UvA-DARE is a service provided by the library of the University of Amsterdam (http://dare.uva.nl)

UvA-DARE (Digital Academic Repository)

Studies on cardiac pacing : emphasis on pacemaker sensors and cardiac resynchronizationtherapy

Erol-Yilmaz, A.

Link to publication

Citation for published version (APA):Yilmaz, A. (2005). Studies on cardiac pacing : emphasis on pacemaker sensors and cardiac resynchronizationtherapy. Amsterdam: Amsterdam University Press.

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

Disclaimer/Complaints regulationsIf you believe that digital publication of certain material infringes any of your rights or (privacy) interests, please let the Library know, statingyour reasons. In case of a legitimate complaint, the Library will make the material inaccessible and/or remove it from the website. Please Askthe Library: https://uba.uva.nl/en/contact, or a letter to: Library of the University of Amsterdam, Secretariat, Singel 425, 1012 WP Amsterdam,The Netherlands. You will be contacted as soon as possible.

Download date: 19 Jun 2019

Studiess on Cardiac Pacing Emphasiss on Pacemaker Sensors and Cardiacc Resynchronization Therapy

Author:: A Erol-Yilmaz

ISBN:: 90-9019108-9

Coverr and Lay-out; Chris Bor, Medische Fotografie en illustratie

AMC,AMC, The Netherlands.

Printedd by: Buijten & Schipperheijn, Amsterdam. The Netherlands

Studiess on pacing. Emphasis on pacemaker sensors and cardiac resynchronization

therapy. .

Thesiss University of Amsterdam, The Netherlands.

Copyrightt © A, Erol-Yilmaz. Amsterdam, The Netherlands. All rights reserved. No

partt of this publication may be reproduced or transmited in any form or by means,

electronicc or mechanical, including photography, recording, or any information

storagee and retrieval system without permission in writing of the copyright owner.

Thiss thesis has been made possible through unrestricted research grants

providedd by (alfabetic order) Guidant BV Nederland. Medtronic BV Nederland

andd Vitatron.

Thee printing of this thesis was financially supported by Jacques H. de Jong Stichting,

J.. E, Jurriaanse Stichting. Guidant BV Nederland, Medtronic BV Nederland,Vitatron,

Hollandd Medical, St Jude Medical Nederland BV. Biotronik Nederland BV. Orbus

International,, Pfizer BV, Sanofi-Synthelabo. Astra Zeneca, Guerbet Nederland BV

Farmasel. .

Studiess on Cardiac Pacing Emphasiss on Pacemaker Sensors and

Cardiacc Resynchronization Therapy

ACADEMISCHH PROEFSCHRIFT

terr verkrijging van de graad van doctor

aann de Universiteit van Amsterdam

opp gezag van de Rector Magnificus

prof.. mr. P.F. van der Heijden

tenn overstaan van een door het college voor promoties ingestelde

commissie,, in het openbaar te verdedigen in de Aula der Universiteit

opp 18 maart 2005. te 10.00 uur

door r

Aytenn Yilmaz

geborenn te C icekdagi (Turkije)

Promotiecommissie: :

promotor: :

co-promotor: :

overigee leden:

Prof.dr.. A.A.M. Wilde

Dr.. R. Tukkie

Prof.dr.. C. Ince

Prof.dr.. J.M.Tde Bakker

Prof.dr.. J.H. Ravesloot

Prof.dr.. N.M. van Hemel

Prof.dr.. M.J. Schalij

Dr.. N. Sulke

Faculteitt Geneeskunde

Financiall support by the Netherlands Heart Foundation for the printing of this

thesiss is gratefully acknowledged.

Anneme e

Voorr mijn moeder

Contents s

Chapterr 1 8

Introduction,, aims and outline of the thesis.

Parti i

Chapterr 2 30

Iss manual rate response optimization necessary in current rate

adaptivee pacemakers?

SubmittedSubmitted for publication

Chapterr 3 56

Heartt rate profiles during two types of exercise testing in healthy

individuals. .

Chapterr 4 74

Individuall optimization of pacing sensors improves exercise

capacityy without influencing quality of life.

PACEPACE 2005: 28:17-24

PresentedPresented at world congress of pacing and electrophysiology.

20032003 Hong Kong and Cardiostim 2004 Nice, Accepted for

presentationpresentation at the American College of Cardiology, Orlando 2005.

Chapterr 5 94

Cerebrall blood flow velocity and cardiac output at the onset of

dynamicc exercise at two settings of pacemaker determined heart rate.

SubmittedSubmitted for publication

PresentedPresented at first joint meeting of the European Federation of Auto-

nomicnomic Societies (EFAS) & American autonomy society {AAS). 2004

Amsterdam.Amsterdam. Finapres Medical System travel fellowship award.

Chapterr 6 112

Directt comparison of a contractility and activity pacemaker sensor

duringg treadmill exercise testing.

PACEPACE 2004: 27:1-7

PresentedPresented at Cardiostim 2004 Nice.

Partt II

Chapterr 7 128

Cardiacc resynchronization induces favorable neurohumoral changes.

AcceptedAccepted for publication in PACE

PresentedPresented at Cardiostim 2004 Nice; 51th annual meeting of the Society

ofof Nuclear Medicine, 2004 (Philadelphia. USA): Accepted for presenta-

tiontion American College of Cardiology. Orlando 2005.

Chapterr 8 144

Cardiacc resynchronization improves microcirculation.

SubmittedSubmitted for publication

AcceptedAccepted for presentation American college of cardiology, Orlando 2005

Chapterr 9 158

Reversedd remodeling of dilated left sided cardiomyopathy after upgrad-

ingg from WIR to WIR biventricular pacing.

EuropaceEuropace 2002; 3: 445-449

Chapterr 10 170

Summaryy and conclusions

Chapterr 11 180

Samenvattingg en conclusies

Dankwoordd 192

Curriculu mm Vitae 200 0

Introduction,, aims and outlinee of the thesis

Chapterr ]

Introductio n n

Thee development of the cardiac pacemaker as an implantable device to sustain heart

rhythmm has revolutionized the treatment of bradyarrhythmia's. After the first

permanentt pacemaker implantation in 1958, the initial developments in pacemaker

technologyy were aimed at improving the longevity and the reliability of their

performance.11 The standard method of pacing in these early years was a single

chamberr ventricular pacemaker, which was effective as a life saving therapy.

Overr the past two decades, however, there have been tremendous advances in

pacemakerr technology resulting in the availability of complex, multiprogrammable,

multii chamber and rate-adaptive pacemakers to meet the haemodynamic needs of

ann individual patient. In addition, recent years generated wide interest in multisite

pacing,, alternative site pacing, preventive pacing for atrial fibrillation, biventricular

pacingg and the development of digital pacemakers.

Inn this chapter, a brief history of cardiac pacing, indications for pacing therapy and

thee rationale for cardiac resynchronization therapy (CRT) are described. At the end

off this introduction, the research aims of this thesis are formulated and the

structuree of this thesis is outlined.

Pacingg history

Ass early as 1580. Mercuriale Geronimo of Padua (Italy) described a syncope

associatedd with a slow pulse, an event now known as the typical Adams Stokes

attack.22 Galvani discovered in 1791 the fundamentals of electrical stimulation of the

heart.33 In 1798, Bichat in Paris examined hearts of decapitated men and showed that

humann hearts reacted to electrical stimuli.4 In a horrific experiment in 1819,

Aldidnii tried to stimulate the heart of criminals soon after their death penalty with

aa pen through the thorax. He realized from these observations that the heart rhythm

couldd be sustained by artificial stimulation.5

Betweenn 1929 and 1932, Albert and Charles Hyman invented an electromechanical

devicee and coined the term pacemaker. It consisted of a magnetic generator

weighingg 7.2 kg. A spring motor could spin the generator for 6 minutes and the

pulsee frequency could be fixed at 30, 60, or 120 impulses per minute.

AA transcutaneously introduced special needle with a noninsulated bare tip

stimulatedd the left ventricle. Paul Zoll was the first who accomplished in 1952

transcutaneouss cardiac pacing and until 1957 transcutaneous closed chest cardiac

stimulationn was the only available approach for cardiac st imulat ion/

Inn 1958. the first pacemaker was implanted by Ake Senning in a 40-year old patient.1

Thee implanted pulse generator had a rechargeable nickel-cadmium battery. Senning

andd his associate Rune Elmqvist had developed and tested this pacemaker between

19566 and 1958.s Although the first pulse generator failed within few hours, a

successorr lasted already for about 6 weeks. This first patient died eventually at the

agee of 86 years after receiving 27 pacemakers during his lifetime.1

Twoo years later, W.M. Chardack carried out the first successful implantation of a

pacemakerr with limited programmable functions.9 The amplitude of the electrical

currentt and the pacing rate could be adjusted using a needle electrode through the

skin.. Engineer Wilson Greatbatch had worked together with Chardack in designing

thiss device.9 This achievement of Chardack and Greatbatch has since been

recognizedd as a defining moment in the history of pacing. A few weeks later, Paul

Zolll and associates implanted a pacemaker of somewhat similar design.10

Clearly,, the idea for an implantable pacemaker was not the exclusive property of any

groupp but was 'in the air,' At that time, implantation required a left anterior

thoracotomyy and exposure of the myocardium: thus training in thoracic or

cardiovascularr surgery was a necessity, Complications from generator failure, rapid

batteryy exhaustion, packaging defects, electrode disruption and infection plagued

thiss endeavor.

Thee problems faced by these pioneers in cardiac pacing have today been largely

overcome,, although some of these problems, but fortunately less frequently, are

stilll encountered.

Comparedd to the early days, cardiac pacing is now a routine procedure in most parts

off the world. A recent world-wide survey of cardiac pacing showed that in The

Netherlandss in one year (2001) 314 new pacemakers per millio n of the population

weree implanted.11 Germany had the highest number of new implants per millio n of

thee population (837/million). followed by The United States with 786 /million. High

degreee atrioventricular block and sick sinus syndrome remain the two major

indicationss for pacemaker implantations. In 2 % of cases, biventricular pacing was

thee mode of pacing in those countries that implanted such systems in 2001. There

hass been an increased use of DDDR systems in most countries.

12 2

Chapterr 1

Pulsee generator

Thee earliest implantable pacemakers delivered stimuli at a fixed rate, regardless of

anyy intrinsic electrical activity. Hence, this early kind of pacing was known as

asynchronouss pacing (later coded VOO) with the objective simply to sustain an

adequatee rhythm. David Nathan approached the Cordis Corporation with the idea

forr an atrial synchronous (or VAT) pacemaker.12 Devices that restored atrial

synchronyy were described as early as 1957, and the first human system was

implantedd by Nathan and co-workers in 1963.12 This device was capable of sensing

intrinsicc atrial electrical activity. Dwight Harken implanted in 1966 the standby

pacer."" today's WI pacemaker.13 Non-competitive pacing was an important step

towardss current dual-chamber pacemakers, which can sense and pace in one or both

cardiacc chambers and are thus capable of mimicking normal physiologic rhythm.

Atrioventricularr (AV)sequential (DVI pacing mode) pacing became popular in the

latterr half of the 1970s and Funke et al. developed in 1977 the first DDD

pacemaker.. Today, the modern rate- responsive DDDR pacemakers have adjustable

sensorss to respond to the patient's metabolic need and contain numerous

programmablee automatic features. We will address in detail the issue of rate

adaptivee pacing by artificial sensors in chapter 2, 4 and 6. Figure 1 illustrate the

evolutionn of pacemakers,

Seymourr Furman and his associates were active from the mid-1960s in designing

instrumentss that assisted the physician to assess the functioning of the pacemaker

att implantation and afterwards.'5 From this work, the Montifore group went on in

19699 to describe the first practical techniques for routine monitoring of pacemaker

ratee by telephone. Transtelephonic monitoring became popular in the United States

duee to the long distances to the hospital. In 1972, Cordis introduced the

Omnicorr line of pacemakers, the first adjustable pacers under noninvasive

electronicc control (see figure l).18 This pacemaker could be non-invasively

reprogrammedd for rate (six choices) and output (four choices). The Omnicor also

containedd a miniature magnetic reed switch. This was the ancestor of today's

externall programmers.

13 3

Figuree 1. The development of pacemakers 1: Demo of the first human pacemaker implanted in 1958:: 2: Chardack Greatbatch pacemaker with adjustable sensitivity (Medtronic Inc. Minnesota. USA).. 3: Siemens Elema (Sollna, Zweden) pacemaker with connection possibilities for a bipolar lead.. 4: Ventricor fixed rate pacer with a rate of 70/min (Cordis. Florida, USA). 5: One of the first WII pacemakers (Medtronic Inc. Minnesota. USA). :6 0mni-Stanicor R-wave inhibited programmable cardiacc pacemaker (Cordis. Florida. USAl 7: Single lead VDR pacemaker (Sorin Biomedica. Sallugia. Italy).. 8: A modern DDDR pacemaker (Guidant Insignia. Guidant. Minneapolis. USA). 9: Cardiac resynchronizationn device (Guidant Contak Renewal TR 2. Guidant. Minneapolis. USA).

M M

Chapterr 1

Pacemakerr leads

Permanentt pacing leads contain five major components: I) the pacing electrode, 2)

thee conductor. 3) the insulation. 4) the connector pin and 5) a fixation mechanism.

Thee pacing lead remains the weakest link of the whole implantable pacing system.

Leadd failures have been a major issue from the beginning of pacemaker therapy.19

Permanentt transvenous pacing first appeared in the early 1960s but did not achieve

generall acceptance in the United States until after 1965. Chardack developed a

helical-coiledd conduction wire, a great advance over earlier lead designs when

introducedd by Medtronic Inc. in 1962,20 Hunter et al. constructed in 1959 a bipolar

myocardiall electrode, which stabilized the resistance over the myocardial-electrode

interfacee after an initial rise.

Thee lead and the Iead-myocardial interface play a major role in pacemaker battery

consumption.. The search for low-threshold, high-impedance leads by fractionate tip

technology,, steroid-eluting designs, and small spherical electrode tip designs are

majorr contributions in this respect. An early approach to improve the handling

characteristicss and downsize the lead diameter was successfully achieved by the

Vitatronn Slimtine high performance silicone-insulated bipolar lead (Vitatron,

Arnhem.. The Netherlands). Intermedics introduced a different approach with the

Thinlinee lead (Angleton, Texas, USA). The leads based on coated wire technology

openedd a new era in lead design. These leads use a single radial conductor coil with

interconductorr electrical insulation provided by a coating of ethylene tetrafluor

ethylenee fluoropolymer on each wire.

Thee earliest transvenous pacing leads had large stimulating electrode surface areas

off approximately 100 mm2 resulting in pacing impedances of about 250 ohms.22 '29

Byy the late 1970s most pacing lead designs had cathodes with reduced surface areas

inn the range of 8-12 mm2 (pacing impedance of 400-800 ohms).23 Today high

impedancee leads (> 1000 ohms) with less than 2 mm2 surface area are available.

Thee conductor of a pacing lead is composed of wire that conducts the electrical

currentt from the pulse generator to the stimulating electrode. Unipolar leads

requiree one conducting coil whereas bipolar leads require two. Conductors are also

responsiblee for transfer of the sensed cardiac signals from the electrodes to the

sensingg amplifier of the pulse generator. The development of coated wire

technologyy enabled that bipolar leads have now outside diameters as small as those

off unipolar leads. The lead insulation extends from the lead connector to the

15 5

cathodee tip and, in a bipolar lead, is interrupted by the anode-ring.

Thee lead connector connects the lead to the pulse generator. The original pacing

leadss had no specialized connector. A major improvement in lead connectors was

thee low profile, in-line bipolar design. An in-line connector places both electrical

terminalss on a single lead pin, with an insulating barrier separating the anode from

thee cathode. The development of the international standard 1 (IS-1), UNI. for

unipolarr leads, and IS-1 BI for bipolar leads as lead connector standard, has been a

majorr improvement in the clinical practice of cardiac pacing. There is a proposal to

developp an international standard connector (IS-IV) with connections for sensing,

pacingg and two shock coils in cardioverter defibrillators. This connector could also

bee used in single lead VDD pacemakers.

Thee ideal pacing lead should have an electrode with a small radius (to increase

currentt density) and a large surface area (to improve sensing). The solution to these

conflictingg considerations for optimal stimulation and sensing characteristics has

beenn sought in the development of electrodes with a small radius but having a

complexx surface structure, The use of electrodes with a textured surface has

resultedd in a dramatic increase in the surface area of the electrode without an

increasee in radius. The textured surface of modern leads minimizes polarization and

improvess sensing and stimulation efficiency. A further advance in permanent pacing

leadd technology has been the development of electrodes that elute small amounts of

thee corticosteroid dexamethason sodium phosphate.24 The steroid-eluting leads are

characterizedd by a minimal change in stimulation threshold from implantation to a

follow-upp period of several years due to the reduction of scar tissue at the lead-tip.

Too provide good tissue contact, two fixation methods of leads are available, so called

passivee and active fixation. Passive fixation mechanisms include tines, fins, helices,

orr conical structures that are extensions of the silicone or polyurethane insulation.

Tiness are the predominant passive fixation mechanism currently used in permanent

pacingg leads. Of the active fixation methods with a screw, barb or hook, the screw

methodd is the one most used. Figure 2 shows the different fixation mechanisms.

Thee recent survey by Mond et al. shows a predominance of transvenous, bipolar

passivee fixation leads. They also observed an increased use of active-fixation leads

inn the atrium.

16 6

Chapterr 1

Figuree 2. Different fixation mechanisms of various types for endocardial and epicardial leads from leftt to right.

2.a.. Atrial endocardial leads.

2.b.. Right ventricular endocardial leads.

I I I I

2.c.. Epicardial leads.

17 7

Pacemakerr batteries

Thee mercury zinc cell, originally developed by Ruben during the Second World War

largelyy for military operations, was patented in 1947.5 Although, initiall y used in

implantablee pacemakers, the longevity limitations of this battery was recognized in

1970.. The mercury zinc cell was replaced by the lithium iodine cell, which continues

too be the prominent battery to date. The litiu m iodine cell was patented in 1972

withh the battery technology largely attributed to Greatbach et al.25 The various

batteriess based on lithium chemistry had somewhat different properties and varied

actuariall survival performance, but all enjoyed significant advantages over mercury.

Thee output voltage of the lithium-iodine cell decreases gradually rather than

abruptlyy as in the mercury zinc cell, giving the physician ample warning of the need

too replace the pulse generator. Finally, the new battery generated no gas as a

chemicall byproduct, so the entire pulse generator could at last be hermetically

sealed. .

Anotherr alternative to the mercury-zinc battery, the nuclear generator, was

conceivedd in the 1960s and brought to clinical function in the early 1970s.

Thee initiative for the nuclear pacemaker came from Parsonnet in 1965-26 For the

youngerr patient likely to live many years with an implanted pacemaker, the nuclear

generatorr had the obvious advantage that the power source would last as long as the

patient.. But by the mid-1970s, the nuclear generator was competing with a new kind

off pulse generator that combined lithium batteries, low current drain, hermetic

encapsulation,, and more advanced circuitry. Nuclear pacemakers worked to

perfection,, but progress in other pacemaker components eventually rendered them

obsolete. .

Pacemakerr indications

Inn the early days of pacing, the primary indication for the implantation of a

pacemakerr was the so-called symptomatic sick sinus syndrome and atrioventricular

block.. In recent years, pacing indications have expanded tremendously. Interest

focusedd on multisite pacing to alleviate dyssynchrony and preventive pacemaker

therapyy to reduce the burden of atrial fibrillation. The efficacy of pacing (especially

atriall septal) in preventing atrial fibrillation has to be established in the future.

18 8

Chapterr 1

Pacingg for neurocardiogenic syncope with application of rate-drop algorithms and

pacingg with short AV delay for symptomatic obstructive hypertrophic

cardiomyopathyy resulted in contradicting data. The AHA and ACC guidelines for

implantationn of cardiac pacemakers updated in 2002, now include several of these

newerr indications.27

Pacemakerr modes

Thee physician has now a wide range of options when selecting a pacemaker for a

givenn patient, depending on a variety of clinical parameters and preferences.

However,, there is uncertainty as to which pacing mode offers most benefit, single

orr dual-chamber. In recent years, in the era of evidence based medicine, several

largee scale randomized trials have tried to answer this question.28

Twoo large-scale randomized trials, the Canadian Trial of Physiologic Pacing (CTOPP)

andd the Mode Selection Trial (MOST), included patients with sinus node

dysfunctionn (MOST) trial and patients with symptomatic bradycardia who needed a

firstt permanent pacemaker without atrial fibrillation (CTOPP) trial.29'31 These two

largee trials demonstrated that AV synchronous dual-chamber pacing does not reduce

thee incidence of stroke or improve survival when compared with ventricular pacing

alone.29"511 However, dual-chamber pacing did reduce the incidence of atrial

fibrillationn and in patients with sinus node dysfunction reduced the incidence of

heartt failure when compared with ventricular pacing.50,32

Otherr studies performed to assess the effect of pacing mode on morbidity and

mortalityy were the Danish study, PAcemaker Selection in the Elderly (PASE), United

Kingdomm Pacing. Cardiovascular Events (UK-PACE) and the Dual Chamber and WI

Implantablee Defibrillator (DAVID) trial.33 34 The treatment arms in the Danish study

weree AAI or 'physiologic'pacing vs. WI pacing in patients with sick sinus

syndrome.355 This trial was the only trial that showed a mortality benefit of

physiologicc pacing over WI pacing. The PASE trial compared DDDR and WIR pacing

inn elderly patients (> 65 years) with a pacing indication for prevention or treatment

off bradycardia. No statistically significant difference was found in quality of life

betweenn DDDR and WIR pacing, although there was a trend towards improved

qualityy of lif e in patients with sinus node dysfunction randomized to dual chamber

pacing.. The UK-PACE trial also compared DDD with WI pacing indicated in elderly

19 9

patientss (>70 years) with AV block. Again, this trial showed no significant

differencee between pacing modes in the primary endpoint of all cause mortality.7

Dataa are accumulating that right ventricular pacing, especially in the right

ventricularr apex, is harmful and care should be taken to minimize the amount of

ventricularr pacing. The DAVID trial supports this concept.54 The DAVID trial tested

thee hypothesis that the atrio-ventricular pacing mode would produce improved

hemodynamicss and total mortality compared to backup ventricular pacing in

patientss indicated for implantable defibrillator therapy. The striking finding of this

triall was that ventricular backup pacing produced less than 3% ventricular pacing.

whilee atrioventricular pacing produced approximately 60% atrial and ventricular

pacedd heart beats, resulting in increased mortality compared to the ventricular

backupp pacing group. The results of the DAVID trial suggest that, especially in

patientss with left ventricular dysfunction, prevention of interventricular

dyssynchronyy is mandatory. Considering the magnitude of the deleterious effects

associatedd with right ventricular pacing in this trial, future studies should explore

thee possibility that left ventricular stimulation may be the only pacing mode

capablee of preventing bradycardia without increasing death and congestive heart

failure.. Two trials {DANPACE and SAVE-PACE) are currently underway that wil l

clarifyy the clinical significance of reducing forced ventricular desynchronization.

Thee results of these trials may direct pacemaker physicians away from the right

ventricularr apical lead toward a new imperative of atrioventricular and right

ventricular-leftt ventricular synchrony.51

Biventricularr pacing history

Wiggerss et al. described in 1925. the importance of the sequence of electrical

activationn of the ventricles for optimal cardiac performance by using artificial

electricall stimulation of the heart.1 Kosowsky et al. showed that right ventricular

apexx pacing reduced the pump function, but that His-bundle pacing maintaining the

normall activation sequence did not.57

AA cardiothoracic surgeon (P. Bakker, from the HLCU of Utrecht, The Netherlands)

presentedd in 1994 preliminary finding of one of the first clinical studies showing

beneficiall effects of biventricular pacing in patients with congestive heart failure.

Inn 5 patients with dilated cardiomyopathy (NYHA HI and IV, left ventricular (LV^

ejectionn fraction 5-23%. complete left bundle branch block and prolonged PR

20 0

Chapterr 1

interval),, a triple chamber pacemaker was implanted. An endocardial right

ventricularr and an epicardial LV lead were connected to the ventricular channel. The

pacemakerr was programmed to DDD-biventricular pacing with an intraoperatively

determinedd optimal AV delay (70-100 msec). Differences in preoperative and 3

monthss postoperative NYHA class, diuretic dose. LV ejection fraction and

echocardiographicc parameters were determined. The investigators showed that

biventricularr pacing significantly improved functional NYHA class from 4 to 2.5 and

exercisee capacity in patients with end-stage dilated cardiomyopathy and LBBB.39

Mosss et al. were one of the first to use of the coronary vein for pacing indications in

1974.400 Later this technique became the preferable way to position a pacing lead for

Figuree 3. Examples of coronary sinus leads for cardiac resynchronization therapy.

3.aa Medtronic Attain ™ OTW Model 4193 iMedtronic Inc.. Minneapolis,USA), b The Aescula LV left-heartt Lead (St Jude Medical, California. USA),c. The EASYTRAK® coronary venous leads (Guidant, Minnesota,, USA).

LVV pacing. Despite advances in percutaneous techniques (lead material, guiding

sheats),, LV lead implantation fails approximately in 10-15% of the patients, and lead

dislodgementt occurrs in an additional 11 %. Therefore lead implantation in the

coronaryy sinus branches remains a challenging procedure (see figure 3. for

exampless of coronary sinus leads).

Heartt failure and rationale for biventricular pacing

Inn heart failure patients, conduction delay progressively develops and is associated

withh a poor outcome. Conduction delay, usually present at different levels, namely

atrioventricular,, interventricular and/ or intraventricular, is associated with

paradoxicall septal movement, presystolic mitral regurgitation, disturbed left

ventricularr AV-timing and reduced diastolic fillin g time. Asynchronous activation

leadss to long-term adaptation of the myocardium, usually referred to as remodeling.

21 1

Rightt ventricular pacing leads to ventricular dilatation and asymmetric hypertrophy,

fiberr disarray, increased myocardial catecholamine concentrations and disturbed

perfusionn tsee figure 4).41 The disturbed delayed electrical activation in patients

withh heart failure results in an abnormal contraction pattern and abnormal

mechanicall loading conditions of the myocardium with increased wall stress and

myocardiall oxygen consumption. The reduced pump function increases

neurohumorall activity, which is beneficial on short term, but on long term this

increasee has many deleterious effects.

Inn this thesis, we report on studies that attempted to further elucidate the

beneficiall effects of CRT on the neurohumoral imbalance (chapter 7) and whether

microcirculatoryy changes could be observed with CRT (chapter 8).

Thee rationale for biventricular pacing in heart failure patients is: 1) to reverse the

mechanicall dyssynchrony by improving the atrio-ventricular, interventricular and

intraventricularr dyssynchrony and 2) to correct the asymmetric activation of the

mitrall papillary muscles with reduction of mitral valve insufficiency. Improved

Neurohumoral l activation n

Regional l Differences s

Mechanicall work

Asynchronouss electrical activation T^

I I —— Dyscoordinate contraction

___J J Reducedd pump function

Rightwardd shift Increased pressuree volume relation wall stress

Ventricularr dilatation

-> Hypertrophy

Longerr conduction pathh length

f t t

Molecular/cellular r changes s

Asymmetric hypertrophy

Figuree 4 Proposed relation between events following onset of asynchronous electrical activation (modifiedd from Vernooy et al. with permission.1~"

22 2

Chapterr I

synchronouss activation of the heart leads to earlier activation, earlier relaxation and

earlierr rapid fillin g with increased left ventricular fillin g time and as consequence

ann increased left ventricular ejection fraction. In responders to biventricular pacing

(700 to 80% of the patients) an acute reduction of mitral regurgitation and long term

reversedd remodeling of the myocardium can be achieved.42

Clinicall trials on CRT

Severall trials provided clinical evidence for the beneficial effects of CRT.43'45 These

beneficiall effects of CRT include: improvement in quality of life, 6-minute walking

distance.. NHYA functional class, peak oxygen consumption, treadmill exercise time,

cardiacc geometry, ejection fraction and reduction of hospitalizations for heart

failure.. In most of these trials, inclusion criteria required heart failure patients to

bee in NYHA III or IV. sinus rhythm. QRS > 120 msec of left bundle branch block type.

andd left ventricular ejection fraction < 35%.

Thee PAcing THerapies for Congestive Heart Failure (PATH-CHF) and MUltisite

STimulationn In Cardiomyopathy (MUSTIC) studies had a cross-over design, whereas

thee Multicenter Insync RAndomized CLinical Evaluation (MIRACLE) and CONTAK

CDD trials are parallel arm studies with controls (device implanted but pacing

inactivated).455 The Comparison of Medical therapy. Pacing aNd defibrillatlON in

heartt failure (COMPANION) trial randomized 1520 patients in a 1:2:2 ratio to receive

optimall pharmacological therapy alone or in combination with CRT with either a

pacemakerr or a pacemaker-defibrillator.46 The primary composite end point was the

timee to death from or hospitalization for any cause. This study showed that in

patientss with advanced heart failure and prolonged QRS interval, CRT decreases the

combinedd risk of death from any cause or first hospitalization and, when combined

withh an implantable defibrillator, also significantly reduces mortality.47 A recent

meta-analysiss suggests that CRT also without implantable defibrillator back-up

reducess mortality from heart failure.48 The ongoing Cardiac Resynchronization in

Heartt Failure (CARE-HF) study is powered to detect a beneficial effect of CRT on all

causee mortality.49

Nowadays,, patients are selected mainly on electrocardiographic criteria. However,

thee standard EKG is less reliable in the characterization of the extent of

dyssynchrony.. since even patients with normal QRS duration on EKG can have

markedd dyssynchrony. while 20-30% of patients with wide ORS complexes do not

respondd to CRT.50 Patient selection using advanced tissue Doppler>maging

techniquess is likely to improve the response rate to CRT.51

Outlin ee of this thesis

Thee first part of this thesis (chapter 2 to 6) concerns the clinical evaluation of

pacemakerr sensors, especially the effect of individual optimization of pacemaker

sensorss on exercise physiology and quality of life. The second part of this thesis

(chapterr 7 to 9) concerns the effect of cardiac resynchronization therapy on cardiac

function,, the neurohumoral system and the microcirculation in patients with heart

failure.. To gain insight into these issues, research concerning rate-adaptive

pacemakerss and cardiac resynchronization therapy has been performed and the

thesiss has been outlined accordingly.

Partt I Studies on pacemaker sensors

Chapterr 2 studied the literature whether manual rate adaptive optimization is

necessaryy in current rate response pacemakers. The various types of sensors, the

effectss of manual sensor optimization, tools for rate response optimization,

pacemakerr automaticity and automatic optimization of sensors are reviewed.

Inn chapter 3 heart rate curves of healthy individuals of various ages are described

duringg the chronotropic assessment exercise protocol and 6- minute hall walk test

withh focus on the achieved time to maximal heart rate, achieved maximal heart rate

andd achieved exercise duration.

Chapterr 4 investigated the influence of individual optimization of sensors on

qualityy of life and exercise tolerance in a randomized, single blind study in patients

withh WIR, DDDR or AAIR pacemakers.

Inn chapter 5 programming the pacemaker to default is compared to optimizing the

pacemakerr sensors with regard to the effects on blood pressure, cardiac output,

strokee volume and mean flow velocity in the middle cerebral artery during graded

ergometryy cycling.

Chapterr 6 provides a direct comparison of the chronotropic functions of two types

off sensors: the peak endocardial acceleration (PEA) and the activity sensor.

24 4

Chapterr ]

Partt II Studies on biventricular pacing

Chapterr 7 examined whether cardiac resynchronization therapy induces

neurohumorall improvements in patients with heart failure and cardiac

dyssynchrony. .

Chapterr 8 investigated the sub-lingual microcirculatory changes in heart failure

patientss as a result to cardiac resynchronization therapy and right ventricular

pacingg by use of orthogonal polarization spectral imaging.

Chapterr 9 described one of the first patients who was upgraded to biventricular

pacingg after worsening of his heart failure with increase of mitral regurgitation due

too WIR pacing after His bundle ablation. This chapter also reviews the literature

concerningg reversed remodeling of the left ventricle after biventricular pacing.

Chapterr 10 and 11 summarizes and discusses the main findings derived from the

studiess described in the previous chapters (english and dutch).

25 5

References s

1.. Senning A. Cardiac pacing in retrospect. Am } Surg. 1983:145:733-9-2.. Mercuriale G. De cognoscendis e curandis humani corporis affectibus. opera postuma.

1606:238.. 242. 243. 3.. Galvani L. De viribus electricitatis in motu musculari commentaris. Bologna Instit Scient.

1771. . 4.. Bichat X. Recherches physiologiques sur la vie et la mort. Paris:Brosson. Gabon & Cie. 1800. 5.. Jeffrey K, Parsonnet V. Cardiac pacing, 1960-1985: a quarter century of medical and industrial

innovation.. Circulation. 1998;97:1978-91. 6.. Hyman A. Resuscitation of the stopped heart by intracardial therapy. Experimental use of an

artificiall pacemaker. Arch Intern Med:50:283-305-7.. Zoll PM. Resuscitation of the heart in ventricular standstill by external electric stimulation. N

EnglEngl J Med. 1952:247:768-71. 8.. Elmqvist R. Review of early pacemaker development. Pacing Clin Electrophysiol. 1978:1:535-6. 9.. Chardack WM, Gage AA. Greatbatch W. A transistorized, self-contained, implantable pacemak-

err for the long-term correction of complete heart block. Surgery. 1960:48:643-54. 10.. Zoll PM, Frank HA. Zarsky IX Linenthal AJ, Belgard AH. Long-term electric stimulation of the

heartt for Stokes-Adams disease. Ann Surg. 1961:154:330-46. 11.. Mond HG, Irwin M. Morili o C, Ector H. The world survey of cardiac pacing and cardioverter

defibrillators:: calendar year 2001. Pacing Clin Electrophysiol. 2004:27:955-64. 12.. Nathan DA, Center S, Wu CY, Keiler W. An implantable synchronous pacemaker for the long

termm correction of complete heart block. Am J Cardiol. 1963:11:362-7. 13-- Zuckerman W. Zaroff LI. Berkovits BV, Matloff ]M. Harken DE. Clinical experiences with a new

implantablee demand pacemaker. Am J Cardiol. 1967;20:232-8. 14.. Funke H. Optimized sequential pacing of atria and ventricles: A new concept in treatment of

bradycardiacc dysrhythmias. Herz Kreislauf. 1978:10:479-483. 15.. Furman S. Schwedel JB. An intracardiac pacemaker for Stokes-Adams seizures. AT Engl J Med.

1959:261:943-8. . 16.. Furman S, Parker B. Escher DJ. Schweldel [B. Instruments for evaluating function of cardiac

pacemakers.. Med Res Eng. 1967:6:29-32. 17.. Furman S, Parker B, Escher DJ. Transtelephone pacemaker clinic. N YState ]Med.

1971:71:1931-2. . 18.. Parsonnet V. Cuddy TE. Escher DJ. Furman S, Morse D, Gilbert L. Zucker IR. A permanent

pacemakerr capable of external non-invasive programming. Trans Am Soc Artif Intern Organs. 1973:19:224-8. .

19.. Ellenbogen KA. Cardiac pacing. 1996:66-88. 20.. Chardack WM. A Myocardial Electrode for Long-Term Pacemaking. Ann N Y Acad Sci.

1964:111:893-906. . 21.. Hunter SW, Roth NA. Bernardez D, Noble JL. A bipolar myocardial electrode for complete

heartt block. J Lancet. 1959:79:506-8. 22.. Ellenbogen K. Wilkoff . Clinical cardiac pacing and defibrillation. 2000:127-147. 23.. Mond HG. Unipolar versus bipolar pacing—poles apart. Pacing Clin Electrophysiol.

1991;14:1411-24. . 24.. de Voogt WG. Pacemaker leads: performance and progress. Am J Cardiol. 1999:83:187D-191D. 25.. Greatbatch W. Lee JH, Mathias W, Eldridge M, Moser JR, Schneider AA. The solid-state lithium

battery:: a new improved chemical power source for implantable cardiac pacemakers. IEEE TransTrans Biomed Eng. 1971:18:317-23.

26.. Parsonnet V, Myers GH, Gilbert L, Zucker IR. Clinical experience with nuclear pacemakers. Surgery.Surgery. 1975:78:776-86.

26 6

Chapterr 1

27.. Gregoratos G, Abrams J. Epstein AE, Freedman RA. Hayes DL. Hlatky MA, Kerber RE. Naccarel-lii GV. Schoenfeld MH, Silka MJ. Winters SL, Gibbons RI. Antman EM. Alpert JS, Hiratzka LF, Faxonn DP. Jacobs AK. Fuster V, Smith SC. Jr. ACC/AHA/NASPE 2002 guideline update for im-plantationn of cardiac pacemakers and antiarrhythmia devices; summary article. A report of thee American College of Cardiology/American Heart Association Task Force on Practice Guide-liness tACC/AHA/NASPE Committee to Update the 1998 Pacemaker Guidelines), J Cardiovasc ElectiophysiolElectiophysiol 2002:13:1183-99.

28.. Hayes DL. Furman S. Cardiac pacing: how it started, where we are. where we are going. Pacing ClinClin Electrophysiol 2004:27:693-704.

29.. Skanes AC, Krahn AD, Yee R. Klein G). Connolly SJ. Kerr CR. Gent M, Thorpe KE Roberts RS. Progressionn to chronic atrial fibrillation after pacing: the Canadian Trial of Physiologic Pacing. CTOPPP Investigators. J Am Coll Cardiol. 2001:38:167-72.

30.. Newman D. Lau C. Tang AS. Irvine J. Paquette M. Woodend K, Dorian P, Gent M, Kerr C, Con-nollyy SJ. Effect of pacing mode on health-related quality of lif e in the Canadian Trial of Physi-ologicc Pacing. Am Heart J. 2003:145:430-7.

31.. Hussein SJ, Hennekens CH, Lamas GA, An update on clinical trials in pacing: is dual chamber pacingg better? Curr Opin Cardiol 2004:19:12-8.

32.. Lamas GA, Lee KL, Sweeney MO, Silverman R, Leon A. Yee R. Marinchak RA, Flaker G, Schron E.. Orav EJ, Hellkamp AS, Greer S, McAnulty J, Ellenbogen K, Ehlert F, Freedman RA. Estes NA, 3rd,, Greenspon A, Goldman L. Ventricular pacing or dual-chamber pacing for sinus-node dys-function,, N Engl J Med. 2002:346:1854-62.

33.. Lamas GA. Orav EJ, Stambler BS. Ellenbogen KA, Sgarbossa EB. Huang SK, Marinchak RA, Estes NA.. 3rd, Mitchell GF, Lieberman EH, Mangione CM, Goldman L. Quality of lif e and clinical out-comess in elderly patients treated with ventricular pacing as compared with dual-chamber pac-ing.. Pacemaker Selection in the Elderly Investigators. N Engl J Med. 1998:338:1097-104.

34.. Wilkoff BL. The Dual Chamber and WI Implantable Defibrillator (DAVID) Trial: rationale, de-sign,, results, clinical implications and lessons for future trials. Card Electrophysiol Rev. 2003;7:468-72. .

35.. Andersen HR, Thuesen L, Bagger JP, Vesterlund T, Thomsen PE. Prospective randomised trial of atriall versus ventricular pacing in sick-sinus syndrome. Lancet. 1994:344:1523-8.

36.. Toff WD, Skehan JD. De Bono DP, Camm AJ. The United Kingdom pacing and cardiovascular eventss (UKPACE) trial. United Kingdom Pacing and Cardiovascular Events. Heart. 1997:78:221-3.

37.. Kosowsky BD, Scherlag BJ, Damato AN. Re-evaluation of the atrial contribution to ventricular function:: study using His bundle pacing. Am J Cardiol. 1968:21:518-24.

38.. Bakker PF dJN, van Mechelen R. Wittkampf F. Mower M, Thomas A. Beneficial effects of biv-entricularr pacing incongestive heart failure. PACE. 1994:17:20.

39.. Bakker PF MH, de Jonge N, van Mechelen R, wittkampf F, Mower M. Thomas A. Beneficial ef-fectss of biventricular pacing in congestive heart failure. PACE. 1994; 17:20,

40.. Moss A, Rivers R. Jr, Kramer D. Permanent pervenous atrial pacing from the coronary vein. Long-termm follow-up. Circulation. 1974:49:222-225-

41.. Prinzen FW, Delhaas T, Arts T, Reneman RS. Asymmetrical changes in ventricular wall mass by asynchronouss electrical activation of the heart. Adv Exp Med Biol. 1993:346:257-64.

42.. Erol-Yilmaz A, Tukkie R. Schrama TA. Romkes HJ. Wilde AA. Reversed remodelling of dilated leftt sided cardiomyopathy after upgrading from WIR to WIR biventricular pacing. Europace. 2002;4:445-9. .

43.. Auricchio A, Stellbrink C, Sack S, Block M. Vogt J, Bakker P. Mortensen P. Klein H. The Pacing Therapiess for Congestive Heart Failure (PATH-CHF) study: rationale, design, and endpoints of aa prospective randomized multicenter study. Am ƒ Cardiol. 1999;83;130D-135D.

44.. Auricchio A. Stellbrink C, Sack S. Block M, Vogt J. Bakker P, Huth C, Schondube F, Wolfhard U, Boekerr D. Krahnefeld O, Kirkels H, Long-term clinical effect of hemodynamically optimized cardiacc ^synchronization therapy in patients with heart failure and ventricular conduction delay.. J Am Coll Cardiol. 2002;39:2026-33.

45.. Lau CP, Barold S, Tse HF, Lee KL, Chan HW. Fan K, Chau E. Yu CM. Advances in devices for car-diacc ^synchronization in heart failure, ƒ Interv Card Electrophysiol. 2003;9:167-81,

21 21

46.. Salukhe TV, Francis DP, Sutton R. Comparison of medical therapy, pacing and defibrillation in heartt failure {COMPANION! trial terminated early: combined biventricular pacemaker-defi-brillatorss reduce all-cause mortality and hospitalization. Int f Cardiol. 2003:87:119-20.

47.. Bristow MR, Saxon LA. Boehmer J, Krueger S, Kass DA. De Marco T. Carson P. DiCarlo L. DeMetss D. White BG. DeVries DW. Feldman AM. Cardiac-resynchromzation therapy with or withoutt an implantable defibrillator in advanced chronic heart failure. NT Engl J Med. 2004:350:2140-50. .

48.. Bradley DJ, Bradley EA, Baughman KL. Berger RD. Calkins H, Goodman SN. Kass DA, Powe NR. Cardiacc resynchronization and death from progressive heart failure: a meta-analysis of rand-omizedd controlled trials. Jama. 2003:289:730-40.

49.. Cleland JG. Ghosh J, Khan N, Hurren S. Kaye G. Ongoing trials of cardiac resynchronisation. MinervaMinerva Cardioangiol. 2003:51:197-207.

50.. Peichl P. Kautzner J, Cihak R, Bytesnik J. The spectrum of inter- and intraventricular conduc-tionn abnormalities in patients eligible for cardiac resynchronization therapy. Pacing Clin Elec-trophysiol.trophysiol. 2004:27:1105-12.

51.. Bax [J. Ansalone G. Breithardt OA, Derumeaux G, Leclercq C, Schalij M], Sogaard P, St John Suttonn M. Nihoyannopoulos P. Echocardiographic evaluation of cardiac resynchronization therapy:: ready for routine clinical use? A critical appraisal. ƒ Am Coll Cardiol. 2004;44:1-9.

52.. Vernooy K, Verbeek XA. Peschar M, Prinzen FW. Relation between abnormal ventricular impulsee conduction and heart failure. / Interv Cardiol. 2003:16:557-62.

x x

Iss manual rate response optimization necessaryy in current rate adaptive

pacemakers? ?

Aytenn Erol-Yilmaz MD and Raymond Tukkie MD PhD

SubmittedSubmitted for publication

Abstract t

Background d

Automaticc functions are increasing in rate adaptive pacemakers. Whether we can

relyy on these automatic functions in daily practice is limited described. We

thereforee review in this article various types of sensors and try to answer the

questionn if manual rate response (RR) optimization improves patient outcome and

iss still necessary with the existing automaticity in RR pacemakers,

Material ss and methods

AA literature search was performed in journal articles published in Medline using the

keywordss rate adaptive pacemakers, sensors, automaticity, optimization, quality of

lif ee (QOL) and exercise.

