcts. The latter con-firms in a clinical setting the early observations by Pfeffer et al..22 They showedin a rat infarct model that the largest treatment effect of captopril occurred in thegroup with medium size infarcts after long-term follow up.

Quantitative 2D-echocardiography has provided important insights into themechanism of left ventricular enlargement and modulation of this enlargementby ACE inhibitors.16 However, most of these studies used patients selected inthe subacute phase9 or were part of larger intervention trials using clinical end-points.17 The design of the present mechanistic study differs in many importantaspects from most of the echocardiographic postinfarction intervention trialswhich evaluated ACE inhibition and have been reported so far. First, patients

Table 8.7. Clinical endpoints

Placebo CaptoprilN=149 N=149

Mortality 10 13Heart failure 53 39PTCA 42 34CABG 11 12Reinfarction 6 15Any of the above 92 86

CABG indicates coronary artery bypass grafting, PTCA, percutaneous transluminal cor o-naryangioplasty. Figures re present numbers of patients.

ONE-YEAR CATS RESULTS

179

were not selected on the basis of their cardiac function. All consecutive first an-terior myocardial infarctions treated with streptokinase were included. There-fore, the CATS population comprised both small and large infarcts and thus,patients with low and high risk of subsequent dilatation and clinical detoriation.This enabled us to evaluate treatment effects in different risk strata. Secondly, inthe present study, captopril was administered early within six hours after onsetof symptoms concomitantly with thrombolytic therapy. Left ventricular en-largement has been shown to start very early after onset of myocardial infarc-tion. Within this early phase stretching of the infarct zone due to slippage of thenecrotic myofibrils occurs.3 Experimental studies have shown that ACE inhibi-tion may improve survival of the ischemic myocyte.18,19 To achieve its benefi-cial effect it is critical that treatment starts very early and that the drug is able toreach the myocytes at risk. Within the clinical setting, this situation may best becreated if treatment is initiated within the thrombolytic window. The early phaseof left ventricular dilatation is a regional phenomenon located predominantly inthe infarcted zone. This is followed by a second phase of dilatation, which in-volves the entire ventricle, and may continue for months after infarction. Thelatter process consists, besides dilatation, of hypertrophy of the noninfarctedzone in response to the loss of contractile elements and increased wall stress. Toevaluate and compare treatment effects in these two periods, the present studyused serial echocardiography with observations in the early phase (day 1, 3 andat discharge from hospital) and in the late phase (3 and 12 months).

What explains the discrepancy between the effect of captopril on occurrenceof dilatation and the lack of effect of captopril on the average left ventricularvolume index? The consecutive enrollment of patients in the CATS studyyielded an broad range of infarct sizes from only minor damage to very large in-farctions. The random effect model revealed that infarct size is the most power-ful determinant of volume expansion with the highest incidence (86%) of dila-tation occurring in the group with large infarcts, and least ventricular enlarge-ment (25%) for the small infarcts. Furthermore, the distribution of left ventricu-lar volumes within each group shows a broad range which is progressively in-creasing towards high values in patients with large infarcts. If the detrimentalprocess of dilatation occurs it continues in the chronic phase as a gradual proc-ess of ongoing dilatation. The difference in average extent of dilatation in thetwo treatment groups combined with the large range of ventricular volumes mayhave been too small for a treatment effect on average ventricular volume indexto appear. The study was powered to detect differences in left ventricular end-diastolic volume index greater than 8 ml/m2 between the treatment groups after12 months. A study population of at least 700 patients would have been neededto establish whether the differences found on average ventricular volumes are

CHAPTER 8

180

due to a true treatment effect. Finally, the analysis of the data was performedaccording to the intention to treat principle. At the end of the 12-month follow-up period, 31% of the patients had developed clinical symptoms of heart failureand were treated with open label captopril.

Clinical endpoints

After 12 months, clinical outcome between the two groups of patients wassimilar, except for a significant reduction in the incidence of heart failure(p=0.036). The magnitude of this reduction is comparable to the figures whichwere reported in SAVE,8 SOLVD,20 CONSENSUS II21 and AIRE.12 This obser-vation, as explained before, is not paralleled by a reduction of average left ven-tricular volumes, i.e. attenuation of left ventricular remodeling in the totalpopulation. However, the time course of individual dilatation and effect of cap-topril on this phenomenon shows a comparable pattern. This suggests that cap-topril prevents the progression to heart failure in low and medium risk patientsby preventing the occurrence of dilatation. A statistically significant effect ofcaptopril on progression of dilatation was only demonstrated in patients with amedium size infarct. If dilatation has already occurred progression to sympto-matic heart failure is not significantly affected by treatment as can be observedin the group with large infarcts. Apparently, in this group the effect of captoprilon dilatation and subsequent progression to heart failure is offset by the magni-tude of myocardial damage.

Implications for clinical practice

Before any implications for clinical practice are derived from this study, itshould be taken into account that the study population represents approximately30-40% of the total infarct population (1995), since only first anterior myocar-dial infarctions receiving streptokinase were randomized. The design of theCATS study focused on intervention additional to thrombolysis, as contrasted tomost other studies, in which ACE inhibitors were studied for secondary preven-tion after myocardial infarction. The outcome of the recently published ISIS-410

and GISSI-311 studies indicate a modest favorable effect on mortality during thefirst weeks after infarction if treatment is started within 24 hours. Based on theoutcome of their studies, the investigators advocated the early use of ACE in-hibitors. However, their study revealed no mechanism or insight into which pa-tient will benefit most from treatment. The present study re-emphasizes thatdilatation is a very early process and that the effects of treatment are already ap-parent at this very early stage. Therefore, we suggest that treatment should start

ONE-YEAR CATS RESULTS

181

as early as possible preferably within the thrombolytic window. Based on the re-sults from CONSENSUS II,21 in which an aggressive intravenous treatmentstrategy with enalapril was used, we suggest that early treatment should startwith oral captopril in low dose in patients who are hemodynamically stable, i.e.with a systolic blood pressure above 100 mmHg. If well tolerated, this dose maybe increased more rapidly thereafter. If during hospitalization infarct size ap-pears to be small and left ventricular volume index remains within the normalrange, discontinuation of treatment may be considered at discharge.

In conclusion, CATS provides important echocardiographical data on themechanism of dilatation and the effects of early oral captopril administration inpatients with acute anterior myocardial infarction receiving thrombolysis. De-spite the outcome of the recently presented large scale trials several questionsremained to be answered. The present data show that ventricular enlargement inthis group of patients is mostly an early phenomenon. Treatment with captoprilprevents the progression to dilatation. Especially patients in this study popula-tion with an intermediate infarct size profit from early treatment. Patients withlarge infarction still dilate despite treatment. However, the present study mayhave been underpowered to demonstrate beneficial effects of captopril in large,but also small, infarctions. Prospective studies using prespecified strata with dif-ferent infarct sizes are needed to address this question.

References

1. Hammermeister KE, DeRouen TA, Dodge HT: Variables predictive of survival inpatients with coronary artery disease: selection by univariate and multivariateanalyses from the clinical, electrocardiographic, exercise, arteriographic, andquantitative angiographic evaluations. Circulation 1979; 59: 421-430.

2. White HD, Norris RM, Brown MA, Brandt PWT, Whitlock RML, Wild CJ: Leftventricular end-systolic volume as the major determinant of survival after recoveryfrom myocardial infarction. Circulation 1987; 76: 44-51.

3. Pfeffer MA, Braunwald E: Ventricular remodelling after myocardial infarction: e x-perimental observations and clinical implications. Circulation 1990; 81: 1161-1172.

4. ISIS-3: a randomised comparison of streptokinase vs tissue plasminogen activatorvs anistreplase and of aspirin plus heparin vs aspirin alone among 41.299 cases ofsuspected acute myocardial infarction. Lancet 1992;339:753 -70.

5. Warren SE, Royal HD, Markis JE, Grossman W, McKay RG: Time course of leftventricular dilatation after myocardial infarction: influence of infarct -related arteryand success of cor onary thrombolysis. J Am Coll Cardiol 1988;11:12-9.

6. GISSI: Effectiveness of intravenous thrombolytic treatment in acute myocardial i n-farction. Lancet 1986;1:397-402.

CHAPTER 8

182

7. ISIS-2: Randomised trial of intravenous streptokinase, oral aspirin, both or neitheramong 17.187 cases of suspected acute myocardial infarction. Lancet1988;2:349-60.

8. Pfeffer MA, Braunwald E, Moyé LA, Basta L, Brown EJ, Cuddy TE, Davis BR,Geltman EM, Goldman S, Flaker GC, Klein M, Lamas GA, Packer M, Rouleau J,Rouleau JL, Rutherford J, Wertheimer JH, Hawkins CM, on behalf of the SAVE i n-vestigators: Effect of captopril on mortality and morbidity in patients with left ve n-tricular dysfunction after myocardial i nfarction. N Engl J Med 1992;327:669 -77.

9. Oldroyd KG, Pye MP, Ray SG: Effects of early captopril administration on infarctexpansion, left ventricular remodelling and exercise capacity after acute myocardialinfarction. Am J Cardiol 1991;68:713-8.

10. ISIS-4: A randomised factorial trial assessing early oral captopril, oral mononitrate,and intravenous magnesium sulphate in 58 050 patients with suspected acutemyocardial infarction. Lancet 1995;345:669-685.

11. GISSI-3: Effects of lisinopril and transdermal glyceryl trinitrate singly and togetheron 6-week mortality and ventricular function after acute myocardial infarction.Gruppo Italiano per lo Studio della Sopravvivenza nell'infarto Miocardico. Lancet1994;343:1115-1122.

12. The Acute Infarction Ramipril Efficacy (AIRE) Study Investigators: Effect oframipril on mortality and morbidity of survivors of acute myocardial infarctionwith clinical evidence of heart failure. Lancet 1993;342:821-828.

13. Kingma JH, van Gilst WH, Peels CH, Dambrink JHE, Verheugt FWA, Wielinga RP:Acute intervention with captopril during thrombolysis in patients with first anteriormyocardial infarction. Results from the Captopril and Thrombolysis Study (CATS)Eur Heart J 1994; 15:898-907.

14. van der Laarse A, Kerkhof PLM, Vermeer F, Serruys PW, Hermens WT, VerheugtFWA, Bär FW, Krauss XH, Wall, vd EE, Simoons ML, for the working group onthrombolytic therapy in acute myocardial infarction: Relation between infarct sizeand left ventricular performance assessed in patients with first acute myocardial i n-farction randomised to intracoronary thrombolytic therapy or to conventionaltreatment. Am J Cardiol 1988;61:1-7.

15. Goldstein H: Efficient statistical modelling of longitudinal data. Ann Human Biol1986; 13: 129-141.

16. Sharpe N, Smith H, Murphy J, Greaves S, Hart H, Gamble G: Early prevention ofleft ventricular dysfunction after myocardial infarction with angiotensin-converting-enzyme inhibition. Lancet 1991;337:1174.

17. St John Sutton M, Pfeffer MA, Plappert T, Rouleau JL, Moyé LA, Dagenais GR,Lamas GA, Klein M, Sussex B, Goldmans S, Menapace FJ, Parker JO, Lewis S, Se s-tier F, Gordon DF, McEwan P, Bernstein V, Braunwald E, for the SAVE investig a-tors: Quantitative two-dimensional echocardiographic measurements are majorpredictors of adverse cardiovascular events after acute myocardial infarction. Theprotective effects of captopril. Circulation 1994; 89: 68-75.

18. van Gilst WH, de Graeff PA, Kingma JH, Wesseling H, de Langen CDJ: Captoprilreduces purine loss and reperfusion arrhythmias in the rat heart after coronary a r-tery occlusion. Eur J Pharmacol 1984;100:113 -7.

ONE-YEAR CATS RESULTS

183

19. de Graeff PA, van Gilst WH, Kingma JH, de Langen CDJ, Wesseling H: Effects ofcaptopril in a closed -chest pig model against ischemia -reperfusion injury. NaunSchmied Arch Pharmacol 1986;280:181 -93.

20. SOLVD Investigators: Effect of enalapril on survival in patients with reduced leftventricular ejection fractions and congestive heart failure. N Engl J Med1991;325:293-302.

21. Swedberg K, Kjekshus J, Rasmussen K, Ryden L, Wedel H, CONSENSUS II StudyGroup: Effects of early administration of enalapril on mortality in patients withacute myocardial i nfarction. N Engl J Med 1992;327:678 -84.

22. Pfeffer MA, Pfeffer JM, Steinberg C, Finn P: Survival after an experimental my o-cardial infarction: beneficial effects of long-term therapy with captopril. Circula-tion 1985; 72: 406-412.

CHAPTER 9

Effects of captopril on early and late arrhythmic events inpatients with thrombolytic therapy for a first anterior

myocardial infarction

J-H.E. Dambrink,1 MD, J.H. Kingma,1,2 MD and W.H. van Gilst,2 PhD for theCATS investigators3

1Department of Cardiology, St Antonius Hospital, Nieuwegein2Department of Clinical Pharmacology, University of Groningen, Groningen3see Appendix

CAPTOPRIL AND VENTRICULAR ARRHYTHMIAS

187

Introduction

In recent years, it has become clear that thrombolytic therapy is one of thefew treatments that can significantly reduce the incidence of life-threateningventricular arrhythmias after myocardial infarction. Combined results of a num-ber of large thrombolysis trials show a nearly 15% reduction of in-hospital VF.1-

5 Sofar, attempts to further reduce life-threatening ventricular arrhythmias in thesetting of thrombolytic therapy have not been very successful. Even beta-blockers, agents with powerful (indirect) anti-arrhythmic properties and verysuccessful in the pre-thrombolytic era,6 do not seem to further reduce the inci-dence of early life-threatening ventricular arrhythmias when thrombolytic ther-apy is used.7,8

ACE inhibitors may provide an interesting alternative to conventional anti-arrhythmic therapy. Several animal experiments have shown a reduction ofventricular arrhythmias during reperfusion after ligation of a coronary artery.9-11

Blunting of the neurohumoral response and reduction of infarct size have beensuggested as underlying mechanisms.12-14 Furthermore, ACE inhibitors areknown to modulate left ventricular remodeling in the later phases of myocardialinfarction.15-18 Both left ventricular dilatation19 and ventricular hypertrophy,20

major components of the remodeling process, are well known arrhythmogenicfactors. Thus, modulation of the remodeling process may also result in an indi-rect reduction of late ventricular arrhythmias. In addition, this anti-arrhythmiceffect may be potentiated by modulating effects of ACE inhibitors on electrolyteabnormalities21 and autonomic imbalance12 during the later stages of myocardialinfarction.

