quantitative radionuclide haemodynamic rest exercise*

Br Heart JT 1980; 43: 125-133

Assessment of function of contractile segments inpatients with left ventricular aneurysms byquantitative first pass radionuclide ventriculographyHaemodynamic correlation at rest and exercise*DUNCAN S DYMOND, JOHN STEPHENS, DAVID STONE, P H JARRITT,tALEX ELLIOTT, K E BRITTON, R A J SPURRELL

From the Departments of Cardiology and Nuclear Medicine, St. Bartholomew's Hospital, London

SUMMARY Thirteen patients with left ventricular aneurysms complicating myocardial infarction werestudied by contrast angiography and by first pass radionuclide ventriculography. The ejection fractionof the contractile segment (EFCS) was measured from both studies using a double hemispheroid model,and the values correlated closely. There was a monotonic relation between EFCS and stroke volumeindex measured from thermodilution cardiac outputs carried out simultaneously with the radionuclidestudy When radionuclide ventriculography was performed at submaximal supine exercise, changesin EFCS paralleled changes in the total left ventricular ejection fraction in 10 of the 13 cases.

In nine patients, changes in EFCS paralleled changes in stroke volume index and the relation betweenEFCS and stroke volume index was maintained at exercise. After administration of the vasodilatorisosorbide dinitrate to 12 patients, repeat exercise radionuclide ventriculography showed an improve-ment in left ventricular ejection fraction and in eight patients EFCS improved.

First pass radionuclide ventriculography can accurately estimate EFCS, which may be animportant factor in predicting the likely response to aneurysmectomy. Changes in EFCS on exerciseare reflected in changes in total left ventricular ejection fraction and stroke volume index. Isosorbidedinitrate may improve contractile segment function on exercise.

Surgical resection of left ventricular aneurysmscomplicating myocardial infarction is a well-established therapeutic technique, though there arewide ranges in both the reported in-hospitalmortality and the long-term mortality,'-5 and theprocedure is by no means certain to produce clinicalimprovement. The crucial factors in the assessmentof patients being considered for aneurysmectomyare the demonstration of segmental rather thandiffuse ventricular dysfunction and the presenceof sufficiently good function of the contractileportion of ventricle. The differentiation betweensegmental and diffuse ventricular disease has formany years been accomplished by contrast left

* This work was supported by the British Heart Foundation.t Present address: Department of Nuclear Medicine, MiddlesexHospital, London WIN 8AA.Received for publication 19 June 1979

ventriculography,6 but contractile segment functionhas often been assessed by visual inspection of thecineangiogram, rather than by objective quantifi-cation. Recently, attempts have been made toquantify contractile segment function angiographi-cally and to relate this function to the risks andresults of aneurysmectomy.7 8 These studies havesuggested that a contractile segment ejectionfraction of approximately 40 per cent is necessaryfor a good result from aneurysmectomy.The ability of radionuclide ventriculography to

detect left ventricular aneurysms and to discriminatebetween aneurysms and diffuse hypokinesis is welldocumented.9 10 This study describes the quanti-fication of contractile segment function by radio-nuclide ventriculography and haemodynamic cor-relates of this function at rest and on exercise. Theeffect of isosorbide dinitrate on left ventricularfunction on exercise is also assessed.

125

on January 25, 2022 by guest. Protected by copyright.