Results s

Despitee sophisticated sensor systems and algorithms, today there is still as yet no

sensorr system that can simulate ideal sinus rhythm behavior. The majority of

pacemakerss are programmed in default sensor setting, although there is evidence

thatt individual optimization of sensors is beneficial. Limited studies are done

clarifyingg the clinical benefit of automatic features. Although exercise tests and

QOLL questionnaires are most used for evaluating of RR optimization internationally,

clearr guidelines for optimization are lacking.

Conclusions s

Automaticc features can be helpful in reducing time during follow-up of pacemakers,

howeverr individually adjustment of pacing sensors is still necessary.

32 2

Chapterr 2

Introductio n n

Thee first rate adaptive pacemaker, capable of altering paced rate on the basis of

inputt from a sensor incorporated in a pulse generator was proposed by Krasner in

1966.11 In 1982, the first rate adaptive (RR) pacemakers were implanted for treatment

off chronotropic incompetence in bradyarrythmias and they became available for

generall use in the United States in 1986.

Onee of the aims of a pacemaker sensor is to simulate normal sinus response.

Despitee large number of implantable sensors to detect exercise that have been

proposedd or instrumented in pacing devices, there is as yet no single sensor that

cann simulate ideal sinus node behavior.2 Many sensors have failed to demonstrate

adequatee chronotropic response after experimental or preliminary clinical studies:

oxygenn saturation, and central venous temperature.3"5 In the surviving artificial

sensorss (activity, OT interval, minute ventilation, peak endocardial acceleration

(PEA)),, several limitations have been observed, related mainly to sensor

characteristics:: speed of response, sensitivity to exercise versus no exercise

stimulation,, specificity to increases in oxygen demand, proportionality and

responsee to different exercise modalities.2 6"9 However, optimal operation of these

pacemakerss depends on correct functioning and programming of the implanted

sensor.. In the future, the role of sensors could expand to include functions other

thann rate augmentation, such as monitoring of cardiac hemodynamics during heart

failure,, which is beyond the scope of this article.

Automaticc functions are increasing in RR pacemakers. Whether we can rely on these

automaticc functions in daily practice is limitedly described. The last review about

RRR pacing dates from 2000,7 The purpose of this review is to outline if manual RR

optimizationn improves patient outcome and is still necessary with the existing

automaticityy in RR pacemakers. The various types of sensors, the effects of manual

sensorr optimization, the tools for RR optimization, automaticity and automatic

optimizationn of sensors are reviewed. Search is performed in journal articles

publishedd between 1966 and 2004 indexed in Medline using the keywords rate

adaptivee pacemakers, sensors, automaticity. optimization, QOL and exercise.

33 3

Algorithmss and rate response

Ratee adaptive pacemakers have two basic components, the sensor Is) and

algorithms.. Sensors detects raw physical and physiological changes and algorithms

transformm these raw sensor data into a corresponding heart rate.2'8 Q The obtained

raww sensor data are first filtered to exclude unwanted signals and then modified

appropriately.. These filtered signals are then converted into a rate response (RR)

curvee e.g. linear, curvilinear or tri-phasic. In accessible pacemakers, the physician

cann modify the RR by changing this filter.

Twoo different algorithms are used in practice: 1) counting energy-peaks, which pass

throughh the threshold according to the programmable thresholds 'low, medium or

high'.. 2) integrating the voltage graph over a standardized interval. Corresponding

too the programmable threshold, a specific value is subtracted from the computed

integral.. The remaining signal is transformed into the stimulation rate.

Theree are two ways (open or closed loop systems) in a RR pacemaker, which can

drivee an algorithm to induce an HR. An open loop system detects a physiological or

physicall change and is used in most currently available sensors. This change in turn

initiatess an algorithm where after a HR is created. Dependent on how sensor

settingss are programmed by the physician a specific HR wil l be produced. An open

loopp system is easily implemented and usually does not require special pacing

electrodes.. An example is an activity sensing system that detects body movements.

AA closed loop system detects a physiological change, for instance blood

temperature,, and reacts with a negative feedback. This negative feedback induces a

physiologicall change in opposite direction, so that the physiological variation

responsiblee for the rate change wil l return to the baseline condition. The

physician'ss role in obtaining a clinically desired rate is in this system minimal.

Activityy based sensors

ActivityActivity sensor

Jacquess and Pierre Curie first discovered piezoelectricity in 1880 when they struck

quartzz with a hammer and noticed an electric signal. Today most piezoelectric

materialss are ceramics like barium titante or lead zirconate, however they are still

calledd crystals, and work in the same way: bending or deforming the crystal

34 4

Chapterr 2

producess an electric signal. By means of a piezoelectric crystal attached to the inside

off the pacemaker can (see figure la), vibrations above a programmable threshold are

treatedd as counts, the frequency of which can then be converted into a pacing rate

byy a series of programmable slope of rate responses. This is a mechanical sensor,

whichh detects physical activity signals in a range of 1-5 Hz so electrical inputs

intoo the enclosure, such as myopotentials. are not sensed. In the LIVIN G I (Sorin

Biomedica.. Sallugia, Italy) is the physical activity sensor of a gravitational type and

thereforee particularly sensitive to vertical displacement of the body. Functioning of

thiss sensor is based on the opening/closing of a mobile contact consisting of a

droplett of mercury of appropriate inertial mass. The physical/dimensional

characteristicss of the sensor and associated electronic circuit are designed to select

signalss with rates typical of physical activity (2-5 Hz).

Humenn et al. reported the initial clinical experience of 6 patients implanted with

thee piezoelectric crystal.' Improvement in cardiac output was observed in all

patientss in the rate responsive mode. However, they were unable to demonstrate

improvementt of exercise capacity in their patients because exercise tolerance of

somee patients was limited by noncardiac factors such as intermittent claudication.

Inn subsequent larger studies, improvement of exercise capacity has been

demonstratedd beside the reported symptomatic benefit.11 '2

Thee main advantage of the activity sensor is the rapid onset of RR at the onset of

isotonicc exercise. A disadvantage of this sensor is that the RR of the sensor does not

correlatee closely with sinus rate (see table I and 2). Especially during walking up an

inclinee or cycling, inadequate low HR are achieved with this sensor while riding a

carr can induce inappropriate high HR.13 Thus although activity-sensing RR pacing

givess a prompt increase in pacing rate and improves maximum exercise tolerance,

furtherr refinement is necessary because detection of vibrations as indicator of

activitiess does not correlate well with the level of exertion. 14 These limitations

aree probably due to direct attachment of the crystal to the pacemaker can and to

nonselectivee processing of vibration signals.

Electronics s

Activity y

Figuree l.a: Activity sensor mounted inside in the pacemakerr can.

Battery y

35 5

Accelerometer Accelerometer

Measurablee acceleration forces applied to the human body during different types of

physicall stress were published for the first time in 1988.1:,Accelerometer-based RR

stimulatorss were then developed using sensors directly mounted on the hybrid

circuitt (plate with electronic circuitry of the pacemaker without battery) coupled

withh low pass filters, see figure lb.1 3 , 1 6'18

Betweenn accelerometers there are differences in: 1) which direction of the body

accelerationn is measured: 2) range of the low pass filters used (1-10 Hz) and 3) the

designn of the accelerometer. The different directions nowadays used for body

accelerationn are anteroposterior, horizontal, vertical or biaxial. There are also

triaxiall accelerometers but as far as we now not yet used in pacemakers.1

Triaxiall accelerometers can have advantages in RR pacing due to improvement in

describingg physical activity and as a consequence the RR. Steele et al. showed that

thee triaxial accelerometer has the potential to provide more precise measurement of

dailyy physical functioning in COPD patients.

Thee response of the accelerometer to exercise is rapid and can accommodate brief

activities,, such as running and stair-climbing in which a brisk increase in HR is

requiredd and adapts pacing rates during treadmill exercises independent of

treadmilll speed or slope better than those controlled by a conventional housing

pressuree or vibration sensor.~' However, physiological activities such as handgrip,

valsalvaa maneuver, are not associated with significant vibrations, therefore the

pacingg rate does not correlate well with the level of exertion. Furthermore,

accelerationn signal amplitudes could vary up to 20% from patient to patient during

comparablee treadmill tests.22 Different walking behaviors or inclinations of the can

onn the prepectoral area could account for these variations.23 Also the nature of

patient'ss footgear could influence acceleration levels. 2A Therefore, individually

calibratingg the accelerometer signals seems crucial.

Figuree l.b: Sensor mounted on hybrid circuit.

Accelerometer r Battery y

Electronics s

36 6

Chapterr 2

Physiologicc sensors

QTQT sensor

Rickardss et al. established the principle of using the paced evoked QT interval

(figuree 3) to determine pacing rate in 1983.2b This parameter has been used in the

developmentt of a QT sensing pacemaker (Vitatron, Arnhem, The Netherlands). The

pacedd QT (stimulus-t) interval is measured from the pacemaker spike to the

maximumm negative deflection of first derivate of the endocardial T wave (see figure

2).. An algorithm converts the change in stimulus-T interval into a change of pacing

rate.. There is a non-linear relationship between pacing and evoked QT intervals.

Thee principle of the evoked QT interval is that the QT interval shortens during

physicall exercise and mental stress. Approximately half of this shortening is

broughtt about by an increase in HR. but the other half is probably related to direct

effectt of catecholamines, such as those induced by emotional stress.27 2& Drugs,

myocardiall ischemia, and electrolyte changes vary also the duration of QT interval

andd hence the HR in the absence of changes on metabolic demand.29

Althoughh response to stress appears physiological, a potential disadvantage of this

sensorr could be undesirable tachycardia mediated by an increased sympathetic

tone,, occurring in stressful situations like angina pectoris or acute myocardial

infarction.30,311 Another disadvantage of this sensor is the slow response especially

inn the first stage of exercise, even though in new QT software with dynamic slopes

theree is a reduction of t ime delay of RR and also an improved correlation between

thee sinus activity and pacing rate in the dynamic algorithm. Close follow-up and

frequentt reprogramming, guided by results of exercise tests and holter monitoring,

mayy be necessary to adjust for chronic changes in QT parameters to ensure that the

sensorr continues to function optimally. Particularly because under identical

conditions,, the HR/QT and HR rate /stimulus-T interval relations vary significantly

inn different patients and in the same subject from one occasion to another,32

Thee QT sensor can also be used for AV- interval optimization which was

demonstratedd by Ishikawa et al.35 They studied the relationship between AV delay

andd evoked QT interval and cardiac function in 13 elderly patients with an

implantedd QT-driven DDDR pacemaker. A special pacemaker software module was

downloadedd into the pacemaker memory for evoked QT interval data logging.

37 7

AVV delay was set at 100, 120, 150, 180, 210 and 240 ms. Cardiac output was

measuredd by continuous wave Doppler echocardiography. The cardiac output was

maximall when AV delay was set at the AV delay at which the evoked QT interval was

maximal,, They conclude that the optimal AV delay can be predicted from the evoked

QTT interval sensed by an implanted pacemaker and automatic setting of the optimal

AVV delay can be achieved by the OT sensor of an implanted pacemaker.

kk >1 QTT interval

Figuree 2. OT-interval after a ventricular stimulus.

MinuteMinute ventilation sensor

Rossii et al. have shown that changes in the respiratory rate correlate significantly

withh changes in HR during exercise. 34 First respiratory pacemakers sensed the

respirationn rate using an auxiliary lead, which was implanted subcutaneous and

weree susceptible to erosion. Nappholtz et al. have shown the possibility of

intracardiacc measurement of MV using impedance measuring. Impedance is a

measuree of electric resistance and derived by measuring the resistance from an

injectedd electric current across a tissue (see figure 3). The impedance principle has

beenn used extensively for measuring respiratory parameters. Minute ventilation

(MV,, the product of respiratory rate and tidal volume) is derived in the pacemaker

byy using current pulses, injected between the proximal ventricular or atrial

electrodee and the pacemaker casing.

Lauu et al. described the initial clinical experience with the MV sensor in the Meta

pacemakerr (Telectronics-Cordis. Colorado, USA). They showed that rate modulated

pacingg by sensing MV resulted in better exercise capacity and symptomatology. ' In

thee new generation pacemakers, for example the Insignia ™ (Guidant Minnesota,

38 8

Chapterr 2

USA).. MV is still measured by transthoracical impedance changes. The sensed MV

changess are translated into a rate change using slope or RR factor. The main

disadvantagee of this senor is the initial delay of 30 seconds at the onset of exercise

beforee RR is observed, although nervous activity results in an almost instantaneous

changee of ventilation.

Differentt conditions with no direct relevance to cardiac output may affect

respiration,, and hence, the RR behavior of MV sensor. For example, Cheyne-Stokes

breathingg may be associated with inappropriate tachycardia during the rapid

breathingg phase. MV may be interrupted if the patient is talking during exercise and

duee to patient activities, such as upper limb and in particular arm movement. Other

factorss which can influence the RR obtained by the MV sensor are: concomitant

heartt failure, pulmonary disease, posture difference, breathing patterns, type of

exercise,, phonation and coughing.'8 Especially in patients with for example

hyperventilation,, pacemakers with other sensors may be preferable because

otherwisee accurate RR adjusting can be difficult .

Figuree 3- Transthoracic impedance mesurement. 1= current. V = voltage

PeakPeak endocardial acceleration sensor

Thee PEA sensor mounted at the tip of the right ventricular pacing lead, can measure

thee mechanical vibrations generated by the left ventricular myocardium during the

isovolumetricc contraction phase. The sensor detects natural heart acceleration,

whichh represents the endocardial acceleration at the interface between the lead tip

andd the myocardial wall (see figure 4). The peak-to-peak amplitude of the

39 9

endocardiall acceleration signal is detected during the isovolumetric ventricular

contractionn phase. Although the PEA parameter is a time period and not an actual

volumee measurement, PEA and stroke volume are closely related, and PEA gives an

indirectt indication of the contractility of the myocardium.39Contractility is sensitive

too adrenergic stimulation that accompanies increase in metabolic needs and can

supplyy relevant information regarding the patient's hemodynamics.39 The PEA

signall is not influenced by HR but is significantly increased by emotional stress,

exercisee stress, inotropic stimulation and follows linearly the changes in maximum

leftt ventricular dP/dt. In addition this sensor appears to measure global left

ventricularr contractile performance rather than regional mechanics.

Dataa from the European multicenter study of 1-month and 1-year confirmed that

PEAA is a stable long-term parameter that can be used for physiological driving of a

RRR pacemaker. The evaluation of the study confirmed the absence of adverse long-

termm effects, and the reliability of the sensor, lead sensing and pacing performances

remainedd within the expected range.6,40,41 Clementy et al. validated the PEA sensor

withh serial standardized exercise testing and showed that PEA sensor successfully

restoredd chronotropic response in a population of paced patients with severe

chronotropicc incompetence.42

Rickardss et. al. suggest that the PEA reflects the underlying contractile state of the

entiree heart and is not sensitive to local myocardial properties at the location of the

accelerometer,, however it must be considered that they worked with favourable

conditionss like normal geometry of the ventricles. Interpreting the results of the

PEAA measurements can be more difficult in patients with heart failure with valve

insufficiencyy or regional contractile dysfunction. Another limitation in general, is

thee use of uniaxial acceleration measurements because ventricle displacement and

thereforee acceleration follows a very complicated three-dimensional trajectory. The

possibilityy that the three-dimensional amplitude response may be different from

thee one dimensional measurement cannot be ruled out although in some studies,

PEAA is described as insensitive to local myocardial properties or the location of the

accelerometerr lead."13

Besidee appropriate use in RR pacing, the PEA signal results in new options such as

automaticc AV delay optimization. Previous studies showed a close relationship

betweenn PEA, first heart sound, and the AV delay. A louder first heart sound is

observedd with short AV delays as the ventricle contracts after atrial emptying.44 45

Dupuiss et al. showed that the optimal AV delay estimates obtained by the PEA

40 0

Chapterr 2

analysiss during automatic AV delay scanning are consistent with those obtained by

echocardiography.. 45 Recent clinical trials have shown that selected patients with

recurrentt vasovagal syncope may benefit from permanent pacing with a PEA sensor.

Thee authors demonstrated using head-up til t testing that increase in sympathetic

activityy preceding syncope could be sensed by a PEA sensor. 4 A1 The PEA sensor is a

promisingg tool for long-term hemodynamic monitoring and serial evaluation of the

effectss of multisite ventricular pacing in heart failure patients.48 Another possibility

withh the PEA signal is the estimation of the aortic diastolic pressure with the PEA II

signall ( = the abrupt deceleration of the moving aortic blood mass).49 Disadvantages

off the sensor are mostly related to the lead characteristics (lead body size) and the

sensorr location. The PEA sensor is located at the lead tip. consequently sensor

failuree necessitates a new lead insertion.

Forcee transducer Electronics s

Pacingg tip Rigidd and hermetic can Microo mass

Inert iall acceleration

Figuree 4. PEA sensor system.

Sensorr combination

Thee achieved rate profile during treadmill exercise testing resulting from dual

sensorr pacing (with complementary properties) is improved over single sensor

pacingg because sensor combination provides improvement in speed,

proportionality,, sensitivity and specificity.9 5 Combinations include association of

ann activity sensor giving an aggressive response for exercise that is light and/or of

shortt duration and a non-activity sensor, e.g., OT interval or minute ventilation,

addedd to provide a delayed but proportional and stable acceleration to sustained

exercisee and deceleration during recovery.2

Thee association of two sensors allows compensation of failure of one by the other.

41 1

Combiningg sensors is theoretically superior, but no clinical study has yet

demonstratedd this superiority in terms of quality of life. As far as proportionality

andd speed of RR are concerned, the current sensor combinations are an

improvementt over single sensor pacing.

Sensorr blending and cross-checking

Combiningg sensors with different RR properties requires adequate blending of

respectivee sensor activities. The sensor blending could be performed in several ways.

Blendingg can be performed at signal production as for example in the blending system

usedused in the Vitatron devices (Vitatron, Arnhem. The Netherlands), combining QT

intervall and activity in 5 different possibilities (activity, activity > QT interval,

activity=QTT interval, activity <QT interval, QT interval). This blended signal is

transmittedd to the algorithm working with a variable automatic slope. The resulting

signall that is transmitted to the algorithm is a mixture of a percentage of activity

sensorr signal with an inverse percentage of QT sensor signal.51

Anotherr possibility for sensor blending is prioritisation as in Medtronic Kappa series.

Thee activity sensor initially accelerates the rate from the lower rate to a daily life

plateauu (programmable 90-95 beats/min). The rate returns to the lower rate if effort

stopss or is accelerated proportionally from the plateau by the MV sensor up to the

maximumm sensor rate, which then is in charge of the recovery rate decrease. Sensor

crosss checking is used to avoid inappropriate rate increase.

Thee insignia I (Guidant, Minnesota, USA) combines quantitative blending and

prioritisationn with cross checking.52 At the beginning of an effort, the signal of the

activityy sensor is usually faster than the MV sensor signal. At lower rates, the

influencee of the acceleration sensor is higher than MV. As long as the activity signal

iss higher than the MV signal, the magnitude of the RR provided by the activity signal

iss still higher then the influence of activity sensor controlled by the MV sensor

(prioritisation).. If. at maximum sensor rate, the activity signal is still higher, then

thee influence of accelerometer is lower than MV (cross checking). When the MV

signall is higher than the activity signal, a steady state or recuperation is assumed

andd RR is based on MV dependence. The physiologic relationship between MV and

ratee is approximately bilinear. During exercise levels up to the anaerobic threshold,

thiss relationship can be approximated by a linear relationship. At exertion levels

abovee the anaerobic threshold, the slope becomes less pronounced. The

42 2

Chapterr 2

relationshipp between the secondary slope and the primar y slope varies from person

too person and depends on several factors such as gender, age, and exercise

frequencyy and intensity.

Thee importance of optimized programming is illustrated by the study of Lau et al.

Hee studied the effect of sensor cross-checking in a dual sensor system with an

activit yy and QT sensor in 4 patients. Rate response setting of each sensor was

individuall yy optimized, and an equal rate contributio n for the OT and activity

sensorss (QT=activity ) was used. Three maximal treadmil l exercise tests were

performedd in random order in three different sensor blending setting, QT only.

QT=activity .. and in activity only. In the dual sensor mode, the time for onset of RR

(delayy time) was reduced compared with OT only. However, the time to reaching

50%% of RR in the low activity threshold dual sensor mode was delayed compared to

sensorr blending setting (QT = activity) and was similar to the QT only mode. Lau et

al.. showed that programming the activity threshold to low. results in sensor

crosscheckingg at rest, which limit s the speed of RR onset.

Tablel.. Advantages and disadvantages of currently available sensors

Sensors s Advantages s Disadvantages s

Activity/accelee rome ter #

QT T

Conventionall lead and implantt procedure Fastt response to changes inn exercise Simplee to program

Standardd bipolar pacing lead Metabolicc demandf Reactss to mental stress

-- Compromised ability to adapt too different workload levels

-- No response to emotion -- Displays not a circadian variation -- Occasional sensing of mechanical signalss unrelated to activity

-- Post-activity response dictated by pacemakerr circuitry

-- Operates only with paced events -- Can be used only in the ventricle -- Slow reaction*

MV V Closee relationship to oxygen - Pulmonary disease alters MV consumptionn response mechanism Metabolicc demandf - Slow reaction*

PEA A Metabolicc demandf Reactss to mental stress

Sensorr at lead tip Sensorr failure necessitates a neww lead insertion Sloww reaction*

#-Aaelerometerr has the same advantages and disadvantages as the activity sensor but with improvedd reaction on physical activity compared with the activity: MV-minute ventilation; PEAA = peak endocardial acceleration: slow reaction* = slow reaction especially at the beginning and cessationn of exercise: Metabolic demandf = Proportional to metabolic demand and circadian variation.

43 3

Tablee 2. Sensor characteristics

Sensorr Proportionality Speed Sensitivity Specificity

Activityy - + + Accelerometerr - + - -QTT - + MVV + + -PEAA + +

•-not :: +=good- + + = very good; = moderate ariation

Manuall rate response optimization

Essentiall factors for successful RR optimization are to inventarize: 1} the individual

needss of the patient, 2) the medical history, 3) the programmable parameters of the

pacemaker,, in particular the sensor behavior and 4) evaluation of the effect of RR

optimization. .

Vitall elements in recording the needs of the patient are to comply gender, age,

medicall history and activity level. Detailed information about the activity level in

dailyy life is of utmost importance. A helpful tool can be the activity log in the new

developedd pacemakers such as in the Insignia plus pacemaker (Guidant, Minnesota.

USA}.. This feature is especially important in patients with heart failure, because

reducedd daily activity has been identified as an independent marker of poor

prognosis. .

P rogrammablee pa r ame te r s

Mainn programmable variables of rate modulation are: pacing mode, responsiveness,

sensorr status, threshold, minimum and maximum HR, slope and recovery time.

Sensorr status informs about the possible programmable variables of the sensors such

ass sensor blending and cross-checking. Threshold determines the minimal level of

senorr input required to produce a HR. Minimum and maximum HR is dictated in a

pacemakerr by the lower rate limit (LRL) and the upper rate limit (URL). Slope governs

thee HR change within a time period for a given sensor input. Recovery time governs

thee time required to return to minimum HR after cessation of work.

Theree is little attention for the LRL in daily practice, however programming the LRL

accuratee can limit unnecessary developments of symptoms and contributes to a

smoothh exercise and energy saving. 28

44 4

Chapterr 2

Thee estimation of the maximal HR has been largely based on the Astrand formula

(maximumm HR-220-age).:n~sq Despite the acceptance of this formula, Robergs and

Landwehrr showed in research spanning more than two decades, the large error

inherentt in the estimation of maximum HR with this formula.60 Considering the

manyy factors which can influence the achieved maximal HR, careful programming of

thee maximal HR is needed.61 It bears repeating that this is only an estimate of an

individuall patient's maximum HR. Better methods need to be developed to estimate

especiallyy the HR responses at sub maximal levels because patients needing a

pacemakerr are of an age group where they may have disabilities limitin g their

physicall work capacity.

Scrutinyy the functioning of the pacemaker sensor(s) is indispensable. Detailed insight

inn direct reaction of sensors and the long- term adjustment of the algorithm to

changess of sensor settings is vital. For example in the dual sensor pacemakers of

Vitatronn (Arnhem, Netherlands) with the QT and the activity sensor there is 4 to 6

weekss needed to achieve maximal effect of the changed sensor blending When

evaluationn is prior to week 4, the physician has to take into account that the maximal

effectt is not yet achieved. Furthermore application of a ramdump of the manufacturer

cann be helpful in understanding the autoslope function from this system.

Evaluat ionn of sensor op t im iza t ion

Effectss of sensor adjustment can be evaluated with serial exercise testing and

qualityy of lif e questionnaires (QOL-q).

ExerciseExercise tests

Forr pacemaker optimization, tests are in use with largely isotonic (dynamic or

locomotory)) exercise. These exercise tests can be divided in: in-hospital and out-

hospitall exercise tests.

Fromm the in-hospital tests, the chronotropic assessment exercise protocol (CAEP)

exercisee test according to Wilkof f and the 6-minute walk test are frequently used.7 62

Otherr in hospital tests, which are not frequently used, are the LITE walk protocol.

shuttlee walk test, and the Kaltenbach test. Main disadvantage of in-hospital testing

iss that the tests poorly represent daily activities. Especially a treadmill test is not

ann ideal exercise test for pacemaker patients, because in pacemaker patients.

45 5

exercisee is often limited by loss of muscle strength and mass rather than

cardiopulmonaryy capacity. This loss of muscle strength is particularly apparent

whenn exercise testing is performed on a bicycle or treadmill. The CAEP exercise test

hass non-linear characteristics. The first 10 minutes requires low metabolic

workload,, beyond which the workload abruptly increases. Patients with preserved

functionall capacity, capable of exercising for more than 10 minutes, may quit before

reachingg maximal 02 uptake, mainly because of excessive increments in workload

nearr the end of the test. thus, being limited by mechanical rather than metabolic

barriers.. Caution is needed in the application of the CAEP protocol which is also

illustratedd by Freedman et al.6^ They showed that specifically failure in attaining

maximumm exercise could create the appearance of sub-optimal pacemaker

performance. .

Thee limited studies evaluating exercise duration and sensor optimization show no

significantt improvement in exercise duration after sensor optimization and

guideliness are lacking how to evaluate pacemaker function with exercise tests.

Thee guidelines of American College of Cardiology on exercise testing, only mention

thatt the optimum protocol for any test should last 6 to 12 minutes and should be

adjustedd to the subject's needs. International guidelines are needed to

standardizee pacemaker sensor optimization in all chronotropic incompetent

patients.. First, we suggest to adjust the pacemaker sensors individually. Second, to

adjustt the sensors separate in dual sensor systems to gain insight in the

contributionn of each sensor. Third, using preferably the holter capacities of the

currentt pacemakers, monitoring sensor function during daily activities at home,

mayy visualize the true activity level of the patients better.

QualityQuality of life

Qualityy of lif e is a relatively new scientific measure to evaluate effectiveness of

treatment.. QOL reflects the patients perceived health and may considerably

improvee the evaluation of optimal RR.

Usingg OOl-q in international studies is a challenge since interpreting the data could

posee several difficulties. Most important is that the generalizability of health

conceptss between different countries and cultures are being assumed without data

basedd evaluations. This complicates comparing studies using different OOL-q's

andd methods for assessing OOL (transtelephonic, self administered at hospital by

46 6

Chapterr 2

researcherr /patient). Therefore, it is advisable to interpret QOL data in pacemaker

patientss cautiously. b 9 ' °

Inn general, there is a lack of validated pacemaker patient specified OOL-q.' To date.

onlyy two questionnaires were specifically developed for pacemaker patients

(Hacettepee Qol-q71, Aquarel QOL-q).72 Oto et al. developed the Hacettepe OOL-q,

whichh is a pacemaker patient specific QOl-q. In his study. 11 patients were

randomlyy assigned to a crossover study in order to assess their overall exercise

capacityy and OOL scores. They concluded that WIR pacing offers a better QOL in

additionn to an improved exercise capacity, compared to WI pacing.

Thee Karolinska QOL-q and the recently developed Aquarel QOL-q are also disease

specific.72'744 The Aquarel Qol-q is specifically designed for the pacemaker patient.

Thee Karolinska QOL -q is in our opinion complex to fil l in because of the use of

manyy visual analogue scales. The Hacettepe OOL-q and the Aquarel QOl-q were only

usedd domestic with few patients and the Hacettepe Ool-q has a lack of questions on

arrhythmias/ /

Thee study of Sulke et al. is yet the only study, which primarily assessed the effect of

RRR pacing on OOL. Appropriate, over and under programming of the RR in both dual

andd single chamber activity sensor rate adaptive pacemakers was evaluated with

visuall analogue scales and specific activity questionnaires. 5 Symptoms were least

inn the appropriate programmed pacemakers after 2 weeks of follow up.75

Inn addition, in most pacemaker studies their is a discrepancy in results attained

fromm exercise tests and QOL-q. ' •" Improvement in exercise capacity is not always

achievedd together with improvements in QOL-q. For example, studies on cardiac

resynchronisationn therapy have shown only minor changes in exercise capacity yet

majorr changes in QOL. This can be explained by several factors: 1) QOl-q used are

nott sensitive enough for measuring the effect of optimization as described above. 2)

typee of in-hospital exercise tests are different compared to the type of exercise in

dailyy activities. We therefore think that evaluating RR optimization using daily

activitiess is superior to the known in-hospital exercise tests.

Effectss of manual vs. automatic RR optimization

Automaticityy in antibradycardia pacemakers started in the early 1960s with the

introductionn of inhibited ventricular pacing mode. '8 Today, numerous new-

automatedd features and algorithms have been developed.'' ' The introduction of

47 47

digitall pacemakers promises new possibilities in the future. Automaticity could

helpp in reducing follow up time and simplify programming. Accessibility of

pacemakerss remains crucial to verify that RR pacemakers respond correctly.

Thee majority of pacemaker sensors remain at the original programmed settings of

thee manufacturer and are never reprogrammed during their entire lifespan, despite

thee fact that several studies have shown the benefit of appropriate programming of

RR.. ^ '83 Table 3 summarizes the studies affecting manually vs. automatic RR

optimization.. It is striking that most studies do not clarify how RR optimization was

performed.. RR optimization was carried out with instructions of the manufacturer

andd or to the discretion of the physician without using standardized protocols for

optimization,, Most studies focus on programming the URL and do not describe how

otherr RR parameters such as threshold and slope were changed. Thereby the

influencee of the programming sequence of RR parameters is also not described.

Inn most trials, in-hospital exercise tests are used for the evaluation of RR. which has

disadvantages,, as described before in this review. Other limitations of these studies

aree the lack of randomised trials and the predominance of single sensor systems.

Onee of the first studies on autoprogrammability was the study of Mahaux et al.24

Hiss group descibed in 11 patients the autoset function of the Sensolog 703 and

evaluatedd after 6 and 10 months of follow-up whether manual adjustment was still

needed.. The autoset function was time consuming (15-48 min) and in most patients

manuallyy optimization was still necessary. Other larger studies comparing

automaticc with manual adjustment confirmed these results that manual

adjustmentss remained needed. Automatic adjustment lowered in the study of

Kloniss follow-up time in contrast to the first study on autoprogrammability of

Mahauxx et al.24'84-85

Schuchertt et al. showed in a study with 37 patients after 1 month follow up. no

differencee in complaints or activity level when randomized to fixed activity RR

programmingg according to the physicians judgment of patient's life style and

automaticc RR programming in the Thera DR (Medtronic. Minneapolis. USA). &b

Leungg et al.' compared automatic sensor adaptation with individually optimization

inn 9 patients with DX2 pacemakers (Medtronic. Inc., Minneapolis, USA) with ACT

andd MV dual sensor systems. ' The accuracy of rate profile optimization versus

manuall programming was assessed at 1 month and there was no significant

differencee between the two methods of programming. The lack of demonstrating

anyy differences in these two studies between automatic vs. manual adjustment of

48 8

Chapterr 2

pacemakerr sensors may be related to some limitation s of the studies. First, the

follow-upp period of 1 month in both studies could be short to evaluate RR

optimization,, longer follow-up may be needed to detect some differences. Second,

Tablee 3. Summary of studies with manual and automatic sensor optimization

Study y

Mahaux, ,

19892'1 1

NN = ] l

Sulke. . 1990 ^

NN = 20

Lau. .

19966 "

NN = 4

Gentzler. .

1996" "

NN = 93

Schuchert t

199S7t t

NN = 37

Leung. .

19988 ^7

N=9 9

Studyy aim

Too study if manually

adjustmentt is

neededd with autoset

functionn of Sensolog

7Ü3 3

Too study die effect of

appropriate,, under

andd over-program-

mingg of RR

Too study the effects

off manually

adjustmentt of

activityy sensor on RR

kineticss of a OTand

activityy dual sensor

systemm with sensor

cross-- checking

algorithm m

Too evaluate the

automaticc sensor

adjustment t

Too compare the

continouss auto-

maticc adjustment of

activityy RR with

fixedd activity RR

programming g

Too study the

accuracyy of auto-

maticc RR optimiza-

tionn vs manual

programming g

Usedd test/

parameter r

AÜL L

Hoo I ter

TT T

Histograms s

TT T

ADL L

QOL L

TT T

ADL L

Briskk 1-

minutee test

Sensor r

indicatedd rate

histogram m

ADL L

Halll walk

'IT T

ADL L

Optimization n

guide e

Automatic c

withh MG

Manual l

withh MG

Manual l

withh DP

Automatic c

withh DP

Manuall with

DP/Automatic c

Manuall with

DP/Automatic c

Parameters s

optimized d

Reactionn time

Threshold d

Slope e

Recoveryy time

Reactionn time

Threshold d

Slopee URL

Sensorr blending

(QT/activity) )

Slope e

Activit yy ra te|

LRL L

Threshold d

Slope e

Recovery y

URL L

Fu u

6.10 0

months s

22 weeks

Direct t

1.3.6 6

months s

11 month

1.3 3 months s

Conclusion n

Automatic c

programmingg is time

consuming. .

Manuallyy adjust-

mentt needed in

mostt patients.

Appropriate e

programmingg is

betterr than under

andd overprogram-

ming g

Programmingg the

activityy threshold to

loww can induce

cross-checkingg at

rest,, which gives

delayedd RR onset

76%% of patients had

appropriate e

autoslope e

Noo difference in

complaintss or

activityy level of

patients s

Automaticc and

Manuall RR

optimizationn are

comparable e

TableTable continued on next page

49 9

Tablee 3. (continued)

Study y

Garrigue. .

20022 "*

N'' = 66

Kloms. .

200 ^ "

KK = S7

Studyy aim

Too study the auto-

maticc calibration

functionn of Opus G*

accelerometer r

comparedd to

manuallyy adjusted

activityy sensors and

healthyy controls

Too compare

automaticc with

manuaii sensor

adjustment t

Usedd test.

parameter r

Rapidd walking

Stairr climbing

Walkk test.

Sensor r

indicatedd rate

Histogram m

Optimization n

tluiJe e

Manuall with

DP-Automatic c

Manuall with

DP/Automatic c

Parameters s

optimized d

Acceleration n

time e

Slope e

Recoveryy time

URL L

Threshold d

Slope e

FF Li eva-

luation n

1.. 3

months s

1,3.6 6

months s

Conclusion n

Automatic c

calibrationn of Opus

GG pacemaker better

thann manual

adiustment t

Autoo sensor

adiustment t

requiress less time

comparedd to

manuall optimized

MG== manufacturers guide: DP= discretion of the physician: ADL= activities daily life; TT =

treadmilll test; MV = minute ventilation: RR=rate response; Fu = follow-up; * -W I R pacemaker. Ela

Medical.. Montrouge. France); Activity ratef = according to patients life style (active, moderate active

orr less active), FU = follow-up.

inn the study of Schuchert et al., the participating physicians were highly

experiencedd in pacemaker follow up and changed already at discharge in most

patientss the nominal RR to an individual setting.86 Therefore, these study results

cannott be applied to patients in whom the activity-triggered RR generally remains at

nominall values.

Garriguee et al, is the only study which showed better daily life performances with

ann automatic sensor function (autocalibration function of the Opus G WIR

pacemaker.. Ela medical) than without.88 In 43 patients with the autocalibration

function,, the provided daily life performances (rapid walking, stair climbing) were

closerr to those of healthy controls, than 23 patients with activity sensor controlled

pacemakerss without the autocalibration function.88

Inn summary these study results, confirm that manual adjustment of pacemaker

sensorss is beneficial and illustrate that automatic RR optimization can reduce

follow-upp time

Chapterr 2

Conclusions s

Thee majority of pacemaker sensors remain at the original programmed settings of

thee manufacturer, however there is evidence that individually adjustment of

pacemakerr sensors improves exercise capacity and quality of life. It is preferable to

matchh the type of sensor with the individual patient before pacemaker

implantation.. The improvements in automaticity with sophisticated sensor systems

cann be helpful in reducing follow-up time of pacemakers. The development of a

sensorr system that can simulate the ideal sinus rhythm behaviour remains a challenge

forr scientists and manufacturers. International guidelines are needed to standardize

pacemakerr sensor optimization in all chronotropic incompetent patients.

Acknowledgments s

Thee authors thank Tim Schrama (pacemaker technician) for the fruitful discussions.

51 1

References s

1.. Voukydis PC. Krasner JL. A physiologically regulated cardiac pacemaker. Br ƒ Exp Pathol. 1967;48:118-23. .

2.. Clementy J. Barold SS, Garrigue S. Shah DC. )ais P. Le Metayer P. Haissaguerre M. Clinical significancee of multiple sensor options; rate response optimization, sensor blending, and trending.. Am J Cardiol, 1999:S3:l66D-171D.

3.. Alt E, Hirgstetter C, Heinz M. Theres H. Measurement of right ventricular blood temperature duringg exercise as a means of rate control in physiological pacemakers. Pacing Clin Eiectrophysiol.Eiectrophysiol. 1986:9:970-7.

4.. Alt E. Hirgstetter C, Heinz M, Theres H, Blomer H. Central venous blood temperature for rale controll of physiological pacemakers. / Cardiovasc Surg (Torino). 1988:29:80-8.

5.. Griffi n JC, Jutzy KR. Claude JP, Knutti JW. Central body temperature as a guide to optimal heartt rate. Pacing Clin Eiectrophysiol. 1983:6:498-501.

6.. Langenfeld H. Krein A, Kirstein M. Binner L. Peak endocardial acceleration-based clinical testingg of the "BEST" DDDR pacemaker. European PEA Clinical Investigation Group. Pacing ClinClin Eiectrophysiol. 1998:21:2187-91.

7.. Strobel JS, Kay GN. Programming of sensor driven pacemakers, Cardiol Clin. 2000:18:157-76. ix. S.. Chew PH, Bush DE, Engel BT. Talan MI, Abell RT. Overnight heart rate and cardiac function in

patientss with dual chamber pacemakers. Pacing Clm Eiectrophysiol. 1996:19:822-8 9-- Benditt DG. Mianulli M, Lurie K, Sakaguchi S, Adler S. Multiple-sensor systems for

physiologicc cardiac pacing. Ann Intern Med. 1994:121:960-8, 10.. Humen DP, Kostuk WJ. Klein GJ. Activity-sensing, rate-responsive pacing: improvement in

myocardiall performance with exercise. Pacing Clin Eiectrophysiol. 1985;8:52-9. 11.. Lau CP, Tse WS. Camm AJ. Clinical experience with Sensolog 703: a new activity sensing rate

responsivee pacemaker. Pacing Clin Eiectrophysiol. 1988:11:1444-55-12.. Frielingsdorf J. Dur P. Gerber AE, Vuilliomenet A. Bertel O. Physical work capacity with rate

responsivee ventricular pacing (WIR1 versus dual chamber pacing (DDD1 in patients with normall and diminished left ventricular function. Int} Cardiol. 1995:49:239-48.

13.. Lau CP. Stott JR. Toff WD, Zetlein MB. Ward DE. Camm A]. Selective vibration sensing: a new conceptt for activity-sensing rate- responsive pacing. Pacing Clin Eiectrophysiol. 1988:11:1299-309. .

14.. Lau CP. Mehta D. Toff WD. Stott RJ. Ward DE. Camm AJ. Limitations of rate response of an activity-sensingg rate-responsive pacemaker to different forms of activity. Pacing Clin Eiectrophysiol.Eiectrophysiol. 1988:11:141-50.

15.. Alt E. Matula M. Thilo R. Theres H, Heinz M, Blomer H. A new mechanical sensor for detectingg body activity and posture, suitable for rate responsive pacing. Pacing Clin Eiectrophysiol.Eiectrophysiol. 1988:11: 1S75-S1.

16.. Alt E. Matula M. Theres H, Heinz M. Baker R, The basis for activity controlled rate variable cardiacc pacemakers: an analysis of mechanical forces on the human body induced by exercise andd environment. Pacing Clin Eiectrophysiol. 1989:12:1667-80.

17.. Lau CP. Tai YT, Fong PC. Li jP. Leung SK. Chung FL, Song S. Clinical experience with an activity sensingg DDDR pacemaker using an accelerometer sensor. Pacing Clin Eiectrophysiol. 1992:15:334-43, ,

18.. Matula M. Holzer K. Zitzmann E, Schon H, Alt E, [Behavior of various activity-controlled cardiacc pacemakers in treadmill stress tests with variable slopes]. Z Kardiol. 1993:82:108-15-

19.. Steele BG, Holt L, Belza B, Ferris S. Lakshminaryan S, Buchner DM. Quantitating physical activityy in COPD using a triaxial accelerometer. Chest. 2000:117:1359-67.

20.. Mayagoitia RE, Lotters ]C. Veltink PH, Hermens H. Standing balance evaluation using a triaxiall accelerometer. Gait Posture. 2002:16:55-9.

21.. Alt E, Matula M, Holzer K. Behavior of different activity-based pacemakers during treadmill exercisee testing with variable slopes: a comparison of three activity-based pacing systems. PacingPacing Clin Eiectrophysiol. 1994:17:1761-70.

22.. Alt E, Matula M. Comparison of two activity-controlled rate-adaptive pacing principles: accelerationn versus vibration. Cardiol Clin. 1992:10:635-58.

Chapterr 2

23.. Schutz Y. Weinsier S. Terrier P. Durrer D. A new accelerometric method to assess the daily walkingg practice. Int } Obes Relat Metab Disord. 2002;26:111-8.

24.. Mahaux V. Waleffe A. Kulbertus HE. Clinical experience with a new activity sensing rate modulatedd pacemaker using autoprogrammability. Pacing Clin Electrophysiol. 1989:12:1362-8.

25.. Mahaux VA, Verboven YJ. Waleffe A. Kulbertus H. Stepwise analysis of the calibration proceduree of an accelerometer-based pacemaker. Pacing Clin Electrophysiol. 1994:17:1955-9.

26.. Rickards AF. Donaldson RM. Thalen HJ. The use of QT interval to determine pacing rate: early clinicall experience. Pacing Clin Electrophysiol 1983:6:346-56.

27.. Baig MW, Boute W, Begemann M, Perrins E]. Nonlinear relationship between pacing and evokedd QT intervals. Pacing Clin Electrophysiol. 1988:11:753-9.

28.. Jordaens L, Backers f. Moerman E, Clement DL. Catecholamine levels and pacing behavior of QT-drivenn pacemakers during exercise. Pacing Clin Electrophysiol. 1990:13:603-7.

29.. Hedman A. Nordlander R, Pehrsson SK. Changes in QT and Q-aT intervals at rest and during exercisee with different modes of cardiac pacing. Pacing Clin Electrophysiol. 1985;8:825-31.

30.. Edelstam C. Hedman A. Nordlander R, Pehrsson SK. QT sensing rate responsive pacing and myocardiall infarction: a case report. Pacing Clin Electrophysiol. 1989:12:502-4.

31.. Singer I. Olash ], Brennan AF, Kupersmith ). Initial clinical experience with a rate responsive pacemaker.. Pacing Clin Electrophysiol. 1989:12:1458-64,

32.. Fananapazir L, Bennett DH, Faragher EB. Contribution of heart rate to QT interval shortening duringg exercise. Eur Heart ƒ. 1983:4:265-71.