In this study, we tested the hypothesis that treatment with the ACE inhibitorcaptopril during thrombolysis results in a reduction of clinically relevant ven-tricular arrhythmias both early and late after myocardial infarction.

Methods

Patients. This study was part of the Captopril And Thrombolysis Study(CATS), in which the effect of captopril treatment, started during thrombolysis,was evaluated in patients with a first anterior myocardial infarction. The patientselection and methods of this study are described elsewhere in detail.22

CHAPTER 9

188

In brief, 298 patients were included in 12 hospitals in The Netherlands. Se-lection criteria included a typical history of chest pain consistent with myocar-dial infarction with onset of symptoms no longer than 6 hours before admission,and electrocardiographic criteria for acute anterior myocardial infarction. Ex-clusion criteria included presence of left bundle branch block and severe heartfailure (Killip class III or IV). Informed consent was obtained by witnessed oralconsent, later confirmed by written informed consent following the acute phaseof myocardial infarction.

Infarct size. Enzymatic infarct size was estimated by cumulative alpha-hydroxybutyrate dehydrogenase values over the first 72 hours after myocardialinfarction (α-HBDH Q72) as described by van der Laarse et al..23 This method isnot influenced by the presence or absence of reperfusion.

Echocardiography. Regional wall motion abnormalities were evaluated usingthe wall motion score recommended by the American Society of Echocardi-ography.24 In this scoring system the left ventricle is divided into 16 segments,

Table 9.1. Early ventricular arrhythmias (within 48 hours)

Nr Age (y) M/F Arrhythmia Treatment / outcome

1 58 M VT Lidocaine23456789101112131415161718

6063646058737270556274546974606347

MMMMMMFMFMMMFFMMM

VTVTVTVTVTVF

VT,VFVTVTVTVFVT

VT,VFVTVTVFVT

LidocaineLidocaineLidocaineLidocaineLidocaineCardioversion, died on day 7 (VT,VF)CardioversionLidocaineLidocaineLidocaineDied, free wall rupture on autopsyLidocaine, died day 212, cardiogenic shockDied, free wall rupture on autopsySotalolLidocaineDuring rescue PTCA, cardioversionLidocaine

19 36 M VT Cardioversion

F indicates Female; M, Male; VF, ventricular fibrillation; VT, ventricular tachycardia; Y,years.

CAPTOPRIL AND VENTRICULAR ARRHYTHMIAS

189

scoring each segment as 1 for normokinesia, 2 for hypokinesia, 3 for akinesia, 4for dyskinesia and 5 for an aneurysmal segment. A Wall Motion Score Index(WMSI) was computed as the sum of scores of all segments divided by thenumber of segments evaluated. Left ventricular end-systolic and end-diastolicvolumes were calculated from a two- and four-chamber view using the modifiedbiplane Simpson’s rule.24 From these volume measurements the ejection fractionwas calculated.Measurements were made off-line from end-diastolic and end-systolic still-frames using a Microsonics cardiac analysis system (Nova Microsonics). Leftventricular volumes were indexed for body surface area. Left ventricular dilata-tion was defined as follows. Left ventricular volumes of all patients with an en-zymatic infarct size of α-HBDH Q72 of less than 730 U/l, which represents the

Time (days)

0 50 100 150 200 250 300 350 400

Pro

babi

lity

0,50

0,60

0,70

0,80

0,90

1,00

Captopril

Placebo

p =0.12

Figure 9.1. The arrhythmia-free survival in patients randomized to captopril or placebo isdepicted. Despite a clear difference during the first days after myocardial infarction in fa-vor of captopril, there is no significant difference between the two groups during the firstyear of follow up.

CHAPTER 9

190

lower tertile of all randomized infarct sizes, were pooled. This group of patientsshowed no change between first and last echocardiographic evaluation. Meanleft ventricular end-diastolic and end-systolic volume index and correspondingstandard deviations of this reference group were considered as the normal valuefor the study population. Subsequently, all individual end-systolic and and-diastolic volume indexes at each time point were compared with this normalvalue. If a patient showed a left ventricular volume index of more than onestandard deviation above normal and end-systolic and end-diastolic volume in-dex were 1.5 standard deviation above normal, the patient was considered tohave left ventricular dilatation. This dilatation criterion appeared to be robustsince variation of the criterion between 0.5 and 1.5 standard deviation generatedcomparable results.

Arrhythmic events and Holter monitoring. In this study, ventricular arrhyth-mias requiring anti-arrhythmic treatment were investigated. These arrhythmiasincluded ventricular tachycardia (VT), ventricular fibrillation (VF) and suddencardiac death. Sudden death was defined as death within one hour of symptoms,but also included unwitnessed death in patients who were previously stable. VTwas defined as three or more ventricular premature beats with a rate exceeding100 beats/min. In addition, Holter recordings before discharge, and at three and12 months were part of the study protocol. Paired ventricular premature beatsand VT were defined as high-grade ventricular arrhythmias, corresponding toLown class 4A and 4B.

Norepinephrine levels. Blood samples were collected for assessment of no-repinephrine levels at 0, 1, 12, 24, 48, 72, and 96 hours after the start of studymedication, which was at the completion of streptokinase infusion. Norepi-nephrine was measured using a sensitive assay with electrochemical detection.Cumulative norepinephrine values were calculated following the trapeziumrule.25

Statistical analysis . Results are presented as means with standard deviation,except when stated otherwise. Differences between groups were examined usingthe Student’s t-test. The Chi-square test was used for discrete data. Fisher’s Ex-act test was used in case of small patient numbers (indicated separately). TheLog-Rank test was used in the case of survival analysis.

CAPTOPRIL AND VENTRICULAR ARRHYTHMIAS

191

Results

Early ventricular arrhythmias

In the first 48 hours after thrombolytic therapy, 19 patients (6%) had ventricu-lar arrhythmias requiring anti-arrhythmic treatment (Table 9.1). Fourteen pa-tients had

CHAPTER 9

192

Table 9.2. Characteristics of patients with- and without early ventricular arrhythmias

Arrhythmias N No arrhythmias N p-value

DemographicsAge (years) Male (%) Onset (hours)Laboratory α-HBDH Q72 (U/l) Potassium (mmol/l) Norepinephrine (pg/ml) at 1 hour cumulative (96 hours)Echocardiography WMSI LVESVI (ml/m2) LVEDVI (ml/m2) LVEF (%) LV aneurysm (%) Dilatation (%)Hemodynamics Heart rate (beats/min) Blood pressure(mmHg) systolic diastolic Rate-pressure productMedication ACE inhibitor (%) randomized open label Diuretics (%) Beta-blocker (%)

62 ± 1079

3.0 ± 1.0

2240 ± 15264.1 ± 0.3

1409 ± 1230731 ± 297

2.2 ± 0.332 ± 1465 ± 1952 ± 11

1883

79 ± 14

131 ± 2085 ± 12

10162 ± 2214

21121818

191919

1617

1510

117771712

16

151515

19171717

59 ± 1075

3.5 ± 1.3

1215 ± 9354.1 ± 0.4

965 ± 648796 ± 435

1.9 ± 0.425 ± 1055 ± 1355 ± 10

1050

83 ± 17

126 ± 2178 ± 1510471 ±

2966

521129

279279278

242261

235207

222174174174241224

196

204204195

279272272272

0.3000.9050.112

< 0.0010.415

0.0170.639

0.0040.1020.0730.3830.5970.047

0.453

0.3450.1090.694

0.0180.0070.7350.471

Digoxin (%) 6 17 5 272 0.706

α-HBDH Q72 indicates cumulative alpha-hydroxy butyrate dehydrogenase over the first72 hours after myocardial infarction; LVESVI, left ventricular end-systolic volume index;LVEDVI, left ventricular end-diastolic volume index; LVEF, left ventricular ejectionfraction; onset, time from onset of symptoms to randomization; WMSI, wall motion scoreindex. Hemodynamics were assessed 8 hours after randomization.

CAPTOPRIL AND VENTRICULAR ARRHYTHMIAS

193

VT, one of whom needed DC cardioversion (case 19); the other patients weretreated with mostly class I anti-arrythmic agents. Patient 13 was treated withlidocaine and also needed temporary pacemaker support for subsequent brady-cardia. This patient died at day 212 due to progressive heart failure. Patient 8and 14 had VT degenerating into VF. Patient 8 had successful DC cardiover-sion. However, patient 14 did not recover and was shown to have a free wallrupture on autopsy. Patients 7, 12 and 17 had primary ventricular fibrillation.Patient 7 initially recovered after DC cardioversion and was treated with amio-darone. However, despite this therapy the patient had several episodes of VTand died on day 7 due to untreatable VF. Autopsy was performed, and a dilatedleft ventricle was found with a 90% stenosis in the left anterior descending ar-tery, and a significant stenosis in the ramus circumflexus. Patient 12 died afterfailure of a CPR procedure; a free wall rupture was found on autopsy. Patient 17had VF during a PTCA procedure, and recovered after cardioversion withoutfurther complications. Thus, four of the 19 patients (21%) with early ventriculararrhythmias requiring therapy died during the first year of follow up, three ofwhom died during the first week. Coronary angiography was performed in an-other 5 patients (case 6, 7, 11, 13 and 16) within 20 days. All of these patientshad a patent infarct-related (left anterior descending) artery.

In Table 9.2, other characteristics of patients with and without early ventricu-lar arrhythmias are given. On average, patients with early ventricular arrhyth-mias received thrombolytic therapy and subsequent study medication 0.5 hoursearlier than patients without these arrhythmias (difference not significant). Inaddition, these patients were characterized by a significantly larger enzymaticinfarct size (α-HBDH Q72), a higher wall motion score, and a trend towards alarger end-diastolic volume. At one hour after thrombolytic therapy, this wasparalleled by increased norepinephrine levels. However, cumulative levels ofnorepinephrine over 96 hours did not differ significantly between patients with-and without early ventricular arrhythmias. During follow up, more left ventricu-lar dilatation was seen in the group with early ventricular arrhythmias. Heart rateand blood pressure did not differ significantly between both groups, althoughthere was a trend towards a higher diastolic blood pressure in patients with earlyventricular arrhythmias. Finally, a significant treatment effect of captopril onearly ventricular arrhythmias was observed, since of patients randomized tocaptopril, 3% had early ventricular arrhythmas compared to 10% in patients al-located to placebo, reflecting a relative risk of 0.27 (95% CI 0.09 - 0.78).

CHAPTER 9

194

Late ventricular arrhythmias

In Table 9.3, patients with late ventricular arrhythmias (after 48 hours up toone year after myocardial infarction) are listed. Approximately half of these latearrhythmic events occurred before hospital discharge (cases 1 to 9, 45%). Thus,including the early ventricular arrhythmias, 28 of all 39 arrhythmic events inthis study (72%) occurred before hospital discharge.

Of all patients with late ventricular arrhythmias, eight had VT, treated withanti-arrhythmic medication as listed. Patient 7 had VT during exercise and wastreated with sotalol. This patient experienced a second myocardial infarction onday 143 and died due to progressive heart failure and untreatable VT. The otherseven patients with VT survived up to one year without further events. Two

Table 9.3. Late ventricular arrhythmias (after 48 hours, up to one year)

Nr Age (y) M/F Arrhyth-mia

Days Treatment / outcome

1 72 M VF 2 During coronary angiography23456789

10111213141516171819

5458736762555869

64685567546263617053

MMMMMMMM

MMMMMMMFMM

VTVT

VT,VFVFVFVTVT

VT,VF

VTVTVT

SCDVF

SCDVT

SCDSCDSCD

3378

11111214

14141926445863171171277

ProcainamideLidocaineDiedCardioversionCardioversion, lidocaineSotalol. Died day 143 (CHF/VT)MexiletineCardioversion

AmiodaronePropafenonLidocaineUnwitnessedReinfarctionData not availableLidocaineUnwitnessedUnwitnessedUnwitnessed

20 64 M SCD 277 VF registered

CHF indicates congestive heart failure; SCD, sudden cardiac death. Other abbrevations asin Table 9.1.

CAPTOPRIL AND VENTRICULAR ARRHYTHMIAS

195

other patients had VT degenerating into VF. Patient 9 was successfully car-dioverted, but patient 4 did not survive despite a CPR procedure. (This patient isthe same as case nr. 7 listed under early ventricular arrhythmias).