http://heart.bmj.com

/B

r Heart J: first published as 10.1136/hrt.43.2.125 on 1 F

ebruary 1980. Dow

nloaded from

http://heart.bmj.com/

Dymond, Stephens, Stone, jarritt, Elliott, Britton, Spurrell

Patients and methods

Thirteen patients with left ventricular aneurysmsoccurring after myocardial infarction were studied.There were 11 men and two women. Their agesranged from 44 to 64 years with a mean age of 56years. The main reason for investigation was cardiacfailure in 10, angina pectoris in two, and recurrentventricular arrhythmias in one. Informed consentwas obtained in all cases. Left heart catheterisationwas performed via a right brachial arteriotomy, leftventriculography being carried out in the rightanterior oblique projection. A flow-directed thermo-dilution catheter was positioned in the pulmonaryartery. In eight patients the haemodynamic measure-ments were made at the time of catheterisation withthe following procedure. Thirty minutes wereallowed to lapse for the effects of contrast media tosubside" and control measurements were made ofcardiac output by the thermodilution technique,'2using a Waters cardiac output computer (COR1OOA), and of heart rate. All measurements weremade in triplicate. The patients then performedsubmaximal supine leg exercise with a bicycleergometer, at a previously determined work ratethat caused a rise in heart rate of between 30 and50 per cent over resting levels, but that did notcause limiting dyspnoea or angina pectoris. Measure-ments were then repeated. Submaximal exercisewas employed in these patients as all were severelysymptomatic. They would have been unable tomaintain exercise for long enough to allow threesets of haemodynamic measurements to be madehad higher workloads been used. After a 10 minuterest period, 5 mg isosorbide dinitrate were adminis-tered sublingually. Fifteen minutes after the drug,by which time resting left ventricular fillingpressure had fallen to 16 mmHg or less in allpatients, exercise was reperformed to the samelevel as previously and measurements repeated.The other five patients were not exercised in the

catheter laboratory but had all haemodynamicmeasurements performed at the time ofradionuclideventriculography.A resting radionuclide ventriculogram was

carried out in the right anterior oblique projection,within six hours of cardiac catheterisation in 10patients and within 24 hours in the other three,using a computerised multicrystal gamma camerawith high count-rate capabilities'3 (Baird-Atomic,System 77). A 10 mCi bolus of high specific-activity technetium99m, as pertechnetate, wasinjected through the right atrial lumen of thethermodilution catheter with a rapid saline flush.One thousand frames of data were collected at 50 msframing intervals to monitor the first pass of the

radionuclide through the central circulation. A1 in parallel-hole collimator was used. Data werecorrected for field non-uniformity and the deadtime of the instrument, and stored on magnetictape. The day after the resting study, radionuclideventriculography was performed at exercise, afteran identical workload in the case of the eightpatients who had exercised in the catheter laboratory.The other five patients were exercised using anidentical procedure, haemodynamic measurementsbeing made immediately before the cessation ofexercise and the injection of the radionuclide. In12 of the 13 patients, a further exercise radionuclideventriculogram was performed 15 minutes after5 mg isosorbide dinitrate were administered sub-lingually, allowing at least 25 minutes between thetwo exercise studies. This period was allowed forthe clearance of the majority of the first bolus oftechnetium99m from the intracardiac blood pool.Before the second exercise study a static backgroundframe was counted to a maximum of 9000 countsper cell and the second study corrected for residualblood pool activity as well as for field non-uniformityand dead time. Since constant patient positioningwas crucial to enable accurate background bloodpool correction, the positions of four light beamsemitted by the detector head were marked on thepatients' chests and adjusted as necessary beforethe injection of the post-isosorbide dinitrateradionuclide bolus.There was no morbidity in any case and the

maximum accumulated dose of technetium99m foreach patient was 35 mCi.For this study, a ventricular aneurysm was

defined angiographically as a clearly demarcatedakinetic segment with or without associated areasof dyskinesis,"4 separated from a contractile segmentof ventricle, in sinus rhythm. Coronary arterystenoses of 70 per cent or more of the arterial lumenwere judged to be significant.For objective evaluation of the contrast angio-

grams end-diastolic and end-systolic frames weretraced by selecting the largest and subsequentsmallest appearing ventricular outlines of a beatthat was not an ectopic, or immediately after anectopic. Analysis of the radionuclide studies wascarried out using software supplied by the manu-facturers. Total left ventricular ejection fraction wascalculated from the change in counts from end-diastole to end-systole, read from the high frequencybackground corrected time-activity curve from theleft ventricular region of interest,'0 1B making noassumptions about ventricular geometry. A repre-sentative cardiac cycle was generated by summingcorresponding frames from the peaks and troughsof the time-activity curve. The perimeters of the

126



/B


ebruary 1980. Dow

nloaded from


Radionuclide ventriculography and function of contractile segments

ventricle at end-diastole and end-systole weregenerated after linear interpolation of the end-diastolic and end-systolic frames from a data matrixof 294 points (14 x 21) to one of 4704 points (56 x84).The computerised edge enhancement technique

used for perimeter generation consisted of high-lighting the 5 per cent count band correspondingto 25 to 30 per cent of the maximum end-diastoliccount for both end-diastole and end-systole. Thismethod of edge enhancement has been found toproduce the most consistent agreement withcontrast angiographic volumes.10EFCS per cent was measured using the double

hemispheroid model of Watson et al.8 In this modela line of demarcation is drawn between the sectionof ventricle that shows definite inward motionduring systole, that is the contractile segment, andthe akinetic or dyskinetic segment. EFCS wascalculated from an area-length method modifiedto fit the double hemispheroid model, with thelong axis of the segment being the longest lengthperpendicular to the line of demarcation. Theareas and lengths were calculated from the tracedperimeters of both contrast and radionuclidestudies, using an electronic pressure pad linked toa computer, and an algorithm analogous to plani-metry. For each area and length, three measurementswere taken and meaned.