33.. Ishikawa T. Sugano T. Sumita S, Toda N. Kosuge M. Kobayashi I. Matsusita K. Ohkusu Y, Kimuraa K, Usui T. Umemura S. Optimal atrioventricular delay setting determined by QT sensorr of implanted DDDR pacemaker. Pacing Clin Electrophysiol. 2002:25:195-200.

34.. Rossi P, Plicchi G, Canducci G. Rognoni G, Aina F. Respiration as a reliable physiological sensorr for controlling cardiac pacing rate. Br Heart ]. 1984:51:7-14.

35.. Lau CP. Ward DE, Camm A[. Rate-responsive pacing with a pacemaker that detects respiratory ratee (Biorate); clinical advantages and complications. Clin Cardiol. 1988:11:318-24,

36.. Nappholtz T, Valenta H. Maloney J. Simmons T. Electrode configurations for a respiratory impedancee measurement suitable for rate responsive pacing. Pacing Clin Electrophysiol, 1986:9:960-4. .

37.. Lau CP, Antoniou A, Ward DE. Camm Af. Initial clinical experience with a minute ventilation sensingg rate modulated pacemaker; improvements in exercise capacity and symptomatology. PacingPacing Clin Electrophysiol. 1988:11:1815-22.

38.. Duru F, Radicke D. Wilkoff BL, Cole CR. Adler S. Nelson L, fensen DN, Strobel U. Portzline G. Candinass R. Influence of posture, breathing pattern, and type of exercise on minute ventilationn estimation by a pacemaker transthoracic impedance sensor. Pacing Clin Electrophysiol.Electrophysiol. 2000:23:1767-71.

39.. Bongiorni MG. Soldati E. Arena G. Quirino G, Vernazza F. Bernasconi A. Garberoglio B. Is local myocardiall contractility related to endocardial acceleration signals detected by a transvenous pacingg lead? PacingClin Electrophysiol. 1996:19:1682-8.

40.. Bennett T. Sharma A. Sutton R, Camm AJ, Erickson M. Beck R. Development of a rate adaptive pacemakerr based on the maximum rate-of- rise of right ventricular pressure iRV dP.'dtmax). PacingClinPacingClin Electrophysiol. 1992;15:219-34.

41.. Clementy }. Dual chamber rate responsive pacing system driven by contractility; final assessmentt after 1-year follow-up. The European PEA Clinical Investigation Group. Pacing ClinClin Electrophysiol. 1998:21:2192-7.

42.. Clementy J, Kobeissi A. Garrigue S, Jais P, Le Metayer P. Haissaguerre M. Validation by serial standardizedd testing of a new rate-responsive pacemaker sensor based on variations in myocardiall contractility. Europace. 2001:3:124-31.

43.. Rickards AF. Bombardini T, Corbucci G. Plicchi G. An implantable intracardiac accelerometer forr monitoring myocardial contractility. The Multicenter PEA Study Group. Pacing Clin Electrophysiol.Electrophysiol. 1996:19:2066-71.

44.. Ritter P, Padeletti L. Gillio-Meina L. Gaggini G. Determination of the optimal atrioventricular delayy in DDD pacing. Comparison between echo and peak endocardial acceleration measurements.. Europace. 1999:1:126-30.

53 3

45.. Dupuis JM. Kobeissi A. Vitali L. Gaggini G, Merheb M, Rouleau F. Leftheriotis G. Ritter P. Victorr J. Programming optimal atrioventricular delay in dual chamber pacing using peak endocardiall acceleration; comparison with a standard echocardiographic procedure. Pacing ClinClin Electrophysiol. 2003:26:210-3.

46.. Deharo JC, Brunetto AB, Bellocci F, Barbonaglia L. Occhetta E. Fasciolo L. Bocchiardo M, Rognonii G. DDDR pacing driven by contractility versus DD1 pacing in vasovagal syncope: a multicenter.. randomized study. Pacing Clin Electrophysiol. 2003:26:447-50.

47.. Deharo JC, Peyre JP. Chalvidan T. Thirion X. Valli M. Ritter P. Djiane P. Continuous monitoring off an endocardial index of myocardial contractility during head-up til t test. Am Heart ƒ. 2000:139:1022-30. .

48.. Bordachar P. Garrigue S, Reuter S. Hocini M. Kobeissi A. Gaggini G, Jais P. Haissaguerre M. Clementyy J. Hemodynamic assessment of right, left, and biventricular pacing by peak endocardiall acceleration and echocardiography in patients with end-stage heart failure. Pacing ClinClin Electrophysiol, 2000:23:1726-30.

49.. Plicchi G. Marcelli E. Pariapiano M. Bombardini T. PEA I and PEA II based implantable haemodynamicc monitor: pre clinical studies in sheep. Europace. 2002;4:49-54.

50.. Avery P. Banning A. Lawson T. McGurk L, Buchalter M. Physiological pacing improves symptomss and increases exercise capacity in the elderly patient. Int J Cardiol. 1994:46:129-33.

51.. Clarity reference guide VMBV. 1999:31-33. 52.. Insignia I Plus rg, Guidant Corporation. 2001. 53.. Lau CP. Leung SK. Lee IS. Delayed exercise rate response kinetics due to sensor cross-checking

inn a dual sensor rate adaptive pacing system: the importance of individual sensor programming.. PacingClin Electrophysiol. 1996:19:1021-5.

54.. Walsh IT, Charlesworth A. Andrews R, Hawkins M, Cowley A]. Relation of daily activity levels inn patients with chronic heart failure to long-term prognosis. Am J Cardiol. 1997:79:1364-9.

55.. Astrand. Aerobic work capacity its relation to age. sex. and other factors, circulation research. 1967:20:1-2111 1-217.

56.. Astrand I, Astrand. PO, Rodahl K. maximal heart rate during work in older men. Journal of appliedapplied physiology. 1959:14:562-566.

57.. Astrand I. Aerobic work capacity in men and women with special reference to age. Acta Physio!Physio! Scand. 1960;49(Suppl 1Ó9H-92.

58.. Astrand I. Aerobic work capacity its realtion yo age, sex. and other factors. Circ Res. 1967:XX:1-211-217. .

59.. Astrand PO. Ryhming I. A nomogram for calculation of aerobic capacity (physical fitness) from pulsee rate during sub-maximal work. J Appl Physiol. 1954:7:218-21.

60.. Robergs R.A. LR. The surprising history of the "Hrmax = 220-age" EQUATTION. JEPonline. 2002;5:1-10. .

61.. Cooper KH PJ. White Sr. et al. Age-fitness adjusted maximal heart rate. Med Sport 1977:10:78-88. 62.. Kay GN. Quantitation of chronotropic response: comparison of methods for rate- modulating

permanentt pacemakers, ƒ Am Coll Cardiol. 1992:20:1533-41. 63.. Freedman RA. Hopper DL. Mah J, Hummel J. Wilkoff BL. Assessment of pacemaker

chronotropicc response: implementation of the Wilkoff mathematical model. Pacing Clin Electrophysiol.Electrophysiol. 2001:24:1748-54.

64.. Page E, Bonnet JL. Durand C. Comparison of metabolic expenditure during CAEP versus a test adaptedd to aerobic capacity (Harbor test) in elderly healthy individuals. Pacing Clin Electrophysiol.Electrophysiol. 2000:23:1772-7.

65.. Sulke N. Dritsas A. Chambers J. Sowton E. Is accurate rate response programming necessary? PacingClinPacingClin Electrophysiol. 1990:13:1031-44.

66.. Fletcher GF. Balady GJ. Amsterdam EA. Chaitman B. Eckel R. Fleg J, Froelicher VF, Leon AS. Pinaa IL. Rodney R. Simons-Morton DA, Williams MA, Bazzarre T. Exercise standards for testingg and training: a statement for healthcare professionals from the American Heart Association.. Circulation. 2001;104:1694-740.

67.. Gibbons R], Balady GJ. Bricker JT. Chaitman BR. Fletcher GF. Froelicher VF. Mark DB. McCallisterr BD, Mooss AN, O'Reilly MG. Winters WL, Jr.. Antman EM. Alpert JS. Faxon DP. Fusterr V, Gregoratos G. Hiratzka LF. Jacobs AK. Russell RO, Smith SC. Jr. ACC/AHA 2002 guidelinee update for exercise testing: summary article: a report of the American College of

54 4

Chapterr 2

Cardiology/Americann Heart Association Task Force on Practice Guidelines (Committee to Updatee the 1997 Exercise Testing Guidelines). Circulation. 2002;106:1883-92.

68.. Linde C. Quality-of-life in pacemaker and implantable cardioverter defibrillator recipients. PacingPacing Clin Electrophysiol, 2000:23:931-3.

69.. Linde C. How to evaluate quality-of-life in pacemaker patients: problems and pitfalls. Pacing ClinClin Electrophysiol. 1996:19:391-7.

70.. Stofmeel MA. Post MW, Kelder JC. Grobbee DE, van Hemel NM. Quality-of-life of pacemaker patients:: a reappraisal of current instruments. Pacing Clin Electrophysiol. 2000:23:946-52.

71.. Oto MA. Muderrisoglu H. Ozin MB. Korkmaz ME, Karamehmetoglu A. Oram A. Oram E, Ugurluu S. Quality of lif e in patients with rate responsive pacemakers: a randomized, cross-overr study. PacingClin Electrophysiol. 1991;14:800-6.

72.. Stofmeel MA. Post MW. Kelder JC. Grobbee DE. van Hemel NM. Psychometric properties of Aquarel,, a disease-specific quality of lif e questionnaire for pacemaker patients, ƒ Clin Epidemiol.Epidemiol. 2001:54:157-65.

73.. Stofmeel MA. Post MW. Kelder JC, Grobbee DE, van Hemel NM. Changes in quality-of-life afterr pacemaker implantation: responsiveness of the Aquarel questionnaire. Pacing Clin Electrophysiol.Electrophysiol. 2001:24:288-95.

74.. Linde-Edelstam C. Nordlander R. Unden AL. Orth-Gomer K. Ryden L. Quality-of-life in patientss treated with atrioventricular synchronous pacing compared to rate modulated ventricularr pacing: a long-term, double-blind, crossover study. Pacing Clin Electrophysiol. 1992:15:1467-76. .

75.. Lopez-Jimenez F, Goldman L, Orav EJ. Ellenbogen K, Stambler B. Marinchak R, Wiikoff BL. Mangionee CM. Yoon C. Vitale K, Lamas GA. Health values before and after pacemaker implantation.. Am Heart]. 2002:144:687-92.

76.. Huth C, Friedl A. Klein H, Auricchio A. [Pacing therapies for congestive heart failure consideringg the results of the PATH-CHF study]. Z Kardiol. 2001:90 Suppl 1:10-5.

77.. Newman D, Lau C, Tang AS. Irvine J. Paquette M. Woodend K. Dorian P, Gent M, Kerr C, Connollyy SJ. Effect of pacing mode on health-related quality of lif e in the Canadian Trial of Physiologicc Pacing. Am Heart J. 2003:145:430-7.

78.. Chawla NP, Rea W, Shapiro W. The use of an implanted demand pacemaker in bradyarrhythmias.. Arch Intern Med. 1969:124:593-9.

79.. Saoudi N. Appl U, Anselme F, Vogiimacci M, Cribier A. How smart should pacemakers Be? Am JJ Cardiol. 1999;83:180D-186D.

50.. Jones BR. Kim J. Zhu Q. Nelson JP, KenKnight BH. Lang DJ. Warren JA. Future of bradyarrhythmiaa therapy systems: automaticity. Am ƒ Cardiol. 1999:83:192D-201D.

51.. Neuzner J. Schwarz T. Sperze! J. Pacemaker automaticity. Am ƒ Cardiol. 2000:86:K104-K110. 82.. Griffi n JC. Schuenemeyer TD. Hess KR. Glaeser D, Anderson BJ. Romans E. Jenkins MA.

Nielsenn AP. Pacemaker follow-up; its role in the detection and correction of pacemaker system malfunction.. PacingClin Electrophysiol. 1986:9:387-91.

83.. Ribeiro AL, Rlncon LG. Oliveira BG. Mota CC. Pires MT. Enhancing longevity of pacemakers throughh reprogramming. Underutiiization and cost-effectiveness. Arq Bras Cardiol. 2001:76:437-44. .

84.. Klonis D, Zhang X. Patel U. Gulamhusein S. Patel J. Hurwit H. Banish D, Marco D. Automatic sensorr algorithms expedite pacemaker follow-ups. Pacing Clin Electrophysiol. 2003:26:225-8.

85.. Gentzier RD, Lucas EH. Automatic sensor adjustment in a rate modulated pacemaker. North Americann Trilogy DR + Phase I Clinical Investigators. Pacing Clin Electrophysiol. 1996:19:1809-12. .

86.. Schuchert A, van Langen H. Michels K, Meinem T. A prospective randomized comparison betweenn fixed rate response programming and automatic rate response optimization in activity-triggeredd DDDR pacemakers. Thera Pacemaker Study Group. Cardiology. 1998:89:25-8.

87.. Leung SK, Lau CP. Tang MO. Leung Z, Yakimow K. An integrated dual sensor system automaticallyy optimized by target rate histogram. Pacing Clin Electrophysiol. 1998:21:1559-66,

88.. Garrigue S. Gentilini C. Hofgartner F. Mouton E. Rousseau A, Clementy J. Performance of a ratee responsive accelerometer-based pacemaker with autocalibration during standardized exercisee and recovery. Pacing Clin Electrophysiol. 2002:25:883-7.

55 5

Heartt rate profiles during two types off exercise testing in healthyy individuals

Aytenn Erol-Yilmaz MD. Raymond Tukkie MD PhD and Arthur Wilde MD PhD

Abstract t

Background d

Thee normal values concerning heart rate (HR) profiles during two frequently used

exercisee tests are limited described in literature, especially in elderly healthy

persons.. Data is lacking about which test is better for pacemaker sensor optimiza-

tionn purposes, Therefore, we studied the HR responses of healthy individuals (HI)

off different age categories using the chronotropic assessement exercise protocol

(CAEP)) and the 6- minute hall walk test (6-HWT).

Materiall and methods

HII of different age categories without co-morbidity and medication were included

andd randomized to either CAEP or 6-HWT exercise test. The age categories were 20-

300 years (groupl). 30-40 years (group 2). 40-50 years (group 3). 50-60 years (group 4)

andd > 60 years (group 5). HR at rest, HR at 1 minute of exercise, time to peak HR,

maximall achieved HR, HR at 1,3 and 10 minutes recovery period, exercise duration,

andd METS or achieved distance (meters) were measured.

Results s

Hundred-seventy-fivee HI (87 men, 88 female) were included in the several age

categories.. Ninety-one HI were randomized to the CAEP and 79 HI to the 6-HWT,

Numberr of HI in each categorie were: 30 in group 1 (mean age 25 2 years), 29 in

groupp 2 (mean age 35 3 years), 40 in group 3 (mean age 44 3 years), 41 in group

44 (mean age 55 2 years) and 30 in group 5 (mean age 66 5 years) respectively.

Thee achieved HR at one minute of exercise was significantly higher and the time to

peakk HR significantly shorter during 6-HWT compared to CAEP. although the

achievedd maximal HR was comparable. There are no gender differences in HI

randomizedd to 6-HWTand minimal gender differences in HI randomized to CAEP.

Thee predicted maximal HR according to the Astrand formula (220-age) was not

significantlyy different compared to the achieved maximal HR in both tests.

Conclusions: :

Thee HR rate profiles can be used to further optimization of the pacemaker sensors.

Thee Astrand formula (220-age) can still be used for prediction of the maximal HR.

Thee 6-HWT is preferable for pacemaker sensor optimization.

58 8

Chapterr 3

Background d

Physicianss are accustomed to decision-making on the basis of knowledge of normal

rangess and normal responses to (diagnostic) tests. Several exercise tests are used for

pacemakerr sensor optimization (the chronotropic assessment of exercise protocol

(CAEP),, 6 minute hall walk test (6- HWT), stair climbing, Kaltebach step test, daily

activities).. The normal values concerning heart rate (HR) profiles during two most

frequentlyy used exercise tests are limited described in literature, especially in the

elderlyy healthy persons.1'31

Thee CAEP protocol according to Wilkoff is designed for pacemaker patients and is one

off the most used exercise test in this patient category beside the 6- HWT. ' In

addition,, data is lacking about which test is better for pacemaker sensor optimization

purposes.. Therefore we studied the HR responses of healthy individuals (HI) of

differentt age categories using the CAEP and the 6-HWT exercise test. These data were

laterr used for pacemaker sensor optimization in chapter 4. We also tried to answer

whichh exercise test is the optimal test for pacemaker sensor optimization.

Material ss and methods

Healthyy individuals

Healthyy individuals (HI) of different age categories without co-morbidity and

medicationn were included and randomized to one of the exercise test protocols.

Thee age categories were 20-30 years (groupl), 30-40 years (group 2), 40-50 years

(groupp 3). 50-60 years (group 4) and > 60 years (group 5)- The two exercise tests used

weree the six minute hall walk test (6-HWT), and the chronotropic assessment

exercisee protocol (CAEP) according to Wilkoff.29 The subjects were exercised until

fatigue,, symptomatic or end of the protocol, but no systematic effort was made to

encouragee exercise to exhaustion. Also the HR response to postural change was

measured.. The HI were recruited by advertising in the local newspaper, on the

severall outward-clinics of our hospital.

59 9

Posturee change

Alll HI were examined in the same conditions (before noon, uniform room tempera-

turee and footgear (subject's own shoes)). After instrumentation, subjects rested

supinee on an examination table with one pillow for 5 minutes. They then elevated

too the sitting position and immediately to the standing position.

Exercisee test protocols

ChronotropicChronotropic assessment of exercise protocol

Healthyy individuals randomized to the CAEP according to Wilkoff underwent a

symptomm limited treadmill test,29 One MET equals 3.5 ml oxygen uptake/kg body

weight/min,, representing the approximate metabolic cost to stand quietly. In his

protocol,, oxygen consumption and carbon dioxide production was not measured,

andd thus metabolic workload (METS ) was not directly measured during exercise.

Rather,, metabolic levels during each stage of exercise were estimated using tread-

milll grade and speed.

SixSix minute hall walk test

Afterr 5 minutes of rest, HI were brought to the parcour. A parcour of 100 m was

createdd by attaching stickers each meter in an oval form showing the walk distance.

Normalss were instructed to walk or run for 6 minutes at the parcour after hearing

thee start sign. After the symptom limited 6-HWT, the HI were brought back to the

testt room for a recovery period of 10 minutes.

Measurements s

Beatt to beat HR was recorded during the physical tests with the Polar advantage

systemm using electrodes mounted in a belt (Polar Electro OY. Kempele. Finland). The

firstt 10 HR of each minute were averaged. During the CAEP exercise test, HI were

alsoo continuously monitored by 12-lead electrocardiographic recordings. HR at rest,

HRR at 1 minute of exercise, time to peak HR, maximal achieved HR, HR at 1,3 and 10

60 0

Chapterr 3

minutess recovery period, exercise duration, and METS or achieved distance (meters)

weree measured.

Statistics s

Firstt the Kruskal-Wallis test was used to analyse differences between the different

agee categories. When the Kruskall-Wallis test was significant, the nonparametric

dataa were analyzed using the Mann- Whitney U test and the parametric data were

analyzedd using the paired sample t test. All data are expressed as mean SD. A P

valuee <0.05 is considered statistically significant.

Results s

Normall population

Hundred-seventy-fivee HI (87 men. 88 female) were included in the several age

categories.. Ninety-one HI were randomized to the CAEP (42 men, 49 female) and 79

HII to the 6-HWT (42 men, 37 female). Number of HI in each categorie were: 30 in

group.. 29 in group 2, 40 in group 3, 41 in group 4 and 30 in group 5 respectively. Of

fourr HI 3 in the HWT (66, 63. 32 years) and 1 in the CAEP (28 year) data derived from

thee Polar advantage system could not be analyzed due to technical disturbance of

thee system. One HI in group 5, a 65 year old female, developed atrial fibrillation

duringg the CAEP and stopped the exercise test.

Exercisee tests

ChronotropicChronotropic assessment exercise protocol

Restt rate

Thee resting HR in HI randomized to the CAEP was not significantly different be-

tweenn group I to 5 (see table 1 and figure 1).

61 1

HRR at one minute of exercise

Thee HR at one minute of exercise was not significantly different between group 1 to

55 (see table 1).

Timee to peak HR

Inn each group the time to peak HR increased. The achieved time to peak HR in group

11 was significantly longer compared to group 4 (p= 0,004) and 5 (p<0.001). The

achievedd time to peak HR in group 2 was significantly longer compared to group 4

(p== 0.07) and 5 (p<0.001). The achieved time to peak HR in group 3 and 4 were

significantlyy longer compared to group 5 (p<0.001. see table 1).

Maximall HR

Thee maximal HR diminished in each group with increasing age. The achieved

maximall HR in group 1 was significantly higher compared to group 3. 4 and 5

(p<0.009.. see table 1, figure ), The achieved maximal HR in group 2 was significant-

lyy higher than in the groups 4 and 5 (p = 0.07. p<0.001). The achieved maximal HR

inn group 3 and 4 was significantly higher compared to group 5 (p=0.004. p = 0,03)-

Thee percentage difference between the achieved and predicted maximal HR was the

largestt in group 1 and 5 (184 10 vs. 195 bpm, A in bpm -6% and 144 36 vs, 154.

AA bpm =6% respectively). In group 2, 3 and 4 the percentage difference between the

achievedd and predicted maximal HR differed minimal (181 15 vs. 185 bpm, A in

bpmm = 2%, 172 13 vs. 176 bpm. A in bpm =3%. 167 10 vs. 165, A in bpm = 2%

respectively). .

Recovery y

Thee HR at 1 minutes of recovery was significantly higher in group 1 compared to

groupp 3, 4 and 5 (p-0.06, p = 0.003, p = 0.03). The achieved HR at 10 minutes was

significantlyy higher in group 2 compared to group 4 (p —0.001). The achieved HR at

1,3.. 10 minutes of recovery in group 3 compared to group 5 was significantly

higherr (p< =0.06).

62 2

Chapterr 3

Exercisee duration and METS

Thee exercise duration of the HI in group 1 was significantly longer than in group 3.

44 and 5 (p = 0.05. p = 0.006, p<0.001). The exercise performance of the HI in group 2,

33 and 4 were significantly higher compared to group 5 (p< 0.03. see table 1).

Thee achieved METS were significantly higher in group 1 and 2 compared to group 4

andd 5 (p<0.001). The achieved METS were also higher in group 3 and 4 compared to

55 (<0.001. see figure 3.d).

6-6- minute hall walk test

Restt rate

Thee resting HR in HI randomized to the 6-HWT was not significantly different

betweenn group 1 to 5 except for group 2 compared to 3 (55 12 vs. 64 7 bpm,

pp = 0.02. see table 2. and figure 2).

HRR at one minute of exercise

Inn contrast to the CAEP test, differences existed in the achieved HR at one minute of

exercisee between the groups. The achieved HR at one minute of exercise was

significantlyy higher in the group 1 compared to 4 (p-0.009). The achieved HR at one

minutee of exercise was significantly higher in group 2 compared to 3.4 and 5

(p<0.05).. The achieved HR at one minute of exercise was significantly higher in

groupp 3 compared to 4 (p-0.006, see table 2 and figure 2).

Timee to peak HR

Thee achieved time to peak HR was only significantly longer in group 1 compared to

44 and 5 (p<0.03, table 2).

63 3

Maximall HR

Thee maximal HR diminished with each age category. The achieved maximal HR in

groupp 1. 2 3. 4 was significantly higher than in group 5 (p<0.009, see table 2, figure 2).

Thee percentage difference between the achieved and predicted maximal HR (Astrand

formule== 220-agel is the largest in group 1 and 2 (179 19 vs. 195 bpm. A in bpm

=8%.. 175 7 vs. 154, A bpm =5%)- In group 3. 4 and 5 the percentage difference

betweenn the achieved and predicted maximal HR minimally differed (175 2 vs. 185

Tablee 1. Heart rate during rest, exercise and recovery with CAEP

Age e

(years) )

266 3 igroup D

355 3 (group 2}

455 3 Igroup 3)

555 2 (group 41

677 5 (group5 )

Resrr rate

(hr/min) )

666 10

600 12

677 12

622 10

655 10

HRatt 1 Ibpml l

911 16

866 13

800 11

833 14

900 18

TIPP HR (mini i

21 2

20 3

19 3

188 3

166 3

Exercise e

MHR R (bpm) )

1844 10

1811 15

1722 13

1677 i 10

1444 + 36

Duration n min n

222 2

211 4

200 3

199 3

177 3

METS S iml/'kg'min) )

16.66 3

16.22 2

14.99 i 3

13-55 2

10.33 3

Tablee 1. CAEP = chronotropic assessment exercise protocol HRR = heart rate: TTPHR = timc to peak heart rate; MHR = maximal heart rate; METS= oxygen uptake/kgg body weigh t/min. min = minute: bpm=beats per minute.

Tablee 2. Heart rate during rest, exercise and recovery with 6-HWT

Age e

(years) )

244 2 (group 11

355 2 (group 21

444 3 \group 31

555 3 '(group 4

666 5 igroup 5)

Restt rate

(hr/min.) )

611 9

555 12

644 7

644 13

622 10

MRR at 1 (bpm) )

1422 19

1522 12

1411 0

1244 17

1311 10

TTPP HR min n

4.33 1.7

33 1.7

377 1.6

2.88 16

2,99 i 1-3

Exercise e

MHR R (bpm) )

33 79 19

1755 7

1755 12

1688 12

1511 17

Duration n min n

66 0

66 0

66 0

5.88 0.7

5.77 0-7

METS S (ml/'kg/min) )

12366 236

1116.. 278

9833 181

10066 178

7811 281

Tablee 2. 6-HWT = 6 minute hall walk test; HR^heart rate: TIP HR = time to peak heart rate: MHRR = maximal heart rate: bpm = beats per minute; min= minutes: m = meters.

64 4

Chapterr 3

bpm.. A in bpm =0.6%. 168 12 vs. 176 bpm. A in bpm = 2%. 151 17 vs. 165, A in

bpmm =2% respectively.

Recovery y

Thee HR at 3 and 5 minutes of recovery was significantly higher in group 1 compared

too 2 (p < 0,03) • The achieved HR at 1 and 3 minutes of recovery was significantly higher

inn group 1 and 3 compared to group 5 (p<0,05). The achieved HR at 3 and 5 minutes

Recovery y

HRR at 1 min HR at 3 min HR at 5 min HR at 10 min (bprrOO (bpm) (bpm) (bpm)

1488 21 105 0 95 16 93 15

1366 6 106 17 97 2 96 9

1311 4 106 22 92 1 89 2

1277 6 98 17 90 4 88 2

1199 6 95 18 89 4 3

Recovery y

HRatt 1 min HR at 3 min HRat 5 min HR at 10 min

(bpm)) (bpm) (bpirO (bpm)

1377 22 108 0 98 13 95 15

1222 8 89 6 86 6 85 10

1355 8 108 15 96 2 93 2

11 104 4 95 2 86 0

1133 2 94 6 92 5 89 6

65 5

CAEP P 200 0

Minute s s

Figuree 1. Changes in heart rate \mean during CAEP of the different age categories \male and female!.

6-HWT T

99 12 Minute s s

Figuree 2. Changes in heart rate imean1 during 6-HWT of the different age categories (malee and female)

wass significantly higher in group 2 compared to 3 (p<0.05). The achieved HRat 1

minutess of recovery was significantly higher in group 4 compared to 5 (p=0.05).

Exercisee duration and distance

Alll HI exercised for 6 minutes except one HI in group 4 and one in group 5. they

stoppedd before the end of the test due to fatigue. The achieved distance was signifi-

ed d

Chapterr 3

cantlyy longer in group 1 compared to 3. 4 and 5 (p<0.01) and also the achieved

distancee was significantly longer in group 2 and 3 compared to 5 (p<0.04).

Genderr differences

HII randomized to 6-HWT showed no differences between females compared to

maless in the different age categories. Females compared to males in group 2 rand-

omizedd to CAEP showed significant differences in resting HR (52 9 vs.68 9 bpm,

p=0.09).. achieved maximal HR (178 4 vs. 182 8 bpm. p=0.025). exercise

durationn (20 2 vs. 21 4 min. p = 0.046) and achieved METS (14 2 vs. 17.2 1.8

METS,, p = 0.028). Within group 3 and 4, the achieved METS significantly differed

(12.55 1 vs. 17.6 1.8. p=0.001 vs. 12.6 1.8 vs. 14.7 2 METS. p=0.03). In group

5.. achieved maximal HR (134 41 vs. 162 16 bpm, p = 0.035), HR at 3 minutes of

recoveryy {88 14 vs. 106 18 bpm, p = 0.042), HR at 5 minutes of recovery (84 4

vs.. 97 12 bpm, p=0.036) and HR at 10 minutes of recovery (82 13 vs. 93 10

bpm,, p=0.066) were significantly different between men and women.

CAEPP vs. 6-HWT

Thee achieved HR at 1 minute of exercise and the achieved time to peak HR were

significantlyy different between CAEP and 6-HWT for all age categories (86 5 vs. 138

5 bpm. p<0.001 and 18.8 0.9 vs. 3-3 0.3 min, p<0.01). The rest rate, maximal

HRR and HR at 1, 3. 5 and 10 minutes of recovery were not significantly different.

Discussion n

Optimall programming of pacing variables is important in rate adaptive pacemakers

too improve exercise capacity and reduce symptoms. Individual programming should

bee adapted to the age of patients and to associated medical conditions, as well as to

overalll functional capacity. For pacemaker optimization, tests are in use with largely

isotonicc (dynamic or locomotory) exercise. These exercise tests can be divided in:

in-hospitall and out-of hospital exercise tests. From the in-hospital tests, the CAEP

exercisee test according to Wilkof f and the 6-minute walk test are frequently

67 7

used.266 ,2° However, recent data are lacking concerning normal values of HR respons-

ess in different age categories with CAEP and 6-HWT. In addition, data are lacking if

duee to the increasing height and weight during the last century, the HR response is

changedd in HI and whether the Astrand formula (220-age) for prediction of the

maximall HR can still be used. We therefore describe the HR response of 170 HI

withoutt apparent disease and medication during CAEP and 6HWT.

Thee disadvantages of in-hospital testing with standardized tests such as treadmill

andd bicycle ergometry is that these tests are time consuming and not always feasi-

blee in patients with pacemakers (as they are often elderly persons, with physical

limitations)) and the tests poorly represent daily activities.32 Especially the treadmill

testt is not an ideal exercise test for pacemaker patients, because in pacemaker

patients,, exercise is often limited by loss of muscle strength and mass rather than

cardiopulmonaryy capacity. This loss of muscle strength is particularly apparent

whenn exercise testing is performed on a bicycle or treadmill. The CAEP exercise test

hass non-linear characteristics. The first 10 minutes requires low metabolic work-

load,, beyond which the workload abruptly increases. Patients with preserved

functionall capacity, capable of exercising for more than 10 minutes, may quit before

reachingg maximal 02 uptake, mainly because of excessive increments in workload

nearr the end of the test, thus, being limited by mechanical rather than metabolic

barriers.. Caution is needed in the application of the CAEP protocol which is also

illustratedd by Freedman et al.32 They showed that failure in attaining maximum

exercisee could create the appearance of sub-optimal pacemaker performance.

Onee of the first studies which performed an in-hospital walk test was done in

patientss with chronic obstuctive pulmonary disease.7 The in-hospital HWT is

describedd in pacemaker patients with different duration (2. 6. 12 minutes) and most

off them used brisk walking, which is in contrast to our study. In our study the HI

weree allowed to run and they stopped because of maximal exhaustion, with symp-

tomss or end of the protocol. Therefore it is difficult to compare the HR response of

HII to the limited existing literature. Provenier et al. validated the six- minute walk

testt in rate response pacemakers. In his study, the six-minute walk test was per-

formedd in a corridor 45 m long with 1 (m) and 5 (m) marks on the floor.2

Thee 13 HI in his study the achieved a maximal rate of 73% of the predicted maximal

HR,, corrected for age (220-age). The HI in our study achieved almost their predicted

maximall HR, The HWT has several advantages when compared to the CAEP. The

08 8

Chapterr 3

HWTT is short, cheap, has a better practicability and uses a natural way of moving

duringg the walk with good correlation to other types of exercise at home,i q

Inn our study, the achieved HR at one minute of exercise was significantly higher and

thee time to peak HR significantly shorter during 6-HWT compared to CAEP, al-

thoughh the achieved maximal HR was comparable. We therefore think, considering

thee advantages of the 6-HWT that the 6-HWT is the prefered test for pacemaker

sensorr optimization.

Thee difference in body mass index of our HI compared to other studies is minimal

(0-11 kg/m2). even when we compared with the dated studies, which can be explained

duee to the increased height together with the increased weight. '-0'10 1213.15.20.24.27.28

Thereforee the achieved peak HR, which is correlated the body mass index, is not

significantlyy different compared to the dated literature, ^ó-1012^-^.20.24.27.2s

Thee estimation of the maximal HR has been largely based on the Astrand formula

(maximumm HR=220-age). In 13-507 healthy men from several studies, the predicted

maximall HR differed only 4.3 beats per minute compared to the achieved maximal

HRR in our study. 1-6.10.12.13,15.20.24.27.28 Considering the many factors which can

influencee the achieved maximal HR, careful programming of the maximal HR is

needed.333 It bears repeating that this is only an estimate of an individual patient's

maximumm HR. Better methods need to be developed to estimate especially the HR

responsess at sub maximal levels because patients needing a pacemaker are of an age

groupp where they may have disabilities limiting their physical work capacity.

Inn a meta-analysis, Londeree and Moeschberger showed that although 73% of the

variabilityy of the peak HR response could be attributed to age. and 5% to some other

factorss that were examined, the remaining variability could not be accounted for. 23

Thee other minor variables, which have been shown to have some influence on the

maximall HR include: gender, fitness and cardiac illness.18 Fitness and training cause

somee slowing of HR for a given external workload,115

Inn our study there were surprisingly no gender differences in HI randomized to the

6-HWT.. HI randomized to the CAEP had minimal gender differences in the groups 2

too 5. This is in contrast to the studies described before, which showed that females

hadd a higher HR at all levels of exercise. J'

InIn chapter 4. we showed that pacemaker sensor optimization improved exercise

capacityy (METS). However, the HI in this study achieved higher HR than patients

despitee individual optimization, probably because current sensors are still hypo-

69 9

chonotropicc and physicians need to program conservative due to concomitant heart

disease.. The elderly patients in the pacemaker sensor optimization study achieved

muchh lower METS (7.6 vs. 10.3) compared to the elderly HI from this study. These

dataa underline that exercise capacity can be improved in pacemaker patients by

pacemakerr sensor optimization, although other factors such as co-morbidity and

fitnesss also plays a major role. The HI were more active and had no co-morbidity

comparedd to the patients in our study.

Conclusions s

Wee describe the heart rate profiles during two types of exercise testing in 170

healthyy individuals. Only the heart rate at 1 minute of exercise and time to peak

heartt rate are significantly different when these two tests are compared. Surprising-

ly,, there are no gender differences in healhy controls randomized to 6-minute hall

testt and minimal gender differences in healthy individuals randomized to the CAEP.

Inn addition, the body mass index is not significantly changed during the last

century,, The Astrand formula (220-age) can still be used for prediction of the

maximall heart rate, however precaution is needed in the elderly pacemaker patients

withh co-morbidity. Furthermore, our data confirm that the 6-minute hall walk test is

preferablee for pacemaker sensor optimization because of the ease of implementa-

tionn and the natural way of moving during the walk with good correlation to other

typess of exercise at home. These heart rate profiles can be used to further optimize

thee pacemaker sensor.

70 0

References s

Chapterr 3

1.. Astrand PO, Ryhming I. A nomogram for calculation of aerobic capacity (physical fitness) from pulsee rate during sub-maximal work, j Appl Physiol. 1954:7:218-21.

2.. Astrand I, Astrand. PO, Rodahl K. maximal heart rate during work in older men. journal of ap-pliedplied physiology. 1959;14:562-566.

3.. Astrand I. Aerobic work capacity in men and women with special reference to age. Acta Physi-olol Scand. 1960;49(Suppl 169):l-92.

4.. Astrand I. Aerobic work capacity its realtion yo age. sex, and other factors. Circ Res. 1967:20:1-211-217. .

5.. Bowyer AF. Thomas RA. Bowyer SR. Heart rate response to exercise in normal women. Meth-odsods InfMed. 1993;32:206-10.

6.. Bruce RA, Fisher LD. Cooper MN, Gey GO. Separation of effects of cardiovascular disease and agee on ventricular function with maximal exercise. Am J Cardiol. 1974;34:757-63.

7.. Butland RJ. Pang ]. Gross ER. Woodcock AA. Geddes DM. Two-, six-, and 12-minute walking testss in respiratory disease. Bi Med j {Clin Res Ed). 1982;284:1607-8.

8.. Cardus D. Spencer WA. Recovery time of heart frequency in healthy men: its relation to age andd physical condition. Arch Phys Med Rehabil. 1967;45:71-7.

9.. Casaburi R. Spitzer S. Haskell R. Wasserman K. Effect of altering heart rate on oxygen uptake att exercise onset. Chest. 1989:95:6-12.

10.. Cooper K. Purdy JG. White SR. Age-fitness adjusted maximal heart rates, medicine sport. 1977:10:78-88. .

11.. Crook B. Nijhof P, van der Kemp P. Jennison C. The chronotropic response of the sinus node too exercise: a new method of analysis and a study of pacemaker patients. Eur Heart}. 1995;16:993-8. .

12.. Ellestad MH. stress testing principles and practice. Tirth edition:567-573-13.. Ellestad M. Allen W. Wan M. Kemp G. Maximal treadmill stress testing for cardiovascular eval-

uation.. Circulation. 1969:39:517-522. 14.. Greig CA, Young A. Skelton DA. Pippet E, Butler FM, Mahmud SM. Exercise studies with elder-

lyy volunteers. Age Ageing. 1994:23:185-9-15.. Hammond HK. Froelicher VF. Normal and abnormal heart rate responses to exercise. Prog Car-

diovascDis.diovascDis. 1985;27:271-96. 16.. Higginbotham MB. Morris KG. Williams RS. Coleman RE. Cobb FR. Physiologic basis for the

age-relatedd decline in aerobic work capacity. Am j Cardiol. 1986;57:1374-9. 17.. Higginbotham MB. Morris KG. Williams RS. McHale PA, Coleman RE. Cobb FR. Regulation of

strokee volume during submaximal and maximal upright exercise in normal man. Circ Res. 1986;58:281-91. .

18.. Jonsson BG. Astrand I. Physical work capacity in men and women aged 18 to 65- Scand j Soc Med.Med. 1979;7:131-42.

19.. Langenfeld H. Schneider B. Grimm W. Beer M, Knoche M, Riegger G. Kochsiek K. The six-minutee walk—an adequate exercise test for pacemaker patients? Pacing Clin Electrophysiol. 1990;13:1761-5--

20.. Lester M, Sheffield LT, Trammell P. Reeves TJ. The effect of age and athletic training on the maxima]] heart rate during muscular exercise. Am Heart], 1968:76:370-6.

21.. Lewalter T, MacCarter D. Jung W. Schimpf R, Manz M. Luderitz B. Heart rate to work rate rela-tionn throughout peak exercise in normal subjects as a guideline for rate-adaptive pacemaker programming.. Am ƒ Cardiol. 1995;76:812-6.

22.. Loeppky )A, Greene ER, Hoekenga DE. Caprihan A, Luft UC. Beat-by-beat stroke volume assess-mentt by pulsed Doppler in upright and supine exercise, ƒ Appl Physiol. 1981:50:1173-82.

23.. Londeree B. Moeschberger ML. Influence of age and other factors on maximal heart rate, ƒ CardiacCardiac Rehabil. 1984:4:44-9-

24.. Master A. a simple exercise tolerance test for circulatory efficiency eith standard tables for normall individuals. American journal of the medical sciences. 1929:177:223-243.

71 1

25-- Page E, Bonnet JL, Durand C. Comparison of metabolic expenditure during CAEP versus a test adaptedd to aerobic capacity (Harbor test) in elderly healthy individuals. Pacing Clin Electro-physiol.physiol. 2000:23:1772-7.

26.. Provenier F, Jordaens L. Evaluation of six minute walking test in patients with single chamber ratee responsive pacemakers. Br Heart J. 1994:72:192-6.

27.. Robinson S. Experimental studies of physical fitness in relation to age. zeitschrift fur die physiologiephysiologie der menschen beiarbeit. 1938:10:251-323.

25.. Sue DY, Hansen JE. Normal values in adults during exercise testing. Clin Chest Med. 1984;5:89-98. .

29.. Wilkoff BL, Miller RE. Exercise testing for chronotropic assessment, Cardiol Clin. 1992:10:705-17. .

30.. Wilkoff B. Corey ], blackburn G. a mathematical model of the Cardiac chronotropic response too exercise. Journal of electrophysiology. 1989:3:176-180.

31.. Meine M, Achtelik M, Hexamer M. Kloppe A. Werner ]. Trappe HJ. Assessment of the chrono-tropicc response at the anaerobic threshold: an objective measure of chronotropic function. PacingPacing Clin Electrophysiol. 2000:23:1457-67.

32.. Freedman RA, Hopper DL. Mah J, Hummel J. Wilkoff BL. Assessment of pacemaker chrono-tropicc response: implementation of the Wilkoff mathematical model. Pacing Clin Electrophys-iol.iol. 2001:24:1748-54.

33.. Cooper KH PJ. White Sr. et al. Age-fitness adjusted maximal heart rate. Med Sport. 1977:10:78-88.

72 2

Individuall optimization of pacing sensorss improves exercise capacity withoutt influencing quality of life

Aytenn Erol-Yilmaz MD. Tim A, Schrama, Jutta Schroeder Tanka*, MD PhD, Jann G. Tijssen PhD, Arthur A. Wilde MD PhD, and Raymond Tukkie MD PhD

Fromm the department of Clinical and Experimental Cardiology, Academic Medical Centerr and the department of Cardiology. Sint Lucas-Andreas hospital*, Amsterdam,

Thee Netherlands.

PACEPACE 2005;28:17-24

Abstract t

Introductio n n

Programmablee pacemaker sensor features are frequently used in default setting.

Limitedd data are available about the effect of sensor optimization on exercise

capacityy and quality of life (QOL). Influence of individual optimization of sensors

onn QOL and exercise tolerance was investigated in a randomized, single blind study

inn patients with WIR. DDDR or AAIR pacemakers.

Materiall and methods

Patientss with >75% pacing were randomized to optimized sensor settings (OSS) or

defaultt sensor setting {DSS). Standardized optimization was performed using three

differentt exercise tests. QOL questionnaires {OOL-q: Hacettepe, Karolinska and

RAND-36)) were used for evaluation of the sensor optimization. One month before

andd after optimization, exercise capacity using CAEP and the three QOL-q were

assessed. .

Results s

Fifty-fourr patients (26 men. 28 female) with a mean age of 65 16 years were

enrolledd in the study. In each group (OSS and DSS) 27 patients were included. One

monthh after sensor optimization the achieved maximal HR and METS were signifi-

cantlyy higher in OSS compared to DSS (124 28 vs. 108 20 bpm, p=0.036; 7.3 4

vs.. 4.9 4 METS, p = 0.045). Highest HRand METS were achieved in patients with

pacemakerss with accessible sensor algorithms. In patients with automatic slope

settingss (33 %). exercise capacity did not improve after sensor optimization. QOL

didd not improve in OSS compared to DSS.

Conclusions s

Afterr 1 month of individual optimization of rate response pacemakers, exercise

capacityy was improved and maximum heart rate increased, although QOL remained

unchanged.. Accessible pacemaker sensor algorithms are mandatory for individual

optimization. .