CHAPTER 9

196

Four patients had VF after 48 hours. Patient 1 had VF during coronary angi-ography and was successfully cardioverted. Patient 5 had VF and was car-

Table 9.4. Characteristics of patients with and without late ventricular arrhythmias

Arrhythmias N No arrhyth-mias

N p-value

DemographicsAge (years) Male (%)Laboratory α-HBDH Q72 (U/l) Potassium (mmol/l) Echocardiography WMSI LVESVI (ml/m2) LVEDVI (ml/m2) LVEF (%) LV aneurysm (%) Dilatation (%)Exercise testing Exercise duration (s) Positive for ischemia (%)Functional class NYHA (%) I II III IV ≥ IIMedication ACE inhibitor (%) randomized open label Diuretics (%) Beta-blocker (%)

62 ± 1095

1725 ± 9494.4 ± 0.5

2.03 ± 0.4733 ± 1361 ± 1147 ± 14

1273

421 ± 11325

50446050

4765029

1919

1812

141212121711

1616

981018

19181818

59 ± 1074

1245 ± 10074.5 ± 0.4

1.81 ± 0.4227 ± 1259 ± 1556 ± 10

1150

428 ± 36826

67284113

5021932

279279

240191

224176176176240225

229229

17672103

261

279253253253

0.312 0.077

0.051 0.496

0.063 0.076 0.655 0.003 0.740 0.250

0.939 0.873

0.209 0.208 0.793 0.469

< 0.001

1.000 0.866 0.004 0.884

Digoxin (%) 17 18 9 253 0.479

NYHA indicates New York Heart Association class, other abbrevations as in Table 9.2. A s-sessments were performed before hospital discharge. Potassium levels were measured atsix months.

CAPTOPRIL AND VENTRICULAR ARRHYTHMIAS

197

dioverted successfully; however, this patient died suddenly at day 26 (case 5and 13 represent the same patient). Patient 6 had VF and was cardioverted andtreated with lidocaine. Patient 14 had VF during a second infarction on day 44and was cardioverted successfully. There were six cases of sudden death afterhospital discharge (2% of all patients). In only one patient, case 20, an ar-rhythmia (VF) was documented. Patient 15 had an unsuccessful CPR attempt ina hospital not participating in CATS; data on (the type of) arrhythmias were notobtained. The other four patients (cases 13,17,18,19) died unwitnessed. Coro-nary angiography was performed in 12 patients with late ventricular arrhythmiasup to 136 days after myocardial infarction. Five patients had an occluded LAD(42%) compared to 30 out of 151 patients (20%) without these arrhythmas(p=0.160).

In Table 9.4, other characteristics of patients with late ventricular arrhythmiasare listed. Similar to patients with early ventricular arrhythmias, those with lateventricular arrythmias were characterized by a relatively large enzymatic infarctsize. There was also a trend towards more wall motion abnormalities and alarger end-systolic volume before hospital discharge. This resulted in a clearlyreduced ejection fraction in patients with late ventricular arrhythmias. Left ven-tricular dilatation was not seen more frequent in the late arrhythmia group, al-though all patients who died suddenly during follow up showed some degree ofdilatation (data previously published).26 The reduction in ejection fraction wasparalleled by more symptoms of heart failure: half of those with late ventriculararrhythmias were in NYHA class II or higher, compared to 13% of patientswithout these arrhythmias. There was no clear treatment effect of captopril,since the fraction of patients randomized to captopril were comparable betweenpatients with- and without late ventricular arrhythmias. There were no clear dif-ferences in the use of open label ACE inhibitors or digoxin, but patients withlate ventricular arrhythmias did use significantly more diuretics at discharge.

Effects of captopril

Table 9.5 shows some of the other effects of captopril, documented in CATS.Already one hour after the first dose of captopril (6.25 mg) a significant 11%reduction of ACE activity was observed in patients allocated to captopril. Thiswas paralleled by a significant 15% reduction of norepinephrine levels com-pared to baseline in the captopril group. At 8 hours after the start of study medi-cation, after a third dose of 6.25 mg, there were still no differences in heart rate,blood pressure or rate-pressure product. Peak α-HBDH and cumulative α-HBDH over 72 hours in large infarcts were significantly reduced by captopril.At day 3, there was no difference in potassium levels between patients random-

CHAPTER 9

198

ized to placebo or captopril. However, six months after randomization, potas-sium levels were significantly higher in patients allocated to captopril. Althoughno significant differences were found in left ventricular end-diastolic and end-systolic volume, these dimensions were consistently larger in the placebo groupduring the first months of follow up.22 Moreover, captopril was shown to pre-vent left ventricular dilatation, especially in patients with moderately sized in-farcts.

In Figure 9.1, the arrhythmia-free survival in patients treated with captopril orplacebo is depicted during the first year of follow up. It is shown that early aftermyocardial infarction ventricular arrhythmias were more frequent in the placebogroup. However, after the first few days, no additional difference in the inci-dence of ventricular arrhythmias was observed, resulting in a non-significantdifference for the complete observation period of one year.

Holter monitoring performed before hospital discharge and after three and 12months showed no differences in high-grade ventricular arrhythmias (Lownclass 4A and 4B). However, when patients were divided into two subgroupsseparated by the median ejection fraction, the following was observed. In pa-tients with an ejection fraction above the median of the population (56%), theprevalence of high-grade ventricular arrhythmias were similar (Figure 9.2A).However, in patients with a reduced ejection fraction, high-grade ventricular ar-rhythmias were consistently reduced in patients allocated to captopril, with a dif-ference reaching statistical difference at 12 months (Figure 9.2B). Patients withan ejection fraction above the median were characterized by the absence of leftventricular dilatation, whereas a significant increase in end-diastolic volume wasseen in those with a reduced ejection fraction.

Discussion

In this double-blind, placebo-controlled study a reduction of early postinfarc-tion arrhythmic events was observed in patients randomized to the ACE inhibi-tor captopril. No effect was found on late arrhythmic events, although high-grade ventricular ectopy during Holter monitoring was reduced in a subgroup ofpatients with pronounced left ventricular dysfunction. These effects may be ex-plained by a reduction of enzymatic infarct size, early blunting of the neurohu-moral response, beneficial effects on potassium levels, and reduction of leftventricular dilatation during follow up.

CAPTOPRIL AND VENTRICULAR ARRHYTHMIAS

199

0

10

20

30

40

50

60

70

80

0

40

45

50

55

60

65

Predischarge 3 Months 12 Months

High-grade VAs LVEDVI(%) (ml/m²)

PlaceboCaptopril

0

10

20

30

40

50

60

70

80

0

50

55

60

65

70

75

80

85

Predischarge 3 Months 12 Months

High-grade VAs LVEDVI

*

**

(%) (ml/m²)

PlaceboCaptopril

Figure 9.2. Effect of captopril on high-grade ventricular arrhythmias in sub-groups of preserved (panel A) and reduced (panel B) ejection fraction. For ex-planation, see text. LVEDVI, left ventricular end-diastolic volume index; VAs,ventricular arrhythmias. *p=0.039, **p=0.001.

CHAPTER 9

200

Early ventricular arrhythmias - previous studies

A limited number of studies have investigated the effects of ACE inhibitionon ventricular arrhythmias early after myocardial infarction. Ray et al.27 studiedthe effect of captopril, administered a mean of 15 hours after the onset ofsymptoms in 99 patients with acute myocardial infarction. None of the patientsreceived thrombolytic therapy; patients with small infarcts were excluded. Therewas no difference in the incidence of ventricular arrhythmias during the first 24hours. Pipilis et al.28 investigated the effects of ACE inhibition 13 hours afteronset of symptoms in 100 patients, 92 of whom received thrombolytic therapy(ISIS-4 pilot study). There was a non-significant reduction of VT and AIVR af-ter captopril treatment. Bussman et al.29 found a significant reduction of ven-tricular arrhythmias after captopril during the first 48 hours of Holter monitoringin 49 patients (50% received thrombolytic therapy). None of the patients in thecaptopril group had VF during this period, compared to seven patients in theplacebo group. Study medication was given intravenously a mean of 10 hoursafter the onset of symptoms. In the present study, study medication was admin-istered during thrombolytic therapy at a mean of 3.5 hours after onset of symp-toms. A significant reduction of early ventricular arrhythmias requiring therapywas observed (3% vs 10%, p=0.018). These data suggest that the time of ad-ministration may be an important determinant of the anti-arrhythmic effect ofACE inhibitors, with a more pronounced effect when ACE inhibition is appliedearlier. Di Pasquale et al.30 addressed this issue using a randomized design. Intheir study, captopril treatment started before thrombolytic therapy was com-pared to treatment initiated several days later. Early treatment resulted in signifi-cantly less ventricular arrhythmias compared to late treatment. This may becaused by the fact that the incidence of ventricular arrhythmias drops rapidlybeyond 12 hours after the start of symptoms.31 When the incidence of arrhyth-mias is low, a treatment effect may be difficult to detect and clinically less rele-vant. In addition, potentially anti-arrhythmic effects of ACE inhibitors may beespecially operative during the early stages of myocardial infarction.

How do ACE inhibitors reduce early ventricular arrhythmias ?

It is generally accepted that ACE inhibitors do not have direct anti-arrhythmiceffects.32 Despite this, substantial reductions in the incidence and duration of VFupon reperfusion have been reported in experimental studies.9-11 In these stud-ies, this was attributed to the observed limitation of myocardial injury and re-duction of catecholamine overflow. This beneficial effect on ventricular ar-

CAPTOPRIL AND VENTRICULAR ARRHYTHMIAS

201

rhythmias, infarct size and norepinephrine levels was abolished by indometha-cin, suggesting a prostaglandin-mediated mechanism.10 In contrast to these ex-perimental studies, where rapid and complete reperfusion was accomplished af-ter a short episode of ischemia (minutes), thrombolysis in man provides gradualreperfusion several hours after the onset of ischemia. After one hour or more ofischemia, AIVR and nonsustained VT become the dominating reperfusion ar-rhythmias,33 both of which are considered relatively harmless. Still, life-threatening ventricular arrhythmias requiring therapy did occur in the earlystages after thrombolytic therapy (Table 9.1), and these arrhythmias appeared tobe reduced by ACE inhibition. This may be explained by modulation of severalarrhythmogenic factors.

Myocardial ischemia is a major causative factor of early ventricular arrhyth-mias.34 It has been shown that patients with early VF are characterized by moreextensive coronary artery disease.35 A reduction of myocardial ischemia may beaccomplished by a reduction of blood pressure and/or heart rate after ACE in-hibition. However, in CATS there were no significant differences in blood pres-sure, heart rate or rate-pressure product between patients allocated to captoprilor placebo during the first few days after myocardial infarction, although hy-potension occurred somewhat more frequent in patients randomized to capto-pril.22 Another mechanism by which ACE inhibition could reduce ischemia maybe an increase in collateral flow to the area at risk, as demonstrated by Ertl etal..36 In this experimental study in dogs, this resulted in a significant reductionof infarct size. Similarly, treatment with captopril in the present study resulted ina reduced enzymatic infarct size, especially in patients with large infarcts. Thismay indicate that after captopril treatment collateral flow to the area at risk is in-creased, although the evidence supporting this hypothesis is indirect.

Neurohumoral activation. Denervating the heart before ligation of a coronaryartery almost completely prevents the occurrence of subsequent ventricular ar-rhythmias.37,38 Pharmacological reduction of sympathetic imput may also resultin an anti-arrhythmic effect. As stated before, a reduction of norepinephrinelevels after ACE inhibition has been described in animal experiments.9-11 Thisreduction was paralleled by a clear-cut reduction of ventricular arrhythmias. Inthe present study, a modest but significant reduction of norepinephrine levelswas observed one hour after the start of captopril. It should be noted that thesevalues represent systemic catecholamine levels, whereas noreprinephrine out-flow in the mentioned animal experiments was measured locally in the coronaryeffluent. Thus, a small reduction in systemic norepinephrine levels may repre-sent a much larger reduction at the site of reperfusion.

Angiotensin II, which is increased in the setting of acute myocardial infarc-tion,39 also has documented pro-arrhythmic effects.40 Although this component

CHAPTER 9

202

of the renin-angiotensin system was not measured directly in CATS, a rapid re-duction in angiotensin II levels is likely since ACE activity was already signifi-cantly reduced one hour after the first dose of study medication (6.25 mg capto-pril). So far, the pro-arrhythmic effects of angiotensin II have not been docu-mented in man. Still, the reduction of this peptide may add to the anti-arrhythmic effect of ACE inhibitors.

Electrolytes. Potassium is also important determinant of early ventricular ar-rhythmias.41 Although clear effects of ACE inhibition on potassium levels havebeen reported,42 no effect was seen three days after start of captopril. However,we did observe a significant increase of potassium levels after six months oftreatment.

In conclusion, the reduction of early ventricular arrhythmias is most likelyprecipitated by a reduction of the ischemic area, resulting in a smaller infarctsize, and beneficial effects on norepinephrine and angiotensin II levels.

Late ventricular arrhythmias - previous studies

Most information concerning the effect of ACE inhibition on late ventriculararrhythmias originates from patients with heart failure. Many of these studieshave reported a beneficial effect on total mortality.43-47 In addition, some45-47 butnot all investigators43,44 have reported an effect on sudden cardiac death. Thismay be explained by differences in the severity of heart failure, design of thestudy, and definition of heart failure. Recently, the SAVE investigators48 re-ported on the effect of captopril in patients without heart failure, but with sig-nificant left ventricular dysfunction (ejection fraction of 40% or less) after myo-cardial infarction. In this study, total mortality was reduced by 19%. In addition,sudden cardiac death was also reduced by 19%.49 Additional Holter recordingsshowed a reduction of VT in the captopril group after one year, compared to arelative increase in the placebo group relative to baseline.49

In the present study, late ventricular arrhythmias were not reduced after cap-topril treatment. In addition, no effect was seen on high-grade ventricular ar-rhythmias (Lown 4A and B) during Holter monitoring. The differences betweenCATS and SAVE are substantial: patients in CATS were not selected, and all re-ceived thrombolytic therapy, resulting in a median ejection fraction of 56% be-fore hospital discharge. In contrast, in SAVE only patients with an ejection frac-tion of ≤ 40% were included. Thus, a certain degree of left ventricular dysfunc-tion may be required for ACE inhibitors to produce an effect on late life-threatening ventricular arrhythmias. Søgaard et al.50 described the effect ofcaptopril on the incidence of ventricular arrhythmias during the first six months

CAPTOPRIL AND VENTRICULAR ARRHYTHMIAS

203

after myocardial infarction in patients with left ventricular dysfunction (EF ≤45%).They found a consistent reduction of ventricular arrhythmias in patients treatedwith captopril. The authors attributed this result to a reduction of left ventriculardilatation and ischemia in the treated group. This further demonstrates the effectof ACE inhibition on ventricular arrhythmias in patients with significant leftventricular dysfunction. In fact, when patients in CATS were divided into twogroups separated by an ejection fraction of 56% (median value), a significant ef-fect of ACE inhibition on high-grade ventricular arrhythmias was observed(Figure 9.2, panel B). In conclusion, it appears that ACE inhibition reduces re-petitive forms of ventricular ectopy during Holter monitoring in patients withsignificant left ventricular dysfunction. This may also result in a reduction ofsudden cardiac death, although at present the evidence is not conclusive.