Stroke volume index was calculated from themean of the thermodilution cardiac outputs, heartrate, and body surface area. Results were analysedby Student's t test and Spearman's rank test.

Results

Aneurysms were anterolateral or apical in 12, onepatient having a discrete inferior aneurysm. No

E 45

o 40

., 35-

> 3)-

- 25-

C 20-

a 15-

E l010

Ln 5LL

L UQ0

0 .

0 n= 13r = 0-86

0 5 10 15 20 25 30 35 40 45 50 55EFCS from contrast angiogram

Fig. 1 Plot of EFCS from contrast angiogram againstEFCS from radionuclide study. EFCS, ejection fractionof contractile segment.

60-

50 -

40

30

20

10Co

* 10

.,

0 0EaE 60-0

LA 5C) 501-LU

40'

30

20

10

0

g

0 v

0

0 *0

0 0

n=13r = 078

10 20 30 40 50 60

b

*

0S

0* *t0

0

.

n = 13r = 072

0 10 20 30 40 50 60SVI (mi/M2)

Fig. 2 (a) Correlation between stroke volume index(SVI) and EFCS at rest. (b) Correlation betweenstroke volume index (SVI) and EFCS at exercise.EFCS, ejection fraction of contractile segment.

patient had significant mitral regurgitation. Eightpatients had one-vessel coronary disease, seven ofthese having an occluded left anterior descendingartery, one an occluded right coronary. One hadtwo-vessel disease and four had three-vessel disease.EFCS from the contrast angiogram ranged from

17 to 53 per cent with a mean of 34-7 per cent,while EFCS from the radionuclide study at restranged from 14 to 46 per cent with a mean of 35 0per cent. Total left ventricular ejection fraction wasreduced in all patients, ranging from 19 to 37 percent, with a mean of 27 0 per cent. The correlationbetween EFCS from contrast and radionuclidestudies is shown in Fig. 1, the correlation coefficientbeing 0 86 (p < 0 001). There was a poor correlationbetween EFCS and total left ventricular ejectionfraction (r=0 3). Thus total left ventricular ejectionfraction could not be used to differentiate patientswith good and bad contractile segment function.Stroke volume index ranged from 20 to 59 ml/m2,with a mean of 36 ml/m2. Fig. 2a shows the relationbetween resting EFCS and resting stroke volumeindex to be monotonic. Analysis of this relation by

127



/B


ebruary 1980. Dow

nloaded from


Dymond, Stephens, Stone, JYarritt, Elliott, Britton, Spuirrell

Spearman's non-parametric rank test shows asignificant correlation with a coefficient of 0.78.This is not surprising as by definition the contractilesegment is the only ventricular segment contributingto cardiac. output.

Six patients had a resting EFCS of 40 per cent ormore. Residual standard error from regression wasused to estimate the predicted value of EFCS fromthe radionuclide angiogram which correspondedto a value of40 per cent from the contrast angiogram.Thus EFCS (radionuclide) = 39-2 ±3 0 per cent atEFCS (contrast) of 40 per cent.

EFFECTS OF EXERCISEFor the whole group, left ventricular ejectionfraction ranged from 16 to 40 per cent with a meanof 28 per cent, not significantly different from rest.EFCS ranged from 17 to 50 per cent with a meanof 35 per cent. All six patients who showed a risein total left ventricular ejection fraction had anincreased EFCS, and of five with a fall in total leftventricular ejection fraction, four had decreasedEFCS. Only one patient had a decreased leftventricular ejection fraction with a rise in EFCS.Fig. 3 shows the changes in left ventricular ejectionfraction on exercise plotted against the changes inEFCS, there being a significant correlation betweenthe changes, whether positive or negative (r=0-70,p < 0 01). Thus changes in ventricular function asassessed from the count-volume method reflectthose from the geometric model using computer-generated ventricular perimeters.Of seven with an increased stroke volume index

on exercise, six had an increased EFCS. Of four

6 LVEF(+)161412 I108642

6 EFCS(-).- .

18 16 14 12 10 8 6 4 22S

0 4

* 681012L

15 LVEF (H

0

0

*. ' '- 6 EFCS(+)2 4 6 8 10 12 14

.

nf= 13r = 0 7

Fig. 3 Plot of change in left ventricular ejection fractionand change in EFCS from rest to submaximal exercise.