6 6

Chapterr 4

Introductio n n

Thee normal heart adapts its rate in response to the body's changing metabolic

demandss and is therefore chronotropically competent. Earlier generation of pace-

makerss provided constant rate and were unable to maintain or restore chronotropic

competence.. Since the first permanent pacemaker implantation in 1958, there have

beenn tremendous advances in pacemaker technology with the availability of com-

plex,, multiprogrammable, dual chamber and rate adaptive pacemakers which could

meett the haemodynamic needs of an individual patient.1"3 Rate adaptive pacing

improvess cardiac output, exercise tolerance and a sense of well-being compared to

fixedd rate pacing.4

Normall sinus node function is used as the golden standard for the development of

thee ultimate pacemaker sensor. Sinus node behavior can be described with several

methods:: exercise testing protocols, the mathematical model according to Wilkoff

andd Holter monitoring in daily life. These methods together with evaluation of the

obtainedd quality of lif e (QOL) associated with a given sensor can all be used to

assesss the efficacy of a pacemaker sensor. The American college of cardiology

guideliness of exercise testing used for evaluating rate adaptive pacing advices

adjustmentt of the exercise test for the subjects need and however several exercise

testss (CAEP, LITE, Bruce, six minute walk test, Kaltenbach step test, stair climbing)

aree described, consensus lacks about the ideal exercise test protocol for pacemaker

sensorr optimization. 8"10' I4"19

Programmablee pacemaker sensor features are frequently used in default sensor

settingg (DSS) in daily practice, although rate responsive pacing has been reported to

improvee the physical capacity and QOL compared to fixed rate pacing. It is un-

knownn whether individual optimization of rate response is a necessary factor to

improvee QOL and exercise capacity with the existing sophisticated sensor technolo-

gyy compared to the default sensor setting (DSS) of the manufacturer.8 Therefore,

wee tested the effect of optimized sensor setting (OSS) with a standardized individu-

all optimization protocol and DSS on exercise capacity and QOL in patients with

AAIR,, W1R and DDDR pacemakers in a prospective single blind randomized trial.

Material ss and Methods

Patientt population

Fifty-fourr consecutive patients with >75% pacemaker sensor driven HR with mean

agee of 65 16 years (range 27 to 89 years), New York Heart Association (NYHA) class

I-- II. stable medication, stable psychosocial conditions and commercially available

pacemakerss (Medtronic Minneapolis USA. Guidant Minnesota USA, Vitatron Arn-

hemm The Netherlands) were included at least 3 months after pacemaker implanta-

tion.. Pacemaker type and pacing indication are described in table 1.

Fromm our existing institutional database of exercise testing in healthy controls (HO

withoutt co-morbidity and medical therapy , we selected nineteen age and exercise

typee matched individuals (mean age 66.5 6). All patients gave written informed

Tablee 1. Demographics and clinical variables

Patients s

Numberr of patients Agee (year) Sexx Imale/femalel [%) BMII ikg/m2} NYHAA class \l-4) LVEFF (normal/reduced. %) History y

CoronaryCoronary artery disease (%' HeartHeart valve insufficiency (%) Atria!Atria! fibrillation {%)

Medication n DigoxwDigoxw (%l ^-blocker^-blocker (%) CalciumCalcium antagonist {%} AmiodaroneAmiodarone {%)

Facemakerr type VitatronVitatron \%) MedtronicMedtronic (!fci GuidantGuidant i%1

Pacemakerr indication SSSW SSSW AVV nodal block (%i AFAF with AV nodal disease [%) AFAF with His bundle ablation iV Other Other

DSS S

27 7 066 16

59/41 1 299 6

1.77 0.9 89/11 1

4 4 6 6 42 2

4 4 10 0 6 6 2 2

41 1 33 3 26 6

15 5 15 5 11 1 52 2 7 7

OSS S

27 7 655 15 37/63 3 266 4 1.88 9 89/11 1

6 6 12 2 58 8

4 4 14 4 10 0 2 2

60 0 29 9 11 1

22 2 15 5 18 8 42 2

3 3

Tablee 1. DSS= default sensor setting: OSS= optimal sensor setting: BMI - body mass index: NYHAA = New York Heart Association: LVEF= left ventricular ejection fraction: SSS~sick sinus syndrome.. AV = atrioventricular: AF = atrial fibrillation.

78 8

Chapterr 4

consentt and the ethics committee of our institutio n approved the study.

Randomizationn protocol and study design

I nn thi s s ingle bl in d prospect ive randomized trial , pat ients were firs t randomized to

OSSS or DSS. Primar y endpoint of the study is exercise capacity. Patients in OSS were

thenn randomized to one of three exercise protocols used for opt imizat ion of the

sensor.. The three exercise tests were the six minut e hall walk test (6-HWT) , six

m inu t ee hall walk test wi t h stair c l imbin g (6-HWT+SC) and the chronotropi c assess-

mentt exercise protocol (CAEP) according to Wilkoff. 11

Al ll pat ients underwent at basel ine a CAEP exercise test wi t h posture change (PC)

andd sui tcase lift in g (SL) wi t h their pacemakers programmed in default sett ing. The

pacemakerss of the pat ients in OSS group were individual l y opt imized 1 month after

basel ine.. The CAEP test was repeated after 2 mon ths in OSS and DSS groups.

Qual i t yy of l i f e quest ionnaires (QOL-q) were assessed in all pat ients, 1 month before

andd after sensor opt imizat ion (see figur e 1).

Patients s

Baseline: : Groupp OSS (n = 271

CAEPP + PC+ SL in default

Groupp DSS (n = 27)

22 months

Afterr I month optimization using ++ QOL

Afterr 2 months: QOL +

t t

HWT T

t t

HWTT + SC

CAEPP + PC+ SL CAEP++ PC+ SL

Figuree 1. Randomization and flowchart of the study. OSS- optimalized sensor setting: DSS-defaultt sensor setting: CAEP- chronotropic assessment exercise protocol; PC- posture change:: SL- suitcase lifting: QOlq = quahty of lif e questionnaires: 6-HWT= 6 minute hall walk lest: SC== staircase ascent and descent.

79 9

Sensorr optimization

Alll sensor optimizations were done by one investigator (A-E.Y). The optimized

pacemakerr sensor setting was individually determined by 3 parameters: 1) detailed

analysiss of HR curve obtained with the exercise test after 1 month (onset of exer-

cise,, slope, total exercise time, time to peak HR, maximal HR). One of the three

exercisee tests (depending on randomization) were used for optimization (CAEP, 6-

HWTT or 6-HWT+SC). HR curves of 19 age matched HC (see figure 2) and the availa-

blee literature about normal HR during exercise were used as reference.12 2) upper

ratee limi t was programmed according to Astrand (220 - age)13 and 3) development of

complaints,, Threshold, lower rate limi t (LRL), upper rate limi t (URL), slope and

sensorr specific settings (sensor blending in dual sensor systems) were adjusted to

obtainn the predicted optimal sensor setting.

HR R

180 0 - * -- 6-HWT

-•-CAEP P

CAEP-HC C

- * -- 6-HWT-HC

i—i—i—i—n—i— rr r r t i i i—i—r~i—i—i—r—i—i—i—n—!—I—i—i—i—i—i—i— r

11 4 7 10 13 16 19 22 25 28 31 34 Minutes s

Figuree 2. Changes in heart rate ImeanV HR-heart rate; 6-HWT= 6 minute hall walk test: CAEP = chronotropicc assessment exercise protocol. HC = healthy controls.

SO O

Chapterr 4

Exercisee test protocols

PosturePosture change and suitcase lifting

Alll patients were examined in the same conditions (before noon, uniform room

temperaturee and footgear (patients own shoes)). After instrumentation, patients

restedd supine on an examination table with one pillow for 5 minutes. They then

elevatedd to the sitting position and immediately to the standing position for 2

minutes.. The patients raised a standard suitcase (9 kg weight and measuring 47 cm

xx 37 cm x 15 cm) from the floor onto the examination couch (a height of 100 cm)

usingg their preferred arm (left, right, both).

ChronotropicChronotropic assessment of exercise protocol

Alll patients underwent a symptom limited treadmill test using the CAEP protocol

accordingaccording to Wilkoff .

Onee MET equals 35 ml oxygen uptake/kg body weight/min. representing the

approximatee metabolic cost to stand quietly. In his protocol oxygen consumption

andd carbon dioxide production was not performed, and thus metabolic workload

(METSS ) was not directly measured during exercise. Rather, metabolic levels during

eachh stage of exercise were estimated using treadmill grade and speed.

SixSix minute hall walk test

Afterr 5 minutes of rest patients were brought to the parcour. A parcour of 100 m

wass created by attaching stickers each meter in an oval form showing the walk

distance.. Patients were instructed to walk or run for 6 minutes at the parcour after

hearingg the start sign. After the symptom limited 6-HWT, the patients were brought

backk to the test room for a recovery period of 10 minutes.

StaircaseStaircase descent and ascent

First,, the patients descended 5 flights of stairs as rapidly as possible (82 steps, with

totall horizontal distance of 41.5 m and vertical distance of 15 m). After 2 minutes of

rest,, the patients then ascended the same flights as fast as they could.

81 1

Qualityy of lif e

Too compensate for the limited pacemaker patient specific OOL-q, we used 3 OOL-q.

Thee Rand-36 consisting 9 domains (physical functioning, social functioning, role

functioningg physical problems, role functioning emotional problems, mental health,

energy,, domains bodily pain, general health perception and change in general

health)) was used as a generic core module. The Rand-36 was evaluated using the

scoree system from 0 to 100 % in each domain (0% =low OOL to 100% = high QOL). H

Thee second questionnaire assessed was the Hacettepe QOL-q. which is a pacemaker

patientt specific questionnaire with 8 domains (general well-being, physical activity

andd symptoms, sleeping dysfunction, appetite, sexual functioning, cognitive

functioning,, social participation and work performance). The scores of each domain

aree between 3-50 points. (3 = low QOL and 50 = high OOL).

Fromm the final questionnaire, the Karolinska OOL-q only A domains (chest pain,

palpitations,, dizziness, dyspnoea) were used as complementary to the other two

questionnaires.. The 4 domains of the Karolinska QOL-q were evaluated using Visual

Analogg Scales and required patients to place a mark along a line of 10 cm in length

fromm a minimum of 0 (no complaints) to a maximum of 10 (maximal grade of

complaints).. The results were expressed as a percentage of the distance from the

discretee minimum point to the position of the mark divided by the length of the

line.155 In the Rand-36 and the Hacettepe QOL-q a high score means a high OOL and

inn the Karolinksa QOL-q a high score means a low QOL.

Measurements s

Beatt to beat HR was recorded during the physical tests with the Polar advantage

systemm using electrodes mounted in a belt (Polar Electro OY, Kempele, Finland).

Duringg the CAEP exercise test patients were also continuously monitored by 12-lead

electrocardiographicc recordings. HR at rest, time to peak HR, maximal achieved HR.

HRR at 10 minutes recovery period, exercise duration and METS were measured, The

OOL-qq was assessed after 1 and 2 months of inclusion.

82 2

Chapterr 4

Statisticall analysis

Powerr analysis was performed on the primary endpoint of the study, exercise

capacityy in METS. Assuming that the common standard deviation is 35 and a mean

differencee in achieved METS is 3 (the difference between mean DSS l and OSS, JJL2

inn METS ), we calculated the power of 80 % using a two group t-test with a 0.05 two-

sidedd significance level. A sample size of 23 patients in each group DSS vs. OSS had

aa power of 80% to detect a difference in means of 3. Parametric data was analysed

usingg the Student's t-test while nonparametric data was analysed using the Mann-

Whitneyy U test. All data are expressed as mean SD. A P value <0,05 is considered

statisticallyy significant.

Results s

Fifty-fourr patients (26 men, 28 female) with a mean age of 65 16 years (range 27

too 89 years) were enrolled in the study. In each group (OSS and DSS) 27 patients

weree included. Demographics and baseline characteristics between both groups

weree not significantly different (see table 1).

Fivee patients (1 in OSS and 4 patients in DSS) withdrew from the study. The patient

inn group OSS stopped because of diagnosis of a pulmonary tumour. In group DSS,

threee patients stopped after programming to DSS because of symptoms of heart

failure.. The forth patient withdrew for psychosocial reasons.

Pacemakerr and sensor type

Thee majority of patients (58% in OSS group vs. 39% in DSS group) had pacemakers

equippedd with dual sensor systems of Vitatron (QT and activity. Vitatron. Arnhem

Thee Netherlands). Activity sensors were the most frequent used sensors in the

singlee sensor systems, The OSS group had 31% of activity sensors of Medtronic

(Medtronic.. Minneapolis USA), while the DSS group had 35% Medtronic and 9%

Guidantt (Guidant. Minnesota USA) activity sensors. Accelerometers of Guidant

(Guidant,, Minnesota USA) were used 17% in the DSS group and 11% in OSS.

83 3

Tciblee 2. Pacemaker optimization settings

Mode e Baseline e

WIRR i%! WII i%* DDDRR i%i ODDD i%l AAIRR i% AAII i%l

4 4

26 6 9 9 4 4 7 7

fticemakerfticemaker settings changed Baseline

Afterr optimization

59 9 0 0 35 5 0 0 6 6 0 0

Afterr optimization Frequency of change (%)

LRMbprrO O URLL ibpm) THH ilow, medium/low. medium,, medium/high, high' S l o pee [*•)

Sensor r

60 0 1211 4.4f medium m

I I

default t QT=ACT T

611 2.7

14544 14.31 low,, medium/low

moree agressive OKACT.. ACT

12 2 92 2 11 1

15 5 31 1

Tablee 2. LRL = lower rate limit ; URL = upper rate limit : TH^threshold: QT= OT sensor: ACT = activityy sensor: \*) dependent of the manufacturer: response factor, response time or acceleration time:: blending OT=ACT-> 50:50. QT<ACT-> 25:75: t URL after optimization is significantly different,, p <0.001.

Sensorr optimization

Ratee response was programmed on in all patients. The LRL was changed in 12% of

thee patients but increased not significant. The URL was programmed higher in 92%

off patients, from 121 4.4 to 145.4 14.3 bpm, (p<0.001). The threshold was

changedd in 11% of the patients: from medium to medium/ low in 4% vs. medium to

loww in 7 % and the slope settings were changed in 15 % of the patients. The re-

sponsee factor vs. response time vs. acceleration of the slope was programmed in a

moree aggressive setting: from factor 8 to 14 in 7 % vs. 30 to 10 seconds in 4 % vs.

standardd to fast in 4 %. In 33% of patients, the slope settings could not be adjusted

duee to the autoslope setting inVitatro n pacemakers (Arnhem, the Netherlands),

despitee the need for a more aggressive slope setting. Sensor blending was changed in

311 %. In 12% the default sensor blending setting (QT = ACT) of the Vitatro n pacemak-

erss (Arnhem, the Netherlands) were programmed to activity only and in 19 % to

OT<ACT ,, see table 2.

84 4

Chapterr 4

Tablee 3. Summary of HR during rest, exercise and recovery

Restingg rate ibpm.1-Timee to peak HR ymui' Programmedd maximum Achievedd maximum HR 100 mill recovery ibpm.1

METS S

Exercisee duration mini. M.tximall HR during SI. Posturee change

HRR \bpm l

^bpm.1 1

DSS S

Baseline e

vn-27^ ^ Default t

677 14

100 5 1211 4

1099 21 744 10 66 4 111 5

711 12 722 11

22 months

li ii = 231 Default t

633 29 100 7

I21 4" 11 OS 20 #

666 6 4.99 4"

100 5 688 10

688 + 6

Baseline e

inn = 27 Default t

711 14

n 5 121 4

1144 15 777 10

7 4

5 5 711 13 700 10

inn = Q CAEP P

700 13 111 +. 0

1477 + I4f I I SS 9 699 15

66 + 4 111 5

7 6 + 17 7

766 5

OSS S

22 months

inn = 81 6-HWT T

666 + 7 100 4

1422 12 S 128++ 34

65 7 7.88 4*

122 5 9 9

67 6

m-Ql l &-HWTT + SC

666 9 1 1 +5 5

1477 18+ * 1233 29 #

62 8

7.99 4* 133 * 4 68 8 67 7

P-value e

ns s ns s

t<0 .001 1 -*1== 0.036

ns s

0.045* * ns s ns s nn s

Tablee 3. HR= heart rate; bpm = beats per minute: min = minutes; DSS = default sensor setting. OSSS = optimalized sensor setting: CAEP= chronotropic assessment exercise protocol; 6-HWT = 6 minute halll walk test; SC= staircase ascent and descent; SL = suitcase lifting; #-11= the achieved maximal heart ratee is significantly higher in the total group of OSS compared to DSS, p = 0.036; * = the achieved METS wass significantly improved in the subgroups 6-HWT and 6-HWT +SC compared to These three exercise testss iCAEP. 6-HWT and ó-HWT +SO were used for optimization. DSS. p = 0.045.

HR R

140 0

1300 -

120 0

110 0

100 0

90 0

80 0

70 0

60 0 50 0

40 0

-*-DSS-B B -•—DSS-- 2 months

Rest t Exercise e Recovery y ~ii i r r T r ~!! T

33 6 9 12 15 18 21 24 27 30 33 36 Minutes s

Figuree 3. Changes in heart rate tinea nV HR = heart rate. DSS-B = default sensor setting at baseline.

S5 S5

HR R 160 0

140 0

120 0

100 0

SO O

60 0

40 0

-è^^ OSS-B 22 months

Rest t Exercise e Recovery y

- x -- -x--x-

11 3 5 7 9 11 13 15 17 19 21 23 25 27 29 31 33 35 Minutes s

Figuree 4. Changes in heart rate (mean). HR-heart rate: OSS-B- optimalized sensor setting at baseline.. *P = 0.036.

Exercisee test protocols

PosturePosture change, suitcase lifting

Thee achieved maximal HR during posture change and suitcase liftin g were compara-

blee in OSS and DSS at baseline and after 2 months (see table3). One patient could

nott lif t the suitcase.

chronotropicchronotropic assessment exercise protocol

Thee achieved maximal HR and METS after sensor optimization were significantly

higherr compared to DSS (124 28 vs. 108 20. p = 0.036: 7.3 4 vs. 4.9 4 METS,

p=0.045). .

Highestt increase in maximal HR, METS and exercise duration was achieved in the

subgroupss whereby the sensor was adjusted with 6-HWT (128 34 vs. 7.8 4 vs.12

86 6

Chapterr 4

5)and6-HWT +SC (123 29 vs. 7.9 4 vs. 13 4) compared to the subgroup

CAEPP (118 19 vs. 6 4 vs. 11 5Ï-

Otherr measured parameters between the groups and subgroups like resting rate,

timee to peak HR, HR at 10 minute recovery, maximal HR during suitcase liftin g and

posturee change were comparable between the two CAEP tests (see table 3, figure 3

andd figure 4).

Qualityy of lif e

Thee achieved scores from all the three QOL-q (RAND-3Ó, Hacettepe and Karolinska)

onee month after optimization compared to DSS were not significantly different.

Alsoo scores attained within the subgroups of the three QOL-q were not significantly

different. .

Discussion n

Exercise e

Ourr study shows that individual optimization of pacemaker sensors results in

improvedd exercise capacity. Also, after sensor optimization the reason for stopping

thee exercise test shifted to concomitant restrictions (e.g. physical activity level of

thee elderly pacemaker patients) rather than cardiovascular limitations. The in-

creasedd METS and HR indicate that patients have an improved exercise capacity

afterr sensor optimization

Thee highest benefit from optimization in achieved METS and maximal HR was

obtainedd with the 6-minute HWK and 6-minute HWK+ SC. In the CAEP subgroup,

theree were 33% more pacemakers with automatic slope settings, which could not be

individuallyy adjusted. Other possible factors, which could influence the exercise

capacityy between the groups. like medical history and medical therapy, were not

significantlyy different. These results underline the importance of individually

adjustmentt of pacemaker sensors and the necessity of accessible pacemaker

algorithms. .

Too date, limited studies investigated the effect of sensor adjustment and to the best

87 7

off our knowledge: our study is the first and largest randomized controlled study,

comparingg DSS with OSS in detail. Previous studies mainly compared different

sensorss and were not primarily designed to evaluate individual adjustment,5

Sulkee et al. showed that appropriate programming of sensors is crucial in rate

responsivee pacing. & In 20 patients the effects on exercise and QOL of appropriate,

overr and under programming of the sensor were assessed. In contrast to our study,

onlyy activity sensors were evaluated and the sensors were optimized according to

thee manufacturer's instructions without detailed insight in these instructions. In

anotherr study, Klonis et al. investigated whether automatic algorithms for sensor

optimizationn could reduce clinical follow up time compared to manual adjustment.

Usingg the activity level of patients (more active, the same, or less active) they

concludedd that automatic adjustment is less time consuming than manual adjust-

ment,, This study supports the current trend in increasing automaticity in pacemak-

ers,, although it is disputable if such a rough guide as three levels of activity can be

usedd as a guide for sensor optimization. In general, sensor optimization methods

aree poorly described, mostly referring to the manufacturers advice.

Inn agreement with previous reports, exercise duration was not significantly im-

provedd after sensor optimization.8 2D This can partly be explained by the category of

patientss in need of permanent pacing (elderly and relatively sedentary). In these

patients,, exercise is often limited by loss of muscle strength and mass rather than

cardiopulmonaryy capacity. This loss of muscle strength is particularly apparent

whenn exercise testing is performed on a bicycle or treadmill. The CAEP exercise test

consistt nonlinear characteristics. The first 10 minutes it requires a low metabolic

workload,, beyond which it increases abruptly. Patients with preserved functional

capacity,, capable of exercising for more than 10 minutes, may quit before reaching

maximall 02 uptake, mainly because of excessive increments in workload near the

endd of the test, thus, being limited by mechanical rather than metabolic barri-

ers.12211 In our study there is a trend to increase of exercise duration with a mean of

11 5 minutes. This increased exercise duration results in a higher achieved speed

andd steeper slope on the CAEP. And this longer exercise duration is achieved due to

thee higher achieved heart rate. The symptom limited 6 minute hall walk test resem-

bless closer daily activities and less time consuming compared to the treadmill or

bicyclee exercise tests. 22

Thee healthy controls achieved higher HR than patients despite individual optimiza-

tion,, probably because current sensors are still hypochonotropic and physicians feel

88 8

Chapterr 4

reluctancee to program such high upper rates due to concomitant heart disease. The

maximall HR in the default setting is generally too low, well below the age predicted

maximall HR of the mean age of a typical pacemaker patient of 70 year, according to

thee Astrand formula (220-age). An inappropriately low programmed maximum sensor

ratee and failure to reach maximum exercise will result in sensor indicated pacing rates

duringg exercise testing below the calculated expected HR. The clinician may attempt

too compensate for this by programming more aggressive rate response parameters

whichh may result in excessive rate response behavior of the pacemaker during the

patient'ss normal ambulatory activities.0

Qualityy of lif e

Qualityy of life did not improve with individual sensor optimization. Specific sub-

groupss that derived benefit were not observed, including stratified to device indica-

tion,, age or specific complaints. Again, only the study of Sulke et al. assessed the

effectt of rate responsive pacing on QOL. Appropriate, over and under programming

off the rate response were evaluated in both dual and single chamber activity sensor

ratee adaptive pacemakers with visual analog scales and specific activity question-

naires.. Symptoms were least in the appropriate programmed pacemakers after 2

weekss of follow up. The found differences between both studies could be explained

duee to the greater programming steps in Sulke's study (rate response off, appropri-

atee programming and aggressive programming). Another explanation for the failure

too improve QOL after sensor optimization could be the relatively good baseline

functionall capacity, because patients with relative preserved functional capacity at

enrollmentt show the lowest improvement in health related values.25

Alsoo in large pacemaker trials, improvements in QOL were minimal. The PAcemaker

Selectionn in the Elderly study (PASE, n=407) showed only in the subgroup with sick

sinuss syndrome a moderate improvement in QOL in patients with dual chamber

pacingg as opposed to ventricular pacing, whereas in the Canadian Trial Of Physiolog-

icc Pacing investigating (CTOPP, n = 172II only the global well being score was better

inn the physiologic vs. WIR mode.

Inn general, there is a lack of validated pacemaker patient specified quality of life

QOL-q.244 To date, only two questionnaires were specifically developed for pacemak-

err patients (Hacettepe Qol-q2\ Aquarel QOL-q). This complicates comparing

studiess using different QOL-q's and methods for assessing QOL (transtelephonic,

89 9

selff administered at hospital by researcher /patient). Therefore, it is advisable to

interprett QOL data in pacemaker patients cautiously.15 A The recent developed

Aquarell Qol-q is an accurate QOL-q for study design as our present study, therefore

wee suggest to use this OOL-q in future studies. We could not use this QOL-q for our

studyy because it was not available.

Inn this study we showed an improvement in exercise capacity without improvement

inn QOL. The reverse is shown in cardiac resynchronization therapy studies with

onlyy minor changes in exercise capacity yet major changes in QOL. The PATH-CHF

studyy there was only a minor improvement of 60 meters after cardiac resynchroniza-

tionn therapy while QOL improved clearly.2' These disconnect between exercise

capacityy and QOL can be explained possibly due to several factors. The pacemaker

implantationn has already a large impact and therefore it is more difficult to obtain

moree additional improvement. In this study we compared in hospital tests with

evaluationn of QOL based on the activities at home. As described above the in- hospital

testss are artificial, therefore it would be theoretically and scientifically better when

wee compared the sensor function at home using holter registration with QOL.

Conclusions s

Afterr 1 month of individual optimization of rate response pacemakers, exercise

capacityy and maximum HR was improved, although QOL remained unchanged.

Accessiblee pacemaker sensor algorithms are mandatory for individual optimization,

althoughh automatic features are indisputably important in the pacemaker sensor

development. .

Developmentt of sensor algorithms which create the possibility for monitoring

sensorr behavior in detail at home is a great challenge to overcome the disadvantages

off hospital exercise tests for sensor optimization.

Acknowledgments s

Thee authors thank Wandena Ramsoekh. pacemaker technician for the technical

supportt and Michael Kortz, MD (department of Cardiology, Flevohospital. Almere.

Thee Netherlands) for helping with the inclusion of patients.

Q0 0

References s

Chapterr 4

1.. Lass J, Kaik J, Meigas K, Hinrikus H, Blinowska A. Evaluation of the quality of rate adaptation algorithmss for cardiac pacing. Europace. 2001:3:221-8.

2.2. Sack S. [Modern pacemaker therapy]. Herz. 2001;26:1. 3.. Elmqvist R. Review of early pacemaker development. Pacing Clin Electtophysiol. 1978:1:535-6. 4.. Leung SK. Lau CP. Tang MO. Cardiac output is a sensitive indicator of difference in exercise

performancee between single and dual sensor pacemakers. Pacing Clin Electrophysiol. 1998;21:35-41. .

5-- Zegelman M. Cieslinski G, Kreuzer J. Rate response during submaximal exercise: comparison off three different sensors. Pacing Clin Electrophysiol. 1988:11:1888-95.

6.. Freedman RA, Hopper DL, Mah J. Hummel ). Wilkoff BL. Assessment of pacemaker chrono-tropicc response: implementation of the Wilkoff mathematical model. Pacing Clin Electrophys-iol.iol. 2001:24:1748-54.

7.. Lau CP. Leung SK. Guerola M, Crijns HJ. Comparison of continuously recorded sensor and si-nuss rates during daily lif e activities and standardized exercise testing: efficacy of automatical-lyy optimized rate adaptive dual sensor pacing to simulate sinus rhythm. Pacing Clin Electro-physiol.physiol. 1996;19:1672-7.

8.. Sulke N. Dritsas A. Chambers J. Sowton E. Is accurate rate response programming necessary? PacingClinPacingClin Electrophysiol. 1990:13:1031-44.

9-- Strobel JS, Kay GN. Programming of sensor driven pacemakers. Cardiol Clin. 2000:18:157-76. ix. 10.. Kay GN. Quantitation of chronotropic response: comparison of methods for rate- modulating

permanentt pacemakers. J Am Coll Cardiol. 1992:20:1533-41. 11.. Wilkoff BL, Miller RE. Exercise testing for chronotropic assessment, Cardiol Clin. 1992:10:705-

17. . 12.. Ellestad MH. stress testing principles and practice.tirth edition:567-573-13.. Astrand. Aerobic work capacity its relation to age, sex. and other factors, circulation research.

1967;20:1-2111 1-217. 14.. Hays RD, Sherbourne CD, Ma2el RM. The RAND 36-Item Health Survey 1.0. Health Econ.

1993:2:217-27. . 15.. Linde C. How to evaluate quality-of-life in pacemaker patients: problems and pitfalls. Pacing

ClinClin Electrophysiol. 1996:19:391-7. 16.. Klonis D. Zhang X, Patel U. Gulamhusein S, Patel J, Hurwit H. Banish D. Marco D. Automatic

sensorr algorithms expedite pacemaker follow-ups. Pacing Clin Electrophysiol. 2003:26:225-8. 17.. Sulke N, Pipilis A, Bucknall C. Sowton E. Quantitative analysis of contribution of rate re-

sponsee in three different ventricular rate responsive pacemakers during out of hospital activi-ty.. Pacing Clin Electrophysiol. 1990;13:37-44.

18.. Bellamy CM, Roberts DH, Hughes S, Charles RG. Comparative evaluation of rate modulation inn new generation evoked QT and activity sensing pacemakers. Pacing Clin Electrophysiol. 1992:15:993-9. .

19.. Mahaux VA. Verboven Y], Waleffe A, Kulbertus H. Stepwise analysis of the calibration proce-duree of an accelerometer-based pacemaker. PacingClin Electrophysiol. 1994;17:1955-9-

20.. Hasegawa A. Hatori M, Amano M, Iijima T. Adachi H, Yamaguchi E. Fukuda T, Murata K. Nagai R.. Adequacy of pacing rate during exercise in rate responsive ventricular pacing. Pacing Clin Electrophysiol.Electrophysiol. 1997:20:307-12.

21.. Page E. Bonnet JL, Durand C. Comparison of metabolic expenditure during CAEP versus a test adaptedd to aerobic capacity (Harbor test) in elderly healthy individuals. Pacing Clin Electro-physiol.physiol. 2000;23:1772-7.

22.. Hayes DL, Von Feldt L, Higano ST. Standardized informal exercise testing for programming ratee adaptive pacemakers. Pacing Clin Electrophysiol. 1991;14:1772-6.

23.. Lopez-Jimenez F, Goldman L. Orav EJ. Ellenbogen K. Stambler B, Marinchak R, Wilkoff BL, Mangionee CM. Yoon C, Vitale K, Lamas GA. Health values before and after pacemaker implan-tation.. Am Heart J. 2002;144:687-92.

24.. Stofmeel MA. Post MW. Kelder JC. Grobbee DE, van Hemel NM. Quality-of-life of pacemaker patients:: a reappraisal of current instruments. Pacing Clin Electrophysiol. 2000:23:946-52.

91 1

25.. Oto MA, Muderrisoglu H, Ozin MB. Korkma2 ME. Karamehmetoglu A. Oram A, Oram E. Ugur-luu S. Quality of lif e in patients with rate responsive pacemakers: a randomized, cross-over study.. Pacing Clin Electrophysiol. 1991:14:800-6.

26.. Stofmeel MA. Post MW, Kelder JC. Grobbee DE, van Hemel NM. Psychometric properties of Aquarel,, a disease-specific quality of lif e questionnaire for pacemaker patients, ƒ Clin Epide-miolmiol 2001:54:157-65.

27.. Huth C. Friedl A, Klein H. Auricchio A. [Pacing therapies for congestive heart failure consider-ingg the results of the PATH-CHF study]. Z Kaidiol. 2001:90 Suppl 1:10-5.

92 2

Cerebrall blood flow velocity and cardiacc output at the onset of

dynamicc exercise at two settings of pacemakerr determined heart rate

Aytenn Erol-Yilmaz MD1, Lysander W.J. Bogert MD,2 Raymond Tukkie MD PhD1, Arthur A.. Wilde MD PhD1. Wouterr Wieling MD PhD2 and Johannes J. van Lieshout MD PhD 2

Clinicall and Experimental Cardiology1, Department of Internal Medicine2. Academicc Medical Center. Amsterdam, The Netherlands.

Dr.. Erol-Yilmaz and Dr. Bogert contributed equally to the article

SubmittedSubmitted for publication

Abstract t

Background d

Inn cardiac patients, the increase in the transcranial Doppler determined middle

cerebrall artery (MCA) mean flow velocity (Vmean) during exercise depends on the

abilityy to increase cardiac output (Q), The aim of this study was in six patients (66

(49-76)) yr.) with complete heart block and ventricle rate adaptive pacemakers to

evaluatee whether variation in heart rate (HR) affects the increase in MCA V m e an and

OO during exercise.

Material ss and methods

Thee effect of setting the pacemaker to "default" (DSS) vs. "optimized" (OSS) on

bloodd pressure (BP).Q, stroke volume (SV1 and MCA Vmean was evaluated during

gradedd ergometry cycling. During the first 3 min. of exercise in OSS vs. DSS at 25

andd 75 W the rise in HR (26 (1-59) vs. 13 (0-25) bpm; P<0.05 and 43 (0-83) vs. 27 (2-

43)) bpm: P<0,05, respectively) was larger. However, with OSS the increment in SV

wass smaller at 75 W( + 33 (-4-71)% vs. +51 (18-88)%: P<0.05) resulting in a compa-

rablee rise in Q (+95 (73-109)% vs. +101 (58-119)%). For both OSS vs. DSS there was

noo significant increase in MCA V m e an during exercise. The manipulation of the

pacemakerr setting had no effect on the maximal workload (133 (100-225) vs. 129

(75-200)) W).

Conclusions s

Thee results indicate that in patients provided with a rate adaptive pacemaker. MCA

VV , changes in cardiac output and work capacity do not depend on the increase

inHR. .

96 6

Chapterr 5

Introductio n n

Inn both animals 8 and humans cerebral blood flow (CBF) 14 A& and middle cerebral

arteryy (MCA) mean flow velocity (Vmean)22 24 increase —25% on the transition from

restt to moderate exercise.14'i q 25 33 However, the increase in MCA Vmean becomes

attenuatedd when the ability to raise cardiac output (Q) is limited as in patients with

atriall fibrillation23 and in patients with heart failure in whom the substantial

reductionn in CBF is reversed by cardiac transplantation15 Normally both chronotrop-

icc and inotropic cardiac adaptive mechanisms are driving Q in the early stages of the

circulatoryy adaptation to exercise heart rate (HR) being responsible for as much as

—75%% of the increase.4

Withh such a large contribution of HR to the increase in Ó at the onset of exercise,

thee setting of pacemakers may be critical for the patient's ability to maintain daily

activities.. Manipulation of HR is possible with rate-adaptive pacing and the ability to

increasee may be improved6 40 Rate adaptive pacemakers are used in patients with

atrioventricularr (AV)-nodal block and the rate-adaptive pacing is considered preferable

forr patients with chronotropic incompetence.31 Patients with complete heart block

followingg curative AV ablation for atrial fibrillation are provided with a rate-modulat-

edd pacemaker, i.e. the pacing rate is modulated based on one or more internal sensors

thatt detect exercise and metabolic need.31 In these patients there are several ways to

programm the pacemaker and evaluate the effects on HR and Q.

Thee increase in Q and blood pressure (BP) during dynamic exercise is preserved

whenn there is littl e change in HR, such as in patients with a heart transplant or in

healthyy individuals following autonomic blockade,30 32 35 We considered in patients

dependentt on permanent rate adaptive ventricular pacing, whether manipulation of

thee HR response would enhance the HR response to exercise and in turn MCA Vmean.

QQ and work capacity.

Methods s

Studyy Population

Sixx ventricular paced untrained patients (66 (49-76) years) with permanent program-

mablee ventricular rate adaptive pacemakers (Vitatron, Arnhem Netherlands, or

97 7

Guidant,, Minnesota USA) participated in the investigation (table 1). Pacemakers were

placedd because of drug resistant atrial fibrillation with His bundle ablation [n = 5). and

sickk sinus syndrome (n=Il. On the basis of history, medical and echocardiographical

examinationn of myocardial function all but one patient were considered without any

evidencee of inotropic dysfunction. One patient with a history of myocardial infarction

usedused amiodarone for recurrent non-sustained ventricular tachycardia and had a left

ventricularr ejection fraction of 25%. Separate analysis of the study data of this patient

revealedd no significant differences in any of the cardiovascular responses to exercise

andd data from this patient were included in the group analysis. All patients received a

verball and written explanation of the objectives and techniques of measurements and

riskss and benefits associated with the study and provided written informed consent

inn accordance with the Helsinki Declaration.

Tablee 1. Patient characteristics and pacemaker settings

Patient t no. .

1 1 2 2 3 3 4 4 5 5 6 6

Age e (yeaa rs)

69 9 49 9 76 6 67 7 58 8 57 7

Sex x IM/Ft t

M M F F M M F F M M F F

Diagnosis s

AF F AF F AF F AF F SSS S AF F

Defa a Sensor r

OT=ACT T OT=ACT T QT=ACT T ACTT only QT=ACT T ACTT only

ultt setting URL L

120 0 120 0 120 0 120 0 120 0 130 0

TH H

Med d Med d Med d Med d Med d Med d

Optimizedd setting Sensor r

OT<ACT T QKACT T OT<ACT T ACTT onlv QT=ACT T ACTT only

URL L

150 0 170 0 150 0 150 0 150 0 170 0

TH H

L L Med d MH H Med d Med d ML L

Tablee 1. URL; upper rate limit ; TH; threshold: M: male; F; female; AF: atrial fibrillation with His bundlee ablation; SSS: sick sinus syndrome; L: low: Med; medium: MH: medium high: ML: medium low;; QT; QT-sensor; ACT: activity sensor OT = ACT: equal contribution of each sensor on the pacing rate;; OT<ACT: 25% contribution of OT and 75% contribution of ACT sensor on pacing rate.

Heartt rate

Duringg consecutive runs of symptom limited exercise, using a chronotropic assess-

mentt exercise protocol ^ the pacemaker sensor setting was optimized in accord-

ancee with a database ^ to produce a physiological HR response to exercise derived

fromm electrodes mounted in a belt (Polar advantage system. Polar Electro OY. Kem-

pele.. Finlandl. The pacemakers were provided with an activity sensor that uses a

piezoelectricc crystal for detection of body movement and also a OT sensor, which uses

ratee independent shortening of the OT interval during increased sympathetic activity

ass input. In the pacemakers with a OT sensor during a learning procedure, lasting

98 8

Chapterr 5

severall weeks, software incorporated in the pacemaker correlates the longest QT

intervall and zero activity counts to an individual at rest, and the shortest OT interval

andd the maximum number of activity counts to the maximum level of exercise.

Ultimately,, the rate moves between the lower rate and the maximum sensor rate.

Thee HR response to exercise was used to vary the level of input between the activity

sensorr and the QT sensor, and the threshold of the activity sensor was adjusted to

thee minimal level of vibration required for producing a rate response. With succeed-

ingg series of exercise, pacemaker output was adjusted in a stepwise manner until

thee derived HR curves (optimized sensor sett ing= OSS) approached the reference

curvess 52. OSS was established further by increasing the URL from 122 (SD 4) to 157

(SDD 10) beats per min {bpm) (table l) with the upper rate limi t (URL) set as

(220-age)) bpm,3

Studyy protocol

Afterr completion of the sensor learning period and at least 2 months following

sensorr optimization, patients were admitted to the laboratory for exercise testing,

afterr being fitted with a headgear for transcranial Doppler (TCD) measurement of

MCAA velocity. After a resting period of 15 min in the seated position, resting data

weree collected and the patients performed graded ergometry cycle exercise with

pacemakerr settings in the manufacturers default setting (DSS) and with OSS.

Uprightt exercise was performed on an electrically braked stationary cycle ergometer

fromm rest to maximal exertion with a constant pedaling speed of 60 rpm. Exercise

startedd at 25 W for 3 min. From then on each stage increased wattage from the

previouss stage by 25 W and lasted 2 min in duration. The exercise was stopped

whenn the patients were no longer able to maintain 60 rpm. Then, they rested in the

supinee position for at least 45 min to allow cardiovascular variables to return to

baselinee levels. Thereafter, the pacemaker settings were switched from OSS to DSS

orr vice versa. The order of the pacemaker settings was fixed in 4 of the 6 patients

accountingg for the requested OT sensor learning time and these patients started

exercisee with the pacemaker in OSS. Both the patients and the observer were

blindedd for the settings of the pacemaker.

99 9

Cerebrall and Central Hemodynamic Measurements

Bloodd pressure was measured non-invasively by photoelectric plethysmography

withh a Finapres [Model 5; Netherlands Organization for Applied Scientific Research,

Biomedicall Instrumentation, TNO-BMI, Amsterdam, The Netherlands) with the

fingerr cuff on the mid-phalanx of the middle finger of the dominant hand. To avoid

hydrostaticc level differences, the hand was held at right-atrial level in the mid-

axillaryy line. In the Finapres device, a built in expert system (Physiocal) was in

operationn to establish and adjust a proper volume clamp set point (17). During

progressivee exercise to fatigue there is no significant difference in the intercept

valuee from zero between systolic [SBP), mean (MAP), and diastolic blood pressure

(DBP)) between radial and finger arterial pressure with validity correlations ranging

fromm 0,93 to 0.99 (9. 11. 44). Left ventricular ejection time (LVET) was the time from

upstrokee to dicrotic notch.

Thee SV was determined by a three-element model of arterial input impedance (53)

whichh provides accurate estimates of changes in SV during postural stress (17).

Modell simulated (modelflow) Qfrom radial or finger arterial pressure follows

changess O in as determined by thermodilution in direction and degree during

cardiacc surgery, postural stress, and shock (17, 29. 53). Changes in Doppler echocar-

diographyy determined SV are tracked by model SV during the Valsalva maneuver,

passivee til t and exercise (38, 45. 51). If accurate absolute values are required, the

methodologyy needs calibration against a standard method (28, 50); otherwise, O is

expressedd as changes from control with the same precision in O (17. 27. 37). Q was

Restt Exercise 25W DSSS OSS DSS OSS

MAPP immHg) PPP tmmHgl HRibpmï ï LVET T SV% % Qi%l l TPRR t%l Pp.,CO,, immHg!

ff i tri in ')

899 173-105) 588 \34-73) 700 160-76)

0.277 (0.24-0.32 100 0 100 0 100 0 333 \29-371 466 137-62) 166 d 2-20)

933 175-1081 599 (37-72) 733 105-87)

0.277 t0.24-0.3r 100 0 100 0 100 0 333 L29-371 466 i35o6) 166 (12-191

92 2

85 5

83 3 0.29 9 136 6

161 1 66 6

35 5 48 8

20 0

\76-112* *

152-106H H 165-9911 t

(0.25-0.35^ ^

i l 2 I - 1 5 D t t

i l41-196) t t (50-7311 j .30-37) )

i83-66^ ^

112-27)) t

103 3 81 1

99 9 0.27 7

115 5 154 4

71 1

37 7 50 0

21 1

186-120)*" " (55-109) )

174-128)) * * f l0.22-0.3411 *

87-13«)** *

1128-170)) t (65-8511 t 132-4211 j

144-63' '

i !! 7-251 t

100 0

Chapterr ~>

SVV times HR. Total peripheral resistance iTPR) was the rati o of BP and Q expressed as

changess from its starting value in the sitting resting position prior to exercise (37).

Changess in cerebral perfusion were evaluated by following the blood flow velocity

inn the MCA by insonating its proximal segment (Multido p X. DWL, Sipplingen,

Germany)) through the posterior temporal "window " (1). When the optimal signal-

to-noisee rati o was obtained the probe was secured with a headband (Mark 600,

Spencerr Technologies. Seattle WA, USA), End-tidal C02 (PETC02) as an estimate of

arteria ll C02 tension (PaC02) and breathing frequency (f) were obtained from the

capnogramm by means of a nose tubing connected to a sampling infrare d C02 analyz-

err (Datex Normocap 200. Helsinki. Finland).

Analysis s

Forr off-lin e analysis signals of arterial pressure, MCA velocity and capnogram were

analogue-to-digitall converted at a sampling rate of 100 Hz and stored on disk. Beat-

to-beatt SBP and DBP values were derived. MAP and MCA V were obtained, mean n

respectively,, as the integral of pressure and velocity divided by the corresponding

beatt interval . Pulse pressure was the SBP- DBP-difference and HR was obtained from

thee pulse pressure interval . The HR responses produced with DSS and OSS were

evaluatedd according to the maximal HR attained and to the slope of the regression

off HR vs. time durin g the first 3 min of exercise.