Table 9.5. Early and late effects of captopril

Captopril Placebo p-value

Heart rate (beats/min) 83 ± 17 82 ± 15 0.474Systolic RR (mmHg)Rate-pressure productα-HBDH (U/l) peak Q72 all infarcts Q72 large infarctsNorepinephrine after 1 hour(% reduction)ACE activity after 1 hour (% reduction)Potassium (mmol/l) day 3 six monthsDilatation (%) small infarcts medium size large infarctsLV aneurysm (%)

126 ± 2110547 ± 3151

669 ± 4541166 ± 8861873 ± 738

15

11

4.1 ± 0.44.6 ± 0.4

306387

126 ± 2010352 ± 2676

876 ± 7201390 ± 11092193 ± 1035

8

3

4.1 ± 0.44.4 ± 0.4

194085

0.7730.631

0.0190.0770.045

< 0.001*

0.001*

0.6600.008

0.3540.0360.955

up to 12 months 19 20 0.960

α-HBDH Q72 indicates cumulative alpha-hydroxybutyrate hydrogenase over the first 72hours after myocardial infarction; RR, blood pressure. *Significant differences in the ca p-topril group compared to bas eline, no differences in placebo group (paired t-test).

CHAPTER 9

204

How does captopril reduce late ventricular arrhythmias ?

Left ventricular dilatation. In this study, patients with late ventricular ar-rhythmias were characterized by a reduced ejection fraction, a significantlyhigher NYHA class, and more frequent use of diuretics (see Table 9.4). Thisgroup of patients with left ventricular dysfunction is characterized by progres-sive left ventricular dilatation (Figure 9.2, panel B). Left ventricular dilatation inthe first few weeks after myocardial infarction is a powerful predictor of life-threatening ventricular arrhythmias during follow up.19 In addition, ongoingdilatation after discharge increases the incidence of ventricular arrhythmias evenmore.26 In CATS, left ventricular dilatation was prevented in patients with mod-erate-sized infarcts (Table 9.5) by treatment with captopril. However, this didnot result in a reduction of late arrhythmic events. Still, high-grade ventricularectopy during Holter monitoring was reduced by captopril in patients with a re-duced ejection fraction, a group of patients characterized by progressive dilata-tion.

Anti-thrombotic effects. Most cases of sudden cardiac death after myocardialinfarction are not related to ischemia.51 In addition, in the present study patientswith late ventricular arrhythmia did not show more exercise-induced ischemiacompared to other patients (Table 9.4). However, recurrent coronary thrombosismay result in life-threatening ventricular arrhythmias late after myocardial in-farction.52 For instance, in the present study patient 14 (Table II) had VF at day44, and proved to have a second myocardial infarction. ACE inhibition mayalso have a beneficial effect on late ventricular arrhythmias related to recur-rences of angina pectoris or myocardial infarction. Recent evidence suggeststhat angiotensin II increases levels of plasminogen activator inhibitor-1, themost important physiological inhibitor of tissue-type plasminogen activator(tPA).53 This implies that in patients with an activated renin-angiotensin systemthe risk for recurrent myocardial infarction (and accompanying ventricular ar-rhythmias) may be increased. Furthermore, ACE inhibition may reduce the inci-dence of these thrombotic events. In the SAVE trial,48 the incidence of reinfarc-tion was reduced significantly in patients treated with captopril. The SOLVD in-vestigators also reported a (nonsignificant) reduction of death due to myocardialinfarction in patients with heart failure.44 In CATS, no effect on thromboticevents or thrombosis-related ventricular arrhythmias was observed. This may bedue to the small number of patients studied and the relatively short duration offollow up. In addition, the renin-angiotensin system may have been activated in

CAPTOPRIL AND VENTRICULAR ARRHYTHMIAS

205

only a limited number of patients, since 50% of the patients had an ejectionfraction of more than 56% at discharge.

Potassium. Most of the anti-arrhythmic effect of ACE inhibitors in patientswith heart failure has been attributed to the potassium-sparing effect of theseagents.54 In the present study, potassium was significantly increased in patientstreated with captopril after six months. However, patients with late ventriculararrhythmias requiring treatment were not characterized by low potassium levels(Table 9.4). Still, since these patients are frequent users of diuretics compared topatients without late ventricular arrhythmias (50 vs 19%, p=0.004), hypoka-lemia is more likely to occur in these patients, which may result in more ar-rhythmic events.

Conclusions

In this study, ventricular arrhythmias requiring therapy early after myocardialinfarction were reduced by captopril administered during thrombolysis. An in-crease of collateral flow to the infarcted area may be the predominant mecha-nism, since heart rate and blood pressure were not reduced after ACE inhibition,and ischemia is known to be the predominant cause of these arrhythmias. Inaddition, a reduction of norepinephrine levels and ACE activity, presumablyleading to lower angiotensin II levels, may add to the anti-arrhythmic effect.Late ventricular arrhytmias that were life-threatening or required treatment werenot reduced by captopril. This lack of effect may be caused by the relativelywell-preserved left ventricular function in CATS patients. This was supported bythe finding that high-grade ventricular ectopy during Holter monitoring was re-duced in a subgroup of patients with a reduced ejection fraction.

References

1. ISIS-2 (Second International Study of Infarct Survival Group) CollaborativeGroup: Randomized trial of intravenous streptokinase, oral aspirin, both, or neitheramong 17,187 cases of suspected acute myocardial infarction: ISIS-2. Lancet1988;ii:349-360.

2. Gruppo per lo studio della streptochinasi nell'inf arto miocardico (GISSI): Effe c-tiveness of intravenous thrombolytic treatment in acute myocardial infarction. Lan-cet 1986;i:397-402.

3. Wilcox RG, von der Lippe G, Olsson CG, Jensen G, Skene AM, Hampton JR: Trialof tissue plasminogen activator for mortality reduction in acute myocardial infar c-tion. Anglo-Scandinavian Study of Early Thrombolysis (ASSET). Lancet1988;2:525-530.

CHAPTER 9

206

4. Anon.: Long-term effects of intravenous anistreplase in acute myocardial infar c-tion: final report of the AIMS study. AIMS Trial Study Group. Lancet1990;335:427-431.

5. The ISAM study group: A prospective trial of intravenous streptokinase in acutemyocardial infarction (I.S.A.M.). N Engl J Med 1986;314:1465-1471.

6. Yusuf S, Peto R, Lewis J, Collins R, Sleight P: Beta-blockade during and after my o-cardial infarction: an overview of the randomized trials. Prog Cardiovasc Dis1985;27:335-371.

7. Heidbuchel H, Tack J, Vanneste L, Ballet A, Ector H, Van de Werf F: Significanceof arrhythmias during the first 24 hours of acute myocardial infarction treated withalteplase and effect of early administration of a beta-blocker or a bradycardiac agenton their incidence. Circulation 1994;89:1051-1059.

8. Roberts R, Rogers WJ, Mueller HS, Lambrew CT, Diver DJ, Smith HC, WillersonJT, Knatterud GL, Forman S, Passamani E, et al: Immediate versus deferred beta-blockade following thrombolytic therapy in patients with acute myocardial infar c-tion. Results of the Thrombolysis in Myocardial Infarction (TIMI) II-B Study. Cir-culation 1991;83:422-437.

9. van Gilst WH, de Graeff PA, Kingma JH, Wesseling H, de Langen CDJ: Captoprilreduces purine loss and reperfusion arrhythmias in the rat heart after coronary a r-tery occlusion. Err J Pharmacy 1984;100:113-117.

10. van Gilst WH, de Graeff PA, Wesseling H, de Langen CDJ: Reduction of reperf u-sion arrhythmias in the ischemic isolated rat heart by angiotensin converting e n-zyme inhibitors: a comparison of captopril, enalapril and HOE 498. J CardiovascPharmacol 1986;8:722-728.

11. Linz W, Scholkens BA, Han YF: Beneficial effects of the converting enzyme inhib i-tor, ramipril, in ischemic rat hearts. J Cardiovasc Pharmacol 1986;8 Suppl 10:S91-S99.

12. Bonaduce D, Petretta M, Morgano G, Attisano T, Bianchi V, Arrichiello P, RotondiF, Condorelli M: Effects of converting enzyme inhibition on baroreflex sensitivityin patients with myoca rdial infarction. J Am Coll Cardiol 1992;20:587-593.

13. Clough DP, Collis MG, Conway J, Hatton R, Keddie JR: Interaction of angiotensin-converting enzyme inhibitors with the function of the sympathetic nervous system.Am J Cardiol 1982;49:1410-1414.

14. Ball SG: The sympathetic nervous system and converting enzyme inhibition. JCardiovasc Pharmacol 1989;13 Suppl 3:S17-S21.

15. Pfeffer MA, Lamas GA, Vaughan DE, Parisi AF, Brauwnwald E: Effect of captoprilon progressive ventricular dilatation after anterior myocardial infarction. N Engl JMed 1988;319:80-86.

16. Sharpe N, Smith H, Murphy J, Greaves S, Hart H, Gamble G: Early prevention ofleft ventricular dysfunction after myocardial infarction with angiotensin-converting-enzyme inhibition. Lancet 1991;337:872-876.

17. St. John Sutton M, Pfeffer MA, Plappert T, Rouleau JL, Moye LA, Dagenais GR,Lamas GA, Klein M, Sussex B, Goldman S, et al: Quantitative two-dimensionalechocardiographic measurements are major predictors of adverse cardiovascular

CAPTOPRIL AND VENTRICULAR ARRHYTHMIAS

207

events after acute myocardial infarction. The protective effects of captopril. Circu-lation 1994;89:68-75.

18. Nabel EG, Topol EJ, Galeana A, Ellis SG, Bates ER, Werns SW, Walton JA, MullerDW, Schwaiger M, Pitt B: A randomised, placebo-controlled trial of combinedearly intravenous captopril and recombinant tissue-type plasminogen activator the r-apy in acute myocardial i nfarction. J Am Coll Cardiol 1991;17:467-473.

19. White HD, Norris RM, Brown MA, Brandt PWT, Whitlock RML, Wildt CJ: Leftventricular end-systolic volume as the major determinant of survival after recoveryfrom myocardial i nfarction. Circulation 1987;76:44-51.

20. Cosin Aguilar J, Hernandiz Martinez A, Andres Conejos F : Mechanisms of ventric u-lar arrhythmias in the presence of pathological hypertrophy. Eur Heart J1993;14(suppl J):65-70.

21. Campbell RW, Higham D, Adams P, Murray A: Potassium: its relevance for a r-rhythmias complicating acute myocardial infarction. J Cardiovasc Pharmacol1987;10 Suppl 2:S25-S28.

22. Kingma JH, van Gilst WH, Peels KH, Dambrink J-HE, Verheugt FWA, WielengaRP: Acute intervention with captopril during thrombolysis in patients with first a n-terior myocardial i nfarction. Eur Heart J 1994;15:898-907.

23. van der Laarse A, Kerkhof PL, Vermeer F, Serruys PW, Hermens WT, VerheugtFW, Bar FW, Krauss XH, van der Wall EE, Simoons ML: Relation between infarctsize and left ventricular performance assessed in patients with first acute myocardialinfarction randomized to intracoronary thrombolytic therapy or to conventionaltreatment. Am J Cardiol 1988;61:1-7.

24. Schiller NB, Shah PM, Crawford M, DeMaria A, Devereux R, Feigenbaum H, Gu t-gesell H, Reichek N, Sahn D, Schnittger I, Silverman NH, Tajik AJ: Recommend a-tions for quantitation of the left ventricle by two-dimensional echocardiography.American Society of Echocardiography Committee on Standards, Subcommittee onQuantitation of Two-Dimensional Echocardiograms. J Am Soc Echocardiogr1989;2:358-367.

25. Altman DG: Relation between several variables, in Practical statistics for medicalresearch, 1st ed. London, Chapman and Hall, 1990, pp. 325-364.

26. Dambrink J-HE, Beukema WP, van Gilst WH, Peels CH, Lie KI, Kingma JH: Leftventricular dilatation and high-grade ventricular arrhythmias in the first year aftermyocardial infarction. J Cardiac Failure 1994;1:3-11.

27. Ray SG, Pye M, Oldroyd KG, Christie J, Connelly DT, Northridge DB, Ford I,Morton JJ, Dargie HJ, Cobbe SM: Early treatment with captopril after acute my o-cardial infarction. Br Heart J 1993;69:215-222.

28. Pipilis A, Flather M, Collins R, Hargreaves A, Kolettis T, Boon N, Foster C, A p-pleby P, Sleight P: Effects on ventricular arrhythmias of oral captopril and of oralmononitrate started early in acute myocardial infarction: results of a randomisedplacebo controlled trial. Br Heart J 1993;69:161-165.

29. Bussmann WD, Micke G, Hildenbrand R, Klepzig H Jr: Captopril bei akutemHerzinfarkt: Einfluss auf Infarctgrösse und Rhythmusstörungen. Dtsch MedWochenschr 1992;117:651-657.