3LVEF, change in total left ventricular ejection fraction(count volume method). &EFCS, change in ejectionfraction of contractile segment.

60

50 1

MeanMSEa ± 1'

M0 SEImMeneSLfl Mean+ SE1i 40

30

20 Rest Exercise Exercise

ISDNFig. 4 Changes in EFCS in six patients with initialEFCS of 40 per cent and more.

with a reduced stroke volume index, three hadreduced EFCS. The monotonic relation betweenEFCS and stroke volume index was maintained onexercise, as shown in Fig. 2b, with a correlationcoefficient of 0-72 using Spearman's non-parametricrank test.Of the six patients with an initial EFCS of 40

per cent or more, four showed an increase and twoa decrease at submaximal exercise. Mean EFCSfor this group did not differ significantly from theresting values (Fig. 4). Of those with an initialEFCS of less than 40 per cent, four improved andthree deteriorated on exercise. Again, the meanvalue did not differ significantly from rest (Fig. 5).

EFFECT OF CORONARY STATUS ONEXERCISE RESPONSEOf the eight with one-vessel disease, three improvedand five deteriorated on exercise. All patients withmore extensive disease improved. No patientexperienced angina during exercise.

EXERCISE AND ISOSORBIDE DINITRATEAt exercise after isosorbide dinitrate, left ventricularejection fraction ranged from 22 to 44 per cent witha mean of 32 per cent. This represents a significantimprovement over the pre-isosorbide dinitratevalues (p < 0.01, Fig. 6). Stroke volume indexranged from 29 to 74 ml/m2, mean 41-4, significantlybetter than before isosorbide dinitrate (p < 0 005).EFCS ranged from 26 to 55 per cent with a

mean of 38 per cent, eight patients showing animprovement, three being unchanged, and oneworse than before isosorbide dinitrate, though forthe whole group the improvement in EFCS failedto achieve statistical significance.

128



/B


ebruary 1980. Dow

nloaded from



Analysis of the responses in relation to the restingEFCS shows that of the five patients with restingEFCS of 40 per cent or more who received isosor-bide dinitrate, all improved, with a mean EFCSfor this subgroup of 46-6 compared with 40 9before isosorbide dinitrate (Fig. 4). This improve-ment achieves statistical significance (p < 0 025). Ofthe seven with an EFCS of less than 40 per cent atrest, three improved after isosorbide dinitrate, oneworsened, and three remained unchanged afterisosorbide dinitrate (Fig. 5). All three who improvedachieved a higher EFCS than at rest. Mean EFCSwas 32-0 after isosorbide dinitrate, compared with29-2 before the drug, an improvement that is notstatistically significant.

Fig 7 relates the changes that occurred in thetwo exercise periods to the changes that occurredfrom rest to exercise. Of the seven patients in whomEFCS improved from rest to exercise, six improvedfurther after isosorbide dinitrate and one worsened.Of the five who deteriorated from rest to exercise,two improved and three were unchanged after iso-sorbide dinitrate. Thus a bad response to exercisedid not preclude a good response to isosorbide dini-trate, though a good response to exercise was morelikely to predict a good response to isosorbide dini-trate.

Fig. 8 shows an example of how contractilesegment function may improve after isosorbidedinitrate. The left-hand image shows the perimetersfor the pre-isosorbide dinitrate study, the rightshows them post-isosorbide dinitrate. Theend-diastolic perimeter is shown in white, the end-systolic perimeter in yellow. It is evident that the

50

40'

' 30'LL.w

20'

10

b Mean vSE

Rest Exercise Exercise

ISDNFig. 5 Changes in EFCS in seven patients with initialEFCS of less than 40 per cent. EFCS, ejection.fractionof contractile segment.