Forr comparison data were transformed to equidistantly re-sampled data at 2 Hz by

polynomiall interpolation and expressed as 10 s averages. Baseline measurements

Tablee 2. Values are expressed as mean and range. DSS: defaultt sensor setting: OSS; optimized sensor setting, MAP:: mean arterial pressure, PP: pulse pressure, HR: heartt rate, LVET: left ventricular ejection time. . SV: strokee volume. ; cardiac output, TPR: total peripheral resistance.. P^CO.,: end-tidal COr f: breathing frequency. Vm.ir]:: cerebral mean blood velocity-. DSS vs. OSS * p<0.055 ** p<0.01 rest vs. exercise f p<0.05, J p<0.01.

94 4 104 4

97 7 0,26 6

151 1 20] ]

56 6

36 6 44 4 22 22

Exercis ee 75W DSS S

171-105) ) (78-123) )

(67-117)) t i0.22-0,33^ ^ (118-186) )

\\\\ 58-2274

i53-8D* * \32-42) )

i41-491 1 (16-294 4

100 0 102 2

116 6 0.24 4 132 2

195 5 56 6

37 7

45 5 23 3

OSS S

(79-1151** * 172-1161 1

(73 -152 ) + n n (0.20-0.3011 * 1 191-172)* *

(176-208)* * 148-66)) t

02-451 1 \42-51) )

(17-33)1 1

101 1

Exercisee in DSS

,0 0

0Ü 0Ü

1755 "

11 50-

1255 "

11 00

755 "

50--

5. .

OS S EC C

1755 -

150 0

125--

100 0

755 "

50 0

-- -

00 300 600

Timee (s)

Exercisee in OSS

300 300 600 0

Timee is)

Figuree 1. HR responses to exercise at two levels of pacemaker rate response. Individual (thin lines) andd group averaged (thick lines). HR during default Ueft panel) and optimized (right panel) settings off the pacemaker during exercise. DSS: default sensor setting, OSS: optimized sensor setting. HR: heartt rate.

aree averaged values over a 1 min resting period seated on the bicycle prior to

exercise.. If data fitted a normal distribution as determined by Kolmogorov-Smirnov

testt with Lilliefors' correction they were examined by paired t-test and otherwise by

Wilcoxonn signed-rank test. Data are expressed as mean and range in tables and as

meann S.E.M in figures. A p-value <0.05 was taken to indicate a difference.

102 2

Chapterr 5

w> > _. . --B B

. .

220 0

140 0

100 0

60 0

, - i . ' '

s s > >

,--fci fci u. . oS S

::: :

40 0

0 0

160 160

140 0

100 0

80 0

60 0

220 0 L80 0 140 0 100 0 60 0

66 54 3 2

I JJ I 1

. . . . . . . . . . . . . . . . . . . . . . .

w i ÉÉ , Ï • • • t • • • • • • • * * 8 * • i

,, 4 . * * * * H oo Q 9 2 9 9 22 5 V 2 f ? ^

< <

a a

2200 -

1800 -

140 0

100100 -

600 -

66 C

--

88 o o 0 0 * ö ö Ó

* * 88 5 8 ? * ? !

OWW 25W 50WW 75W

Workloadd (Watt)

Figuree 2. Cerebral and central hemodynamic response at two levels of heart rate. Meanss and SEMs for systemic and peripheral hemodynamics. Arrows indicate a patient that had to stopp exercising. White circles indicate values obtained in optimized settings, black in default settings.. BP: arterial pressure (systolic, mean and diastolic). V: cerebral velocity (systolic, mean and diastolic).. HR: heart rate. SV: relative stroke volume from baseline. Q. relative cardiac output from baseline. .

Results s

Inn the resting sitting condition prior to exercise in DSS and OSS, HR 70 (60-76) vs.

733 (65-87) bpm and MAP 89 (73-105) vs. 93 (75-105) mmHg were comparable. Also

Vmean .. PETCO, and breathing frequency were not affected by the pacemaker setting

(tablee 2). The increase in HR with OSS vs. DSS during the first 3 min of exercise at

255 W was larger and this difference remained at 75 W (Fig. 1). SV increased less with

103 3

50WW 100W 150W 200W 25W W 50WW 75 W ÏOOW

25W W 50W W 75W W 25W W 55 OW

25W W 50WW 75W ÏOOW 125W 25WW 50W 75W ÏOOW 125W

Figuree 3- Individual difference ^optimized - default curves for hemodynamic variables during exercise.. Differences in HR. SV and O during exercise for the individuals. The steeper increase in HRR is compensated by a less steeper increase in SV in the optimized settings. HR: heart rate. SV: strokee volume, O: cardiac output. O: optimized. D: default.

104 4

Chapterr ~>

OSSS with a comparable rise in Q (table 2). Resting LVET was comparable for OSS and

DSS,, while during exercise the decline with OSS was larger at 25 and 75W.

Withh OSS the increase in BP was larger while MCA Vmean did not change (table 2 and

fig.. 2). The individual difference curves for Q and SV at two levels of HR are given in

fig.. 3. The HR response did not modify maximal achieved exercise workload or

exercisee duration in OSS vs. DSS: 133 (100-225) vs. 129 (75-200) W, 684 (355-1476)

vs.. 642 (360-1384) s.

Discussion n

Thee purpose of this study was to determine in patients with a rate adaptive pace-

makerr the contribution of two settings of pacemaker determined HR to the cardiac

outputt and cerebral blood velocity response during exercise. The data indicate that

inn these patients with generally normal heart function, the ability to elevate cere-

brall perfusion during exercise is affected. Enhancing the HR response to exercise

didd not augment O with a larger HR response balanced by a proportional attenua-

tionn of the increase in SV The following discussion details the assumptions and

evidencee that underlie these conclusions.

Normally,, in healthy individuals during exercise an increase in Q provides for the

increasedd tissue oxygen demand with a graded increase in local and regional cere-

brall perfusion corresponding to the MCA territory8- Ï A 18 1Q- 22 24- 2b " • 4 8 In turn,

pharmacologicall reduction of the exercise related increase in Q by cardioselective

bj-adrenergicc blockade reduces the increase in MCA Vmean without affecting MAP or

PETC02 .255 This suggests that during dynamic exercise a reducedQ has an effect on

thee cerebral circulation and in patients with heart failure physical exercise may

evenn reduce cerebral perfusion.20

Wee considered that during dynamic exercise diastolic cardiac volume is secured by

venouss return43 and that in pacemaker dependent patients manipulation of the HR

responsee during rate adaptive pacing allows to study the contribution of HR to the

cardiacc and cerebral response to exercise. Our study has, however, several potential

limitations. .

Rightt ventricular apical pacing results in an asynchronous ventricular contraction'1

andd Q may possibly have been lower due to asynchronous activation. However,

105 5

sincee each patient was his/her own reference this possible restriction applies for

bothh HR responses. During exercise, Q is tightly controlled by and linearly related to

oxygenn uptake and in this study workload, duration of exercise and attained Q

weree comparable for the two levels of HR response. By design the two exercise

protocolss were performed in a fixed order to allow for the required QT sensor

learningg period and although both patients and observer were blinded for the actual

pacemakerr setting we cannot exclude an influence on the outcome.

Thee increase in MAP during exercise with OSS was larger. Increasing the pacing rate

enhancess peripherally measured systolic, diastolic and mean arterial pressures

relatedd to alterations in the timing of pressure wave reflection without modifying

centrall systolic pressure.54

Wee assumed that changes in MCA Vmean reflect changes in cerebral blood flow,

whichh is true only in so far the diameter of the vessel does not change. The large

cerebrall arteries are conductance rather than resistance vessels and changes in MAP

withi nn the physiological range appear to have negligible effects on the diameter of

thee insonated artery.42 49 More direct observations made during craniotomy reveal

thatt the vessel diameter does not change significantly during variations in MAP

andd changes in MCA V„ seem to follow cerebral Xe clearance. ' Also, in °° mean

healthyy adults who performed at 60% of their maximum exercise capacity, cerebral

norepinephrinee spillover was not increased making an increase in brain sympathet-

icc activity as a cause of constriction of the MCA unlikely.24

PaC022 is an important denominator of CBF. From the supine to the upright position,

aa reduction in O and a gravity-induced V/O mismatch both affect PETC0213 and from

restt to peak exercise the arterial to end-tidal gradient for CO, increases with exer-

cisee intensity47 questioning PETC02 as an estimate of PaC02. We consider that the

exercise-relatedd increase in P£ TC02 was comparable for both settings of HR making a

significantt contribution of this gradient to the MCA Vmean responses at the moder-

atee exercise intensity of this study less likely.

Inn response to exercise intensities of about 40-60% of maximal oxygen uptake the

increasee in O results from both chronotropic and inotropic cardiac responses, i.e. an

instantaneouss increase in HR and a more gradual rise in SV. 2 21 39 At higher work-

loadss the increase in SV reaches a limi t or SV may even fall and especially older

endurance-trainedd patients have an impaired ability to maintain SV at high levels of

exercise.. Such a decline in SV during prolonged exercise is influenced by an

106 6

Chapterr 5

increasee in HR, i.e. with the rise in HR inhibited by b-adrenergic blockade the

declinee in SV is less conforming that during exercise at higher HR, SV may no longer

bee maintained.10 These findings, taken together with the present data may indicate

ann intrinsic relationship between HR and SV during exercise exposed by pacing at

twoo levels of pacemaker determined HR response.16

Thee finding that Vmean did not increase when exercising from 75 Watt to the maxi-

mumm workload indicates a subnormal increase in Q 2t and a steeper increase in HR

duringg exercise was not translated in an enhanced Q. Several mechanisms may be

responsiblee for this phenomenon. At first, while training allows for a larger increase

inn SV during exercise Al the patients were sedentary and untrained. Second, there was

noo atrial contribution to Q which is probably of significance to restingQ. 3I 4 At an

elevatedd HR level an increase in SV would be caused by a decrease in end-systolic

volumee (increased contractility). We attribute the larger increase in SV at the lower HR

levell during exercise to enhanced ventricular fillin g time (Frank-Starling mechanism).

Inn conclusion, in older sedentary pacemaker dependent patients with complete

heartt block, the ability to elevate cerebral perfusion during exercise is affected by

ann impaired capacity to increase cardiac output. Enhancing the heart rate response

too exercise does not augment cardiac output by a proportional offset of the exercise-

inducedd increase in stroke volume.

Acknowledgments s

LWJJ Bogert is a research fellow supported by the Netherlands Heart Foundation

(grantt 99182). A Erol-Yilmaz is a research fellow supported by unrestricted research

grantss supplied by pacemaker companies Guidant BV. Nieuwegein. The Netherlands,

Metronicc BV. Heerlen, The Netherlands and Vitatron, Arnhem. The Netherlands.

107 7

References s

1.. Aaslid R. Markwalder TM. and Nornes H, Noninvasive transcranial Doppler ultrasound re-cordingg of flow velocity in basal cerebral arteries. J Neurosurg 57: 769-774. 1982.

2.. Astrand PO, Cuddy TE, Saltin B, and Stenberg J. Cardiac output during submaximal and maxi-mall work. J Appl Physiol 19: 268-274, 1964.

3.. Astrand.P.O., H.A.Dahl. and K.Rodahl. Textbook of Work Physiology: Physiological Bases of Ex-ercise.. Human Kinetics Publishers, 2003.

4.. Benditt DG. Milstein S, Buetikofer f. Gornick CC, Mianulli M. and Fetter J. Sensor-triggered, rate-variablee cardiac pacing. Current technologies and clinical implications. Ann Intern Med 107:: 714-724. 1987.

5.. Bishop CC, Powell S, Rutt D. and Browse NL. Transcranial Doppler measurement of middle cerebrall artery blood flow velocity; a validation study. Stroke 17: 913-915. 1986.

6.. Burns CA, Sperry RE, Arrowood JA. Wood MA. Nixon IV. and Ellenbogen KA. Doppler echocar-diographyy assessment of an impedance-based dual- chamber rate-responsive pacemaker. Am J Cardioll 73: 569-574. 1993.

7.. Clark JM. Skolnick BE, Gelfand R, Farber RE, Stierheim M. Stevens WC. Beck GJ. and Lambert-senn CJ. Relationship of mX e cerebral blood flow to middle cerebral arterial flow velocity in menn at rest. ] Cerebr Blood Flow Metab 16: 1255-12Ó2. 1996.

S.. Delp MD, Armstrong RB, Godfrey DA. Laughlin MH. Ross CD. and Wilkerson MR. Exercise in-creasess blood flow to locomotor, vestibular, cardiorespiratory and visual regions of the brain inn miniature swine. ] Physiol 533: 849-859. 2001.

9.. Eckert S and Horstkotte D. Comparison of Portapres non-invasive blood pressure measure-mentt in the finger with intra-aortic pressure measurement during incremental bicycle exer-cise.. Blood Press Monit 7: 179-183. 2002.

10.. Fritzsche RG, Switzer TW, Hodgkinson BJ, and Coyle EF. Stroke volume decline during pro-longedd exercise is influenced by the increase in heart rate. ] Appl Physiol 86: 799-805. 1999.

11.. Gabriel A. Lindblad LE, and Angleryd C. Non-invasive vs. invasive beat-to-beat monitoring of bloodd pressure. Clin Physiol 12: 229-235. 1992.

12.. Giller CA, Bowman G. Dyer H, Mootz L, and Krippner W. Cerebral arterial diameters during changess in blood pressure and carbon dioxide during craniotomy. Neurosurgery 32: 737-41, 1993. .

13.. Gisolf J. Wilders R, Immink RV, Van Lieshout JJ, and Karemaker JM. Tidal volume, cardiac out-putt and functional residual capacity determine end-tidal C02 transient during standing up in humans.. J Physiol 554: 579-590, 2004.

14.. Gonzalez-Alonso J, Dalsgaard MK. Osada T. Volianitis S. Dawson EA, Yoshiga CC, and Secher NH.. Brain and Central Hemodynamics and Oxygenation During Maximal Exercise in Humans. JJ Physiol 557: 331-342, 2004,

15.. Gruhn N. Larsen FS. Boesgaard S. Knudsen GM, Mortensen SA, Thomsen G. and Aldershvile J. Cerebrall blood flow in patients with chronic heart failure before and after heart transplanta-tion.. Stroke 32: 2530-2533. 2001.

16.. Haennel RG, Logan T. Dunne C. Burgess ], and Busse E. Effects of sensor selection on exercise strokee volume in pacemaker dependent patients. Pacing Clin Electrophysiol 21: 1700-1708. 1998. .

17.. Harms MPM, Wesseling KH, Pott F, Jenstrup M, Van Goudoever J. Secher NH, and Van Lie-shoutt JJ. Continuous stroke volume monitoring by modelling flow from non-invasive meas-urementt of arterial pressure in humans under orthostatic stress. Clin Sci 97: 291-301, 1999-

18.. Hedlund S. Nylin G. and Regnstrom O. The behaviour of the cerebral circulation during mus-cularr exercise. Acta Physiol Scand 54: 316-324. 1962.

19-- Hellstrom G. Fischer-Colbrie W, Wahlgren NG, and Jogestrand T. Carotid artery blood flow andd middle cerebral artery blood flow velocity during physical exercise. J Appl Physiol 81: 413-418.. 1996.

20,, Hellstrom G. Magnusson G, Wahlgren NG. and Saltin B. Physical exercise may impair cerebral perfusionn in patients with chronic heart failure. Cardiol Elderly 4: 191-194. 1996.

108 8

Chapterr ~>

21.. Hermansen L. Ekblom B. and Saltin B. Cardiac output during submaximal and maximal tread-milll and bicycle exercise. ) Appl Physiol 29: 82-86, 1970.

22.. Huang SY. Tawney KW, Bender PR. Groves BM, McCullough RE. McCullough RG. Micco A). Manco-Johnsonn M, Cymerman A, and Greene ER. Internal carotid flow velocity with exercise beforee and after acclimatization to 4.300 m. I Appl Physiol 71; 1469-1476. 1991.

23.. Ide K. Gullov AL, Pott F. Van Lieshout JJ, Koefoed BG. Petersen P, and Secher NH. Middle cere-brall artery blood velocity during exercise in patients with atrial fibrillation. Clin Physiol 19: 284-289.. 1999.

24.. Ide K. Horn A. and Secher NH. Cerebral metabolic response to submaximal exercise. J Appl Physioll 87: 1604-1608. 1999.

25.. Ide K, Pott F, Van Lieshout JJ. and Secher NH. Middle cerebral artery blood velocity depends onn cardiac output during exercise with a large muscle mass. Acta Physiol Scand 162: 13-20, 1998. .

26.. Ide Kand Secher NH. Cerebral blood flow and metabolism during exercise. Prog Neurobiol 61: 397-414.. 2000.

27.. Immink.R.V, B.J.H.Van den Born, G.A.Van Montfrans. R.P.Koopmans. J.M.Karemaker, and J.J.Vann Lieshout. Impaired cerebral autoregulation in patients with malignant hypertension. Circulationn 2004.(In Press!

28.. Jansen JRC, Schreuder JJ. Mulier JP. Smith NT, Settels JJ. and Wesseling KH. A comparison of cardiacc output derived from the arterial pressure wave against thermodilution in cardiac sur-geryy patients. Br J Anaesth 87: 212-222. 2001.

29.. Jeilema WT. Wesseling KH. Groeneveld AB. Stoutenbeek CP. Thijs LG. and Van Lieshout JJ. Continuouss cardiac output in septic shock by simulating a model of the aortic input imped-ance:: a comparisonn with bolus injection thermodilution. Anesthesiology 90: 1317-1328. 1999.

30.. Kent KM and Cooper T. The denervated heart. A model for studying autonomic control of the heart.. N Engl} Med 291; 1017-1021. 1974.

31.. Lamas GAand Ellenbogen KA. Evidence base for pacemaker mode selection: from physiology too randomized trials. Circulation 109: 443-451. 2004.

32.. Lewis SL. Taylor WF, Bastian BC. Graham RM. Pettinger WA, and Blomqvist CG. Haemodynam-icc responses to static and dynamic handgrip in normal patients and after autonomic blockade. ClinSci64:: 593-599. 1983-

33.. Linkis P. Jorgensen LG, Olesen HL, Madsen PL. Lassen NA. and Secher NH. Dynamic exercise enhancess regional cerebral artery mean flow velocity. J Appl Physiol 78: 12-16. 1995-

34.. McGuire DK, Levine BD. Williamson JW. Snell PG. Blomqvist CG, Saltin B. and Mitchell JH. A 30-year30-year follow-up of the Dallas Bedrest and Training Study: II. Effect of age on cardiovascular adaptationn to exercise training. Circulation 104: 1358-1366,2001,

35.. Nobrega AC, Williamson JW. Garcia JA, and Mitchell JH. Mechanisms for increasing stroke vol-umee during static exercise with fixed heart rate in humans. J Appl Physiol 83; 712-717. 1997.

36.. Ogawa T, Spina RJ. Martin WH, III . Kohrt WM. Schechtman KB. Holloszy JO, and Ehsani AA. Effectss of aging, sex. and physical tiaining on cardiovascular responses to exercise. Circula-tionn 86: 494-503. 1992.

37.. Ogoh S, Volianitis S. Nissen P. Wray DW, Secher NH. and Raven PB. Carotid baroreflex respon-sivenesss to head-up tilt-induced central hypovolemia: effect of aerobic fitness. ) Physiol 551: 601-608.. 2003.

38.. Pott F. Van Lieshout JJ. Ide K. Madsen P. and Secher NH. Middle cerebral artery blood velocity duringg intense static exercise is dominated by a Valsalva maneuver. J Appl Physiol 94: 1335-1344.2003. .

39.. Proctor DN. Beck KC. Shen PH. Eickhoff TJ. Halliwil l JR, and Joyner MJ. Influence of age and genderr on cardiac output-V02 relationships during submaximal cycle ergometry. J Appl Physi-oll 84: 599-605. 1998.

40.. Proctor EE, Leman RB, Mannn DL, Kaiser J. Kratz J. and Gillette P. Single- versus dual-chamber sensor-drivenn pacing: comparison of cardiac outputs. Am Heart J 122: 728-732. 1991,

41.. Saltin B, Blomqvist G. Mitcheli JH, Johnson RL, Wildenthal K. and Chapman CB. Response to exercisee after bed rest and after training. Circulation 38: V1I1-78. 1968.

42.. Serrador JM. Picot PA, Rutt BK. Shoemaker JK. and Bondar RL. MRI measures of middie cere-brall artery diameter in conscious humans during simulated orthostasis. Stroke 31: 1672-1678, 2000. .

109 9

43.. Sheriff DD. Zhou XP. Scher AM. and Rowel] LB. Dependence of cardiac fillin g pressure on car-diacc output during rest and dynamic exercise in dogs. Am J Physiol 265: H3I6-H322.I993-

44.. Shi X. Potts JT. Foresman BH, and Raven PB. Carotid baroreflex responsiveness to lower body positivee pressure-induced increases in central venous pressure. Am J Physiol 265: H918-H922.1993--

45.. Sugawara ], Tanabe T, Miyachi M, Yamamoto K, Takahashi K, Iemitsu M, Otsuki T, Komma S. Maedaa S. Ajisaka R. and Matsuda M. Non-invasive assessment of cardiac output during exer-cisee in healthy young humans: comparison between Modelflow method and Doppler echocar-diographyy method. Acta Physiol Scand 179: 361-366. 2003.

46.. Sundblad P and Wranne B. Influence of posture on left ventricular. Am J Physiol Heart Ore Physioll 283: H1302-H13OÓ.2OO2.

47.. Tanabe Y, Hosaka Y, Ito M, lto E. and Suzuki K. Significance of end-tidal PlCO(2)l response to exercisee and its relation to functional capacity in patients with chronic heart failure. Chest 119:811-817.. 2001. .

48.. Thomas SN. Schroeder T. Secher NH. and Mitchell JH. Cerebral blood flow during submaximal andd maximal dynamic exercise in humans. J Appl Physiol 67: 744-748, 1989.

49.. Valdueza JM. Balzer JO. Villringer A. Vogl TJ, Kutter R. and Einhaupl KM. Changes in blood floww velocity and diameter of the middle cerebral artery during hyperventilation: assessment withh MR and transcranial Doppler sonography. AfNR Am J Neuroradiol 18: 1929-1934. 1997.

50.. Van Lieshout JJ, Pott F. Madsen PL, Van Goudoever J, and Secher NH. Muscle tensing during standing:: effects on cerebral tissue oxygenation and cerebral artery blood velocity. Stroke 32: 1546-1551.. 2001.

51.. Van Lieshout JJ, Toska K. Van Lieshout E]. Eriksen M, Walloe L, and Wesseling KH. Beat-to-beatt noninvasive stroke volume from arterial pressure and Doppler ultrasound. Eur J Appl Physio]] 90: 131-137.2003.

52.. Vergbese AH and Ellestad MH. Stress Testing: Principles and Practice. Oxford University Press. 2003. .

53-- Wesseling KH. Jansen JRC, Settels JJ. and Schreuder JJ. Computation of aortic flow from pres-suree in humans using a nonlinear, three-element model. J Appl Physiol 74: 2566-2573. 1993.

54.. Wilkinson IB. MacCallum H, Flint L. Cockcroft JR. Newby DE. and Webb DJ, The influence of heartt rate on augmentation index and central arterial pressure in humans. J Physiol (Lond) 525:: 263-270. 2000.

55-- Wilkoff BL and Miller RE. Exercise testing for chronotropic assessment. Cardiol Clin 10: 705-717.. 1992.

]] 10

Directt comparison of a contractility andd activity pacemaker sensor during

treadmilll exercise testing

Aytenn Erol-Yilmaz MD, Raymond Tukkie MD PhD, Job de Boo* MD, Tim Schrama andd Arthur Wilde MD PhD

Fromm the department of Clinical and Experimental Cardiology. Academicc Medical Center Amsterdam and the department of Cardiology

Oosterscheldehospitall Goes. The Netherlands.

PACEPACE 2004; 27:1-7

Abstract t

Introductio n n

Theree are limited data about the chronotropic capacity of the peak endocardial

accelerationn (PEA) sensor. This study directly compared the chronotropic function

fromm the PEA and the activity (ACT) sensor.

Methods s

Thee study included 18 patients (73 7 years) with > 75% pacemaker driven heart

ratee (HR) and a PEA sensor and 11 healthy individuals (HI, 67 7 years) underwent a

CAEPP exercise test with the pacemaker patients in WIR mode after programming

thee sensors in the default setting, with adjustment of the upper sensor rate as an

agee related maximum value (220-age). The ACT sensor was externally strapped on

thee thorax.

Results s

Achievedd exercise duration for the patients and controls was, respectively 92 3

vs.. 18.4 4 min. (p<0.001). The maximal achieved HR with the PEA sensor 124

255 beats/min, versus the ACT 140 23, versus the controls with 153 26 bpm,

(p<0.0011 between the groups). For the PEA, ACT and controls, the time to peak HR

was,, respectively, 11 3. 7 36, and 18 4 (p<0,001 between the groups) and HR

afterr 10 minutes recovery was, respectively. 80 20, 65 15, and 82 4 beats/min

(p<0.0011 between the groups).

Conclusions s

Thee PEA sensor functions hypochonotroop during exercise programmed as a single

sensorr system. It is, therefore, preferable to combine the PEA sensor with an

activity-basedd sensor in a dual sensor system. Although both groups had normal left

ventricularr functions, the exercise capacity of pacemaker patients is significantly

lowerr than in healthy individuals.

114 4

Chapterr ó

Introductio n n

Manyy pacemaker sensors have been developed in the last decade with the ultimate

aimm to imitate the sinus node function.1 From these sensors are the non-physiolog-

ic,, activity based (activity, accelerometer) and the physiologic based (QT, minute

ventilation,, peak endocardial acceleration (PEA)) still in use. The activity-based

sensorr is still in use because of the obvious advantages (sensitive, no necessity for

ann extra lead, housing in the pacemaker can, ease in implementation) although

oftenn combined with one of the physiological- based sensors in a dual sensor

system. .

Thee PEA sensor (Sorin Biomedica, Salugia, Italy) incorporated in the tip of a pacing

leadd is a unique sensor, which measures the peak-to-peak amplitude of the endocar-

diall acceleration signal detected during the isovolumic ventricular contraction

phase.. The PEA values differ from patient to patient, the values reflect the varia-

tionss in sympathetic activity due to physical and mental stress in each patient and

cann be changed by medication or pathological events, 2 Previous experiments

showedd that PEA is related to dP/dt. increases during adrenergic stimulation along

withh dP/dt and reflects whole heart contractility. There is limited data about the

chronotropicc capacity of the PEA sensor and wether the sensor alone is capable to

mimicc sinus node function.2"4 No studies are available which compares the PEA

sensorr with another sensor in the same patient, This study, directly compared the

chronotropicc function offered with the PEA sensor and ACT sensor in WIR mode

duringg treadmill exercise test in default setting with adjustment for the maximum

heartt rate (HR) and compared this with healthy individuals (HI).

Patientss and Methods

Patientt population and healthy individuals

Eighteenn (S men. 10 women; mean age 73 8 years) consecutive patients with > 75%

pacemakerr driven rhythm and commercially available pacemakers with PEA sensor

weree included after informed consent. The pacemaker driven patients had minimal

co-morbidityy but a significant part used some medication. From the existing institu-

115 5

tionall database of exercise testing with HI without co-morbidity and medical

therapy.. 11 (6 men, 5 women; mean age 67 ) age- and exercise type- matched

(chronotropicc assessment exercise protocol (CAEP)) HI were selected. The patients

weree in New York heart association (NYHA) class I- II . The activity levels of patients

andd HI were divided into three grades of activity. Minimally active was defined as

havingg a sedentary lif e style, moderately active when one sports ones a week, and

veryy active when one sports twice a week. Pacemaker implantation was indicated

duee to sick sinus syndrome, atrioventricular (AVI block, atrial fibrillation with His-

bundlee ablation or chronotropic incompetence (see tablel).

Tablee 1. Pacemaker indication, mode and sensor settings

Indication n

Mode e

DDDRR t%) WIRR {%)

ssss w AVV blok [%)

Hiss bund le ablat ion i%)

Otherr (%Ï

Baseline e

28 8

72 2

3Q Q

33 3

12 2

16 6

Testt set t ing

W I R R

W I R R

PEAA and ACT Sensor set t ings

LRMPEAandd ACT1 60

USRR iPEAand ACT) 220-age

PEAA sensit iviy med ium

PEAA upward slope med ium

PEAA downward slope med ium

ACTT threshold Low

ACTT accelerat ion 15 sec

ACTT decelerat ion ] 0 min

Tablee 1,Values are given as mean SD or percentage.; SSS= sick s inus syndrome: PEA— peak

endocardiall accelerat ion; ACT = activity sensor: LRL = lower rate l imit : USR upper sensor rate

Pacemakerr devices and optimization

Alll patients underwent a symptom limited CAEP exercise test according to Wilkof f

andd Miller ' in WIR mode after programming the sensors in the default setting.

withh exception of the upper sensor rate that was programmed 220-age in years. 5

Thee Sorin Living 1 and the Minilivin g D (Sorin Biomedical with a PEA sensor were

used.. These pacemakers are DDDR systems with two sensors: along with the micro

accelerometerr of the PEA sensor there is also a gravimetric sensor, which was

116 6

Chapterr ó

switchedd Off during this study to avoid any interference (which is also in accord-

ancee with the default setting of the pacemakers). The pacing leads were the Best

andd the Minibest leads (Sorin Biomedical. The Best leads had a total length of 62 cm

andd a cylindrical shaped tip. The surface area of this tip is 6 mm2 with a 2,1 mm tip

diameter.. The Pt/Ir electrodes are coated with porous platinum. The sensor sensitiv-

ityy is 50 Hz, 6.8 1.3 mV/g. The lead was inserted through a 13 Fr introducer. The

Minibestt leads differ only in a thinner diameter tip of the lead (1.85 mm) compared

too the Best leads. The lead position in relation to the diaphragm was computed in

thee anteroposterior view on the chest x-ray.

Inn eight patients with the Sorin Living 1, beat-to-beat values of PEA and HR were

telemetricallyy transmitted and recorded by an external PEA recorder. These data

weree analyzed with the Living 1 tool software (Sorin Biomedica). In the other ten

patientss with the Minilivin g D pacemaker, PEA was recorded with an internal Holter

becausee an external PEA recorder is not connectible with this type of pacemaker.

Thee HR derived from the PEA signal was analyzed thoroughly for HR driven due to

intrinsicc versus paced beats.

Simultaneously,, a Medtronic Kappa DR 733 (Medtronic, Minneapolis, MN, USA)

withh an ACT sensor (piezoelectric crystal) was strapped to the contralateral side of

thee internal pacemaker on the upper anterior thorax region in exactly the same

mannerr in each patient. Particular emphasis was placed on uniformly stable attach-

mentt of the external pacemaker; large area adhesive strips were used for this

purpose.. It has been previously shown that behaviors of implanted and externally

strappedd conventional activity-based pacemakers do not differ significantly,6 The HR

drivenn by the PEA and the ACT sensor was monitored with a 12 lead electrocardio-

graphh (ECG). The two lead ports of the externally attached ACT sensor were connect-

edd with the ECG leads V1 and V2, which then reflect the ACT sensor induced rate

responsee and the other ECG leads reflect the PEA sensor induced rate response.

Posturee change and exercise protocol

Alll patients and HI were examined in the same conditions (before noon, uniform

roomm temperature and foot gear (patients own shoes)). After instrumentation,

patientss rested supine on an examination table with one pillow for 5 minutes. Then

thee patients and the HI underwent a CAEP protocol.7' Total exercise time. PEA

117 7

signalss and for both sensors: time to peak HR and maximal achieved HR. HR at 10

minutess recovery were measured.

Statisticall Analysis

Alll results are presented as the mean SD of the group. The pacemaker patients

wit hh the PEA and ACT sensor were compared to HI with the paired sample t-test.

AA P value <0.05 was considered significant.

Results s

Clinicall parameters

Al ll patients were in NYHA class 1 and 2 wit h various activity levels and most

patientss used medication for supraventricular tachyarrytmias. The HI had a higher

activit yy level then the patients and used no medication. The comorbidity of the

Tablee 2. Demographics and clinical variables

Agee (year Sexx (male/female) BMII ikg/m^ NHYAA class [\-4) Activit yy level

MinimallyMinimally active i%) ModerateModerate active i%i VeryVery active {%]

LVEF;%I I

History y

MI(%) MI(%) CABG(%) CABG(%) AtrialAtrial fibrihtion (%)

Medication n DigoxinDigoxin {%) ji-blockerji-blocker {%) calciumcalcium antagonist [%<l

amiodaroneamiodarone \%)

Patients s

733 6.8 8/10 0

25.44 3.6 1.66 0.6

6 6 67 7 27 7

63.11 4

11 1 5.6 6 22 2

5.6 6 50 0 11 1

5.6 6

HI I

6776/5 5

24.55

1 1

75 5 24 4 NA A

0 0 0 0 0 0

0 0 0 0 0 0 0 0

Tablee 2 Values arc given as mean SD or percentage = healthy individuals; BMI= body mass index;; NYHA = New York Heart Association; I.VEF = left ventricular ejection fraction: NA- not available:: Ml= myocardial infarction: CABG = coronary artery bypass grafting; MR= mitral regurgitation. .

i l 8 8

Chapterr 6

patientss was minimal and the HI had no comorbidity. The patients versus HI

stoppedd with the exercise test, respectively. 71 % versus. 90 % due to fatigue, in 17 %

versuss 10 % due to dyspnea, in 6% versus. 0 % due to discomfort to their legs and in

6%% versus 0 % due to dizziness (table 2).

PEAA sensor

Duee to the strict inclusion criteria intrinsic beats were minimized. At rest 2 % and

duringg exercise 8% of the total HR of 18 patients were driven due to intrinsic beats.

Duringg the recovery period 2.7 % of the total beats were fusion beats.

Thee pacemaker leads were all located in the right ventricular apex. The angle of the

leadd tip in relation to the diaphram on the anteroposterior view of chest-x-ray was

30.33 13 degrees. During exercise, the PEA increased from 0.24 0.05 at rest to

0.65 0.14 (g) at maximal load, whereas HR increased from 65 7 to 120 25 beats/

min.. The PEA decreased during recovery to 0.45 0.10 (g), whereas HR decreased to

800 19 beats/min (figure 1. figure 2 table 3).

Onee patient showed an inappropriately low PEA value (0.1 g), which resulted in

pacingg at the lower rate limi t of 60. In this patient, the PEA sensor remained pro-

grammedd On with faster slope settings and also the gravimetric sensor was pro-

grammedd On.

Inn three other patients, an inappropriately high PEA value was measured (3.4 0.3

g),, which resulted in pacing at the upper sensor rate limit of 140. In these 3 pa-

tients,, it was necessary to program the rate response Off to stop the high sensor

rates,, because of palpitations. In the recovery period, HR increases in the PEA group

firstt (while normally the HR decreases immediately) partly due to the inappropriate

highh sensor rates of the above-mentioned patients.

CAEPP exercise test

Restt rate

Thee achieved resting HR with the PEA sensor was significantly higher compared to

thee ACT sensor and HI (80 8 vs. 65 7 vs. 8 beats/min, respectively, see

figuree 1 and table 3).

119 9

Timee to peak HR

Thee ach ieved t ime to peak HR w i t h PEA sensor was s igni f icant ly longer compared to

thee ACT sensor b ut s igni f icant ly sho r ter than in HI 111 3 vs. 7 3-6 vs. 18 4

min. respect ive ly,, see f igure 1 and table 3).

Tablee 3. Summary of HR of normals, the activity and PEA sensor during rest, exercise and recovery.

Restingg rate (bpm.) Timee to peak pacing rate (min) Maximumm HR (bpm.) 100 min. recovery (bpm.) Exercisee duration (min.)

PEA A

800 8*t 111 3 * t

1244 25 * t 800 20 *f 9-22 3 *

ACT T

655 7 # 77 3.6 #

1400 23 # 655 15 #

HC C

655 8 18 8

1533 26 822 4

18.44 4

P-value e

** t = s. # = ns

** t # = s ** + # = s

** # = s. t = n s '' - s

Tablee 3.Values are given as mean SD.HC = healthy controls: ACT = activity sensor-, PEA= peak endocardiall acceleration: HR= heart rate: bpm=beats per minute: min= minutes. * = PEA vs ACT t == PEAvs HC; # =ACT vs HC; s = p<0.001: ns = p>0.05-

Maximall HR

Thee ach ieved max imal HR w i t h PEA sensor was s igni f icant ly lower compared to the

ACTT sensor and HI (124 25 vs. 140 23 vs. 153 26 beats /min . respect ive ly, see

f iguree 1 and table 3).

HRR (bpm)

160 0

140 0

120 0

100 0

80 0

60 0

40 0

20 0

'PEA A

'ACT T

HC C

Rest t Exercise e Recovery y

<< MIC << > I '

11 3 5 7 9 11 13 15 17 19 21 23 25 21 2Q 31 33 Timee (min)

n(P)) = 18 17 16 14 12 9 8 6 4 3 2 1

n(HC)== ii x e 4 2

Figuree 1. Changes in heart rate (mean SD) PEA vs ACT vs HC. ACT = activity sensor; PEA= peak endocardiall acceleration sensor: HC= healthy controls-. P= patients: n (p) = number of patients remainingg during exercise: n (HC) number of healthy controls remaining during exericse.

120 0

Chapterr 6

90 0

22 ss Figure 2. Changes in PEA vs. heart ££ rate. PEA= peak endocardial activity.

155 20 255 33

Timee imin)

Recovery y

Thee achieved HR at 10- minute recovery with the PEA sensor was significantly

higherr compared to the ACT sensor but comparable to the HI (80 20 vs. 65 8 vs.

822 4 beats/min, respectivelysee figure 1 and table 3).

Exercisee duration

Thee exercise performance of the pacemaker patients with the PEA sensor was signifi-

cantlyy lower compared to HI (9-2 3 vs. 18.4 4 min, see figure 1 and table 3).

Discussion n

Normall sinus node behavior is the ultimate tool to evaluate sensor characteristics.

Severall methods have been used to describe the sinus node behavior and thus the

efficiencyy of different sensors: 1), by measuring the chronotropic response with

differentt exercise testing protocols (CAEP, Bruce, six minute walk test, Kaltenbach

stepp test, stair climbing),2). by measuring the rate response with Holter monitoring

inn daily lif e and 3) by evaluation of the obtained quality- of- lif e (OOL) associated

withh a given sensor.

Inn limited previous studies, the PEA sensor was validated with supposed high

12: :

correlationn with normal sinus rhythm.49"11 Therefore, the PEA sensor is appropriate

too use in rate adaptive pacing2'4'9 10 Also, the PEA signal results in new options; 1)

automaticc AV delay optimization, The studies before showed a close relationship

betweenn PEA, first heart sound, and the AV delay. A louder first heart sound is

observedd with short AV delays as the ventricle contracts after atrial emptying12'13

Dupuiss et al. 13 showed that the optimal AV delay estimates obtained by the PEA

analysiss during automatic AV delay scanning are consistent with those obtained by

echocardiography;; 2) recent clinical trials have shown that selected patients with

recurrentt vasovagal syncope may benefit from permanent pacing with a PEA sensor.

Thee authors demonstrated using head-up til t testing that increase in sympathetic

activityy preceding syncope could be sensed by a PEA sensor.1415 and 3) the PEA

sensorr is a promising tool for long-term hemodynamic monitoring and serial

evaluationn of the effects of multisided ventricular pacing in heart failure patients.16

Anotherr possibility with the PEA signal is the estimation of the aortic diastolic

pressuree with the PEA II signal (the abrupt deceleration of the moving aortic blood

mass).1'' Beside these advantages, a number of critical remarks can be made, disad-

vantagess of the sensor are mostly related to the lead characteristics (lead body size)

andd the sensor location. The PEA sensor is located at the lead tip; consequently

sensorr failure necessitates a new lead insertion. Langenfeld et al.2 showed in a

multicenterr trial with 105 patients uncomplicated lead implantation in all patients.

Att a mean follow-up of 22.1 months no pacing or sensing failure was reported.

Uncomplicatedd function of the sensor was reported in 98% of the patients after 1-

yearr follow- up. Although Vitali et al.18 showed that the lead size is significantly

reducedd in size from 11 to 9 Fr introducer, placement of the lead could pose some

difficultyy in small patients, when multiple leads are used or when using the cephal-

icc vein. Long-term lead survival has to be followed carefully. The lead localization,

size,, and the relative short follow -up period described in the restricted studies of

thee PEA sensor limits the benefit of the PEA sensor.2

Noo studies are available that directly compare the PEA sensor with another sensor

typee in the same patient during exercise. The present study clearly demonstrates

thee hypochonotropic function of the PEA sensor compared with the ACT sensor and

withh HI. 5 1922 in addition, the programmed upper rate limi t is not achieved with

thee PEA sensor in contrast to maximum achieved HR by the ACT sensor which came

closee to the achieved maximal HR of the HI, The findings are in accordance with the

studyy of Clementy et al.11 They compared the chronotropic function offered by the

122 2

Chapterr 6

Bestt Living system in patients suffering from chronotropic incompetence with that

off normal sinus rhythm measured in matched normal subjects (n= IS). They also

showedd that a single PEA sensor successfully restored chronotropic response in a

populationn of paced patients (n= 14) with severe chronotropic incompetence after

optimall programming of the sensor. They showed a physiological chronotropic

modulationn of the paced rate with the PEA sensor during exercise, although the

overalll rate response remains hypochonotropic (the configuration of the HR curve

derivedd from the HI was comparable but lower HR were achieved with the PEA

sensor).. As previously described, 23"25 the present study also showed the main

characteristicss of the hyperchonotropic ACT sensor namely: prompt rise of pacing

ratee with the onset of physical exercise, higher achieved maximal HR which is less

proportionall to exercise and a relative rapid recovery compared with the HI and the

PEAA sensor. In the commercially available pacemakers with the PEA sensor, the ACT

iss programmed Off in the default setting. The disadvantages of the PEA induced rate

responsee can be compensated by programming the ACT sensor On, in combination

withh the PEA sensor and adequate programming of the sensor parameters (e.g.

sensorr cross-checking, sensor blending). The advantages of sensor combination

(ACTT and another existing physiological sensors) considering improved rate re-

sponsee curves during exercise compared with one of the sensors alone, is described

before.. 26 For example, Leung et al. 27 showed an improved cardiac output with

combiningg the ACT sensor with the QT sensor compared to the sensors alone.

Itt remains mandatory to individually optimize the PEA sensor because PEA values

differr from patient to patient, the values reflect the variations in sympathetic

activityy due to physical and mental stress in each patient and can be changed by

medicationn or pathological events.2 28

Itt remains to be proven if all efforts in optimizing pacemaker sensors are expressed

inn an improvement in QOL.29

Thee exercise capacity of patients in the present study were significantly lower

comparedd to HI. The lower exercise capacity of the pacemaker patients compared to

thee HI were possibly due to interaction with the used negative inotropic drugs (a-

blockerr 50%, calcium antagonist 11 %), asynchronous activation of the myocard with

WIRR pacing and absence of AV synchrony.

Surprisingly,, the authors found a high percentage of abnormal PEA signals (22 %

comparedd to 3% in the study of Langenfeld).2 One patient had a very low PEA value

andd three patients had high PEA values and inappropriately high sensor signals

123 3

withoutt efficient rate response function.

Thee low values are possibly explainable due to mechanical contact disturbances of

thee PEA lead tip since the patient had a previous inferior myocardial infarction. The

veryy high PEA values in the present study are possibly induced by overshoot of the

sensor.. In one patient a different acceleration of the heart and thus a different first

heartt sound due to a mechanical mitral valve could explain the overshoot of the

sensor.. Other reasons to explain the overshoot of the sensor, like catch up of the

intrinsicc HR, deviation of the lead position, and the type of the lead, can be exclud-

edd for several reasons. First, the patients in the present study had pacing induced

HRR and usually such high PEA values were only demonstrated in studies after

dobutaminee infusion.2,11 17 Second, all leads were located in the right ventricular

apexx with a comparable angle of insertion of the lead tip. Third, the study of

Rickardss et al.4 showed that the PEA signal is insensitive to local myocardial proper-

tiess or the location of the lead, although other investigators have suggested that

theree are possible limitations in the use of an uniaxial acceleration transducer

becausee ventricular displacement follows a three- dimensional trajectory.