CHAPTER 9

208

30. Di Pasquale P, Barone G, Paterna S, Cannizzaro S, Giubilato A: Efficacy of capt o-pril before thrombolysis in acute myocardial infarction: preliminary findings.Drugs Exp Clin Res 1990;16:581-589.

31. Northover BJ: Ventricular tachycardia during the first 72 hours after acute my o-cardial infarction. Cardiology 1982;69:149-156.

32. Hemsworth PD, Pallandi RT, Campbell TJ: Cardiac electrophysiological actions ofcaptopril: lack of direct antiarrhythmic effects. Br J Pharmacol 1989;98:192-196.

33. de Graeff PA, van Gilst WH, Bel K, de Langen CDJ, Kingma JH, Wesseling: Co n-centration-dependent protection by captopril against myocardial damage duringischemia and reperfusion in a closed chest pig model. J Cardiovasc Pharmacol1987;9(Suppl. 2):S37-S42.

34. Dorian P, Langer A, Morgan C, Casella L, Harris L, Armstrong P, for the TissuePlasminogen Activator: Toronto (TPAT) study group: Importance of ST-segmentdepression as a determinant of ventricular premature complex frequency afterthrombolysis for acute myocardial i nfarction. Am J Cardiol 1994;74:419-423.

35. Kyriakidis M, Petropoulakis P, Antonopoulos A, Barbetseas J, Georgiakodis F,Aspiotis N, Kourouclis C, Toutouzas P: Early ventricular fibrillation in patientswith acute myocardial infarction: correlation with angiographic findings. Eur HeartJ 1993;14:364-368.

36. Ertl G, Kloner RA, Alexander RW, Braunwald E: Limitation of experimental infarctsize by an angiotensin-converting enzyme inhibitor. Circulation 1982;65:40-48.

37. Schaal SF, Wallace AG, Sealy WC: Protective influence of cardiac denervationagainst arrhythmias of myocardial infarction. Cardiovasc Res 1969;3:241-244.

38. Ebert PA, Vanderbeek RB, Allgood RJ, Sabiston Jr DC: Effect of chronic cardiacdenervation on arrhythmias after coronary artery ligation. Cardiovasc Res1970;4:141-147.

39. McAlpine HM, Morton JJ, Leckie B, Rumley A, Gillen G, Dargie HJ: Neuroend o-crine activation after acute myocardial infarction. Br Heart J 1988;60:117-124.

40. de Langen CD, de Graeff PA, van Gilst WH, Bel KJ, Kingma JH, Wesseling H: E f-fects of angiotensin II and captopril on inducible sustained ventricular tachycardiatwo weeks after myocardial infarction in the pig. J Cardiovasc Pharmacol1989;13:186-191.

41. Nordrehaug JE, Johannessen KA, von der Lippe G: Serum potassium concentrationas a risk factor of ventricular arrhythmias early in acute myocardial infarction. Cir-culation 1985;71:645-649.

42. O'Keeffe S, Grimes H, Finn J, McMurrough P, Daly K: Effect of captopril therapyon lymphocyte potassium and magnesium concentrations in patients with conge s-tive heart failure. Cardiology 1992;80:100-105.

43. The CONSENSUS trial study group: Effects of enalapril on mortality in severecongestive heart failure. N Engl J Med 1987;316:1429-1435.

44. The SOLVD investigators: Effect of enalapril on survival in patients with reducedleft ventricular ejection fractions and congestive heart failure. N Engl J Med1991;325:293-302.

45. Newman TJ, Maskin CS, Dennick LG, Meyer JH, Hallows BG, Cooper WH: Effectsof captopril on survival in patients with heart failure. Am J Med 1988;84:140-144.

CAPTOPRIL AND VENTRICULAR ARRHYTHMIAS

209

46. Cohn JN, Johnson G, Ziesche S, Cobb F, Francis G, Tristani F, Smith R, DunkmanWB, Loeb H, Wong M, et al: A comparison of enalapril with hydralazine-isosorbidedinitrate in the treatment of chronic congestive heart failure. N Engl J Med1991;325:303-310.

47. Fonarow GC, Chelimsky-Fallick C, Stevenson LW, Luu M, Hamilton MA,Moriguchi JD, Tillisch JH, Walden JA, Albanese E: Effect of direct vasodilationwith hydralazine versus angiotensin-converting enzyme inhibition with captopril onmortality in advanced heart fai lure: the Hy-C trial. J Am Coll Cardiol 1992;19:842-850.

48. Pfeffer MA, Brauwnwald E, Moyé LA, Basta L, Brown EJ, Cuddy TE, Davis B R,Geltman EM, Goldman S, Flaker GC, Klein M, Lamas GA, Packer M, Rouleau J,Rouleau JL, Rutherford J, Wertheimer JH, Hawkins CM, on behalf of the SAVE i n-vestigators: Effect of captopril on mortality and morbidity in patients with left ve n-tricular dysfunction after myocardial infarction. Results of the Survival and Ve n-tricular Enlargement Trial. N Engl J Med 1992;327:669-677.

49. Packer M, Rouleau J-L, Moyé LA, Rouleau JR, Bernstein V, Cuddy TE, Lewis S,Sussex SA, Sestier F, Goldman S, Jacobson K, Lamas G, McCans J, Randall OS,Wertheimer JH, Davis BR, Brauwnwald E, Pfeffer MA: Effect of captopril on ve n-tricular arrhythmias and sudden death in patients with left ventricular dysfunctionafter myocardial infarction: SAVE trial. J Am Coll Cardiol 1993;21:130A.

50. Søgaard P, Gotzsche CO, Ravkilde J, Norgaard A, Thygesen K: Ventricular a r-rhythmias in the acute and chronic phases after acute myocardial infarction. Effectof intervention with capt opril. Circulation 1994;90:101-107.

51. Bayes de Luna A, Vinolas Prat X, Guindo J: Ventricular arrhythmias in left ve n-tricular hypertrophy and heart failure. Eur Heart J 1993;14 Suppl J:62-64.

52. Verheugt FWA, Brugada P: Sudden death after acute myocardial infarction: theforgotten thrombotic view. Am J Cardiol 1991;67:1130-1134.

53. Ridker PM, Gaboury CL, Conlin PR, Seely EW, Williams GH, Vaughan DE:Stimulation of plasminogen activator inhibitor in vivo by infusion of angiotensin II.Evidence of a potential interaction between the renin-angiotensin system and fibr i-nolytic function. Circulation 1993;87:1969-1973.

54. Poquet F, Ferguson J, Rouleau JL: The antiarrhythmic effect of the ACE inhibitorcaptopril in patients with congestive heart failure largely is due to its potassiumsparing effects. Can J Cardiol 1992;8:589-595.

207

CHAPTER 10. Summary and concluding remarks

The Captopril And Thrombolysis Study (CATS) investigated the effects of theACE inhibitor captopril administered during thrombolytic therapy on left ven-tricular remodeling, neurohumoral activation, and ventricular arrhythmias in298 patients with a first anterior myocardial infarction. In this thesis, focus wason the relation between dilatation of the left ventricle, activation of neurohu-moral systems, and the occurrence of ventricular arrhythmias in the CATSpopulation. These phenomena were assessed using quantitative echocardiogra-phy, determination of neurohormones, and ambulatory electrocardiography.Underlying mechanisms were explored more in depth by signal-averaged elec-trocardiography, body surface mapping and assessment of heart rate variability.

Early observations (within 48 hours after thrombolytic therapy)

The design of CATS was based on a series of experimental studies in whichthe effect of ACE inhibition during ischemia and reperfusion was investigated.These studies demonstrated that ACE inhibition resulted in limitation of myo-cardial injury upon reperfusion, reduced release of catecholamines, and a re-duction of reperfusion-related ventricular arrhythmias. However, there wereconsiderable differences between these experimental studies and the clinicalsetting in which CATS was conducted (Table 10.1). First, the experimentalstudies used animals without preexisting disease, whereas patients participatingin CATS all had coronary artery disease. In addition, the duration of ischemiapreceding reperfusion was considerably longer in CATS compared to the ex-perimental setting. Furthermore, reperfusion was immediate and complete in theexperimental studies compared to gradual restoration of flow during hours todays in 79% of patients, with 21% of study subjects showing no reperfusion atall. This resulted in a different spectrum of ventricular arrhythmias. Ventriculartachycardia and Accelerated IdioVentricular Rhythm (AIVR) were the predomi-nant repetitive ventricular arrhythmias during the first 48 hours after throm-bolytic therapy. When an ischemic period of 15-30 minutes is applied in the ex-perimental setting, ventricular fibrillation is the most frequently observed ar-rhythmia. In contrast to the experimental setting, only five patients (2%) inCATS had ventricular fibrillation early after thrombolytic therapy. Furthermore,the dosage of captopril used in CATS was considerably lower than the doseadministered in many experimental studies. Still, this lower dosage did reducemyocardial injury, quantified by α-HBDH values (Chapter 7). In parallel to thisfinding, systemic norepinephrine levels were reduced in patients treated with

CHAPTER 10

208

captopril. Finally, treatment with captopril resulted in a reduction of early ven-tricular arrhythmias requiring anti-arrhythmic therapy.

In chapter 9, the characteristics of patients with early ventricular arrhythmiasin CATS were investigated. A large enzymatic infarct size and a high wall mo-tion score as a measure of left ventricular dysfunction proved important deter-minants of patients with early ventricular arrhythmias. This was paralleled by atrend towards a larger end-systolic- and end-diastolic volume. In addition, no-repinephrine levels 1 hour after thrombolytic therapy were increased in patientswith these arrhythmias. Vice versa, patients with increased cumulative norepi-nephrine levels up to 96 hours after thrombolytic therapy showed more earlyventricular arrhythmias during Holter monitoring in the early phase. However,increased cumulative catecholamine levels did not predict the occurrence of lateventricular arrhythmias (Chapter 5).

SUMMARY AND CONCLUSIONS

209

Mechanisms by which ACE inhibition may reduce early ventricular arrhythmias

Similar to the preclinical studies, infarct size and norepinephrine levels werereduced by ACE inhibition early after thrombolytic therapy (Chapter 7). A rapidreduction of ACE activity within one hour suggests a reduction of angiotensin IIlevels, a potentially arrhythmogenic component of the renin-angiotensin system.An effect on left ventricular volume was not detectable at this early stage. Inaddition, heart rate and blood pressure were not influenced significantly by ACEinhibition in the first hours after randomization. This would lead to the conclu-

Table 10.1. Comparison of CATS with preceding experimental studies

Experimental studies# CATS

Study subjects Animals withoutpreexisting disease

Patients with coronaryartery disease

Reperfusion Preceding ischemia mode of reperfusion

5 - 60 minutesimmediate in all animals(100% reperfusion)

mean 3.5 hoursgradual in 79%, noreperfusion in 21%*

Dominating arrhyt h-mia after 5 min ischemia after 60 min ischemia > 1 hour ischemia

VF 100%AIVR 92%-

--VT 54%, AIVR 59%¥

Effects of ACE inhib i-tion Dosage Myocardial injury Catecholamines

0.9 to 9 mg/kg/2 hourspurine-loss ↓norepinephrine outflow ↓

0.08 mg/kg/2 hours¶

α-HBDH ↓norepinephrine at 1 hour ↓

Arrhythmias VF ↓ after 5min ische-mia

VT§ + VF ↓

# See Chapter 1, paragraph 2.1, * Data from 163 patients with angiography after a medianof 7 days, ¥ incidence in patients with Holter monitoring, ¶ mean dose in the first 48 hoursfor a patient of 70 kg, § VT requiring anti-arrhythmic treatment. α-HBDH indicates α-hydroxybutyrate dehydrogenase; ACE, angiotensin-converting enzyme; AIVR, acceleratedidioventricular rhythm; VF, ventricular fibrillation; VT, ventricular tachycardia.

CHAPTER 10

210

sion that the reduction of myocardial injury and neurohumoral activation, ratherthan possible effects on wall stress and/or myocardial ischemia, appeared to bethe mechanisms by which early ventricular arrhythmias were reduced. However,reduction of ischemia by increase of collateral flow as a possible mechanismcan not be excluded.

Late observations (after 48 hours up to one year)

In recent years, it has become clear that an enlarged end-systolic volume as-sessed 4-8 weeks after myocardial infarction is an important determinant for theoccurrence of late ventricular arrhythmias. In Chapter 2, we studied this relationbetween left ventricular dilatation and ventricular arrhythmias in more detail.Patients who died suddenly, presumably due to ventricular arrhythmias, werecharacterized not only by a large end-systolic volume at discharge, but also byprogressive left ventricular dilatation during follow up. In addition, ongoingdilatation during the first year after myocardial infarction predicted the preva-lence of nonsustained ventricular arrhythmias during Holter monitoring at oneyear independently of end-systolic volume at discharge. This implies that the as-sociation between left ventricular dilatation and ventricular arrhythmias is a dy-namic one: left ventricular dilatation at discharge is not only associated witheven more dilatation, but also leads to an increased incidence of ventricular ar-rhythmias during follow up. In other words, dilatation begets dilatation, but thisin turn also begets late ventricular arrhythmias.

In Chapters 3 and 4, the underlying electrophysiological mechanisms of therelation between dilatation and ventricular arrhythmias were investigated. Previ-ous data suggested that an acute increase in left ventricular volume results in anincreased dispersion in refractoriness. However, the effects of chronic dilatationon repolarization characteristics were not well known. Body surface mapping in78 CATS patients (Chapter 3) revealed that chronic left ventricular dilatationwas paralleled by increased nondipolarity of QRST integral maps, a measure ofdispersion in refractoriness. In addition, filtered QRS duration, which is alsoconsidered an electrical expression of left ventricular dilatation, was prolongedin patients with an increased end-diastolic volume (Chapter 4). However, bymeans of multiple regression analysis we demonstrated that early dilatationwithin 24 hours, and not subsequent dilatation during follow up, determined thisprolongation of QRS duration. These data suggest that early dilatation, largelydetermined by acute functional dilatation and expansion of the infarcted area, isassociated with a delay in conduction time, whereas late dilatation, occurring ininfarcted and noninfarcted areas, contributes to an increased dispersion in re-fractoriness.

SUMMARY AND CONCLUSIONS

211

Another factor which may add to the arrhythmogenic effects of a dilated leftventricle is concomitant neurohumoral activation. Sympathovagal imbalance isa well-known risk factor for the occurrence of life-threatening ventricular ar-rhythmias. This state can be detected indirectly by heart rate variability assess-ment. In CATS, we measured heart rate variability in 175 CATS patients beforedischarge, and in a smaller group of 120 patients after three months (Chapter 6).At discharge, end-systolic volume and end-diastolic volume were comparable inpatients with and without a reduced heart rate variability. However, during thefirst year of follow up the increase in end-diastolic and end-systolic volume wasmore pronounced in patients with a reduced heart rate variability. In addition,patients with left ventricular dilatation were characterized by a reduced heartrate variability at discharge. Surprisingly, heart rate variability was no longer re-duced in patients with dilatation after three months of follow up. These datasuggest that persistant neurohumoral activation can contribute to left ventriculardilatation in the early phase, when scar tissue formation is incomplete and theinfarcted area is still very vulnerable to straining forces. However, this relationbetween neurohumoral activation and ventricular dilatation does not last untilthree months, when heart rate variability is no longer different in patients withand without dilatation. Consequently, it is not likely that the late arrhythmogeniceffects of left ventricular dilatation can be fully explained by concomitantneurohumoral activation. However, this does not exclude the possibility of achange in local sympathetic activity when the left ventricle is dilated, leading toan increased dispersion in refractoriness. Other studies have recently suppliedinformation to support this hypothesis. Furthermore, in several studies the inde-pendent predictive value of a reduced heart rate variability for the occurrence oflate ventricular arrhythmias was demonstrated. We only found a trend in this di-rection (Chapter 6), which may have been caused by the small number of lateventricular arrhythmias observed in this particular group of patients.

Treatment with captopril resulted in a reduction of the number of patients de-veloping left ventricular dilatation (Chapter 8). This was paralleled by a reduc-tion in the proportion of patients developing heart failure. In contrast to the earlypostinfarction period, ventricular arrhythmias late after myocardial infarctionwere not reduced in patients treated with ACE inhibition (Chapter 9). Recently,other studies have reported a reduction of late postinfarction ventricular ar-rhythmias by ACE inhibition. In these studies only patients with a clearly re-duced left ventricular function were investigated. It may well be that patients inthese studies had a larger infarct size with more left ventricular dilatation andsubsequent ventricular arrhythmias. In CATS, approximately one third of pa-tients did not show any dilatation at all, which may explain the low incidence oflate ventricular arrhythmias and sudden cardiac death (2%) found in this study.

CHAPTER 10

212

In fact, when patients with a reduced ejection fraction were selected, we did ob-serve a reduction of nonsustained ventricular arrhythmias in patients treatedwith captopril.

Implications for clinical practice

The findings of CATS are confirmed by much larger studies like GISSI-3 andISIS-4, in which it is shown that ACE inhibition can safely be applied early afteracute myocardial infarction in patients without severe heart failure, hypotension,or renal failure. Data from our study suggest that the early reduction of mortalityobserved in these large trials may well be based on a reduction of ventricular ar-rhythmias, at least in patients with anterior wall myocardial infarction. After theearly phase of myocardial infarction, patients with moderate-sized infarcts aremost likely to benefit from ACE inhibition, since in these patients left ventriculardilatation and the occurrence of heart failure were significantly reduced bytreatment with captopril. In addition, in this group of patients an effect on non-sustained ventricular arrhythmias, and possibly also on sustained ventricular ar-rhythmias, may be expected.

These findings support a treatment strategy in which all patients with anteriormyocardial infarction, and without signs of severe heart failure, hypotension, orrenal failure receive an ACE inhibitor during thrombolytic therapy or at leastwithin 24 hours after onset of symptoms. When no left ventricular dilatation isobserved, left ventricular function is (close to) normal, and there are no signs ofpersistant neurohumoral activation (e.g., increased heart rate, reduced heart ratevariability), the likelihood of benefit is low and treatment can be stopped. Con-versely, in patients showing left ventricular dilatation and/or neurohumoral acti-vation, a beneficial effect on left ventricular remodeling and accompanyingventricular arrhythmias may be expected.

Indications for future research

Many clinical studies have demonstrated a beneficial effect of ACE inhibitionon left ventricular remodeling. In CATS, an effect on infarct size, one of themajor determinants of remodeling, was also observed. Since the treatment effectobserved was not very large, and limited mostly to patients with large infarcts,this result needs further investigation. Furthermore, some studies suggest thatthe incidence of ventricular fibrillation is increased after very early thrombolytic

SUMMARY AND CONCLUSIONS

213

therapy (i.e., within one hour after onset of symptoms). Since experimental evi-dence of an anti-arrhythmic effect of ACE inhibition was especially observedafter a short duration of ischemia, ACE inhibition during this very early phasemay help to reduce the high incidence of life-threatening arrhythmias in this pe-riod. In addition, immediate and complete reperfusion, as opposed to gradualreperfusion may also reduce the threshold for ventricular arrhythmias. Thismode of reperfusion is typical for coronary angioplasty (PTCA), a procedurewhich is increasingly used in the acute phase of myocardial infarction withoutpreceding thrombolytic therapy. Some studies have already reported an in-creased incidence of ventricular fibrillation after direct angioplasty for acutemyocardial infarction. It should be investigated whether ACE inhibition mayfurther increase the safety of this procedure when used in the setting of acutemyocardial infarction.

Final conclusions

Neurohumoral activation is still a major determinant of early ventricular ar-rhythmias in the setting of thrombolytic therapy for acute myocardial infarction.Increased sympathetic activity and early activation of the renin-angiotensinsystem after myocardial infarction are modulated significantly within one hourafter administration of the ACE inhibitor captopril. Together with the observedreduction of infarct size after captopril treatment, these effects may explain theobserved reduction of early postinfarction ventricular arrhythmias.

Progressive dilatation of the left ventricle proves an important determinant oflate ventricular arrhythmias after thrombolytic therapy. Left ventricular dilata-tion may be promoted by persistant neurohumoral activation, and is associatedwith distinctive changes in electrophysiology, which include prolonged QRSduration and increased dispersion in refractoriness. Treatment with captopril re-sults in a reduction of left ventricular dilatation and accompanying signs of heartfailure, especially in patients with moderate-sized infarcts. A reduction of lateventricular arrhythmias is observed only in patients with significant left ventricu-lar dysfunction.

These clinical observations confirm results from experimental studies andsupport the early use of ACE inhibition during thrombolytic therapy in patientswith anterior myocardial infarction. In the later phases of myocardial infarction,patients with moderate but not severe left ventricular dysfunction are most likelyto benefit from this treatment.

215

HOOFDSTUK 11. Samenvatting en conclusies

In de Captopril And Trombolysis Study (CATS) zijn de effecten van de ACEremmer captopril, toegediend tijdens trombolytische therapie, onderzocht bij298 patiënten met een eerste voorwandinfarct. Het effect van ACE remming ophet proces van vormverandering van de linker kamer na een myocardinfarct, inde Engelstalige literatuur als ‘remodeling’ aangeduid, is onderzocht met behulpvan kwantitatieve echocardiografie. Het effect op kamerritmestoornissen enneurohumorale activatie is bepaald door gebruik te maken van Holtermonitoring en door het meten van een aantal neurohormonen. In dit proefschriftis bovendien aandacht besteed aan de onderlinge relaties tussen dezeeindpunten, met name aan het belang van dilatatie van de linker kamer enneurohumorale activatie als risicofactoren voor het optreden vankamerritmestoornissen. Mogelijke verklarende mechanismen zijn onderzochtdoor gebruik te maken van signal-averaged elektrocardiografie, body surfacemapping en het meten van de heart rate variability.

Vroege waarnemingen (tot 48 uur na trombolytische therapie)

Het ontwerp van CATS was gebaseerd op een aantal eerder verrichteexperimentele studies waarin het effect van ACE remming tijdens ischemie enreperfusie werd onderzocht. Deze studies hadden aangetoond dat onderexperimentele omstandigheden ACE remming tijdens reperfusie resulteert inafname van de schade aan het myocard, verminderde afgifte vancatecholamines en reductie van aan reperfusie gerelateerdekamerritmestoornissen. Er bleken echter aanzienlijke verschillen te bestaantussen deze experimentele studies en CATS (Tabel 11.1). Ten eerste werden ingenoemde studies dieren of preparaten zonder atherosclerose van decoronairvaten onderzocht. Dit in tegenstelling tot de CATS patiënten, die allencoronairsclerose hadden. Ten tweede, in vergelijking met het diermodel was deduur van de ischemie voorafgaand aan reperfusie in CATS aanzienlijk langer.Tevens werd in het diermodel zeer plotseling volledige reperfusie tot standgebracht, terwijl bij de CATS patiënten de reperfusie slechts langzaam (uren totdagen) of in het geheel niet (21%) tot stand kwam. Dit resulteerde ook inverschillende typen kamerritmestoornissen. Kamertachycardie en ‘AcceleratedIdioVentricular Rhythm’ (AIVR) waren de meest voorkomende repetitievekamerritmestoornissen tijdens de eerste 48 uur na trombolytische therapie.Slechts 5 patiënten (2%) hadden in deze vroege fase kamerfibrilleren. Dit integenstelling tot het diermodel, waar reperfusie over het algemeen tot stand

HOOFDSTUK 11

216

werd gebracht na 15 tot 30 minuten ischemie. Hier was kamerfibrilleren demeest frequent voorkomende ritmestoornis. Een ander verschil was de doseringvan captopril, die in CATS aanzienlijk lager was dan in de meesteexperimentele studies. Desondanks bleek deze dosering de myocardschade,uitgedrukt in een stijging van het enzym α-hydroxybutyraat dehydrogenase (α-HBDH), te beperken (Hoofdstuk 7). Ook bleken de noradrenaline spiegelsverlaagd bij patiënten die behandeld werden met captopril. Tot slot leiddebehandeling met captopril ook tot een reductie van het aantalkamerritmestoornissen waarvoor anti-aritmische therapie noodzakelijk was.

In Hoofdstuk 9 zijn de specifieke kenmerken van de patiënten in CATS metvroeg optredende kamerritmestoornissen geanalyseerd. Een groot (enzymatischbepaald) myocardinfarct en een hoge wall motion score als maat voor linkerkamer dysfunctie bleken belangrijke karakteristieken te zijn bij patiënten metvroege kamerritmestoornissen. Daarnaast bestond er bij deze patiënten een trendin de richting van een groter eind-systolisch en eind-diastolisch volume van delinker kamer. Tevens waren bij deze patiënten de noradrenaline spiegels,gemeten één uur na trombolytische therapie, verhoogd in vergelijking metpatiënten zonder deze ritmestoornissen. Omgekeerd bleken patiënten methogere cumulatieve noradrenaline spiegels, gemeten tot 96 uur na trombolyse,vaker vroege kamerritmestoornissen te hebben. Verhoogde catecholaminespiegels in deze fase hadden echter geen voorspellende waarde voor latekamerritmestoornissen (Hoofdstuk 5).

Mechanismen waardoor ACE remming het optreden van vroegekamerritmestoornissen zou kunnen verminderen

Evenals in de experimentele studies had ACE remming, toegediend vlak natrombolytische therapie, een gunstige invloed op de grootte van het infarct ende noradrenaline spiegels (Hoofdstuk 7). Tevens nam de gemeten ACE activiteitaf, hetgeen suggereert dat door deze behandeling ook de angiotensine IIspiegels worden verlaagd. Van angiotensine II is bekend, althans onderexperimentele omstandigheden, dat hogere spiegels gepaard gaan met meerkamerritmestoornissen.

Een effect op de volumina van de linker kamer werd niet waargenomen in devroege fase na het myocardinfarct. Hartfrequentie en bloeddruk werden nietsignificant beïnvloed door ACE remming in de eerste uren na randomisatie.Deze resultaten doen vermoeden dat het beperken van de schade aan hetmyocard en het afremmen van de neurohumorale activatie, en niet de effectenop wandspanning of myocardischemie de belangrijkste mechanismen zijnwaardoor vroege kamerritmestoornissen worden voorkomen. Reductie van

SAMENVATTING EN CONCLUSIES

217

ischemie door toename van collaterale flow als verklarend mechanisme is echterniet uitgesloten.

Late waarnemingen (na 48 uur tot 1 jaar na het infarct)

Eerder onderzoek heeft aangetoond dat een toegenomen eind-systolischvolume 4-8 weken na het myocardinfarct een belangrijke risicofactor is voor hetoptreden van kamerritmestoornissen laat na het myocardinfarct. In Hoofdstuk 2wordt de relatie tussen dilatatie van de linker kamer en kamerritmestoornissen in

Tabel 11.1 Vergelijking van CATS met de voorafgaande experimentele studies

Experimentele studies# CATS

Studie onderwerp Dieren zonder pre-existente afwijkingen

Patiënten met coronair-sclerose

Reperfusie duur van ischemie type reperfusie

5 - 60 minutenonmiddellijk envolledig bij alle dieren

gemiddeld 3.5 uurgeleidelijk bij 79%, geenreperfusie bij 21%*

Meest voorkomenderitmestoornis na 5 min ischemie na 60 min ischemie > 1 uur ischemie

VF 100%AIVR 92%-

--VT 54%, AIVR 59%¥

Effecten van ACEremmingdosering

0.9 tot 9 mg/kg/2 uur 0.08 mg/kg/2 uur¶

myocardiale schade purine verlies ↓ α-HBDH ↓ catecholamines noradrenaline afgifte ↓ noradrenaline na 1 uur ↓ ritmestoornissen VF ↓ na 5 min ischemie (VT§+ VF) ↓

# zie hoofdstuk 1, paragraaf 2.1, * gegevens van 163 patiënten met angiografie na mediaan7 dagen, ¥ incidentie bij patiënten met Holter monitoring, ¶ gemiddelde dosis tijdens deeerste 48 uur voor een patiënt van 70 kg, § VT waarvoor anti-aritmische therapie nodigwas; α-HBDH, α-hydroxybutyraat dehydrogenase; ACE, angiotensin-converting enzyme;AIVR, accelerated idioventricular rhythm;. VF, ventrikelfibrilleren; VT, ventriculairetachycardie.