Conclusions s

Thee PEA sensor functions hypochonotroop during exercise programmed as a single

sensorr system, therefore, it is preferably to combine the PEA sensor with a hyper-

chonotroopp functioning activity-based sensor in a dual sensor system.

Acknowledgments s

Thee authors thank Robert Pilage (Holland Medical, Kerkrade. The Netherlands) for

thee technical support, the Department of Cardiology from the Oosterschelde

hospitall for creating facilities to perform the study and Jan Paul Van Mantgem from

thee Amstelveen hospital for selecting patients for the study.

124 4

References s

Chapterr 6

1.. Leung SK, Lau CP. Developments in sensor-driven pacing. Cardiol Clin. 2000:18:113-55. ix. 2.. Langenfeld H, Krein A, Kirstein M, Binner L. Peak endocardial acceleration-based clinical test-

ingg of the "BEST" DDDR pacemaker. European PEA Clinical Investigation Group. Pacing Clin ElectrophysiolElectrophysiol 1998:21:2187-91.

3.. Clementy J. Dual chamber rate responsive pacing system driven by contractility: final assess-mentt after 1-year follow-up. The European PEA Clinical Investigation Group. Pacing Clin Elec-trophysiol.trophysiol. 1998;21:2192-7.

4.. Rickards AF, Bombardini T, Corbucci G. Plicchi G. An implantable intracardiac accelerometer forr monitoring myocardial contractility. The Multicenter PEA Study Group. Pacing Clin Elec-trophysiol.trophysiol. 1996:19:2066-71.

5.. Wilkoff BL. Miller RE. Exercise testing for chronotropic assessment. Cardiol Clin. 1992:10:705-17. .

6.. Mianulli M BD. Markowitz T. A comparison of strap-on versus implanted activity based rate responsivee pacemakers: Are strap-on studies valid? PACE:iA-.732.

7.7. Oto A Ts. Ozin B et al. The use of rate response function during daily lif e in patietns with ratee adaptive pacemakers with two different sensors. EUR. j.P.E. 1993:1:59-62.

&.&. Stofmeel MA, Post MW, Kelder JC. Grobbee DE, van Hemel NM. Ouality-of-life of pacemaker patients:: a reappraisal of current instruments. Pacing Clin Electrophysiol. 2000:23:946-52.

9-- Greco EM, Ferrario M. Romano S. Clinical evaluation of peak endocardial acceleration as a sensorr for rate responsive pacing. Pacing Clin Electrophysiol. 2003;26:812-8.

10.. Occhetta e PA. Rognoni G, et al. Experience with a new myocardial acceleration sensor during dobutaminee infusion and exercise test. Eur JCPE. 1995:5:204-209.

11.. Clementy J. Kobeissi A, Garrigue S. Jais P. Le Metayer P. Haissaguerre M. Validation by serial standardizedd testing of a new rate-responsive pacemaker sensor based on variations in myo-cardiall contractility. Europace. 2001:3:124-31-

12.. Ritter P. Padeletti L, Gillio-Meina L. Gaggini G, Determination of the optimal atrioventricular delayy in DDD pacing. Comparison between echo and peak endocardial acceleration measure-ments.. Europace. 1999:1:126-30.

13.. Duputs JM. Kobeissi A, Vitali L. Gaggini G. Merheb M, Rouleau F. Leftheriotis G, Ritter P, Vic-torr ]. Programming optimal atrioventricular delay in dual chamber pacing using peak endocar-diall acceleration: comparison with a standard echocardiographic procedure. Pacing Clin Elec-trophysioltrophysiol 2003;26:210-3.

14.. Deharo JC, Brunetto AB. Bellocci F, Barbonaglia L. Occhetta E. Fasciolo L, Bocchiardo M. Rog-nonii G, DDDR pacing driven by contractility versus DDI pacing in vasovagal syncope: a multi-center,, randomized study. Pacing Clin Electrophysiol. 2003:26:447-50.

15.. Deharo JC. Peyre [P. Chalvidan T, Thirion X, Valli M, Ritter P, Djiane P. Continuous monitoring off an endocardial index of myocardial contractility during head-up til t test. Am Heart J. 2000:139:1022-30. .

16.. Bordachar P. Garrigue S. Reuter S. Hocini M, Kobeissi A, Gaggini G. Jais P. Haissaguerre M. Clementyy J. Hemodynamic assessment of right, left, and biventricular pacing by peak endocar-diall acceleration and echocardiography in patients with end-stage heart failure. Pacing Clin Electrophysiol.Electrophysiol. 2000:23:1726-30.

17.. Plicchi G, Marcelli E. Pariapiano M, Bombardini T. PEA I and PEA II based implantable haemo-dynamicc monitor: pre clinical studies in sheep. Europace. 2002:4:49-54.

18.. Vitali L, Gaggini, G, Ceron, C. Improved hemodynamic sensor and lead. 2002. 19.. Freedman RA. Hopper DL. Mah J, Hummel ], Wilkoff BL. Assessment of pacemaker chrono-

tropicc response: implementation of the Wilkoff mathematical model. Pacing Clin Electrophys-ioliol 2001:24:1748-54.

20.. Kay GN, Quantitation of chronotropic response: comparison of methods for rate-modulating permanentt pacemakers. J Am Coll Cardiol 1992:20:1533-41.

21.. Page E. Defaye P. Bonnet JL, Durand C. Amblard A. Comparison of the cardiopulmonary re-sponsee to exercise in recipients of dual sensor DDDR pacemakers Versus a Healthy control group.. Pacing Clin Electrophysiol. 2003;26:239-43.

125 5

22.22. Ellestad MH. stress testing principles and practice.tirth edition:567-573-23.. Mehta D. Lau CP, Ward DE. Camm AJ. Comparative evaluation of chronotropic responses of

QTT sensing and activity sensing rate responsive pacemakers. Pacing Clin Electrophysiol. 1988:11:1405-12. .

24.. All E. Kriegler C, Fotuhi P, Willhaus R. Combs W, Heinz M. Hayes D. Feasibility of using intrac-ardiacc impedance measurements for capture detection. Pacing Clin Electrophysiol 1992;15:1873-9. .

25.. Alt E. Matula M. Holzer K. Behavior of different activity-based pacemakers during treadmill exercisee testing with variable slopes: a comparison of three activity-based pacing systems. Pac-inging Clin Electrophysiol 1994:17:1761-70.

26.. Werner J, Hexamer M. Meine M. Lemke B. Restoration of cardio-circulatory regulation by rate-adaptivee pacemaker systems: the bioengineering view of a clinical problem. IEEE Trans Bi-omedomed Eng. 1999:46:1057-64.

27.. Leung SK, Lau CP. Tang MO. Cardiac output is a sensitive indicator of difference in exercise performancee between single and dual sensor pacemakers. Pacing Clin Electrophysiol. 1998;21:35-41. .

28.. Sulke N, Dritsas A. Chambers J, Sowton E. Is accurate rate response programming necessary? PacingPacing Clin Electrophysiol. 1990:13:1031-44.

29.. Clementy J. Barold SS. Garrigue S. Shah DC. Jais P. Le Metayer P, Haissaguerre M, Clinical sig-nificancee of multiple sensor options; rate response optimization, sensor blending, and trend-ing.. Am } Cardiol 1999:83:l66D-171D.

30.. Wood JC, Festen MP. Lim M], Buda AJ, Barry DT. Regional effects of myocardial ischemia on epicardiallyy recorded canine first heart sounds, f AppI Physiol. 1994:76:291-302.

126 6

Cardiacc synchronization induces favorablee neurohumoral changes

Aytenn Erol-Yilmaz MD.1 Hein J, Verberne MD,2 Tim A. Schrama,1 Jana Hrudova MD,1

Robbertt J. de Winter MD PhD,1 Berthe L.F van Eek Smit MD PhD,2

Riannee de Bruin.' Jeroen ]. Bax MD PhD,3 Martin }. Schalij MD PhD.3

Arthurr A. Wilde MD PhD1 and Raymond Tukkie MD PhD1

Fromm the departments of Clinical and Experimental Cardiology ' and Nuclear Medicine2,, Academic Medical Center, Amsterdam: and the department of Cardiologyy , Leiden University Medical Center.3Leiden, the Netherlands.

AcceptedAccepted for publication in PACE

Abstract t

Aims s

too examine whether Cardiac resynchronization therapy (CRT) induces improve-

mentss in the neurohumoral system.

Methodss and methods

Thirteenn patients with heart failureHF (left ventricular (LV) ejection fraction <35%)

weree included. Before and after 6 months of CRT, myocardial 123I-metaiodobenzyl-

guanidinee (123I-MIBG) uptake indices, used as an index of neural norepinephrine

reuptakee and retention, and brain natriuretic peptide (BNP) levels, used as an index

off LV end-diastolic pressure. NYHA classification and echocardiographic indices

weree assessed.

Results s

Sixx months of CRT resulted in significant improvement in 1) NYHA classification

andd reduction in QRS width (p<0.001, 2) decrease of LV end diastolic diameter

(p=0.005),, LV end systolic diameter (p= 0.005), septal to lateral delay (p = 0.01) and

mitrall regurgitation. (MR, p=0.04), 3) Delayed 123I-MIBG heart/mediastinum ratios

improvedd (p=0.03) and 123I-MIBG washout decreased (p=0.001) 4), and BNP levels

decreasedd (p=0.001).

Conclusions s

Parallell to significant functional improvement and echocardiographic reverse

remodelingg and resynchronization. our data indicate that CRT induces favorable

changess in the neurohumoral system.

130 0

Chapterr 7

Background d

Heartt failure (HF) is a complex clinical syndrome resulting from any structural or

functionall cardiac disorder that impairs the ability of the ventricle to fil l with or eject

blood'' The reduction in cardiac performance induces a series of neurohumoral

adjustments,, including an increase in atrial natriuretic peptide, brain natriuretic

peptidee (BNP), and activation of the renin-angiotensin-aldosterone system. Brain

natriureticc peptide is mainly synthesized in the ventricles of the heart and reflects the

leftt ventricular (LV) end-diastolic pressure, LV end-systolic and LV end-diastolic

volumee index.2 Initially , the activation of the neurohumoral system functions as a

compensatoryy mechanism but long term activation has many detrimental effects. The

activatedd neurohumoral system is reversibly correlated with the worsened prognosis

off (HF) patients. Also, elevated catecholamines have a direct cardiotoxic effect and are

importantt predictors of HF mortality and morbidity.2

Largee scale pharmacological trials in HF patients have shown that treatment with ACE-

inhibitors.. p-blockers and aldosterone inhibitors improves the imbalance in the

neurohumorall system and thereby improve the prognosis of patients with HF.

Cardiacc resynchronization on top of optimal pharmacological therapy has already

beenn shown to improve the symptoms of HF. reduce hospital admission and improve

LVV function.0 A recent meta-analysis suggests that cardiac resynchronization therapy

(CRT)) also reduces mortality from HF. ' This is confirmed by theCOMPANION study,

thee largest prospective trial yet.s However, limited data are available whether func-

tionall improvement, induced by CRT. is related to changes in the neurohumoral

system.55 6

Cardiacc adrenergic nerve activity can be estimated using several techniques: first,

usingg iodine-123metaiodobenzylguanidine (125 I MIBG) as a noradrenaline ana-

logue.00 123I-MIB G is a guanethidine analogue that shares the same uptake, storage

andd release pathway as norep inephr ine and has the potential to reflect the whole

myocardiall adrenergic pathway.0, i0 123I-MIB G myocardial uptake indices can be used

ass a predictor of prognosis in patients with chronic HF.11"14 Second, BNP may have a

sympathoinhibitoryy effect on the sympathetic nerve activity.

Wee prospectively studied a group of chronic HF patients to determine the 6 months

effectss of CRT on the neurohumoral system using 123 I MIBG scintigraphy and BNP.

131 1

Methods s

Patientt population

Thirteenn consecutive patients with chronic symptomatic HF referred for implantation

off a CRT device were included in the present study after informed consent. Patients

receivedd CRT according to the accepted criteri a of ejection fraction < 35%. QRS width

>> 120 ms, and severe HF in NYHA functional classes II I or IV despite maximally

toleratedd medical treatment for HF (American Heart Association Pacing / ICD guide-

lines).. 5 Patients were studied before pacemaker implantation (baseline) and 6

monthss after biventricular pacing. NYHA classification, QRS width . LV functional

parameterss using echocardiography. 125 I MIBG myocardial uptake indices and BNP

weree assessed. Medical therapy was unchanged durin g the follow up of 6 months.

Tablee I . Demographics and clinical variables

Agee (year) 63.6 12,8 Sexx (male/female) {%) 61/39 lHD/non-IHDD 1%) 23/77 NYHAA 37 0.4 QRSS duration vmsl 182.4 18.2 Medication n

ACEACE inhibitor [%) 92 B-biocket[%)B-biocket[%) 50 DiureticsDiuretics i%) 69 SpironolactoneSpironolactone (%J 54 DigoxinDigoxin (%) 38

Valuess are given as mean SD or percentage: IHD - ischemic heart disease: NYHA - New York Heartt Association.

Pacemakerr system and implantation technique

Commerciallyy available biventricular pacemakers were used (Guidant. Minnesota.

USA;; Medtroni c Minneapolis USA: St Jude Medical. Californi a USA). The right atrial

leadd was positioned in the right atrial appendix and the ventricular lead was posi-

tionedd in the right ventricular apex. The LV pacing lead was positioned in a posterior

orr lateral branch of the coronary sinus in all patients. A biventricular DDDR system

wass used in 10 patients. In 3 patients the implantation of an internal defibrillato r was

alsoo indicated.l 5 In these patients a combined device was implanted.

132 2

Chapterr 7

Cardiacc MIBG imaging

Patientss underwent myocardial scintigraphy to determine 123I-MIBG uptake, used as

ann index of neural norepinephrine reuptake and retention. The analysis of the

myocardiall scintigraphy at baseline and after 6 months was blinded. To block

thyroidd uptake of free 123I. all patients received 100 mg potassium iodide orally, one

hourr prior to the injection of I23I-MIBG. After a subsequent resting period of at least

300 minutes, patients were injected intravenously with approximately 185 MBq of 123"I-MIB GG (Amersham Health, Eindhoven, the Netherlands). Twenty minutes (early

imaging)) and 4 h (delayed imaging) after MIBG administration, a 10-min planar

anteriorr image of the chest was acquired using a dual-head gamma-camera (e-cam,

Siemens,, Hoffman Estate, Illinois, USA). A 20% energy window was centered on the

1599 keV photopeak of 123I . Images were acquired using a medium energy collimator

andd stored in 128 xl28 matrix. 123I-MIBG myocardial activity was measured using a

manuallyy drawn region of interest around the LV. Mediastinal activity was measured

usingg a fixed 20 x 20 pixel region of interest placed over the upper mediastinum.

Mediastinall and myocardial values are expressed as mean counts/pixel. To evaluate 123I-MIBGG myocardial uptake the H/M ratio was calculated from the early and

delayedd images. Myocardial I23I-MIB G WO was defined as the percentage of change

inn activity from the early and delayed images: ((H early-H delayed) / H early) x 100,

HH values being corrected for 123I physical decay.

Echocardiography y

Patientss were imaged in the left lateral decubitus position using 2 D, M mode and

color-dopplerr and tissue doppler echocardiography {Vivi d 7, GE-Vingmed Ultra-

sound.. Horten Norway). The left ventricular end-diastolic diameter (LVEDD) and left

ventricularr end-systolic diameter (LVESD) were measured from the parasternal long

axiss view. Mitral regurgitation was classified semi-quantitatively in eight degrees

(0== none, 1= minimal. 1.5= minimal to mild, 2= mild. 2.5= mild to moderate. 3 =

moderate.. 3-5= moderate to severe, 4 = severe). The left ventricular ejection frac-

tionn is echocardioghraphically measured.

Whenn available tissue doppler data were obtained in the apical four-chamber view

att frame rates above 100 fps. The resulting digital cine loops where transferred to a

computerr for offline analysis (Echo-Pac software. Ge-Vingmed Ultrasound. Horten

133 3

P== 0.001

Norway).. The simultaneous velocity curves of 2 selected (septal, lateral) LV seg-

mentss were compared (using the 4-chamber view) for a mean of two consecutive

beatss to minimize the variability between measurements. The LV tissue velocities

WOO (%}

700 -

60 0

50 0

40 0

30 0

20 0

10 0

0 0 Baselinee 6 months CRT

Figuree 1. '^ MIBG Cardiac washout (wo) before and after 6 months of CRT.

weree measured during systole. Indicator of LV asynchrony is considered as the

differencee between the timing of the peak systolic velocities of the septum and

laterall LV. For each patient the atrioventricular interval was individually adjusted

too maximize the mitral inflow duration using pulsed doppler echocardiography on

thee first postoperative day.

BNP-concentration n

Bloodd for determination of the plasma BNP concentration was sampled from an

intravenouss canula. Patients were resting in supine position for at least 30 minutes

beforee blood collection. The samples were withdrawn into evacuated ethylenedi-

aminetetraaceticc acid (EDTA) tubes and centrifuged at 4°C at 4000 rpm for 10 min.

Thee samples were stored at -70CC until assay. Plasma BNP (pg/mL) was measured

withh the Triage BNP Test (fluorescence immunoassay device, Biosite San Diego,

USA).1''' Using the Triage BNP test, the upper level of normal BNP was 100 pg/mL.

134 4

Chapterr 7

Statisticall analysis

Thee data are expressed as mean SD. The two-paired sample t-test was applied

comparingg the 123I-MIBG myocardial uptake indices and echocardiography parame-

terss of LV function before and after 6 months of CRT. Changes in BNP measure-

mentss were expressed as percentage change from baseline. The one sample t-test

wass used to compare changes in BNP levels. A probability value of P < 0.05 was

consideredd to indicate a significant difference, The Pearson test (2-tailed) was used

too test the correlation between MIBG indices and BNP.

Results s

Patientt Population

Thirteenn patients were included in the study, 77 % with non-ischemic heart disease.

Tablee 1 shows the demographics and clinical variables, including medication. All

patientss were in sinusrythm except one patient who was in chronic atrial fibrilla-

tionn with slow atrioventricular conduction.

Clinicall Parameters

NYHAA classification improved after 6 months CRT at least one class in all patients

fromm 3-7 0.4 to 2.2 0.6 (p< 0.001). The QRS duration decreased significantly

fromm 182.7 17 to 155-0 14,6 msec {p<0.001). Medication was not altered during

thee pacing period but for furosemide, which had to be withdrawn in one patient

andd low dose digoxin. which was started in another patient.

Echocardiography y

Biventricularr pacing resulted in significant reduction of LVEDD from 73-7 8,9 to

67.66 9.9 mm (p=0.005) and LVESD from 65-8 7.5 to 58.5 12.2 mm (p = 0.005).

Alsoo mitral regurgitation decreased from 2.9 0.6 to 2.2 0.9 (p = 0.04). The left

ventricularr ejection fraction increased from 14 7 to 21 8 % (p = 0.007). There

135 5

Tablee 2. Echocardiography data,ir?l-MIB G data and BNP before and after CRT.

Patients s

1 1 2 2 3 3 4 4 5 5 6 6 7 7 8 8 Q Q

10 0 11 1 12 2 13 3

Total l {mean} } l ii SD)

LVEDD D

(mm) )

83 3 84 4 66 6 76 6 68 8 69 9 79 9 72 2 56 6 71 1 74 4 88 8 78 8

73.7 7 7 7

LVESD D

{mm) )

76 6 76 6 57 7 60 0 63 3 64 4 73 3 64 4 52 2 68 8 66 6 75 5 68 8

65.8 8

7.5 5

Valuess are given as mean SD. . endd diastolic diameter; LVESD = EF-- ejection fraction: S-L-

MI I

(1-4) )

3.5 5 2.5 5 2.5 5 0 0

3.5 5 2 2

3.5 5 3.5 5 3,5 5 3 3 0 0 2 2 3 3

2.9 9 0.6 6

EF F

(%) )

10 0 10 0 20 0 21 1 10 0 15 5 15 5 15 5 15 5 10 0 13 3 10 0 15 5

14 4 7 7

Baseline e

SL L delay y (msec) )

95 5

97 7 93 3 ----

90 0 90 0 97 7

933 3 3.3 3

CRTT = cardiac ^synchronization tr == left ventricular end

septall to lateral. dinee heart/ mediastinum: WO -

weree no sign

non ischemic c

laterall delay

(p=0 .01.. see

i ff i cant clif f ere

pat ients. .

decreased d

tablee 2).

[ n o o

-- washout. --

BNP P

(pg/ml) )

316 6 2140 0 576 6 110 0

2140 0 1040 0 529 9 152 2 614 4 2420 0 1013 3 1024 4

1006.2 2 806 6

wo o

i%) )

55.8 8 51.4 4 373 3 35.8 8 36.5 5 64.0 0 43.8 8 27.3 3 48.1 1 24.1 1 30.0 0 30.3 3 38.3 3

40.2 2 12.0 0

Latee ]2\ MIBGG H/M

ratio o

1.2 2 1.1 1 1.6 6 1.3 3 1.3 3 1.1 1 13 3 1.4 4 1.1 1 1.3 3 1.5 5 13 3 1.1 1

1.3 3 0.2 2

ïerapy;; LVEDD - left ventricular systolicc diameter: MR -mitrall regurgitation;

r;,l-MIBGH/MM = 123-iodine-metaiodobenzylguam-== not available.

ncess be tween the ejectionn fr a : t io nn for the ische

pat ientss t issue doppler data were obtained. Se

afterr 6 months off pac ingg fro m 93.3 39.3 to 26

;mi cc and

ptall to

26.55 msec

--- —

Cardiacc MIBG imaging

Afterr 6 months of biventricular pacing, early l23I-MIB G H/M ratio remained unchanged

(fromm 1.4 0.1 to 1.5 0.1, p=0,224). Late 1231-M1BG H/M ratio increased from 1.3

0.22 to 1.4 0.2 (p = 0.03). 1231-MIBG washout significantly decreased from 40.2 12.0

too 28.3 6.1 (p=0.001, see figure 1 and table 2). The change in H/M ratios and

myocardiall washout were not significantly different for ischemic (H/M earlv: 0.0: H/M

l a t e:0.1:: washout:12 %) and nonischemic (H/M early: 0.1; H/M l a t e;0.1; washout: 11 %)

heartt failure patients.

136 6

Chapterr 7

LVEDD D

(mm! !

82 2 85 5 59 9 59 9 67 7 61 1

59 9 56 6 63 3 68 8 74 4 78 8

67.6 6 9.9 9

LVESD D

imml l

75 5 77 7 51 1 37 7 61 1 52 2

52 2 46 6 54 4 58 8 70 0 69 9

58.5 5 12.2 2

BNP(pg/ml) )

25000 J

2000 ^

1500 ^

1000 ^

5000 -

u u

Ml l

11-41 1

3.5 5 25 5 2 2 0 0

3.5 5 1 1

1 1

2.5 5 2.5 5 0 0 2 2 2 2

2.2 2 0.9 9

Baseline e

Figuree 2. BNP before and after 6

Afterr 6

EF F

i%) )

10 0 15 5 35 5 29 9 10 0 30 0

25 5 25 5 20 0 13 3 20 0 20 0

21 1 8 8

monthss of CRT

monthss of CRT

SL L delay y (msec) )

20 0

25 5 26 6

35 5 30 0 20 0

23.8 8 5 5

BNP P

\pg/mll l

275 5 1340 0 133 3 58 8

1440 0 110 0 236 6 41 1 474 4 180 0 131 1 115 5

377.8 8 487.2 2

WO O

{%) {%)

38.0 0 38.6 6 22.3 3 27.0 0 34.9 9 26.4 4 30.2 2 23.6 6 34.8 8 24.5 5 18.6 6 292 2 24.1 1

28.3 3 6.1 1

PP <0.001

66 months CRT

Latee 123I-MIBGG H/M

ratio o

1.3 3 1.2 2 I I 1 1 1 1 1 1 1 1 1 1 I I 1 1

4 4 5 5 3 3 3 3 4 4 7 7 4 4 4 4

1.4 4 1.8 8 1.4 4

1.4 4 0.2 2

137 7

BNP-concentration n

BNPP levels decreased significantly 1006.2 806 to 377.8 487.6 pg/mL (p<0.001)

att 6 months. There was a great interindividual variability in the measured BNP

values,, expressed by the large SD. The normalized BNP level (baseline value set at

100%)) decreased at least 13 %, not exceeding 93% with a mean of 53-4 29.7%

(p<0.001)) after 6 months of biventricular pacing (see figure 2), In one patient

BNPP could not be measured for technical reasons. We tested with the Pearson test

(2-tailed)) whether a correlation is between MIBG indices and BNP and could not

showw any correlation.

Discussion n

Thiss study showed that CRT induced improvement of functional parameters in a

patientt group with severe HF and broad QRS is accompanied by favorable changes in

neurohumorall system. These favorable changes are illustrated by a significant

decreasee of BNP combined with a reduction of 123I-MIBG washout and improved

delayedd H/M ratio. In accordance with previous studies clinical improvement was

alsoo obtained, indicated by the lower NYHA class, increase of LV ejection fraction

decreasee of mitral regurgitation, echocardiographic reversed remodeling and «syn-

chronizationn of the LV. 6 J 6 1 8 19 All patients responded clinically to CRT, This was

predictedd by the large septal to lateral delay and the very broad QRS complexes.

CRTT and 123I-MIB G

Previouss studies have shown that 123I-MIBG is capable of visualizing the improved

neurohumorall status in HF patients treated with b- blockers and ACE- inhibitors.3'4

Inn these studies, 123I-MIBG WO decreased and late 123I-MIBG H/M ratio increased.

Agostinii et al. showed an increased cardiac 123I-MIBG uptake after 6 months treat-

mentt with carvedilol in HF patients (NYHA 2 or 3)3 Somsen et al. observed a similar

increasee in myocardial 123I-MIBG uptake after 6 weeks of treatment with the ACE-

inhibitorr enalapril.20 In addition, Fukoka et al. found a decrease in I23I-MIB G WO

afterr therapy with metoprolol."1 Our findings show analogue to these studies an

increasee of cardiac 123I-MIBG uptake and less competition between plasma no-

138 8

Chapterr 7

radrenalinee and 123I-MIBG reflected by a decreased washout after 6 months of CRT.

Thiss may also reflect a restoration of pre-synaptic cardiac sympathetic nerve func-

tion.. In addition, also the decrease of isolated, nonischemic mitral regurgitation can

inducee the favorable neurohumoral changes.19 The favorable changes found in this

studyy are at least on top of ACE-inhibitors and in half of the patients also on top of

B-blockers,, In contrast to these favorable changes after CRT, previous studies

showedd worsening of 123I-MIBG parameters after sole right ventricular pacing.22

Ourr data supports that CRT induces reduced sympathetic activity as expressed by an

improvedd 123I-MIBG WO. The recent study of Adamson et al. reported that CRT

inducess a relative shift in cardiac autonomic balance toward a more favorable

profilee that is less dependent on sympathetic activation.25 However, they failed to

showw a clear reduction in sympathetic activity since they did not find a change in

plasmaa catecholamines. This difference in findings might be explained by differ*

encee in patient population and used measurements of sympathetic activity. Moreo-

ver,, more severe asynchrony may have been present in our patient population given

thee wider ORS-complex and larger LV diameters. Another striking difference be-

tweenn our study and the study of Adamson et al is the longer pacing period of 6

monthss compared to 3 months. Whether this difference can explain the differences

inn plasma catecholamines, remains unclear.

Thee improvement in 125I-MIBG parameters may have prognostic implications.

Decreasee of MIBG uptake is probably a late event in the history of the disease,

explainingg its prognostic value in end-stage HF. It has been shown that cardiac 123I-MIBGG imaging, especially cardiac 123I-MIBG WO, has prognostic value in patients

withh chronic HF.U1A 123I-MIBG WO is more useful as an index of adrenergic nervous

activityy because it is independent of the number of neurons available, whereas the 123l-MIBGG H/M activity ratio is not. Therefore, although the 1Z3I-MIBG H/M activity

ratioo and 123I-MIBG WO represent different functions of adrenergic presynaptic

activity,, washout may be a more accurate marker of severity of HF.

CRTT and plasma BNP

Inn our study, plasma BNP levels decreased significantly after 6 months of CRT

(p<0.001)) in all patients, although there were important interindividual differences,

ass frequently observed in HF patients. Pressure and volume overload of the cardiac

chamberss stimulate enhanced production and release of natriuretic peptides. These

139 9

peptidess result in natriuresis, diuresis, vasodilatation and the suppression of the

renin-angiotensin-aldosteronee system in the cardiovascular system and are the

commonn denominator present in patients with systolic or diastolic dysfunction,

volumee overload, and HF, regardless of the underlying cardiovascular disease.25

Atriall and brain natriuretic peptides are elevated in patients with HF, and these

peptides,, especially BNP, are independent prognostic parameters.26"28 Not only are

thee natriuretic peptides important predictors of HF mortality and morbidity, but

changess in these neurohormones over time are associated with corresponding

changess in mortality and morbidity.2 Among the neurohumoral parameters (nore-

pinephrine,, renin activity, aldosterone, arginine-vasopressin and endothelin) BNP is

thee strongest marker for LV dimensions and LV ejection fraction in patients with

systolicc HF.29

Ass has been previously shown, short term CRT results in a significant decrease in

BNPP levels in HF patients, but long term data about the effects of CRT on BNP are

limited.23,300 31 Sinha et al. studied the usefulness of BNP release as a surrogate

markerr of the efficacy of 6 months effect of CRT in 17 patients. After already 6

monthss of pacing, CRT was inactivated using a backup mode (WI 40 beats/min) for

100 days 3 days and then reinitiated using the same stimulation parameters. BNP

levelss were measured after 6 months of pacing just before discontinuation CRT, at

thee last day before and 10 days after CRT reinitiating. Although in the study of Sinha

ett al., the baseline BNP levels (before CRT) was not measured, they showed that BNP

levelss significantly increased immediately after short-term CRT termination and

decreasedd again after CRT reinitiating. In addition to the data of Sinha. we showed

thatt already in the first 6 months the BNP level is reduced. These data indicate that

BNPP might be a reliable marker to assess the effect of CRT and also guide medical

interventionn in case of worsening HF during CRT therapy. Furthermore, we could

nott show a correlation between MIBG indices and BNP levels. Possibly this lack of

correlationn between the two tests may be synergistic, with each test perhaps

examiningg a different aspect of the sympathetic nervous system.

Limitations s

Theree are some potential limitations to the present study. It has been shown that

somee beta-adrenoreceptorblockers and it has been suggested that some ACE-

inhibitorss may influence myocardial 123I-MIBG uptake. As medication was obviously

140 0

Chapterr 7

nott withdrawn in the present study, this could have influenced absolute myocardial 123I-MIBGG uptake measurements. On the other hand as medication was not altered,

measuredd differences in I23I-MIBG uptake will still reflect the effect of CRT.

Anotherr possible limitation might be that in the present study we did not perform

tomographicc imaging. However, especially in patients with end-stage HF it is

extremelyy difficult to obtain adequate tomographic acquisitions, due to poor

contrastt between myocardium and surrounding tissue.

Furthermore,, the present study might be limited by the lack of a control group.

However,, in light of clinical evidence, it is not ethical to withheld CRT in a control

groupp with severe HF and clear indications for CRT.

Conclusions s

Cardiacc resynchronization therapy induces favorable changes in the neurohumoral

systemm already within the first 6 months in patients with severe HF.

Acknowledgments s

Wee thank Irma de Groot, Jim Vleugels and Denise Samson for their technical assist-

ancee with the echocardiography. The authors also thank Christel Feenstra for her

technicall assistance with the myocardial 123I-MIBG scintigraphy.

141 1

References s 1.. Hunt SA, Baker DW. Chin MH, et al. ACC/AHA Guidelines for the Evaluation and Management

off Chronic Heart Failure in the Adult: Executive Summary A Report of the American College off Cardiology/American Heart Association Task Force on Practice Guidelines (Committee to Revisee the 1995 Guidelines for the Evaluation and Management of Heart Failure): Developed inn Collaboration With the International Society for Heart and Lung Transplantation; Endorsed byy the Heart Failure Society of America. Circulation 2001:104:2996-3007.

2.. Anand IS, Fisher LD. Chiang YT, et al. Changes in brain natriuretic peptide and norepine-phrinee over time and mortality and morbidity in the Valsartan Heart Failure Trial iVal-HeFT1. Circulationn 2003:107:1278-83.

3.. Agostini D. Beltn A. Amar MH. et al. Improvement of cardiac neuronal function after carvediloll treatment in dilated cardiomyopathy: a 123I-MIBG scintigraphic study. ] Nucl Med 2000:41:845-51. .

4.. Kawai H, Fan TH, Dong E. et al. ACE inhibition improves cardiac NE uptake and attenuates sympatheticc nerve terminal abnormalities in heart failure. Am ] Physiol 1999;277:H160Q-i7.

5-- Coletta AP Clark AL. Seymour AM, Cleland JG. Clinical trials update from the European Socie-tyy of Cardiology Heart Failure meeting: COMET, COMPANION, Tezosentan and SHAPE. Eur J Heartt Fail 2003:5:545-8.

6.. Auricchio A. Stellbrink C. Sack S, et al. Long-term clinical effect of hemodynamically opti-mizedd cardiac ^synchronization therapy in patients with heart failure and ventricular con-ductionn delay. J Am Coll Cardiol 2002;39:2026-33.

7.. Bradley D]. Bradley EA, Baughman KL, et al. Cardiac ^synchronization and death from pro-gressivee heart failure: a meta-analysis of randomized controlled trials. Jama 2003:289:730-40.

5.. Bristow MR. Saxon LA. Boehmer ]. et al. Cardiac-resynchronization therapy with or without ann implantable defibrillator in advanced chronic heart failure. N Engl J Med 2004:350:2140-50. .

9,, Zhao C, Shuke N, Yamamoto W, et al. Comparison of cardiac sympathetic nervous function withh left ventricular function and perfusion in cardiomyopathies by (123U-MIBG SPECT and (99mlTc-tetrofosminn electrocardiographically gated SPECT. J Nucl Med 2001:42:1017-24,

10.. Somsen GA, van der Wall EE, van Vlies B. Borm ]], van Royen EA. Neuronal dysfunction in heartt failure assessed by cardiac 123-iodine metaiodobenzylguanidine scintigraphy. Int 1 Card Imagingg 1996:12:305-10.

11.. Cohen-Solal A. Esanu Y, Logeart D, et al. Cardiac metaiodobenzylguanidine uptake in patients withh moderate chronic heart failure; relationship with peak oxygen uptake and prognosis, f Amm Coll Cardiol 1999:33:759-66.

12.. Ebina T, Takahashi N. Mitani I. et al. Clinical implications of cardiac il23U-meta-iodobenzyI-guanidinee scintigraphy and cardiac natriuretic peptides in patients with heart disease. Nucl Medd Commun 2002:23.795-801.

13.. Yamada T. Shimonagata T. Fukunami M, et al. Comparison of the prognostic value of cardiac iodine-1233 metaiodobenzylguanidine imaging and heart rate variability in patients with chronicc heart failure: a prospective study. J Am Coll Cardiol 2003;41:231-8.

14.. Ogita H. Shimonagata T, Fukunami M. et al. Prognostic significance of cardiac Ll23)I metaio-dobenzylguanidinee imaging for mortality and morbidity in patients with chronic heart fail-ure;; a prospective study. Heart 2001:8b:656-60.

15.. Gregoratos G. Abrams J. Epstein AE, et al. ACC/AHA/NASPE 2002 guideline update for implan-tationn of cardiac pacemakers and antiarrhythmia devices: summary article, A report of the Americann College of Cardiology/American Heart Association Task Force on Practice Guidelines ACC,, AHA/NASPE Committee to Update the 1998 Pacemaker Guidelines J Cardiovasc Electro-

physioll 2002:13:1183-99. 16.. Bax J]. Marwick T.H. Molhoek S.G. Left ventricular dyssynchrony predicts benefit of cardiac

^synchronizationn therapy in patients with end-stage heart failure before pacemaker implan-tation.. Am ] Cardiol 2003. in press.

17.. Shapiro BP. Chen HH. Burnett JC. Jr.. Redfield MM. Use of plasma brain natriuretic peptide concentrationn to aid in the diagnosis of heart failure Mayo Clin Proc 2003:78:481-6.

142 2

Chapterr 7

18.. Bax J]. Molhoek SG. van Erven L, et al. Usefulness of myocardial tissue Doppler echocardiog-raphyy to evaluate left ventricular dyssynchrony before and after biventricular pacing in pa-tientss with idiopathic dilated cardiomyopathy. Am J Cardiol 2003:91:94-7.

19.. Tsutsui H. Ando S, Kubota T. et al. Abnormalities of cardiac sympathetic neuronal and left ventricularr function in chronic mitral regurgitation: assessment by iodine-123 metaiodoben-zylguanidinee scintigraphy. Am ] Card Imaging 1996:10:14-22.

20.. Somsen GA, van Vlies B, de Milliano PA. et al. Increased myocardial [12311-metaiodobenzyl-guanidinee uptake after enalapril treatment in patients with chronic heart failure. Heart 1996:76:218-22. .

21.. Fukuoka S. Hayashida K. Hirose Y. et al. Use of iodine-123 metaiodobenzylguanidine myocar-diall imaging to predict the effectiveness of beta-blocker therapy in patients with dilated cardi-omyopathy.. Eur J Nucl Med 1997:24:523-9.

22.. Simantirakis EN, Prassopoulos VK. Chrysostomakis SI, et al. Effects of asynchronous ventricu-larr activation on myocardial adrenergic innervation in patients with permanent dual-chamber pacemakers;; an H123)-metaiodobenzylguanidine cardiac scintigraphic study, Eur Heart J 2001;22:323-32. .

23.. Adamson PB. Kleckner KJ. VanHout WL, Srinivasan S, Abraham WT. Cardiac ^synchronization therapyy improves heart rate variability in patients with symptomatic heart failure. Circulation 2003:108:266-9. .

24.. Merlet P, Valette H, Dubois-Rande JL, et al. Prognostic value of cardiac metaiodobenzylguani-dinee imaging in patients with heart failure. J Nucl Med 1992:33:471-7.

25.. Levin ER. Gardner DG, Samson WK. Natriuretic peptides. N Eng] J Med 1998;339:321-8. 26.. Tsutamoto T. Wada A, Maeda K, et al. Attenuation of compensation of endogenous cardiac

natriureticc peptide system in chronic heart failure: prognostic role of plasma brain natriuretic peptidee concentration in patients with chronic symptomatic left ventricular dysfunction. Cir-culationn 1997;96:509-16.

27.. Swedberg K. Eneroth P. Kjekshus J. Wilhelmsen L. Hormones regulating cardiovascular func-tionn in patients with severe congestive heart failure and their relation to mortality, CONSEN-SUSS Trial Study Group. Circulation 1990:82:1730-6.

2&.2&. Gottlieb SS. Kukin ML, Ahern D, Packer M. Prognostic importance of atrial natriuretic peptide inn patients with chronic heart failure. J Am Coll Cardiol 1989:13:1534-9.

29.. Groenning BA. Nilsson JC. Sondergaard L. Kjaer A, Larsson HB, Hildebrandt PR, Evaluation of impairedd left ventricular ejection fraction and increased dimensions by multiple neurohu-morall plasma concentrations. Eur J Heart Fail 2001:3:699-708.

30.. Filzmaier K SA. Breithardt OA, Stiegler H. Short-term effects of cardiac ^synchronisation on brainn natriuretic peptide release in patients with systolic heart failure and ventricular conduc-tiunn disturbance. J Am Coll Cardiol 2002:39:104.

31.. Sinha AM, Filzmaier K, Breithardt OA, et al. Usefulness of brain natriuretic peptide release as aa surrogate marker of the efficacy of long-term cardiac ^synchronization therapy in patients withh heart failure. Am f Cardiol 2003;91:755-8.

143 3

Cardiacc resynchronization improvess microcirculation

Aytenn Erol-Yilmaz MD,1 Bektas. Atasever MD,2 Keshen Mathura.2

Jeromee Lindeboom MD.5 Arthur Wilde MD PhD.1 Can ince PhD.2

andd Raymond Tukkie MD PhD1

Fromm the Department of Clinical and Experimental Cardiology,1 the Department of Physiologyy 2 and the Department of Oral and Maxillofacial Surgery ? Academic

Medicall Center, University of Amsterdam, Amsterdam, The Netherlands,

SubmittedSubmitted for publication

Abstract t

Background d

Althoughh it is known that CRT in HF patients improves systemic circulation, its

acutee effects on microcirculation are as yet unknown. Therefore we investigated the

sub-linguall microcirculatory changes in heart failure (HF) patients due to cardiac

^synchronizationn therapy (CRT) and right ventricular pacing (RV) by use of orthogo-

nall polarization spectral (OPS) imaging.

Material ss and methods

Twelvee consecutive HF patients with a CRT device and 10 healthy individuals were

included.. Acute microcirculatory changes were assessed by functional capillary

densityy (FCD) and capillary velocity (CV) after previous six months of CRT. FCD and

CVV were measured in HF patients sub-lingually after 15 minutes of programming in

onee of the pacing modalities (no pacing, RV pacing, and CRT) in dual chamber mode.

Results s

FCDD was significantly higher in healthy individuals (10.9 1.9. cm/cm2) compared

too HF patients with RV pacing (8.9 1,9. cm/cm2, p = 0.025) and no pacing

(8.33 2.4, cm/cm2, p = 0.008), CRT (12.1 2.2. cm/cm21 significantly increased FCD

inn HF patients compared to RV pacing (8.9 1.9 cm/cm2, p=0.03) and no pacing

(8.33 2.4, cm/cm2. p=0.018). CV was normal in all patients with or without pacing.

Conclusions s

CRTT improves microcirculatory function as assessed by OPS imaging.

146 6

Chaprerr S

Introductio n n

Heartt failure (HF) is a complex clinical syndrome resulting from any structural or

functionall cardiac disorder that impairs the ability of the ventricle to fil l with or

ejectt blood.1 The symptoms of HF can be explained in terms of microvascular

dysfunctionn within the tissues, either inadequacy of nutritive flow or failure to

preservee tissue fluid economy,23

Cardiacc resynchronization therapy (CRT) on top of optimal pharmacological therapy

hass already been shown to improve cardiac and systemic hemodynamics. CRT

inducess reduction of symptoms of HF. reduction of hospital admission, improves

exercisee capacity, V02 max and left ventricular (LV) function,4

Soo far. imaging of the human microcirculation in patients with HF has been limited

too observations in nailfold and conjunctiva,235 Orthogonal polarization spectral

(OPS)) imaging is a recent technique for non-invasive observation and quantification

off microcirculatory function in patients. The OPS imaging, which uses green

(wavelengthh 550 nm) polarized light that is guided through a set of lenses, was

originallyy introduced by Slaaf et al. The OPS technique was validated in previous

studiess with septic shock, brain tumors, nailfold skin and comparing OPS with

conventionall capillary microscopy in healthy volunteers. In addition, other studies

weree performed to assess the effects of no rep inephr ine and dopamine on human

intestinall mucosal perfusion. 7 1 0 13

Thee OPS imaging technique provides the unique opportunity to quantify measure-

mentss of relevant physiological parameters in the microcirculation such as function-

all capillary density (FCD) and capillary flow (CV). FCD is defined as the length of

capillariess perfused with red blood cells per observation area and is given in cm/

cm2.. Since red blood cells are taken into account in the FCD measurements. FCD is a

directt measure of nutritional tissue perfusion and an indirect measurement of

oxygenn delivery to tissue.1^

AA recent study using OPS imaging, identified abnormal microcirculatory flow

alterationn in patients with severe HF.7 However, studies examinating the effect of

CRTT on microcirculation in patients with severe HF are lacking. In this study, we

testedd the hypothesis that in patients known to respond to CRT, CRT also improves

microcirculationn in these patients by observation of sub-lingual microcirculation

withh OPS.