HOOFDSTUK 11

218

detail beschreven. Patiënten die plotseling overleden, naar verondersteld doorkamerritmestoornissen, werden niet alleen gekenmerkt door een vergroot eind-systolisch volume bij ontslag uit het ziekenhuis, maar ook door progressievedilatatie van de linker kamer tijdens follow-up. Bovendien bleek progressievedilatatie na ontslag het optreden van kamerritmestoornissen te voorspellen, enwel onafhankelijk van het eind-systolische volume bij ontslag. Dit betekent datde relatie tussen dilatatie van de linker kamer en kamerritmestoornissen continuaan verandering onderhevig is: dilatatie van de linker kamer vastgesteld bijontslag is niet alleen geassocieerd met nog meer dilatatie nadien, maar leidt ooktot een hogere incidentie van kamerritmestoornissen tijdens follow-up. Metandere woorden: ‘van dilatatie komt dilatatie’, en deze verdere toename involume brengt op zijn beurt weer meer kamerritmestoornissen met zich mee.

In Hoofdstuk 3 en 4 worden twee mogelijke onderliggendeelektrofysiologische mechanismen van de relatie tussen dilatatie enkamerritmestoornissen besproken. Volgens eerdere studies leidt een acutetoename van het volume van de linker kamer tot een toename van regionaleverschillen in repolarisatie, ook wel ‘dispersie’ in repolarisatie genoemd. Degevolgen van chronische dilatatie voor de repolarisatie waren echter nietbekend. Met behulp van body surface mapping (62-kanaals elektrocardiografie)werd aangetoond dat chronische dilatatie van de linker kamer gepaard gaat meteen toegenomen nondipolariteit van QRST integraal map patronen, een maatvoor dispersie in repolarisatie (Hoofdstuk 3).

Daarnaast bleek de QRS duur, nauwkeurig gemeten met behulp van signal-averaged elektrocardiografie, verlengd te zijn bij patiënten met een vergrooteind-diastolisch volume (Hoofdstuk 4). Bij multipele regressie analyse bleek eenverlenging van de QRS duur vooral bepaald te worden door vroege dilatatie,d.w.z. binnen 24 uur, en niet door late dilatatie. Deze gegevens suggereren datvroege dilatatie, met name bepaald door dilatatie en expansie van hetinfarctgebied, samen gaat met een tragere geleiding van de elektrische activiteitin de hartspier. Late dilatatie van de linker kamer, optredend in zowelgeïnfarceerde als niet geïnfarceerde gebieden, gaat vooral gepaard met eentoegenomen dispersie in repolarisatie.

Een andere factor die mogelijk bijdraagt aan de aritmogene effecten van eengedilateerde linker kamer wordt gevormd door begeleidende neurohumoraleactivatie. Naast activatie van het renine-angiotensine systeem is dysbalanstussen sympathische en parasympathische activiteit hiervan een belangrijkonderdeel. Bovendien is dit een belangrijke risicofactor voor het optreden vanlevensbedreigende ritmestoornissen. Het vóórkomen hiervan kan indirectworden vastgesteld door middel van bepaling van de ‘heart rate variability’, eenmaat voor de variatie van de hartfrequentie in de tijd. Deze bepaling werd

SAMENVATTING EN CONCLUSIES

219

verricht bij 175 CATS patiënten voor ontslag uit het ziekenhuis, en bij eenkleinere groep (120 patiënten) nogmaals na 3 maanden (Hoofdstuk 6). Bijontslag was het gemeten eind-systolische en eind-diastolische volumevergelijkbaar bij patiënten met een normale heart rate variability en patiëntenmet verminderde heart rate variability. Tijdens follow-up was de toename vanbeide volumina echter significant groter bij patiënten met een verminderde heartrate variability. Bovendien werd er bij patiënten met dilatatie van de linkerkamer een significant lagere heart rate variability gezien bij ontslag.Opmerkelijk is dat het verschil in heart rate variability tussen patiënten met enzonder dilatatie na 3 maanden was verdwenen. Dit zou betekenen dataanhoudende neurohumorale activatie na het infarct met name bijdraagt aandilatatie van de linker kamer in de vroege fase, wanneer de vorming vanlittekenweefsel nog niet gereed is en het infarctgebied nog zeer gevoelig is voortrekkrachten. In deze groep patiënten verdwijnt na deze periode de relatie tussenneurohumorale activatie en dilatatie, waardoor er na 3 maanden geen verschil inheart rate variability meer aantoonbaar is tussen patiënten met en zonderdilatatie. Op basis van deze gegevens is het onwaarschijnlijk dat de aritmogeneeffecten van linker kamer dilatatie volledig verklaard kunnen worden uitbijkomende neurohumorale activatie. Aan de andere kant sluit dit demogelijkheid van een lokale verandering van sympathische activiteit, mogelijkleidend tot een toename van dispersie in repolarisatie, niet uit. Recentelijkhebben gegevens van andere onderzoekers aanwijzingen in deze richtinggegeven. Daarnaast is in de literatuur meermalen de onafhankelijkvoorspellende waarde van een verminderde heart rate variability voor hetoptreden van late kamerritmestoornissen aangetoond. In CATS werd slechts eentrend in deze richting gezien (Hoofdstuk 6), mogelijk veroorzaakt door hetkleine aantal late kamerritmestoornissen in deze subgroep.

Behandeling met captopril resulteerde in een vermindering van het aantalpatiënten met dilatatie van de linker kamer (Hoofdstuk 8). Dit ging gepaard meteen reductie van het aantal patiënten met hartfalen. In tegenstelling tot devroege postinfarct periode leidde behandeling met ACE remming niet tot eenreductie van late kamerritmestoornissen (Hoofdstuk 9). Andere studies hebbenechter wel een reductie van kamerritmestoornissen laat na het myocardinfarctlaten zien. In deze studies werden uitsluitend patiënten met een duidelijkverminderde functie van de linker kamer onderzocht. Het is goed mogelijk datdeze patiënten een groter myocardinfarct hadden met meer dilatatie van delinker kamer en bijkomende kamerritmestoornissen. In CATS werd bij eenderdedeel van de patiënten in het geheel geen dilatatie van de linker kamer gezien,wat mogelijk de lage incidentie van late kamer ritmestoornissen en plotselingedood (2%) kan verklaren. Dit wordt gesteund door de bevinding dat er wel een

HOOFDSTUK 11

220

reductie van kortdurende kamerritmestoornissen werd gezien bij patiënten meteen verminderde ejectiefractie.

Betekenis voor de klinische praktijk

De uitkomsten van CATS zijn inmiddels bevestigd door veel grotere studiesals de GISSI-3 en ISIS-4 studie, die eveneens hebben aangetoond dat ACEremming veilig kan worden toegepast in de vroege fase na het myocardinfarctbij patiënten zonder ernstig hartfalen, hypotensie of nierinsufficiëntie. De CATSgegevens doen vermoeden dat de in deze grote trials vastgestelde reductie vanvroege mortaliteit verklaard kan worden door een reductie vankamerritmestoornissen, althans bij patiënten met een voorwandinfarct. Effectenop de volumina van de linker kamer werden in deze fase niet waargenomen. Nade vroege fase van het myocardinfarct zijn het met name de patiënten met eenmiddelgroot infarct die het meeste baat hebben bij ACE remming, aangezien bijdeze patiënten het optreden van dilatatie van de linker kamer en hartfalensignificant verminderde door behandeling met captopril. Daarnaast mag bij dezegroep patiënten ook een effect op kortdurende, en wellicht ook op aanhoudendekamerritmestoornissen worden verwacht. Deze bevindingen rechtvaardigen eenbehandelingsstrategie, waarbij alle patiënten met een voorwandinfarct, maarzonder ernstig hartfalen, hypotensie of nierinsufficiëntie met ACE remmingbehandeld worden tijdens trombolytische therapie, of tenminste binnen 24 uurna het begin van de symptomen. Als er geen dilatatie van de linker kamer wordtgezien, de linker kamer functie vrijwel normaal is en er geen tekenen zijn vanaanhoudende neurohumorale activatie (bijvoorbeeld een persisterende hogehartfrequentie of verminderde heart rate variability), dan is het voordeel vanvoortzetten van behandeling met een ACE remmer dermate gering, dat dezegestopt kan worden. Bij patiënten met dilatatie van de linker kamer en/ofneurohumorale activatie mag echter een gunstig effect op remodeling van delinker kamer en bijkomende kamerritmestoornissen verwacht worden.

Richtlijnen voor toekomstig onderzoek

Uitkomsten van een groot aantal klinische studies hebben inmiddels degunstige uitwerking van ACE remming op remodeling van de linker kamerbeschreven. In CATS is eveneens een gunstig effect op de grootte van hetinfarct, een van de belangrijkste determinanten van remodeling, waargenomen.Aangezien dit effect niet erg groot was en zich met name leek te beperken tot degrotere infarcten, zal dit door verder onderzoek bevestigd dienen te worden.Sommige onderzoekers hebben een hoge incidentie van kamerfibrilleren na

SAMENVATTING EN CONCLUSIES

221

zeer vroege trombolytische therapie (binnen 1 uur na het begin van de klachten)beschreven. Aangezien het anti-aritmische effect van ACE remming inexperimentele studies vooral gezien werd na een korte periode van ischemie,zou ACE remming tijdens zeer vroege trombolyse de incidentie vankamerfibrilleren mogelijk verder kunnen reduceren. Daarnaast kan, integenstelling tot geleidelijke reperfusie, door plotselinge en complete reperfusiede incidentie van levensbedreigende kamerritmestoornissen ook toenemen.Deze wijze van reperfusie is kenmerkend voor de ‘directe’ Percutane Trans-luminale Coronair Angioplastiek (PTCA), een techniek die bij hoogrisicogroepen wordt toegepast in de acute fase van het myocardinfarct, zondervoorafgaande trombolytische therapie. Door sommigen is reeds eentoegenomen incidentie van kamerfibrilleren gerapporteerd na directe PTCA. Devraag of ACE remming de veiligheid van deze procedure uitgevoerd in de acutefase van het myocardinfarct kan vergroten, verdient verder onderzoek.

Slotconclusie

Neurohumorale activatie na het myocardinfarct blijft een belangrijkedeterminant voor vroege kamerritmestoornissen, ook wanneer trombolytischetherapie wordt toegepast. Dit proces bestaat onder andere uit toegenomensympathische activiteit en vroege activatie van het renine-angiotensine systeem,en wordt significant afgeremd binnen 1 uur na toediening van de ACE remmercaptopril. Verminderde neurohumorale activatie en reductie van de grootte vanhet infarct na behandeling met captopril vormen waarschijnlijk de verklaringvoor de waargenomen reductie van vroege kamerritmestoornissen. Progressievedilatatie van de linker kamer blijkt een belangrijke voorspeller van latekamerritmestoornissen. Linker kamer dilatatie kan worden versterkt dooraanhoudende neurohumorale activatie en gaat gepaard met karakteristiekeelektrofysiologische veranderingen, waaronder een verlengde QRS duur en eentoegenomen dispersie in repolarisatie. Behandeling met captopril resulteert ineen vermindering van het optreden van linker kamer dilatatie en bijkomendetekenen van hartfalen, in het bijzonder bij patiënten met een (enzymatischbepaald) middelgroot myocardinfarct. Een afname van latekamerritmestoornissen wordt alleen gezien bij patiënten met een significantverminderde functie van de linker kamer.

Deze klinische observaties bevestigen de resultaten van eerdereexperimentele studies en ondersteunen het gebruik van ACE remmers al tijdenstrombolytische therapie bij patiënten met een voorwandinfarct. Tijdens de laterefasen van het infarct zijn het met name de patiënten met matige (maar niet zeer

HOOFDSTUK 11

222

ernstige) dysfunctie van de linker kamer die het meeste baat hebben bij dezebehandeling.

223

DANKWOORD.

Dit proefschrift is grotendeels gebaseerd op de resultaten van de CaptoprilAnd Thrombolysis Study. CATS zou onuitvoerbaar zijn geweest zonder deinspanningen van allen die aan deze studie hebben bijgedragen. In de Appendixzijn de CATS medewerkers van de verschillende centra en hun bijdragen apartgenoemd. In het bijzonder gaat mijn dank uit naar die onderzoekers en hunpatiënten die bereid waren naast het CATS studieprotocol mee te werken aanhet vervaardigen van extra registraties (signal-averaging elektrocardiografie,body surface mapping). Deze metingen waren voor de totstandkoming van ditproefschrift van essentiëel belang.

CATS zou er in het geheel niet geweest zijn zonder zijn ontwerpers: prof. drWiek H. van Gilst en dr J. Herre Kingma, die met dierexperimenteel onderzoekaan de Rijksuniversiteit Groningen, gevolgd door klinische waarnemingen inhet St Antonius Ziekenhuis te Nieuwegein, de wetenschappelijke basis voordeze studie hadden gelegd.