147 7

Material ss and methods

Studyy population

Twelvee consecutive patients (9 male, 3 female. 64.5 5 years) with chronic

symptomaticc HF referred for implantation of a CRT device were included in the

presentt study after informed consent. Ten age matched healthy individuals (HI)

withoutt co-morbidity and medical therapy (10 women: 61.4 2.3 years) were

includedd to confirm the hypothesis that HF patients have an impaired microcircu-

lation.. Patients received CRT to the criteria of device implantation according to

ACC/AHAA criteria (ejection fraction < 35%. QRS width > 120 ms, and severe HF in

NYHAA functional classes III or IV despite maximally tolerated medical treatment for

HF)) and measured echocardiographic LV asynchrony (septal to lateral delay >60 ms)

ass previously described.' l5 Between 2000 and september 2004. in 53 patients CRT

devicess were implanted in our hospital. Due to the criteria used in our academic

centerr for implantation of CRT devices lower than 10% of the patients with a CRT

devicee are non-responders. So the majority of the patients are responders. Five

patientss had a history of managed hypertension and 2 patients had also type II

diabetess mellitus.

Randomizationn protocol and study design

Beforee pacemaker implantation (baseline) and 6 months after CRT, NYHA classifica-

tion.. QRS width, left ventricular (LV) functional parameters using echocardiography

parameterss were assessed. Standard biventricular pacemaker implantation was

performedd using available biventricular pacemakers (Guidant, Minnesota, USA:

Medtronicc Minneapolis USA). The right atrial lead was positioned in the right atrial

appendagee and the ventricular lead was positioned in the right ventricular apex. The

LVV pacing lead was positioned in a posterior or lateral branch of the coronary sinus

inn all patients. A biventricular DDDR system was used in 11 patients. In 1 patient

thee implantation of an internal defibrillator was also indicated and a combined

devicee was implanted. The echocardiographically optimized sensed AV delays were

betweenn 100-120 msec.

Echocardiographyy parameters were measured with a Vivid 7 ultrasound device

14S S

Chapterr 8

(GE-Vingmedd Ultrasound, Horten Norway). Patients were imaged in the left lateral

positionn using 2 D, M mode, color-doppler and tissue doppler echocardiography.

Thee left ventricular end-diastolic diameter (LVEDD) and left ventricular end-systolic

diameterr (LVESD) were measured from the parasternal long axis view. Mitral regur-

gitationn was classified semi-quantitatively in four degrees (I = mild, II - moderate,

IIII = moderate to severe. IV = severe).

Alll patients responded beneficially to CRT. FCD and CV were measured after 6

monthss of CRT in an acute setting using OPS imaging. In each patient FCD was

measuredd in one session at random with 1) no pacing in HF patients, 2) RV pacing

andd 3) biventricular pacing. The heart rates in the RV and biventricular pacing mode

weree programmed to fixed 60 beats/min. Measurements were done after fifteen

minutess of pacing in the at random programmed mode and with no pacing. In each

patient,, the systolic and diastolic blood pressure was measured once with a mercury

sphygmomanometerr 15 minutes after each pacing modality. In 10 healthy individu-

alss also the FCD and the CV were measured.

Orthogonall polarization spectral imaging

Thee OPS imaging device uses the absorption of hemoglobin to visualize the micro-

circulationn using a polarized light technique as described before.59,12 16 The subject

mediumm is illuminated with light that has been linearly polarized in one plane,

whilee imaging the remitted light through a second polarizer oriented in a plane

preciselyy orthogonal to that of the illumination. In order to make the image, the

lightt is collected , passed through a spectral filter to isolate the wavelength region,

andd linearly polarized.

Functionall capillary density and capillary velocity

Alll patients and HI were examined in the resting sitting condition in the afternoon

withh uniform room temperature. The sublingual microcirculation was observed by

OPSS imaging (Cytoscan A/RII (Cytometrics. Philadelphia, Pa), with a 5 X objective

providingg a 166 X final magnification as previously described.0 l2Ab For FCD meas-

urements,, three sublingual regions of interest were selected. From each region two

imagess with the best visual quality in every subject were analyzed. Measurements

weree digitally videotaped for 1 minute. All computed aided analyses was performed

149 9

usingg Cap-image computer computer programme (Cap-image; H Zeintl, Heidelberg,

Germany99 10). FCD was measured as the length of capillaries perfused with red

bloodd cells per observation area given in cm/cm2 as described previously. '

Twoo independent blinded investigators analyzed the perfusion in the microcircula-

toryy bed semi-quantitatively as described elsewhere.12 The perfusion was scored in

low.. normal and high flow at each measurement.

Statisticall analysis

Thee non-parametric data were analyzed using the Mann- Whitney U test. All data were

expressedd as mean SD. A P value <0.05 was considered statistically significant.

Results s

Patientt Population

Twelvee patients were included in the study, of which 8 had non-ischemic heart

disease.. All patients were in sinus rhythm except one patient who was in chronic

atriall fibrillation with slow atrioventricular conduction. All of the patients used

diuretics.. 83% ace-inhibitors. 42% p-blocker. 25% amiodarone and 8% calcium-

antagonist.. All patients were responders to CRT indicated by a significant decrease

inn NYHA classification from 3.6 0,5 to 2.3 0.6 (p< 0.001), decrease in QRS width

fromm 179 17 to 156.0 14 msec (p<0.001). increased LV ejection fraction from 14

too 22% (p<0.001), reduction of LVEDD from 73 9 to 67 9 mm (p = 0.005) and

LVESDD from 64.8 7 to 57.5 12.2 mm (p = 0.005). In addition, mitral regurgitation

decreasedd from 3 0.5 to 2.2 0.8 (p = 0.04). Mean blood pressure in the HF

patientss with biventricular pacing was 125 13 mm Hg, with RV pacing 121 9 mm

Hgandd no pacing 114 7 mm Hg (p = 0.786).

Functionall capillary density and capillary velocity

Inn four patients pacing could not be terminated because of inadequate intrinsic

rhythm.. During intrinsic rhythm FCD was the lowest [&. 2.4 cm/cm2). Biventricular

150 0

Chapterr S

Figuree 1, Functional capillary density off heart failure (HF) patients with intrinsicc rhythm (INT), with cardiac ^synchronizationn therapy (CRTV right ventricularr pacing (RV) and of healthy individualss (HI). HF vs. CRT = pp = 0.018: HF vs. HI =p = 0.008; CRT vs. RVV = p = 0.03: RVvs. HI =p = 0.025.

HF-INTT HF-CRT HF-RV HI I

Figuree 2.a.: Images of the microcircu-lationn with the CRT device pro-grammedd Off.

Figuree 2.b.: Images of the microcircu-lationn with the CRT device pro-grammedd On. Notice that the capillary densityy is higher when the CRT device iss programmed On.

- : :

TableTable 1, Functional capillary density of HF patients, duringg right ventricular pacing and CRT.

Patients s

1 1 2 2 3 3 4 4 5 5 6 6 7 7 8 8 9 9

10 0 11 1 12 2

HF-INT T

8.7 7 11.5 5 7.1 1 8.8 8 3-8 8 7.2 2 9,5 5 8.9 9

RV V

8.8 8 11.4 4 6.4 4 7.7 7 6.2 2 8.8 8 7,7 7 8 8

12.3 3 10.6 6 8.8 8

10.5 5

CRT T

93 3 14.1 1 12 2 9.3 3

14.5 5 10.1 1 14 4 13 3 14.9 9 138 8 10.4 4 10.1 1

HF-INTT = heart failure patients with intrinsic rhythm; RV== right ventricular pacing: CRT = cardiac ^synchronizationn therapy: -notavailable.

pacingg significantly increased FCD compared to RV pacing (12,1 2.2. cm/cm2) vs.

(8.99 1.9 cm/cm2, p = 0.03) and no pacing in HF patients (8.3 2.4, cm/cm2.

p=0.018,, see table 1, figure 1). Figure 2.a and 2.b are representative figures showing

thee capillaries during pacemaker programmed off and with CRT on, Age matched HI

hadd significantly higher FCD than HF with the intrinsic rhythm; 10.9 2.0 vs. 8.3

2.44 cm/cm2, p = 0.008. The FCD in age matched HI compared to FCD induced by CRT

weree not significantly different {10.9 2.0. vs. 12.1 2.2, cm/cm2). Capillary

velocityy was normal in all patients in each pacing modality and also when the

pacemakerr was programmed off.

Discussion n

Inn this study, we showed for the first time that cardiac resynchronization therapy

acutelyy improves microcirculatory perfusion. Today, the beneficial effects of CRT on

topp of optimal pharmacological therapy, regarding HF symptoms, hospital admission

ratess and LV function together with systemic circulation, are well establised. The

MUSTICC study showed the maintenance of clinical and hemodynamic benefits after

122 months of CRT and an improvement of left ventricular ejection fraction.1' Acute

hemodynamicc studies with CRT showed improvement in systolic blood pressure,

pulmonaryy capillary wedge pressure, dp/dt, V02 max and cardiac output.18 19 In the

152 2

Chapterr S

PATH-CHFF study, CRT decreased the left ventricular end diastolic pressure signifi-

cantlyy and increased dp/dt.4 Stroke volume and ejection fraction increased within 2

dayss after CRT in the recently published study by Gorscan et al. 18 The microcircula-

toryy changes in our study were already reached within 15 minutes, while previous

studiess showed also that the macrocirculatory changes could be observed within 20

minutes.44 20 The improvement in microcirculation likely contributes to the ob-

servedd beneficial clinical respons and may explain why some patinets with marked

functionall benefit despite limited objective increase in LV ejection fraction im-

prove,, since improvement in microcirculation means more oxygen delivery to the

tissuess (like muscles).

Thee present study confirmed recent observations by De Backer et al, 7 of decreased

microcirculationn in HF patients imaged with the OPS technique. Main changes in

microcirculatoryy in patients with HF are reduction of capillary density and blood

velocityy as described in previous reports using several other techniques,3 21 22 The

mechanismm for the changes is not clear yet. Influencing factors for these changes

couldd be various inflammatory mediators, increase of tumor necrosis factor, reactive

oxygenn species, increase of adrenergic agents, and increase of sympathovagal

imbalance.. 3' 5 Whether long term CRT can undo these negative changes by e.g.

reducingg sympathovagal imbalance and by improving cardiac output has to be

investigatedd in the future.

Inn the present study, the FCD increased significantly with CRT. A possible explanation

forr the direct increase of the FCD with CRT is the improvement of tissue perfusion by

ann increase in dp/dt and stroke volume, because microcirculatory control mechanisms

activelyy respond to changes of cardiac output as the heart similarly reacts to altera-

tionss in pre- and afterload. However, clinical studies showed that improvement of

macrocircualtionn does not necessarily result in improvement of microcirculation.7 21

Nevertheless,, in our opinion improving microcirculation might be more important

thann improving macrocirculation as illustrated in the study of De Backer et al. These

investigatorss showed that in patients with acute severe HF and cardiogenic shock, the

proportionn of especially perfused small vessels was higher in patients who survived

thann in patients who did not survive, although the perfusion of large vessels was

preservedd in all patient groups.

Similarr improvement in FCD was observed in a recent OPS study by Creteur et al. in

septicc patients where correction of systemic hemodynamic parameter also led to an

improvedd FCD. Treatment by catecholamines of septic shock patients also resulted

153 3

inn capillary recruitment.

Thee capillary velocity in our patients were normal and did not change with CRT and

RVV pacing. Which is in contrast to the study of the Backer et al,7The difference in

resultss could be explained by the different methods used for the analysis. They

recordedd 20 seconds, whereas we recorded 1 minute. Furhtermore, the patient

populationn in the study of De Backer et al. have more severe heart failure compared

too the patients in our study. Namely 77% of the patients were in cardiogenic shock

usingg inotropics Idobuatmine, dopamine, noradrenaline) with dilating effect on

capillariess with as consequence lower flow. However, in our patients there was

alreadyy normal flow but new capillaries are recruited.

Limitations s

AA limitation to the present study might be the fact, that we were not able to investi-

gatee stroke volume and dp/dt simultaneously with the OPS measurements. These

dataa could provid more detailed information about the correlation of systemic

hemodynamicss and microcirculatory changes. Furthermore, measurements were

donee for 15 minutes only in responders to CRT and only after device implantation.

Futuree research should investigate whether non-responders show no or reduced

microcirculatoryy changes. In addition, it would be of interest to know whether

microcirculatoryy changes at implant could predict beneficial response to CRT and

whetherr CRT induces beneficial long-term microcirculatory changes. The study

populationn is too small to directly correlate the improvement in microcirculation to

thee observed clinical benefit.

Itt is known that hypertension and diabetes mellitus can induce microcirculatory

alterations.277 2S However, in our opinion the role of these possible alterations have

noo significant role in our study for several reasons. First, the patients had no severe

hypertensionn or complicated diabetes. Second, each patient was his/her own

reference.. Third, patients with hypertension and diabetes mellitus responded with

aa identical microcirculatory changes to the switch in pacing mode compared to the

otherr patiens.

154 4

Chapterr 8

Conclusions s

Ourr data confirm that heart failure patients have a disturbed microcirculation

comparedd to healthy individuals. Cardiac resynchronization therapy induced an

acutee improvement of microcirculation as assessed by OPS.

Acknowledgement t

B.Ataseverr has been supported by Netherlands Heart Foundation (gr.2001B142).

References s 1.. Gregoratos G. Abrams ]. Epstein AE. Freedman RA, Hayes DL. Hlatky MA. Kerber RE. Naccarel-

lii GV Schoenfeld MH. Silka MJ. Winters SL. ACC/AHA/NASPE 2002 Guideline Update for Im-plantationn of Cardiac Pacemakers and Antiarrhythmia Devices—summary article: a report of thee American College of Cardiology/American Heart Association Task Force on Practice Guide-liness iACC/AHA/NASPE Committee to Update the 1998 Pacemaker Guidelines). J Am Coll Car-dioldiol 2002:40:1703-19.

2.. Tassi G. Maggi G. de Nicola P. Microcirculation in the elderly. Int Angiol 1985:4:275-83. 3-- Houben AJ. Beljaars JH. Hofstra L, Kroon AA, De Leeuw PW. Microvascular abnormalities in

chronicc heart failure: a cross-sectional analysis. Microcirculation. 2003:10:471-8. 4.. Auricchio A. Stellbrink C, Sack S. Block M. Vogt J, Bakker P. Huth C, Schondube F. Wolfhard U,

Boekerr D. Krahnefeld O, Kirkels H. Long-term clinical effect of hemodynamically optimized cardiacc resynchronization therapy in patients with heart failure and ventricular conduction delay.. J Am Coll Cardiol 2002:39:2026-33.

5.. Duprez D, De Buyzere M, Dhondt E. Clement DL. Impaired microcirculation in heart failure. IntInt j Microcirc Clin Exp. 1996:16:137-42.

6.. Slaaf DW. Tangelder G], Reneman RS, Jager K, Bollinger A. A versatile incident illuminator for intravitall microscopy. Int J Microcirc Clin Exp. 1987:6:391-7.

7.. De Backer D, Creteur J, Dubois MJ, Sakr Y, Vincent JL. Microvascular alterations in patients withh acute severe heart failure and cardiogenic shock. Am Heart J. 2004;147:91-9-

8.. Lindert J. Werner J. Redlin M, Kuppe H. Habazettl H. Pries AR, OPS imaging of human micro-circulation:: a short technical report, J Vase Res. 2002:39:368-72.

9.. Mathura KR, Ince. C. First clinical use of orthogonal polarization spectral imaging. Prog Appl Microcirc.Microcirc. 2000;24:94-101.

10.. Mathura KR. Vollebregt KG Boer K, De Graaff JC. Ubbink DT. Ince C. Comparison of OPS imag-ingg and conventional capillary microscopy to study the human microcirculation, J Appl Physi-ol.ol. 2001:91:74-8.

11.. Mathura KR, Bouma GJ. Ince C. Abnormal microcirculation in brain tumours during surgery. Lancet.Lancet. 2001:358:1698-9-

12.. Spronk PE, Ince C, Gardien MJ, Mathura KR. Oudemans-van Straaten HM. Zandstra DF. Nitro-glycerinn in septic shock after intravascular volume resuscitation. Lancet. 2002:360:1395-6.

13-- Spronk PE, Zandstra DF, Ince C. Norepinephrine compromises intestinal microvascular per-fusion?? Intensive Care Med. 2004;30:173-4: author reply 175.

14.. Harris AG, Leiderer R, Peer F, Messmer K. Skeletal muscle microvascular and tissue injury af-terr varying durations of ischemia. Am J Physiol. 1996:271 :H2388-98.

15.. Bax JJ. Molhoek SG, van Erven L. Voogd PJ. Somer S. Boersma E. Steendijk P. Schalij MJ. Van derr Wall EE. Usefulness of myocardial tissue Doppler echocardiography to evaluate left ven-tricularr dyssynchrony before and after biventricular pacing in patients with idiopathic dilated cardiomyopathy.. Am J Cardiol 2003:91:94-7.

16.. Groner W, Winkelman JW. Harris AG, Ince C. Bouma GJ, Messmer K, Nadeau RG, Orthogonal polarizationn spectral imaging: a new method for study of the microcirculation. Nat Med. 1999:5:1209-12. .

17.. Linde C, Leclercq C, Rex S. Garrigue S, Lavergne T, Cazeau S, McKenna W, Fitzgerald M, Deha-roo JC. Alonso C. Walker S, Braunschweig F, Bailleul C, Daubert JC. Long-term benefits of biven-tricularr pacing in congestive heart failure; results from the MUltisite STimulation in cardio-myopathyy (MUSTIC) study. J Am Coll Cardiol. 2002;40:111-8.

18.. Gorcsan J. 3rd, Kanzaki H. Bazaz R, Dohi K. Schwartzman D. Usefulness of echocardiography tissuee synchronization imaging to predict acute response to cardiac resynchronization thera-py.. Am ƒ Cardiol. 2004:93:1178-81.

19.. Blanc JJ. Etienne Y. Gilard M. Mansourati J, Munier S. Boschat J. Benditt DG. Lurie KG. Evalua-tionn of different ventricular pacing sites in patients with severe heart failure: results of an acutee hemodynamic study. Circulation. 1997:96:3273-7.

20.. Leclercq C. Cazeau S. Le Breton H, Ritter P. Mabo P. Gras D. Pavin D, Lazarus A, Daubert JC. Acutee hemodynamic effects of biventricular DDD pacing in patients with end-stage heart fail-ure.. J Am Coll Cardiol. 1998:32:1825-31-

156 6

Chapterr 8

21.. Bittner HB, Diemel KD. Friedel N. Stengele B, Hetzer R. Skin microcirculation and laser Dop-plerr blood flow in patients supported by the Berlin heart biventricular assist device. Vasa. 1992:21:149-53--

22.. Wrobiewski H. Kastrup J, Norgaard T, Mortensen SA. Haunso S. Evidence of increased microv-ascularr resistance and arteriolar hyalinosis in skin in congestive heart failure secondary to id-iopathicc dilated cardiomyopathy. Am } Cardiol. 1992;69:769-74.

23.. Farquharson CA. Butler R, Hill A. Belch JJ. Struthers AD. Allopurinol improves endothelial dysfunctionn in chronic heart failure. Circulation. 2002:106:221-6.

24.. Andersson SE, Edvinsson ML. Edvinsson L. Cutaneous vascular reactivity is reduced in aging andd in heart failure; association with inflammation. Clin Sci [Lond). 2003:105:699-707.

25.. Levine B. Kalman J. Mayer L. Filïi t HM. Packer M. Elevated circulating levels of tumor necrosis factorr in severe chronic heart failure. N Eng! ƒ Med. 1990:323:236-41.

26.. Creteur J D8D, Sakr Y, Koch M. Vincent JL.. Determinant of sublingual PC02 in patients with septicc shock. Critical Care MEd. 2003:31 :A116:419.

27.. Farkas K KE. JaraiZ. Nemcsik f, Farsang C. Non-invasive assessment of microvascular endothe-liall function by laser doppler flowmetry in patients with essential hypertension. Atheroscle-rosis.rosis. 2004;173:97-102.

28.. Jaap AJ, Pym CA, Seamark C. Shore AC. Tooke JE. Microvascular function in type 2 (non-insu-lin-dependent)) diabetes; improved vasodilation after one year of good glycaemk control. Dia-betMed.betMed. 1995:12:1086-91.

157 7

Reversedd Remodelling of Dilated Leftt Sided Cardiomyopathy After

Upgradingg From WIR to WIR Biventricularr Pacing

Case-reportt and review of the literatur e

Aytenn Erol-Yilmaz MD. Raymond Tukkie MD PhD. Tim Schrama, Hans Romkes MD

andd Arthur Wilde MD PhD

EuropaceEuropace 2002: 4:445-449.

Abstract t

Aims s

Too show reversed remodelling of the left ventricle and left atrium after upgrading to

biventricularr pacing in a patient with worsening of heart failure due to WIR pacing.

Too review the literature concerning reversed remodelling after upgrading to biven-

tricularr pacing.

Methodss and results

Inn a 61-year-old man, with cardiomyopathy and worsening of heart failure and

mitrall insufficiency following conservative pacing after His-bundle ablation, upgrad-

ingg to biventricular pacing showed left sided reversed remodelling of the heart with

decreasee of mitral regurgitation in 6 months. The literature is reviewed with search

inn medline between 1966-2001,

Conclusions s

Thee occasionally observed progressive heart failure after RV pacing following His-

bundlee ablation can be reversed by upgrading to W1R biventricular pacing and

underr these circumstances upgrading to biventricular pacing should be considered.

160 0

Chapterr 9

Introductio n n

Promisingg results for the treatment of heart failure patients with cardiac dyssyn-

chronyy have been described with simultaneous right (RV1 and left ventricular (LV)

(biventricular)) pacing in order to obtain improved synchronisation of ventricular

contraction.1"100 The usual indication for biventricular pacing are patients with dilated

cardiomyopathyy without conventional pacing indications, broad QRS and severe

symptomss despite optimal drug therapy. This article describes to our knowledge for

thee first time reversed remodelling of the LV and left atrium after upgrading to

biventricularr pacing in a patient with worsening of heart failure due to WIR pacing.

Thee literature concerning remodelling of the heart due to pacing is discussed.

Casee Report

AA 61-year- old resting butcher was admitted to our hospital with progressive in-

creasee of mitral regurgitation (MR) after WIR-pacing of the right ventricle.

Inn 1978 he was treated because of endocarditis of the mitral valve with positive

bloodculturess (staphylococcus aureus). The chest radiograph documented mild

cardiomegalyy and mild mitral regurgitation was described. The ECG showed atrial

fibrillationn with high ventricular response, which was converted to sinus rhythm by

electricc cardioversion. Therapy with cardioquin and Coumadin was started. With

controll of sinus rhythm, he had minimal dyspnoe on effort, his heart size remarka-

blee decreased and the mitral regurgitation resolved, Between 1978 and 1983 the

patientt was free of atrial fibrillation. From 1983 to 1998 recurrent episodes of atrial

fibrillationn occurred with again increase of heartsize and increase in dyspnoe on

effort.. Several drugs were tested (including amiodarone) without success. Extensive

analysiss did not reveal a cause for the cardiomyopathy (heart catheterisation,

coronaryy angiography, gallium-scintigrafy, cardiac biopsy). We postulated that during

thiss period the increase in left ventricular enddiastolic diameter (LVEDD), from 59

too 69 mm, was due to tachycardiomyopathy (mean resting heart rate was > 100

beats/min).. In an attempt to stop the LV-dilatation and control ventricular rate. His-

bundlee ablation was performed with implantation of a WIR pacemaker in 1998

(Biotronicc actros SR. Biotronik GMBH. Berlin, Germany), The heart rates before and

afterr His-bundle ablation with RV pacing showed improvement. Unexpectedly

161 1

withinn a year after RV pacing. MR increased from moderate to severe (fig. 3. large jet

area,, considerable prevalvular acceleration, calculated regurgitant volume measured

withh Proximal Isovelocity Surface Area (PISA) 62 ml. and systolic flow reversal in the

leftt upper pulmonary vein) with increase in systolic pulmonary artery pressure to

400 mm Hg and decrease of LV ejection fraction to 29 % (table I). These findings

suggestt that WIR pacing of the right ventricle with resultant ventricular dyssyn-

chronyy had increased the MR. Therefore he was admitted to upgrade the pacemaker

systemm from WIR right ventricular to WIR (2V) biventricular. The old pacemaker

(Biotronicc Actros) was explanted and the right ventricular bipolar lead (Slimtime S)

disconnectedd from the pacemaker. A new pulsegenerator DDDR (Guidant Contak TR

1241)) with an accelerometer sensor was subcutaneously implanted. The atrial port

off the pulsegenerator was connected to the right ventricular electrode. The left

ventricularr port was connected to a new unipolar lead (Guidant Easytrak 4512).

whichh was placed in the posterolateral vein over the left ventricular free wall

throughh the coronary sinus (figure 1). The RV port of the pacemaker was sealed.

Biventricularr pacing was established by programming the pacemaker to the DDDR

modee with the shortest A-V stimulation interval (10 ms). Direct after biventricular

pacingg QRS duration was clearly shortened (0,18 too 0.14 sec. figure 2). After 6

monthss follow up, echocardiography revealed obvious signs of reverse remodelling.

namely:: reduction of LVEDD from 69 to 62 mm and left atrial diameter from 54 to

499 mm, improvement of LV ejection fraction to 34 %, the MR diminished from

severee to moderate (calculated regurgitant volume measured with PISA 19 ml) and

pulmonaryy arterial pressure normalised (22 mmHg). The patient experienced

markedd increase in his exercise tolerance: the NYHA functional class after biven-

tricularr pacing reduced from III to II similar to before WIR pacing .

162 2

Chapterr 9

Tablee 1. Echocardiography before and after upgade

Yearr LVEDD FS MR grade LA diam RVSP EF tmm11 \%D U-41 imm1 (mmHgl t'V

6Q Q

62 2

s s ns s nd d

23* *

21 1

nd d s s

ns s

4 4 ipisaa 63 ml)

2 2 ipisaa 19 ml)

nd d s s

nd d

54 4

49 9

nd d nd d nd d

40 0

22 2

nd d nd d nd d

29 9

32 2

s s s s s s

Presentt Case 19977 35 1998.. just before His-bundle 6Q 17 3 53 29 ablationn and RV pacing 2000.. just before upgrade to BV-pacing g 2001.. 6 months after upgrade too BV pacing

Studies s Miracle"" " C.PLauetal .2--Medtronicc Insync1

Companion.. Contak CD, Path-CIIF. Mustic. Ventak CHF and Vigor CHF = none of these parameters is described.11 ^ """""

Tablee 1. RV- right ventricle; BV = biventricular: LVEDD = left ventricular enddiastolic diameter: MR== mitral regurgitation: PISA = proximal isovelocity surface area: LA diam = left atrial diameter: RVSP== right ventricular systolic pressure: F S- fractional shortening: EF- nuclear LV ejection fraction:: * = reduced afterload due to severe MR: s = significant; n s- not significant; nd= not described. .

Figuree 1. Upper panel: Twelve EKG leads before His-bundlee ablation; Atrial fibrillation, ventricular response 80-110/min.,, left heart axis. ORS duration 0.10 sec. ST-segmentt abnormalities (digoxin). Middle panel: after Hisbundlee ablation: Wl pacing 70/min., left heart axis, left bundlee branch block pattern, QRS duration 0.18 sec Lower panel:: during biventricular pacing, marked reduction in the pacedd QRS duration from 0.18 to 0 14 sec.

163 3

Figuree 2. Anteroposterior (left panel) and left lateral (right panel) chest X-rays showing the ventricu-larr pacing leads. Note the posterior position of the left ventricular lead.

Figuree 3- Apical four-chamber view with severe MR during RV pacing (left panel) and moderate MR duringg biventricular pacing (right panel).

Discussion n

Thee beneficial effects of His ablation and pacemaker therapy (improvement of LV

ejectionn fraction, exercise tolerance. Quality of Lif e and fewer hospitalisations) are

welll documented.11" In this case, however with tachycardia-induced dilated

cardiomyopathy,, catheter ablation of the His-bundle and permanent RV pacing

becausee of drug refractory atrial fibrillation, resulted in a marked increase in MR.

Mitrall regurgitation develops in a small number of patients who have His bundle

ablation,, and this occurs immediately after right ventricular pacing is initiated.14

Thee mechanism whereby right ventricular pacing results in MR appears to be

164 4

Chapterr 9

complex.. It affects the competence of the mitral valve due to changes in timing of

activationn of the mitral valve apparatus and/ or due to altered tension generation of

thee papillary muscles. This results in loss of coaptation and an increase orifice area

inn systole.14,15

Thee altered sequence of activation associated with RV apex pacing results in altered

mechanicall stress vectors, heterogeneity of regional coronary blood flow, and

elevatedd tissue catecholamine levels. These changes induce remodelling of the

globall ventricular shape with progressive changes in left ventricular performance.11

15188 Although the whole mechanism of remodelling muscle fibres is not clear yet,

manyy (experimental) studies, show that asynchronous electrical activation plays a

centrall role.3"6 19 In a study of de Cock et al. chronic (up to 3 months) ventricular

pacingg caused about 30% decrease of wall mass in early activated regions but did not

changee wall mass in late activated regions.20 This indicates that chronic asynchro-

nouss activation induces asymmetrical structural adaptations which gives an abnor-

mall LV wall motion pattern, in particular the left ventricular lateral free wall

contractss late in systole while the septum exhibits paradoxical movement. Thus not

onlyy the synchrony between both ventricles is important but also the synchrony in

differentt parts of each ventricle is important. The LV remodelling will lead to higher

volumess and worsening of the MR due to dilatation of the mitral valve annulus. Van

derr Heyden et al. showed in retrospective analyses that especially patients with

highh left ventricular end-diastolic dimensions (64 6 mm in this study) and

moderatee MR {our patient: LVEDD 69 mm and MR grade III) before initiation of RV

pacingg are prone to this complication after AV node ablation.15

Sincee mid 1990s, there is increasing evidence that the synchrony of the ventricles

cann be restored by biventricular pacing.21 Pacemakers capable of simultaneous

stimulationn of both ventricles improve ventricular systolic and diastolic properties

byy a variety of mechanisms including restoration of ventricular contractile synchro-

ny,, resynchronisation of ventricular septal motion, pacing induced decreased

atrioventricularr valve regurgitation and pacing related increases in ventricular

diastolicc fillin g time,

Lauu et al were the first to demonstrate regression of LV remodelling by synchronous

biventricularr pacing in patients with advanced heart failure.22 In this study, there

wass a statistically nonsignificant decrease of LVEDD within 1 week. The fractional

shorteningg of the LV increased within 1 week and was significant after 1 month,

165 5

whichh is comparable with the results of the larger Medtronic Insync study.9 The

Miraclee study showed a significant decrease in LVDD with a significant increase in

thee LV ejection fraction within 6 months.24 The LV ejection fraction was significant

increasedd within 1 month in the study of Lau and within 3 months in the

Medtronicc Insync trial. In the other trials these parameters are not described (see

tablee 1). ' However these patients were all without previous pacing.

Nuöezz et al. recently described a similar case of progressive deterioration after His

ablationn and WI pacing due to severe MR. which was reversed by biventricular

pacing,, but this patient had a mildly hypertrophic well contracting LV Earlier

echocariographicc segmental shortening of the free LV wall resulted in improved

papillaryy muscle function and decrease in MR.29 Also in our case 6 months of

biventricularr pacing resulted in clear reverse remodelling of the deleterious effects

off RV pacing.

Althoughh we cannot exclude continuation of cardiomyopathy in the natural history

off the patient, the time sequence of deterioration clinically and echcardiographical-

lyy after His ablationn makes RV pacing the most likely aetiological factor.

Conclusions s

Thiss case illustrates that the right ventricular apex is not the optimal pacing site in

humanss especially in patients with pre-existent dilated cardiomyopathy and mitral

regurgitation.. The occasionally observed progressive heart failure after RV pacing

followingg His-bundle ablation can be reversed by upgrading to WIR biventricular

pacingg which should be considered for such patients.

Acknowledgement t

Thee authors thank R. }. de Winter MD. PhD, for his skilful support during

implantation. .

16b b

References s

Chapterr 9

1.. Saxon LA. Boehmer JP. Hummel J. et al. Biventricular pacing in patients with congestive heart failure;; two prospective randomised trials. Am J Cardiol I999;S3:120D-123D.

2.. McVeigh ER, Prinzen FW, Wyman BT. Tsitlik JE. Halperin RH, Hunter WC. Imaging synchro-nouss mechanical activation of the paced heart with tagged MRI. MRM 1998:39:507-513.

3-- Prinzen FW, Augustijn CH, Arts T. Delhaas T, Reneman RS. The time sequence of electrical andd mechanical activation during spontaneous beating and ectopic stimulation. Eur Heart J 1992;13:535-543. .

4.. Van Oosterhout MF. Prinzen FW. Arts T. et al. Asynchronous electrical activation induces asymmetricall hypertrophy of the left ventricular wall. Circulation 1998:98:588-595-

5-- Prinzen FW, Van oosterhout MF. Vanagt WY, Storm C, Reneman RS. Optimization of ventricu-larr function by improving the activation sequence during ventricular pacing. PACE 1998:21:2256-2260. .

6.. Prinzen FW. Cheriex EC. Delhaas T, et al. Asymmetric thickness of the left ventricular wall re-sultingg from asynchronous electric activation study in dogs with ventricular pacing and in pa-tientss with left bundle branch block. Am Heart J 1995:130:1045-1053.

7.. Mansourati J. Etienne Y. Gilard M. et al. Left ventricular-based pacing in patients with chronic heartt failure; comparison of acute hemodynamic benefits according to underlying heart dis-ease.. Eur J Heart fail 2000;2:195-199.

8.. Chanda J, Kuribayashi R, Abe T. Ventricular remodelling in dilated cardiomyopathy. Lancet 1997:350:1705. .

9-- Gras D, Mao P. Tang T, et al. Multisite pacing as a supplemental treatment of congestive heart failure:: preliminary results of the Medtronic Inc. Insync study. PACE 1998:21:2249-2255-

10.. Daubert JC. Ritter P. Lebreton H. et al. Permanent left ventricular pacing with transvenous leadss inserted into the coronary veins. PACE 1998:21:9-245-

11.. Brown CS. Mill s RM. Conti JB, Curtis AB. Clinical Improvement after atrioventricular nodal ab-lationn for atrial fibrillation does not correlate with improved ejection fraction. Am J Cardiol. 1997:80:1090-1091--

12.. McComb JM. Gribbin GY. Effect of pacing mode on morbidity and mortality: update of clinical pacingg trials. Am J Cardiol 1999:83:211D-213D.

13-- Leclerq C. Victor F. Alonso C. et al. Comparitive effects of permanent biventricular pacing for refractoryy heart failure in patients with stable sinus rhythm or chronic atrial fibrillation. Am J Cardioll 2000:85:1154-1156.

14.. Twidale N. Manda V. Holliday RT, et al. Mitral regurgitation after atrioventricular node cathe-terr ablation for atrial fibrillation and heart failure; Acute hemodynamic features. Am Heart ] 1999:138:1166-1175--

15.. Vanderheyden M, Goethals M, Anguera I, et al. Hemodynamic deterioration following radiof-requencyy ablation of the atrioventricular conduction system. PACE 1997:20:2422-28.

16.. Mera F, Delurgio DB. Patterson RE. Merlino JD, Wade ME, Leon AR. A comparison of ventricu-larr function duringhigh right ventricular function during high right ventricular septal and apicall pacing after His-bundle ablation for refractory atrial fibrillation. PACE 1999:22:1234-1239. .

17.. Rosado A. Lamas GA. left ventricular remodelling clinical significance and therapy. Basic Res Cardioll 1997:92:66-68.

18.. Moe GW. Armstrong P. Pacing-induced heart failure a model to study the mechanism of dis-easee progression and novel therapy in heart failure. Cardiovasc Res 1999;42:591-599-

19-- Vioral G. Florea. Left ventricular remodelling: the chicken and an egg story of structure and function.. Int J Cardiol 1999:71:207-208.

20.. Cock CC, Meyer A, Kamp O. Visser CA. Hemodynamic benefits of right ventricular out-flowtractt pacing: comparison with right ventricular apex pacing. PACE 1998:21:536-541.

21.. Bakker PF. Meiburg H. de Jonge. Beneficial effects of biventricular pacing in congestive heart failuree (abstract). PACE 1994:17:820.

22.. Lau CP. Cheuj-man Y Chau E, et al. Reversal of left ventricular remodeling by synchronous biventricularr pacing in heart. PACE 2000:23:1722-1725.

167 7

23.. Cazeau S. Leclerq C. Lavergne T, et al. Effects of multisite biventricular pacing in patients withh heart failure and intraventricular conduction delay. N Engl } Med 2001:344:873-880.

24.. Abraham WT. Rationale and design of a randomized clinical trial to assess the safety and effi-cacyy of cardiac resynchronisation therapy in patients with advanced heart failure: the multi-centerr Insync randomized clinical evaluation (MIRACLE). J Card Fail 2000:6:369-380.

25.. Bristow MR. Feldman AM. Saxon LA. Heart failure manamgment using implantable devices forr ventricular resynchronisation: comparison of medical therapy, pacing, and defibrillation inn chronic heart failure (COMPANION) trial. ) Card Fail 2000:6:276-285.

26.. Auricchio A, Stellbrink C. Sack S, et al. The pacing therapies for congestive heart failure (PATH-CHF)study:: rationale, design, and endpoints of a prospective randomized multicenter study.. Am J Cardiol 1999;83:I30D-135D.

27.. Ferrari R, CRT a new dimension in the treatment of heart failure: Landmark trials. Up- to-date onn cardiac resynchronization therapy (editorial), HF- Contdxkt 2001:1:8015.

28.. Up- to-date on cardiac resynchronization therapy. HF- Cpnt(Vkt 2001:1:8015-8022. 29.. Nunez A, Alberta MT. Cosio FG et al. Severe mitral regurgitation with right ventricular pacing

succesfullyy treated with left ventricular pacing. Pacing Clin Electrophysiol 2002:25:226-30.

168 8

Summaryy and conclusions

Chapterr 10

Thee purpose of this thesis was to evaluate the effect of manual adjustment of rate

adaptivee pacemakers on exercise physiology and quality of lif e and to study some

neww aspects of cardiac resynchronization therapy. The first part of the thesis

(Chapterr 2 to 6) concerned studies in pacemaker patients on the effect of individual

optimizationn of pacemaker sensors on exercise physiology and quality of life. The

secondd part of the thesis (Chapter 7 to 9) described the effect of cardiac resynchro-

nizationn therapy on cardiac function, the neurohumoral system and microcircula-

tionn in patients with heart failure.

Partt I Studies on pacemaker sensors

Automaticc functions are increasing in rate response pacemakers. Whether we can

relyy on these automatic functions in daily practice is limited described. We there-

foree reviewed in chapter 2 the various types of sensors used in current pacemakers

andd tried to answer the question whether manual rate response optimization

improvess patient outcome and is still necessary given the existing automaticity in

ratee response pacemakers.

Inn chapter 3 heart rate curves of healthy individuals (HI) from different age catego-

riess are described during the chronotropic assessment exercise protocol (CAEP) and

6-- minute hall walk test (6-HWT). Heart rate (HR) at rest. HR at 1 minute of exercise,

timee to peak HR. maximal achieved HR. HR at 1,3 and 10 minutes recovery period,

exercisee duration, and METS or achieved distance (meters) were measured. The

achievedd HR at one minute of exercise was significantly higher and the time to peak

HRR significantly shorter during 6-HWT compared to CAEP, although the achieved

maximall HR was comparable. There were no gender differences in HI randomized to

6-HWTT and minimal gender differences in HI randomized to CAEP. The predicted

maximall HR according to the Astrand formula (220-age) was not significantly

differentt compared to the achieved maximal HR in both tests. Thus, the Astrand

formulaa (220-age) can still be used for prediction of the maximal HR. The HR rate

profiless described in this chapter can be used to further optimize of the pacemaker

sensors.. The 6-HWT is preferable for pacemaker sensor optimization.

Chapterr 4 described the influence of individual optimization of sensors on quality

off life (QOU and exercise tolerance in a randomized, single blind study in patients

withh WIR. DDDR or AAIR pacemakers. Patients with > 75% pacing were randomized

173 3

too optimized sensor settings (OSS) or default sensor setting (DSS). Standardized

optimizationn was performed using three different exercise tests. QOL question-

nairess (QOL-q: Hacettepe. Karolinska and RAND-36) were used for evaluation of the

sensorr optimization. One month before and after optimization, exercise capacity

usingg CAEP and the three OOL-q were assessed. We showed that one month after

sensorr optimization the achieved maximal HR and METS were significantly higher

inn OSS compared to DSS. Highest HR and METS were achieved in patients with

pacemakerss with accessible sensor algorithms. In patients with automatic slope

settings,, exercise capacity did not improve after sensor optimization. Surprisingly.

QOLL did not improve in OSS compared to DSS.

Inn chapter 5 we investigated whether variation in HR during exercise affects the flow

velocityy in the middle cerebral artery and cardiac output. We therefore evaluated in

patientss with complete heart block and rate responsive pacemakers, the effect of DSS

andd OSS on blood pressure, stroke volume and mean flow velocity in the middle

cerebrall artery during graded ergometry cycling. For both OSS and DSS there was no

significantt increase in flow velocity in the middle cerebral artery during exercise.

Strokee volume and cardiac output increased minimal with OSS compared to DSS.

Wee directly compared the chronotropic function of the peak endocardial accelera-

tionn (PEA) sensor to the activity sensor in chapter 6. Patients with > 75% pacemaker

drivenn HR and a PEA sensor and HC underwent a CAEP exercise test. The pacemak-

erss were programmed in the default setting and WIR mode, with adjustment of the

upperr sensor rate as an age related maximum value (220-age). The activity sensor

wass externally strapped on the thorax. We showed that the PEA sensor functions

hypochonotropicc during exercise programmed as a single sensor system. Although

bothh groups had normal left ventricular functions, the exercise capacity of pacemak

err patients was significantly lower than in HC. It is therefore preferable to combine

thee PEA sensor with an activity-based sensor in a dual sensor system.

174 4

Chapterr 10

Partt II Studies on biventricular pacing

Too examine whether cardiac resynchronization therapy (CRT) induces favourable

changess in the neurohumoral system, we measured in chapter 7 in patients with

heartt failure, myocardial 123I-metaiodobenzylguanidine (123I-MIBG) uptake indices and

brainn natriuretic peptide (BNP) before and after 6 months of CRT. Furthermore, NYHA

classificationn and echocardiographic indices were assessed. Six months of CRT

resultedd in 1) significant improvement in NYHA classification, 2) reduction in QRS

width,, LV end- diastolic diameter, LV end systolic diameter, septal to lateral delay and

mitrall regurgitation, 3) improvement of delayed !23I-MIBG heart/mediastinal ratios

andd decrease of 123I-MIBG washout and A) decreased BNP levels. From these data we

concludedd that CRT induces favorable changes in the neurohumoral system.

Althoughh it is known that CRT in heart failure patients improves systemic circula-

tion,, its acute effects on microcirculation are as yet unknown and improvement of

thee macrocirculation does not necessarily result in improvement of the microcircu-

lation.. Chapter 8 describes the sub-lingual microcirculatory changes in heart failure

patientss due to CRT and right ventricular pacing by use of orthogonal polarization

spectrall (OPS) imaging. Six months of CRT resulted in a reduction in NTHA class and

echocardiographicc reverse remodeling. Acute microcirculatory changes were as-

sessedd by functional capillary density (FCD) and capillary velocity (CV) after previ-

ouss six months of CRT. FCD and CV were measured sublingual after pacing 15

minutess in one of three pacing modalities (no pacing, RV only pacing, and biven-

tricularr pacing with a sensed AV interval of 100-120 msec). FCD was significantly

higherr in healthy controls compared to heart failure patients with right ventricular

pacingg and no pacing. CRT significantly increased FCD in heart failure patients

comparedd to right ventricular pacing and no pacing. CV was normal in all patients

withh or without pacing.

Inn chapter 9 we described in one of the first case reports, reversed remodeling of

thee left ventricle and left atrium after upgrading to biventricular pacing in a patient

withh worsening of mitral regurgitation and heart failure due to WIR pacing after His

bundlee ablation and reviewed the literature concerning reversed remodeling after

upgradingg to biventricular pacing.

175 5

Interpretation ss and conclusions

Thee studies concerning individual optimization of pacemaker sensors on exercise

physiologyy and quality of lif e that are described in this thesis lead to the following

conclusions: :

1.. The majority of pacemaker sensors remain at the original programmed settings of

thee manufacturer, although there is evidence that individually adjustment of

pacemakerr sensors improves exercise capacity and quality of life.

2.. Automatic features can be helpful in reducing the time needed to perform a

follow-upp of pacemakers, however individually adjustment of pacing sensors is

stilll necessary.

3.. The development of a sensor system that can simulate ideal sinus rhythm behav-

iorr remains a challenge for scientists and manufacturers.

4.. International guidelines are needed to standardize pacemaker sensor optimiza-

tionn in all chronotropic incompetent patients,

5.. Individual optimization of rate response pacemakers improves exercise capacity

andd increases maximum HR. although OOL remained unchanged.

6.. Accessible pacemaker sensor algorithms are mandatory for individual optimization.

7.. Pacemaker sensor optimization increased stroke volume and cardiac output

minimallyy during exercise and had no measurable effect on flow velocity in the

middlee cerebral artery.

8.. The PEA sensor functions hypochonotropic during exercise programmed as a

singlee sensor system.

Inn conclusion, individual optimization of pacemaker sensors is necessary. Therefore

accessiblee pacemaker sensor algorithms and clear international guidelines are of

utmostt importance. Combination of a physiologic and an activity based sensor in dual

sensorr systems mimics the sinus node behavior better than a single sensor system.

Thee conclusions of the second part of the thesis concerning the effects of cardiac

resynchronizationn therapy on cardiac function, neurohumoral system and microcir-

culationn in patients with heart failure are:

1.. CRT induces favorable changes in the neurohumoral system in patients with

176 6

Chapterr 10

heartt failure (BNP levels decreased, delayed 123I-MIB G heart/mediastinal ratios

improvedd and 123I-MIB G washout decreased), parallel to significant functional

improvementt and echocardiographic reverse remodeling.

2.. In CRT responders. CRT improves sub-lingual microcirculation in heart failure

patientss as assessed by OPS imaging.

3-- The occasionally observed progressive mitral regurgitation and heart failure after

rightt ventricular pacing following His-bundle ablation can be reversed by upgrad-

ingg to WIR biventricular pacing and in these circumstances upgrading should be

considered. .

Inn conclusion, cardiac resynchronization therapy improves the macrocirculation. the

microcirculationn and induces favorable changes in the neurohumoral system in

patientss with heart failure, Reversed remodeling of the left ventricle can be

achievedd with long- term CRT.

Recommendationss and futur e directions

Parti i

Ourr results might have a number of implications for the clinician. Today, most pacemak-

erss are programmed in the default setting of the manufacturer. We recommend individ-

uall optimization in all chronotropic incompetent patients. After the implantation of a

pacemaker,, carefully follow- up of the patients is needed. The function of the pacemak-

err sensors has to be evaluated and optimized using an exercise test. We feel that in-

hospitall exercise tests are artificial, especially the treadmill and bicycle tests. We

recommendd therefore the use of e.g. a hall walk test with stair climbing, which resem-

bless closer daily lif e activities than other in- hospital tests. Ideally, the sensor function

shouldd be measured directly during daily living activities with for example holter

registration.. In addition, investigating the effects of programming sequence (e.g. first

thresholdd adjustment followed by slope adjustment) of rate response parameters is very

important,, although time consuming and strenuous for the patients. Development of

toolss for home monitoring of the separate rate response parameters can be helpful to

gainn more insight in this issue.

177 7

Thee achieved rate profile during treadmill exercise testing resulting from dual sensor

pacingg (with complementary properties) is improved over single sensor pacing

becausee sensor combination provides improvement in speed, proportionality, sensi-

tivit yy and specificity. As consequence, we recommend to use dual sensor systems with

aa combination of an accelerometer and a physiologic sensor (minute ventilation, QT,

peakk endocardial acceleration) in chronotropic incompetent patients.

Inn our study described in chapter four. OOL-q remained unchanged after one month

off individual optimization of rate response pacemakers. An explanation for the

failuree to improve OOL after sensor optimization could be the relatively good

baselinee functional capacity, because patients with relative preserved functional

capacityy at enrollment show the lowest improvement in health related issues.

Anotherr reason could be the partly inadequate questioning of specific pacemaker

patientt related symptoms with the used OOL-q in this study. Therefore, we advise

too use the Aquarel OOL-O. which is a recently developed pacemaker patient specific

OOL-qq and we feel that future research to improve QOL-q related to pacemaker

patientss has still several challenges. Insight in the interpretation of the OOL-q

scoress in relation with classical used clinical parameters such as NYHA classification

iss important.

Partt II

Furthermore,, the automaticity in rate response pacemakers is increasing tremen-

dously.. Although, automaticity can reduce pacemaker follow- up time, accessible

pacemakerss remain needed for individually adjustment of the rate response param-

eters.. Whether automatic adjustment is better than manual adjustment should be

investigatedd in randomized trials.

Thee results of our studies concerning the effects of CRT on neurohumoral system

andd microcirculatory changes form a step forward in the unraveling of the working

mechanismss of CRT in heart failure patients. To get more insight in this issue we

suggestt to replicate these studies in both responders and non-responders to CRT. It

wouldd be of interest to know whether microrocirculatory changes at implant could

predictt beneficial response to CRT and whether CRT induces beneficial long-term

microcirculatoryy changes. In addition, nowadays patients are selected for CRT

mainlyy on electrocardiographic criteria. The standard EKG is less reliable in the

178 8

Chapterr 10

characterizationn of the extent of dyssynchrony. since even patients with normal QRS

durationn on EKG can have marked dyssynchrony, while 20-30% of patients with wide

QRSS complexes do not respond to CRT. We therefore recommend to select patients

usingg advanced tissue-doppler imaging techniques, because this technique is likely

too improve the response rate to CRT.

Furthermore,, the deleterious effects of right ventricular pacing (RV) are now well

documented.. Considering the large magnitude of the deleterious effects of RV

pacingg we strongly suggest' to go away of the RV' or to minimize RV pacing using

algorithmss such as search hysteresis and mode switching. We think that left ven-

tricularr pacing using the coronary sinus is a more physiologic pacing site also for

conservativee pacing indications (sick sinus syndrome, AV-block) as for the modern

pacingg indications (resynchronization therapy) in patients with already reduced left

ventricularr function and/or mitral valve insufficency. More studies are needed to

identifyy patients who are at high risk for deterioration with RV pacing. Improving

thee technique for pacemaker lead implantation in the coronary sinus (to reduce lead

implantt time) wil l widen the indication for left ventricular pacing.

Finally,, extrapolating from recent developments, J Warren Harthorne (professor in

medicinee Massachusetts General Hospital, Boston, USA, writer of the book 'the future

off cardiac pacing), has suggested, that the future wil l give us "implantable computers

thatt wil l serve as an electronic service center" able to communicate with various

organn systems " to rouse flagging performance of heart, cerebral, respiratory,

gastrointestinal,, genitourinary, and musculoskeletal function" Electrostimulation has

evolvedd so far since its beginning that this description does not seem entirely outside

thee realm of the possible.

179 9

Samenvattingg en conclusies

182 2

Chapcerr I ]

Hett doel van dit proefschrift was om de effecten van manuele optimalisatie van

frequentie-adapterendee pacemakers op inspanningsfysiologie en kwaliteit van leven

tee onderzoeken en bepaalde nieuwe aspecten van cardiale resynchronisatietherapie

tee bestuderen. Het eerste deel van dit proefschrift (hoofdstuk 2 tot en met 6)

behandeltt studies in pacemakerpatiënten naar het effect van individuele optimali-

satiee van pacemakersensoren op inspanningsfysiologie en kwaliteit van leven. Het

tweedee gedeelte van dit proefschrift (hoofdstuk 7 tot en met 9) beschrijft de effect-

enn van de cardiale resynchronisatietherapie op de hartfunctie. het neurohumorale

systeemm en de microcirculatie in patiënten met hartfalen,

Deell I Studies met pacemakersensoren

Automatischee functies nemen toe in frequentie-adapterende pacemakers. Of we in

dee dagelijkse praktijk kunnen rekenen op deze automatische functies is beperkt

beschreven.. In hoofdstuk 2 worden de verschillende typen sensoren die gebruikt

wordenn in huidige pacemakers beschreven en werd geprobeerd om de vraag te

beantwoordenn of manuele optimalisatie van frequentie-adapterende pacemakers

resulteertt in klinische verbetering van patiënten en ook wordt de vraag beantwoord

off het nog steeds noodzakelijk is. gezien de bestaande automaticitieit in frequentie

adaptievee pacemakers.

Hoofdstukk 3 beschrijft hartfrequentieprofielen van gezonde proefpersonen in

verschillendee leeftijdscategorieën gedurende de treadmill test. het zogenaamde

chronotropicc assessment exercise protocol (CAEP) en de 6-minute Hall walk test (6-

HWT.. de 6 minuten looptest). Hart frequentie (HF) gedurende rust. HF na de eerste

minuutt van de inspanning, tijd tot maximale HF. maximaal behaalde HF. HF na 1. 3

enn 10 minuten van de herstelfase van de inspanning, inspanningsduur. inspanning-

scapaciteitt (metabolic equivalent, MET) en behaalde afstand (meters) zijn gemeten.

Dee bereikte HF na 1 minuut van de inspanning was significant hoger en de tijd tot

maximalee HF was significant korter tijdens de 6-HWT vergeleken met de CAEP,

alhoewell de bereikte maximale HF vergelijkbaar was in beide inspanningstesten.

Err waren geen geslachtgebonden verschillen tussen gezonde proefpersonen die

gerandomiseerdd waren naar de 6-HWT en minimale geslachtsgebonden verschillen

tussenn gezonde proefpersonen gerandomiseerd naar de CAEP. De voorspelde

183 3

maximalee HF volgens de bekende Astrand formule (220-leeftijd) was niet significant

verschillendd vergeleken met de behaalde maximale HF in beide inspanningstesten.

Hieruitt kunnen we concluderen dat de Astrand formule (220-leeftijd) nog steeds

gebruiktt kan worden om de maximale HF te voorspellen. De hartslagprofielen

beschrevenn in dit hoofdstuk kunnen als leidraad gebruikt worden voor de optimali-

satiee van de pacemakersensoren. De 6-HWT heeft de voorkeur om als leidraad

gebruiktt te worden voor de pacemaker sensor optimalisatie.

Hoofdstukk 4 beschrijft de invloed van individuele optimalisatie van sensoren op de

kwaliteitt van leven en inspanningstolerantie in een gerandomiseerde, single-

blindedd studie in patiënten waarvan de pacemakers geprogrammeerd zijn in WIR,

DDIRR of AAIR mode. Patiënten met > 75% pacing zijn gerandomiseerd naar optimale

sensorr setting (OSS) of default sensor setting (DSS). Gestandaardiseerde optimalisatie

iss uitgevoerd met behulp van drie verschillende inspanningstesten. Drie verschillende

kwaliteitt van leven vragenlijsten (Hacettepe. Karolinska, RAND-36) zijn gebruikt voor

dee evaluatie van de sensoroptimalisatie. Een maand voor-en-na optimalisatie is de

inspanningscapaciteitt tijdens CAEP bepaald en zijn de drie kwaliteit van leven vragen

afgenomen.. Aangetoond werd dat 1 maand na sensoroptimalisatie de behaalde

maximalee HF en METS significant hoger waren in de OSS vergeleken met de DSS,

Hoogstee HF en METS waren behaald in patiënten met pacemakers met toegankelijke

herprogrammeerbaree sensor algoritmen voor manuele afstelling- Bij patiënten

waarvann de pacemakersensor een automatische slope setting hadden, was de inspan-

ningscapaciteitt niet verbeterd na sensoroptimalisatie. Opvallend was dat sensoropti-

malisatiee de kwaliteit van leven niet verbeterde.

Inn hoofdstuk 5 werd onderzocht of de optimalisatie van de HF tijdens inspanning

dee snelheid van bloeddoorstroming in de arteria cerebri media en de cardiac output

beïnvloedt.. Om deze vraag te beantwoorden werd in patiënten met derde graads AV-

blockk en frequentie-adapterende pacemakers het effect van DSS en OSS op bloed-

druk,, slagvolume, en gemiddelde bloedstroomsnelheid in de arteria cerebri media

tijdenss fietsergometrie bepaald. De bloedstroomsnelheid in de arteria cerebri media

tijdenss inspanning nam niet toe.

Slagvolumee en cardiac output namen minimaal toe in OSS vergeleken met DSS.

Inn hoofdstuk 6 werd de chronotrope functie van de piek endocardiale acceleratie

tPEA)) sensor direct vergeleken met de activiteitssensor. Patiënten die > 75% pace-

makerr gestuurde HF hadden met een PEA sensor en gezonde proefpersonen hebben

184 4

Chapterr 11

dee CAEP inspanningstest ondergaan. De pacemakers waren geprogrammeerd in de

defaultt setting en de WIR mode met uitzondering van de maximale HF (upper

sensorr rate). De maximale HF werd aan de hand van de leeftijd (220-leeftijd) gepro-

grammeerd.. De activiteitssensor werd uitwendig op de thorax geplakt. Aangetoond

werdd dat de PEA sensor geprogrammeerd als single sensor systeem hypochonotroop

functioneertt gedurende inspanning. Ondanks dat de patiënten een normale linker

ventrikell functie hadden, was de inspanningscapaciteit van pacemakerpatiënten

significantt lager dan die van de gezonde proefpersonen. Het is daarom aan te

bevelenn om de PEA sensor te combineren met een activiteit gebaseerde sensor in

eenn zogenaamde dual sensor systeem.

Deell II Studies met biventriculai r pacing

Omm te onderzoeken of cardiale resynchronisatietherapie (CRT) gunstige neurohu-

morall en structurele veranderingen induceert, werd in hoofdstuk 7 bij patiënten

mett hartfalen voor-en-na 6 maanden CRT, de myocardiale 123I-metaiodobenzylguani-

dinee (123I-M1BG) opname het brain natriuretic peptide (BNP), de New York Heart

Associationn (NYHA) klasse en echocardiografische parameters bepaald. Zes maanden

CRTT resulteerde in 1) significante verbetering van de NYHA klasse, 2) reductie van

QRSS duur, linker ventrikel einddiastolische diameter, linker ventrikel eindsysto-

lischee diameter; vermindering van het septale tot laterale linkerventrikel activatie

verschill en vermindering van mitralisklepinsufficiëntie, 3) verbetering van de

delayedd 123I-MIBG hart/mediastinale ratio en afname van 123I-MIBG washout en 4)

afnamee van de BNP spiegel. Vanuit deze data werd geconcludeerd dat CRT gunstige

veranderingenn induceert in het neurohumorale systeem.

Alhoewell het bekend is dat CRT in hartfalen patiënten systemische circulatie

verbetert,, is het directe effect op microcirculatie nog niet bekend. Verder betekent

eenn verbetering van macrocirculatie niet noodzakelijkerwijs eveneens verbetering

vann microcirculatie. Hoofdstuk 8 beschrijft de sub-linguale microcirculatoire

veranderingenn in hartfalen patiënten door CRT en rechter ventrikel pacing met

orthogonall polarization spectral (OPS) imaging. Zes maanden CRT resulteerde in

reductiee van NYHA klasse en echocardiografische reverse remodeling. Na zes

maandenn CRT werden acute microcirculatoire veranderingen onderzocht middels

185 5

hett meten van de functionele capillaire densiteit (FCD) en capillaire stroomsnelheid

(capillaryy velocity, CV). Na pacing in een van de drie pacing modaliteiten ( niet

pacen.. rechter ventrikel pacing, en biventriculair pacing met een gesensde AV

intervall van 100-120 msec). De FCD was significant hoger in gezonde proefpersonen

vergelekenn met hartfalen patiënten met rechter ventrikel pacing en hartfalen

patiëntenn met eigen ritme. CRT verhoogde significant de FCD in hartfalen patiënten

vergelekenn met rechterventrikel pacing en het eigen hartritme. De CV was normaal

inn alle patiënten ongeacht wel of geen pacing. Hoofdstuk 9 beschrijft een van de

eerstee case reports waarbij reversed remodeling aangetoond wordt van de linker

ventrikell en linker atrium na upgrading tot biventriculair pacing in een patiënt met

verslechteringg van mitralisklepinsufficientie en hartfalen veroorzaakt door WIR

pacingg en His bundel ablatie. Tevens wordt in dit hoofdstuk de literatuur betref-

fendee reversed remodeling na upgrading tot biventriculair pacing beschreven.

Interpretatie ss en conclusies

Dee studies welke in dit proefschrift zijn beschreven betreffende het effect van

individuelee optimalisatie van pacemaker sensoren op inspanningsfysiologie en

kwaliteitt van leven, leiden tot de volgende conclusies:

1.. De meeste pacemaker sensoren blijven geprogrammeerd in de nominale setting

vann de fabrikant, ondanks dat er bewijs is dat individuele afstelling van pacemak-

ersensorenn de inspanningscapaciteit verbetert.

2.. Automatische pacemaker functies kunnen de tijd die nodig is bij de poliklinische

follow-upp van pacemakerpatiënten reduceren, desalniettemin is individuele

afstellingg van pacemakersensoren nog steeds noodzakelijk.

3.. De ontwikkeling van sensorsystemen die de ideale sinusknoop functie kunnen

simulerenn blijf t een uitdaging voor wetenschappers en fabrikanten.

4.. Internationale richtlijnen die aangeven hoe individuele optimalisatie van pace-

makersensorenn uitgevoerd dient worden zijn noodzakelijk voor het standaardis-

erenn van pacemaker sensoroptimalisatie in chronotroop incompetente patiënten.

5-- Individuele optimalisatie van frequentie adapterende pacemakers verbetert de

inspanningscapaciteitt en verhoogt de maximale HF. alhoewel de kwaliteit van

levenn onveranderd blijft .

186 6

Chapterr 1 1

6.. Toegankelijke pacemaker sensor algoritmen voor manuele afstelling zijn belangri-

jkk voor individuele optimalisatie,

7.. Pacemaker sensor optimalisatie verhoogt minimaal het slagvolume en de cardiac

ouputt gedurende de inspanning en heeft geen meetbaar effect op de bloeddoor-

stromingssnelheidd in de arteria cerebri media.

8.. De PEA sensor functioneert indien geprogrammeerd als single sensor systeem

hypochonotroopp gedurende de inspanning.

Concluderend,, individuele optimalisatie van pacemaker sensoren blijf t vooralsnog

noodzakelijk,, waarbij, toegankelijke pacemaker sensor algoritmen voor manuele

afstellingg en duidelijke eenduidige richtlijnen voor de dagelijkse praktijk van groot

belangg zijn. Combinatie van een fysiologische en een activiteit gerelateerde sensor

inn een 'dual sensor systeem' benadert de sinusknoopfunctie beter dan een 'single

sensorr systeem.'

Dee conclusies van het tweede deel van dit proefschrift betreffende de effecten van

cardialee resynchronisatie therapie op de hartfunctie. het neurohumorale systeem en

dee microcirculatie in patiënten met hartfalen zijn als volgt:

1.. CRT induceert gunstige veranderingen in het neurohumorale systeem in patiënt-

enn met hartfalen (daling van BNP. de delayed 123I-MIBG hart/mediastinale ratio

warenn verbeterd en de 123I-MIBG washout nam af), parallel aan functionele

verbeteringg en echocardiografische reverse remodeling.

2.. In CRT responders verbetert CRT bij patiënten met hartfalen de sub-linguale

microcirculatiee (gemeten met OPS imaging).

3.. Het incidenteel voorkomen van progressieve mitralisklepinsufficiëntie en

hartfalenn bij patiënten met rechter ventrikel pacing en His-bundel ablatie kan

doorr upgrading tot een WIR biventriculair pacing systeem aanzienlijk geredu-

ceerdd worden. Indien een dergelijke casus zich voordoet dient upgrading overwo-

genn te worden.

Concluderend,, cardiale resynchronisatie therapie verbetert de macrocirculatie de

microcirculatiee en induceert gunstige veranderingen in het neurohumorale systeem

bijj hartfalen patiënten. Reversed remodelering van de linker ventrikel kan bereikt

wordenn met lange termijn CRT

187 7

Aanbevelingenn en toekomstperspectieven

Deell l

Onzee resultaten kunnen een aantal implicaties hebben voor de clinicus in de

dagelijksee praktijk. Tegenwoordig staan nog een groot aantal pacemakers in de

nominalee setting van de fabrikant. We bevelen bij alle patiënten met chronotrope

incompetentiee individuele optimalisatie van de sensor aan. Na de pacemaker

implantatiee is zorgvuldige poliklinische follow-up van de patiënten noodzakelijk

ookk ten aanzien van de pacemakersensor. Het functioneren van de pacemaker

sensorenn dient geëvalueerd en geoptimaliseerd te worden aan de hand van een

inspanningstest.. We vinden dat de huidige inspanningstesten (met name de testen

opp de loopband en de fietst) een artificieel karakter hebben. We bevelen het gebruik

vann de zogenaamde hall walk test aan. liefst gecombineerd met traplopen, daar de

halll walk test wat betreft het type inspanning het meeste lijk t op de dagelijkse

activiteitenn thuis.

Idealiterr zou de sensor functie direct gemeten moeten worden in de thuissituatie

doorr bijvoorbeeld het gebruik van holterregistratie.

Daarbijj is het onderzoeken naar het effect van de volgorde van de afstelling van

frequentie-adapterendee parameters (de threshold, de slope) erg belangrijk, ondanks

datt het tijdrovend is en vermoeiend kan zijn voor de patiënten. Ontwikkeling van

techniekenn om de frequentie-adapterende parameters separaat thuis te monitoren

kunnenn behulpzaam zijn bij het verkrijgen van meer inzicht daarin.

Dee bereikte HF profielen gedurende inspanning op de loopband van dual sensor

systemenn (met complementaire eigenschappen) is beter in vergelijking met die van

singlee sensor systemen, doordat het combineren van sensoren verbetering geeft in

snelheid,, proportionaliteit, sensitiviteit en specificiteit. Daarom raden we aan om

inn chronotroop incompetente patiënten dual sensor systemen te gebruiken.

Dee automaticiteit in frequentie-adapterende pacemakers neemt enorm toe. On-

dankss dat automatisch gestuurde functies in een pacemaker voor een tijdreductie

kann zorgen tijdens poliklinische controles van de patiënten, zijn de pacemakers die

toegankelijkk zijn voor manuele afstelling van frequentie adapterende parameters erg

belangrijk.. Of automatische individuele afstelling van frequentie adapterende

parameterss beter is dan manuele afstelling zou in de toekomst onderzocht moeten

wordenn met gerandomiseerde trials.

188 8

Chapterr 11

Inn de studie beschreven in hoofdstuk vier, bleef de kwaliteit van leven Quality of

life.. QOL) onveranderd 1 maand na individuele optimalisatie van frequentie adapt-

erendee pacemakers. Een mogelijke verklaring voor het niet verbeteren van QOL na

sensorr optimalisatie zou de reeds goede functionele capaciteit (verkregen na de

pacemakerr implantatie) ats uitgangswaarde kunnen zijn. gezien de patiënten met de

relatieff goede functionele capaciteit bij de inclusie nauwelijks verbetering laten zien

inn gezondheidsgerelateerde uitkomsten. Een andere mogelijke verklaring zou de

inadequatee ondervraging van pacemaker patiënt specifieke symptomen kunnen zijn

mett de gebruikte QOL vragen in deze studie. Daarom adviseren we het gebruik van

dee Aquarel QOL vragenlijst, welke recent specifiek is ontwikkeld voor de pacemaker

patiënt.. Desalniettemin denken we dat het nog steeds uitdagend is om onderzoek

tee doen naar mogelijkheden ter verbetering van pacemaker patiënt gerelateerde

QOLL vragen. Verkrijgen van inzicht in de interpretatie van de QOL scores in relatie

mett de inmiddels gangbare parameters zoals de NYHA klasse is erg belangrijk.

Deell II

Dee studies betreffende de effecten van CRT op het neurohumorale systeem en de

microcirculatoiree veranderingen vormen een voorwaartse stap bij de ontrafeling van

dee gunstige werkingsmechanismen van CRT in patiënten met hartfalen. Om meer

inzichtt te verkrijgen in deze materie stellen we voor om in de toekomst studies naar

dee effecten van CRT op het neurohumorale systeem en de microcirculatoire verand-

eringenn te herhalen in zowel responders als non-responders van CRT. Het is belangr-

ijkk om te achterhalen of de eventuele microcirculatoire veranderingen door de CRT

bijj de implantatie een voorspellende waarde hebben voor de selectie van patiënten

welkee baat zullen hebben van CRT en of CRT ook op lange termijn gunstige microcir-

culatioree veranderingen zou kunnen induceren. Daarbij is het zo dat tegenwoordig

patiëntenn grotendeels geselecteerd worden op grond van echocardiografische

criteria.. Het standaard ECG is minder betrouwbaar in het bepalen van de uitgebreid-

heidd van de dissynchronie van het hart, daar patiënten met normale QRS duur op

hett ECG aanzienlijke echocardiografische dissynchronie kunnen hebben, terwijl 20-

30%% van de patiënten met verbrede QRS complexen niet verbeteren na CRT. Om

dezee redenen raden we aan om bij de patiëntenselectie gebruik te maken van

189 9

geavanceerdee tissue-doppler echocardiografie, aangezien daarmee de kans op De

nadeligee effecten van rechter ventrikel (RV) pacing zijn uitgebreid beschreven in de

literatuur.. Beschouwende de nadelige effecten van RV pacing willen we ten sterkste

aanbevelenn 'om de RV te verlaten' of algoritmen te gebruiken ter bevordering van

minimalee RV pacing zoals search hysteresis and mode switching van AAI naar DDD.

Wee denken dat linker ventrikel pacing via de sinus coronarius een fysiologisch

gezienn betere pacing site is zowel voor de conservatieve pacing indicaties (sick sinus

syndrome,, AV-Block) dan voor de moderne pacing indicaties (resynchronisatie

therapie)) in patiënten met afgenomen linker ventrikel functie en/of mitralisklepin-

sufficientie.. Meer onderzoek is nodig naar de identificatie van patiënten met een

hoogg risico voor deterioratie door RV pacing. Verbeteringen in de technieken voor

pacemakerr lead implantatie via de sinus coronarius zullen de indicaties voor linker

ventrikell pacing verruimen.

Extrapolerendd vanuit recente ontwikkelingen heeft J. Warren Harthoren (professor

inn de geneeskunde, Massachusetts General Hospital, Boston,USA, schrijver van het

boekk 'the future of cardiac pacing') gesuggereerd dat de toekomst ons 'implanteer-

baree computers zal geven die kunnen dienen als elektronische service centers' die

inn staat zijn om met verschillende organen te communiceren, die in staat zullen

zijnn om het cardiale, cerebrale, respiratoire, gastro-intestinale, uro-genitale en

muskulo-skeletalee functioneren weer te geven, Elektrostimulatie is sinds zijn

beginjarenn dusdanig ontwikkeld dat deze beschrijving niet onmogelijk lijkt .

190 0

Dankwoord d

194 4

Dankwoord d

Allenn die een bijdrage hebben geleverd bij de totstandkoming van dit proefschrift

wi ll ik hierbij bedanken.

Allereerstt wil ik alle patiënten en gezonde vrijwilligers, die meegedaan hebben aan

dee verschillende studies bedanken voor hun medewerking, geduld en vertrouwen in

klinischh wetenschappelijk onderzoek.

Dee volgende personen wil ik in het bijzonder bedanken.

Dr.Dr. R. Tukkie, Beste Raymond, mijn co-promotor. Je enthousiasme over de verschil-

lendee onderzoeksmogelijkheden en de daarop aansluitende rondleiding over de

afdelingg cardiologie en het halve AMC maakte bij mij tijdens onze eerste kennis-

makingg een dusdanig positieve indruk dat ik terplekke meteen besloot om te

beginnenn aan dit promotietraject. Ik heb je in de afgelopen 4 jaar leren kennen als

eenn erg betrokken, stabiele, en gemotiveerde begeleider, Je wist mij altijd goed te

stimulerenn door mij meteen, van het begin af aan, overal bij te betrekken, Ik mocht

meteenn mee naar de NASPE en naar allerlei cursussen waardoor ik in de gelegen-

heidd kwam om de 'pacing wereld' snel te leren kennen. Je bijzondere manier van

evaluerenn van de lopende onderzoeken, kritisch maar altijd met een positieve

benaderingg en je heldere open visie hebben mij altijd gemotiveerd. Daarnaast gaf je

ookk ruimte voor het bediscussiëren van nieuwe ideeën.

Bestee Raymond ook wil ik je bedanken voor je menselijk inlevingsvermogen, het

mooistee kraam-cadeau kreeg ik van jou. Je stelde zelf voor om 1 dag in de week

thuiss te werken waardoor ik mijn zoontje vaker zag. Je betrokkenheid bij je eigen

familiee bleek uit je leuke verhalen over je vrouw Lena en je zoontjes Marijn en

Wietse.. Daarnaast waren er natuurlijk de gezellige kerstborrels met de ep-leden bij

jouu thuis met echte cadeautjes onder de kerstboom voor iedereen. Ik ben je zeer

dankbaarr voor de manier waarop je mij al die jaren fantastisch hebt begeleid. Ik

hoopp je in de toekomst nog vaak te zien en op de hoogte te blijven van zowel je

werkzaamhedenn als de ontwikkelingen binnen je gezin.

Prof.Prof. Dr. AAM. Wilde, mijn promotor. Beste Arthur, door je heldere kritische blik

hebb je een grote bijdrage geleverd aan de voltooiing van de manuscripten. Ook als je

hett erg druk had werden de manuscripten in korte tijd nagekeken (binnen 1 dag

terug)) waardoor het mogelijk was om de manuscripten tijdig op te sturen. Ik heb

onzee besprekingen, die kort maar krachtig waren, altijd als erg stimulerend en

195 5

prettigg ervaren. Ik wil je bedanken voor het in mij gestelde vertrouwen.

Dee overig leden van de promotiecommissie, prof. dr. Ir.C. Ince. prof. dr. J. H Raves-

loot.loot. prof. dr .ir. J.M. T. de Bakker, prof. dr. M.J. Schalij, prof dr. N.M. van Hemel en

dr.dr. N. Sulke, dank ik voor de bereidheid plaats te willen nemen in de commissie ter

beoordelingg van mijn proefschrift.

Prof.Prof. dr. N. Sulke. I'm grateful for your willingness to participate in the evaluation

committeee of my thesis and to attend its defense here in Amsterdam.

Prof.Prof. dr. K.I. Lie. Beste professor Lie. ik ben u zeer erkentelijk dat u samen met Dr,

A.A. V. la Rivière mij de mogelijkheid heeft geboden om mijn opleiding tot cardioloog

inn het AMC te volgen. Dr. R.B.A. van den Brink dank ik voor het getoonde begrip in

mijnn gezinssituatie, ondersteuning en toestemming om te mogen beginnen met

mijnn vooropleiding dichtbij huis in Den Bosch.

ProfProf dr. J.G.P. Tijssen. Beste Jan, bedankt voor je hulp bij de randomisatie van de

STAPP en de POCASS studie.

ProfProf dr. J.J. Piek. Beste Jan, elk jaar was er minstens een keer sprake van een even-

tuelee verhuizing naar de cubicals van onderzoekers die geen onderzoek deden wat

directt verbonden was aan de cathkamers. Officieel hoorde ik daar ook bij maar

dankzijj jou mocht ik toch op B2 blijven. Jan. dankjewel.

DeDe pacemaker technici

TimTim Schrama. Beste Tim. je was een onmisbare spil voor mij de afgelopen vier jaren.

Jee had altijd tijd voor me. Ik heb veel van je geleerd maar ook veel met je gelachen

onderr een lekker bakje koffie in de meest luxuese pacemakerkamer die ik ken. Ik

benn blij dat je mijn paranymf wil zijn.

WandenaWandena Ramsoekh. Beste Wandena bedankt voor je actieve inzet bij de inclusie van

dee STAP studie en je vrolijkheid (ik heb je nog nooit chagrijnig meegemaakt).

MaritMarit Pfeiffer. Beste Marit, bedankt voor de samenwerking in de SOFA trial

Dee cardiologen wil ik bedanken voor het verlenen van toestemming voor de inclusie

vann patiënten in de verschillende studies. Martin Meesterman en de verpleegkundi-

genn op de cathkamer (Paul Soedarso), de dagverpleging (Truus Verboom. Trudy

]Qo o

Dankwoord d

Stapelkamp),Stapelkamp), en de research [ïndeke Radder) wil ik bedanken voor de gezelligheid

enn het meedoen als gezonde proefpersoon of zoeken daarvan.

HenkHenk de Weert (huisartsgeneeskunde), dr. Schïingeman (oogheelkunde)wil ik

bedankenn voor de inclusie van gezonde vrijwilligers.

DeDe echolaboranten {Irma, Rianne. Denise en Jim) wil ik bedanken voor hun bereid-

heidd van het maken van echo's bij de verschillende onderzoeken.

DeDe ECG laboranten op de polikliniek wil ik bedanken voor hun flexibiliteit bij het

plannenn van patiënten.

Drs.Drs. B.J Verberne, beste Hein het was me een waar genoegen om samen met jou het

artikell over de CRT en MIBG te schrijven. Je was altijd bereid om het artikel op-

nieuww te bespreken en als het stukje weer eens niet was geaccepteerd was je altijd

bereidd om een peptalk te geven. Maar je had gelijk in 'we moeten de bladen een

voorr een afwerken' nu is het toch nog in een goed blad terechtgekomen.

DeDe secretaressen Regina. Anita, Mary Ellen, Monique, en Gerdie wil ik bedanken

voorr hun interesse en meer dan secretariële steun.

Drs.Drs. M. Kortz. Beste Michael, door jou hulp bij de inclusie kwamen ruim eenderde van

dee patiënten van de STAP studie vanuit het Flevoziekenhuis. Bedankt voor je inzet.

Drs.Drs. de Boo. Beste Job. je gastvrijheid en die van de functielaboranten op de afdeling

cardiologiee van het Oosterschelde ziekenhuis bij de uitvoering van het onderzoek

naarr de PEA sensor heeft tot een publicatie in PACE geleid. Bedankt.

KarlaKarla Mulder wil ik bedanken voor de grote bijdrage die ze geleverd heeft door alle

databasess te bouwen voor mijn studies. Lieve Karla met het Acces software program-

maa dat je hebt gebouwd om hartslagen te kunnen samplen vanuit het Polar pro-

grammaa heeft mij heel veel werk uit handen genomen.

LieveLieve Margriet, ik heb veel lol met je gehad toen we samen naast de vroegere

koffiekamerr met Sianos woonden en regelmatig Fokke op bezoek hadden. Al die

verhalenn over je bijzondere vrienden zal me bij blijven.

197 7

Inn het onderzoeksstraatje van B2 richting de koffie kamer kwam ik regelmatig mijn

lotgenotenn tegen; de senior onderzoekers van de kloppende spier {Steven Cham-

uleau,uleau, Martijn Meeuwissen, Michiel Voskuil, Radha Bolangsing, Niels van Royen) en

dee rechter kamer {Anoek van Alem, Gijs Nollen, Igor Tïilevski) julli e zijn altijd een

groott voorbeeld voor me geweest. De overig (ex)-gangbewoners {Fons Windhausen,

MareMare van der Zee. Bektas Atasever, Bart-Jan Verhoeff, Alexander Hirsch, Pieter Bot,

PeterPeter Engel friet, Lilian Meiboom en Tomas Oosterhof), wil ik bedanken voor de

gezelligee dagelijkse, gezamenlijke lunch.

Dee onderzoekers van F3 en F4. Lieve Arno. wij blijven elkaar achtervolgen, eerst was

jee mijn mentor in Nijmegen en inmiddels zijn we collega's in Amsterdam. Mathijs

BoekholdBoekhold en Niek Bijsterveld bedankt voor julli e hulp bij het invriezen van de

neurohormonen. .

Lievee Nancy ik vond het altijd een heerlijke rust moment om met jou bij Antoinne

Petitt te"koffiëen" en bij te praten.

Lievee Gerlind. Je was een belangrijke steun en toeverlaat de afgelopen jaren op mijn

werk.. Ik vond het erg prettig dat ik iemand had waar ik echt alles aan kwij t kon en

diee ook echt luisterde. Met de gezellige bijklets momenten met jou onder een kopje

koffie,, koekjes met een wandelingetje in het AMC kon ik mijn gedachten goed

"resetten."" Ik zal deze momenten erg missen. Bedankt dat je mijn paranymf wil zijn.

Mij nn kamergenoten {Jacobijne, Saskia en Lea).

Lievee Jacobijne, Door jou aanwezigheid werd ik dagelijks op de hoogte gehouden

vann mijn horoskoop en zat ik nooit zonder drank. Lieve Saskia, ik hoefde maar een

engelss of nederlands woord te roepen en ji j wist het wel te vertalen. Mij n blindheid

voorr de lidwoorden de en het kon ik dankzij jou aanwezigheid ook goed vermom-

men.. Lieve Lea, je kennis over de statisitiek en over het algemeen over de comput-

erss vond ik erg waardevol bij het schrijven van de artikelen. Naast de inhoudelijke

supportt heb ik ook veel met julli e kunnen lachen met name tijdens onze 'evaluati-

emomentenn van deze en gene'. Ladies, ik zal julli e straks zeker missen.

FamilieFamilie Krouwel, lieve Aartje en jan dankzij julli e kon ik dagelijks met een gerust

hartt doorwerken. Bedankt voor julli e leuke opvang van Emre.

1QS S

Dankwoord d

LieveLieve pap, jij hebt de moedige stap gezet om naar Nederland te emigreren omm onze familie een beter bestaan te geven. Dat is je gelukt, alleen heb je err zelf te kort van kunnen genieten. Ik weet zeker dat je toch ergens meekijktt en erg trots op ons bent.

LieveLieve mam, canim annecigim. Je bent de sterkste vrouw in mijn leven. Dee wijze levenslessen heb ik van jou geleerd en niet op de schoolbanken. Jee hebt in je eentje ons opgevoed en je leven opgeofferd voor ons. Ik ben jee daar zeer dankbaar voor.

Lieve,Lieve, zus en broers canim ablacigim Sultan, canim abilerim Ersin ve EkremEkrem ve canim kardesim BattaL Jullie zijn mij erg dierbaar. Een voor een hebb ik al die jaren erg veel steun van julli e gehad op alle fronten mijn dankk daarvoor is moeilijk in woorden uit te drukken.

Dee andere leden van de familie mijn neefjes en nichtjes, de kleine sterren vann onze familie (Ethem, Zühal. Meva, Esra. Muhamed, Ekrem, Sükriye. Safa)Safa) en de aangetrouwde familie (Ömer, Selma, Ümran, Mehtap) bedankt voorr de gezelligheid.

Mij nn schoonouders Mehmeten Sebahat Erolwil ik bedanken voor de hartelijkheidd waarmee ze ons altijd ontvangen met de heerlijke uitgebreidee diners. De andere leden van mijn schoonfamilie (Suat, Burcu, Seray,Seray, Selcuk, Hasibe, Fuat, Fatih, Meral) wil ik tevens bedanken voor de gezelligee family meetings.

Lievee Sedat je bent voor mij en mijn idealen verhuisd, je hebt je eigen opleidingsplaatss verplaatst en bent daardoor later begonnen met je opleiding. . Jee was er altijd voor mij en Emre, zeker op de momenten dat ik door wilde werkenn om mijn deadlines te halen of weer vast zat in de file. Lieve schat, ikk hoop nu het proefschrift af is ook meer tijd voor jou te hebben.

Mij nn lieve schattige bambi. Emre, jij bent het mooiste wat mij ooit is overkomen.. Annen seni c,ok seviyor. Ja, mama zal elke dag al je teentjes eenn voor een blijven kussen. Dankzij de ontspannende momenten met jou hebb ik de files en de drukte van alle dag overleefd. Ik zal er altijd voor je zijn. .

199 9

Curriculu mm Vitae

Aytenn Erol-Yilmaz werd geboren op 15 april 1971 te Cicekdagi (Turkije). Na het

behalenn van de HAVO diploma aan de Notre Dame Des Anges in Ubbergen, stu-

deerdee zij eerst een jaar HBOV aan de hoge school van Nijmegen en volgde daarnaast

avondd VWO voor de vakken natuurkunde en scheikunde. Vervolgens studeerde ze

geneeskundee aan de Katholieke Universiteit van Nijmegen. De doctoraalfase werd

behaaldd in 1995 en eind 1997 werd door haar cum laude het artsexamen behaald.

Tijdenss haar studie heeft ze gewerkt als student -assistent bij de vakgroep

Anatomiee & Embryologie en als gids van het museum anatomicum.

Haarr interesse voor de cardiologie werd al vroeg gewekt tijdens klinische stages in

19944 en 1995 op de afdeling cardiologie aan de Hacettepe University (Ankara,

Turkije)) bij prof. dr. S. Kes en prof. dr. A. Oto.

Daarnaast,, heeft ze wetenschappelijk onderzoek gedaan naar acute en late

complicatiess bij percutane transluminal coronaire angioplastiek bij drs. Meursing,

dr.. Aengevaeren en prof. dr. F.W.A. Verheugt (afdeling cardiologie Canisius Wil-

helminaa Ziekenhuis en UMC St Radboud te Nijmegen).

Naa het behalen van de artsexamen heeft ze klinische ervaring opgedaan op de

afdelingg cardiologie van CWZ, UMC St Radboud te Nijmegen en UMC Utrecht bij

prof.. dr. Robles de Medina,

Vanaff begin 2001 is zij werkzaam op de afdeling cardiologie van het Academisch

Medischh Centrum (AMC) in het kader van haar promotie-onderzoek onder

begeleidingg van dr. R. Tukkie en prof, dr. A.A.M. Wilde. Daar heeft zij zich de afgelo-

penn jaren verdiept in pacemakersensor optimalisatie en cardiale resynchronisatie

therapie.. In 2004 schreef ze een LOI cursus over pacemakersensoren voor

pacemakertechnicii en ontving zij de Finapres Medical System travel fellowship

awardd gezamenlijk uitgereikt door de European Federation of Autonomie Societies

enn American Autonomy Society.

Vanaff 1 april 2005 start zij met haar opleiding tot cardioloog te beginnen met de

vooropleidingg interne geneeskunde aan het Bosch Medisch Centrum (opleider:

dr.. P.M. Netten). De opleiding tot cardioloog zal worden afgerond op de afdeling

cardiologiee van het AMC (opleider: dr. R.B.A. van den Brink).

Aytenn Erol-Yilmaz is getrouwd met Sedat Erol en heeft een zoontje Emre van

tweee jaar.

200 0

Stellingen n

1.. Ondanks verbetering van inspanningscapaciteit en verhoging van de maximale

hartfrequentiee resulteert individuele optimalisatie van frequentie-adapterende

pacemakerss niet tot verbetering van de kwaliteit van leven (dit proefschrift).

2.2. Pacemaker sensor optimalisatie leidt tot minimale toename in het slagvolume

enn de cardiac output gedurende de inspanning zonder verandering in de bloed-

doorstromingssnelheidd in de arteria cerebri media (dit proefschrift).

3.. Toegankelijke pacemaker sensor algoritmen voor manuele afstelling zijn erg

belangrijkk voor individuele optimalisatie (dit proefschrift).

4.. De piek endocardiale acceleratie (PEA) sensor functioneert indien geprogram-

meerdd als single sensor systeem hypochronotroop gedurende de inspanning

(ditt proefschrift),

5.. Zes maanden cardiale resynchronisatietherapie leidt tot gunstige veranderingen

inn het neurohumorale systeem (dit proefschrift)-

6.. Cardiale resynchronisatietherapie verbetert de sub-linguale microcirculatie (dit

proefschrift). .

7.7. Als we wisten wat we deden dan heette het geen onderzoek (Albert Einstein)

S.. Emancipatie van de vrouw betekent het waarborgen van de keuzevrijheid van

dee vrouw met behoud van vrouwelijke kenmerken en niet het verkrijgen van

dezee vrijheid dankzij het overnemen van mannelijk gedrag.

9.. De enige ware gids is de wetenschap (M. K. Atatürk)

10.. De toetredingscriteria voor het uiteindelijke lidmaatschap van Turkije bij de

Europesee Unie is net zo veranderlijk als het weer in Nederland.

11.. Het is een morele plicht van de mens om optimistisch te zijn (Sir Karl Popper,

filosoof). .