Beste Herre, jou dank ik in het bijzonder voor de mogelijkheid die je mij hebtgeboden om binnen de afdeling R&D Cardiologie van het St AntoniusZiekenhuis onderzoek te doen als deelnemer aan de CATS studie. Naast hetscheppen van onderzoeksmogelijkheden was jij steeds degene die met de‘helicopter-view’ het overzicht wist te behouden, noodzakelijk voor hetuitzetten van de hoofdlijnen van het onderzoek. Mijn hartelijke dank voor jewaardevolle commentaren en aanwijzingen.

Beste Wiek, ondanks de afstand Groningen - Nieuwegein ben jij steeds nauwbetrokken geweest bij de totstandkoming van dit proefschrift. Met name jeheldere analyse van de gegevens en nuttige aanwijzingen bij de verslaglegginghebben mij vaak geholpen. Bovendien was jij de ‘modulator’ van hetgezelschap, die er voor zorgde dat ook op het persoonlijke vlak alles soepelverliep. Mijn oprechte dank hiervoor.

Prof. dr K.I. Lie wil ik graag bedanken voor het feit dat hij mijn promotor heeftwillen zijn. Geachte professor, het ging hier om onderzoek dat op aanzienlijkeafstand werd uitgevoerd, maar waar u desondanks uw vertrouwen in heeftgesteld. Mijn hartelijke dank hiervoor.

224

Dr Arne Sippens Groenewegen wil ik enorm bedanken, niet alleen voor zijnbijdrage aan het body surface mapping artikel maar ook voor zijn morele steuntijdens de afgelopen jaren. Beste Arne, vooral door jouw enorme enthousiasmeen optimisme hebben we uiteindelijk het de novo fenomeen kunnenoverwinnen! Ik dank je voor je gedegen en gedetailleerde commentaren enpersoonlijke steun. Ik hoop op een hernieuwde samenwerking na je terugkomstuit de Verenigde Staten.

Samen met Kathinka H. Peels werden honderden CATS echo’s beoordeeld.Beste Kathinka, ik dank je voor je geduld en enthousiasme waarmee je me debeginselen van de echocardiografie hebt bijgebracht. Ook na CATS zal ik hiernog veel plezier van hebben. Ik wens je veel sterkte met het afronden van jeeigen onderzoek, en ik hoop dat we in de toekomst nog eens samenwerken.

De voormalig opleider dr C.A.P.L. Ascoop en de huidige opleider dr N.M. vanHemel, evenals de overige maatschapsleden van de afdeling Cardiologie vanhet St Antonius Ziekenhuis te Nieuwegein (E.T.Bal, dr J.M.P.G. Ernst, dr W.Jaarsma, dr J.H. Kingma, E.G. Mast, dr H.W.M. Plokker en dr M.J. Suttorp),dank ik hartelijk voor de ruimte die ik heb gekregen voor het doen vanwetenschappelijk onderzoek. Met name noem ik dr Plokker, die in zijnspaarzame vrije uren de beoordeling van een groot aantalcoronairangiogrammen heeft gesuperviseerd.

Verder ben ik veel dank verschuldigd aan:

De afdeling Hartbewaking en Medium Care (afdeling G3) van het St AntoniusZiekenhuis en de hartbewakingsafdelingen van de andere studiecentra waarCATS patiënten opgenomen werden. In de vroege fase van het studieprotocolheeft het verplegend personeel veel extra metingen gedaan en administratiefwerk verricht.

Het personeel van de Hartfunctie afdeling van het St Antonius Ziekenhuis(hoofd: mevr. J. Bakker-de Graaf) en de andere studiecentra voor alle ECG’s,fietstesten, echo’s en andere onderzoeken die zij in het kader van de CATSstudie verricht/begeleid hebben (zie ook de Appendix).

De dames van R&D Cardiologie: mw A.C.A.M. van der Sluijs-Verweij, mw R.Ciarrocca, en met name mw J.C. Helwig-Vernooij, die met ongekende snelheidaltijd weer de artikelen de deur uit kreeg en het manuscript drukklaar heeftgemaakt. Guusje, Rosemary en Joke: enorm bedankt voor jullie hulp ! Dank

225

ben ik ook verschuldigd aan mw Gerda J.L. van der Kuijl, voor het kritischlezen van de Nederlandstalige passages.

Mijn collegae-assistenten in het St Antonius Ziekenhuis, en met name Pascalvan Dessel, die, toen de nood hoog was, in razend tempo nog een aantalpatiënten in kaart heeft gebracht.

‘De AMC’ers’, en met name André Linnenbank en Mark Potse, die altijd tijdleken te hebben voor nog even een plaatje of nog even een berekeningetjeextra. Hulde !

De (toenmalige) medewerkers van de TCC Groningen, Hans Hillege en Pieter-Jan de Kam, die de ‘Amerikaanse’ database gekraakt hebben en ons vanessentiële gegevens hebben voorzien.

De afdeling Interne Geneeskunde van het Diakonessenhuis te Utrecht (opleider:dr J.B.L. Hoekstra) en mijn huidige collegae-assistenten voor het begrip en deruimte die ik heb gekregen tijdens de laatste loodjes, onder het motto: vóór alleshet Hoofdnummer !

De paranimfen, Jurriën ten Berg en Pieter Pasma voor hun inzet bij deorganisatie van alle festiviteiten.

Mijn zus Brenda, die heeft geholpen met de lay-out van het manuscript toen dedrukker ons op de hielen zat. Lieve Bren, bedankt voor je hulp op een momentdat deze erg nodig was !

Mijn ouders, die mij als altijd hun onvoorwaardelijke steun hebben gegeven inde afgelopen periode. In tijden van wetenschap leek de afstand naar Friesland afen toe onoverbrugbaar. Maar, lieve Pa en Ma, volgend seizoen wordt er weersamen gezeild !

Lieve Trudy, jij hebt alle ups en downs de afgelopen jaren van dichtbijmeegemaakt en intens meegeleefd met alle gebeurtenissen. Dit boekje isdaardoor ook deels het produkt van jouw inspanningen geworden. Ik ben jedaarvoor, zoals je weet, enorm dankbaar.

226

APPENDIX.

In this appendix the investigators and all others who contributed to CATS arelisted. All participating centers were situated in The Netherlands.

Study DirectorsW.H. van Gilst, PhD, Groningen.J.H Kingma, MD, PhD, Nieuwegein.

Participating CentersDiakonessenhuis, Utrecht: 21 patients. A.C. Bredero, MD, principal investiga-tor; L. Bellersen, MD, center-coordinator. CAG, SA-ECG*. Medisch SpectrumTwente, Enschede: 11 patients. P.H. van der Burgh, MD, principal investigator;W. Kiers, MD, center-coordinator. BSM, CAG, SA-ECG. St Sophia Ziekenhuis,Zwolle: 25 patients. R. Enthoven, MD, PhD, principal investigator; R.A. Tio,MD, PhD, center-coordinator. BSM, CAG, SA-ECG. Academisch Ziekenhuis,Groningen: 22 patients. J.P.M. Hamer, MD, PhD, principal investigator; A.Aerts, MD, Y.S. Tuininga, MD, PhD, center-coordinators. BSM, CAG, SA-ECG. Elisabeth Ziekenhuis, Tilburg: 1 patient. N.J. Holwerda, MD, principalinvestigator. Academisch Ziekenhuis V.U., Amsterdam: 52 patients. O. Kamp,MD, PhD, principal investigator; M. Bijl, MD, J. Langeveld, MD, center-coordinators. BSM, CAG, SA-ECG. St Antonius Ziekenhuis, Nieuwegein: 42patients. J.H. Kingma, MD, PhD, principal investigator; F.A.M. Jonkman, MD,PhD, J-H.E. Dambrink, MD, center-coordinators. BSM, CAG, SA-ECG. Aca-demisch Medisch Centrum, Amsterdam: 2 patients. K.T. Koch, MD, principalinvestigator. St Geertruiden Ziekenhuis, Deventer: 23 patients. D.J.A. Lok, MD,principal investigator; M.J.C. Osinga, MD, center-coordinator. BSM, CAG, SA-ECG. Catharina Ziekenhuis, Eindhoven: 23 patients. C.H. Peels, MD, principalinvestigator. CAG, SA-ECG. St Ignatius Ziekenhuis, Breda: 58 patients. R.P.Wielenga, MD, principal investigator; A.F.M. van den Heuvel, MD, center-coordinator. BSM, CAG, SA-ECG. Ziekenhuis “De Lichtenberg”, Amersfoort:18 patients. J.W. Wisse Smit, MD, principal investigator; H. van Ojik, center-coordinator. CAG, SA-ECG.

Coordination and Data MonitoringSt Antonius Hospital, Nieuwegein. M. Butijn, MD; E. Dawson, PhD; I. Hütter-Gering; J.C. Helwig-Vernooij; K. Marinissen, MD. Bristol-Myers Squibb Phar-maceutical Research Institute, Princeton, NJ, USA. C. Maskin, MD; K. Moul-

* Abbreviations indicate participation in one of the following projects: BSM, body surfacemapping; CAG, coronary angiography; SA-ECG, signal-averaged electrocardiography.

227

ton, RN, MS; H. DeRuyter, MD, MRCP. Bristol-Myers Squibb B.V., Woerden.P.A.W.Edgar, MD; P.A. Leijten, MD.

Statistical AnalysisTrial Coordination Center, Academic Hospital, Groningen. H.L. Hillege, MD;P.J. de Kam, MS. Bristol-Myers Squibb Pharmaceutical Research Institute,Princeton, NJ, USA. A. Ona, MS. Academisch Medisch Centrum, Amsterdam.J.G.P. Tijssen, PhD (advisor).

EchocardiographySt Antonius Hospital, Nieuwegein, and Catharina Hospital, Eindhoven. C.H.Peels, MD; W. Jaarsma, MD, PhD; J.C.A. Gadellaa; J-H.E Dambrink, MD; M.St John-Sutton, MD, PhD (advisor); C.A. Visser, MD, PhD (advisor).

Holter MonitoringSt Antonius Hospital, Nieuwegein. J.H. Kingma, MD, PhD; W.P. Beukema,MD; N.M. van Hemel, MD, PhD; A ten Hove; J.J.M. Grob-van Burik (heart ratevariability assessment).

Body Surface MappingSt Antonius Hospital, Nieuwegein. P.F.H.M. van Dessel, MD; W.A.M. Dolman;A.J. Dijkstra; J-H.E. Dambrink, MD; Department of Medical Physics, Universityof Amsterdam. A.C. Linnenbank, MSc; M. Potse, BSc; C.A. Grimbergen, PhD.Heart-Lung Institute, Academic Hospital Utrecht, Utrecht. A. Sippens Groene-wegen, MD, PhD.

Signal-Averaged ElectrocardiographySt Antonius Hospital, Nieuwegein. R. van Ruyven; W.A.M. Dolman; A.J.Dijkstra; J-H.E. Dambrink MD; St Sophia Hospital, Zwolle. T.J.M. Tobé, MD,PhD.

Coronary AngiographySt Antonius Hospital, Nieuwegein. H.W.M. Plokker, MD, PhD; J-H.E. Dam-brink, MD. Academic Hospital VU, Amsterdam. O. Kamp, MD, PhD. CatharinaHospital, Eindhoven. C.H. Peels, MD.

Neuroendocrine AssessmentsDepartment of Clinical Pharmacology, University of Groningen. W.H. vanGilst, PhD; K. Wolters.

228

Cardiac Enzyme AnalysisDepartment of Cardiology, Academic Hospital Leiden. A. van der Laarse, PhD.

Policy Advisory BoardP.A. Poole-Wilson, MD, FRCP (Chairman), London, UK; V. Dzau, MD, PhD,Stanford, USA; H.J. Just, MD, PhD, Freiburg, FRG; J. Roos, MD, PhD, Am-sterdam; J.G.P. Tijssen, PhD, Amsterdam.

Steering and Publication CommitteeJ.H. Kingma, MD, PhD, Nieuwegein; W.H. v Gilst, PhD, Groningen; F.W.A.Verheugt, MD; PhD, Amsterdam; C.H. Peels, MD, Eindhoven; P.H.J.M. Dun-selman, MD, PhD, Breda.

Data Quality CommitteeC.A.P.L. Ascoop, MD, PhD, Nieuwegein; A.J. Fünke Kupper, MD, PhD,Haarlem; P.A. de Graeff, MD, PhD, Groningen.

229

CURRICULUM VITAE

The author of this thesis was born on April 23, 1963 in Utrecht, theNetherlands. In 1981 he obtained his Gymnasium β diploma at the ChristelijkGymnasium in Leeuwarden. From September 1981 to September 1982 hestudied Biology at the State University of Groningen. In September 1982 heentered Medical School at the same university. In 1987 he participated inresearch on rate-responsive pacing at the department of Cardiology of theUniversity Hospital in Groningen (chairman: prof. K.I. Lie). From March toMay 1989 he investigated alternative modes of exercise testing in patients withheart failure at the National Heart Hospital and the Cardiothoracic Institute inLondon, U.K., supervised by prof. P.A. Poole-Wilson. He received his medicaldegree in June 1990. From August 1990 until August 1991, he worked as aresident at the department of Cardiology of the St Antonius Hospital,Nieuwegein (head: dr C.A.P.L. Ascoop). In the following two years, he workedon the Captopril And Thrombolysis Study at the R&D section of the departmentof Cardiology (chairman: dr J.H. Kingma). In july 1993 he started his formaltraining as a cardiologist at the department of Cardiology of the St AntoniusHospital in Nieuwegein, by that time presided by dr N.M. van Hemel. As part ofthis training he is currently working at the department of Internal Medicine ofthe Diakonessenhuis, Utrecht (chairman: dr. J.B.L. Hoekstra). Presumably hewil complete his residency program in July 1998.