-l 50

C-

N 40

C0

.c

3: 30

a 20c0

.S! 10uJ

~~~~zZ.~~~~0I

S

ExercisebeforeSDN

Mean i SEMbefore after

n=12P'0 01

ExerciseafterISDN

Fig. 6 Exercise left ventricular ejection fraction (LVEF)before and after isosorbide dinitrate (ISDN). SEM,standard error of the mean.

contractile segment shows greater inward motion inthe right-hand image, and the calculated EFCS hascorrespondingly increased. The contractile segmentand the aneurysmal segment are both clearly seenin each image.

EFFECT OF CORONARY STATUS ONISOSORBIDE DINITRATE RESPONSEOf the eight patients with one-vessel disease, fiveimproved after isosorbide dinitrate and threeremained unchanged. Of those who improved, threehad shown an improvement from rest to exercise.Four of the five who improved had a resting EFCSof 40 per cent or more. Of those with three-vesseldisease, three improved and one worsened. Allhad shown an improvement from rest to exerciseand only one had a resting value of EFCS of 40per cent or more.These results suggest that the response to isosor-

bide dinitrate is not governed entirely by coronarystatus, inasmuch as the four non-responderscontained three with one-vessel disease.

Discussion

Resection of left ventricular aneurysms is now awell-recognised surgical procedure, but it hasbecome evident that not only is surgical mortalityhigh in some series,4 5 but clinical improvement isby no means the rule.5 Key et al.'16 commented onthe necessity for 'good left ventricular deep con-strictor activity' to increase the likelihood of asuccessful operation. In 1971, Kluge et al."7described an unquantified relation between 'thequality of basal myocardial contractility' andimmediate postoperative survival. More recently,these subjective relations have been superseded by

129



/B


ebruary 1980. Dow

nloaded from


Dymond, Stephens, Stone, J7arritt, Elliott, Britton, Spurrell

objective methods for quantification of basal orcontractile segment function from contrast angio-grams.7 8The importance of a method for objective

quantification of contractile segment function isevident, as total left ventricular ejection fraction atrest was unable, in this study, to differentiate badfrom good contractile segment function, anobservation in agreement with Watson et al.8 Allpatients had reduced total left ventricular ejectionfraction, however. The double hemispheroid modelis easily applicable to the image produced bysuperimposition of the ventricular perimeters(Fig. 8) but is not the only method available.Arthur et al.7 calculated EFCS by excludingthe aneurysmal portion of ventricle from the cine-angiogram and redrawing the cardiac border as afixed elliptical line equidistant from the longi-tudinal axis as the opposite contracting border.Lee et al.18 expressed the contractility of the non-aneurysmal ventricular segment as the differencebetween actual ejection fraction and that predictedfrom the spherical model of Feild et al.19 Thedouble hemispheroid technique was chosen for thisstudy as it has been validated against geometricmodels of known volumes, and the sums of sub-chamber volumes have been shown to correlatewith total ventricular volumes.8The relation between EFCS and stroke volume

index (Fig. 2) is further testimony to the validityof the model as applied to the radionuclide study.Though others have reported no correlationsbetween EFCS and cardiac index,8 this latterdepends on another variable, heart rate, as well ason stroke volume.The rationale behind separating those patients

with an EFCS of 40 per cent or more from those

z0

LfO

'0

,n -"v

mo C)w_ (A

EL-o

+14 -

+12+10+ 8 -+ 6+ 4 -+ 2 -

0

- 2 -- 4 -

- 6 -- 8--10 -

-12

0

0

0

0

0

0

n = 12

t EFCSRest toexercise0

I EFCSRest toexercise

Fig. 7 Changes in EFCS from exercise before to afterisosorbide dinitrate (ISDN) related to changes in EFCSfrom rest to exercise. EFCS, ejection fraction ofcontractile segment,

Fig. 8 Superimposed end-diastolic (white) and end-systolic (yellow) perimeters generated from theradionuclide ventriculogram by edge enhancement(see text). Left-hand image, exercise before isosorbidedinitrate (EFCS 24%). Right-hand image, exerciseafter isosorbide dinitrate (EFCS improved to 36%).

with a value of less than 40 per cent is based onthe findings that an EFCS of less than 44 per centin one study,8 or less than 38 per cent in another,7predicted a poor surgical result. It is implied,therefore, that an EFCS of below approximately 40per cent means a critical impairment of residualventricular function.The short data acquisition time required for a

first pass radionuclide ventriculogram means thatstudies on exercise are not limited by changes in thestability of patients' circulatory status, all datacollection being completed within one minute fromthe cessation of exercise.

Borer et al.20 have previously shown that nosignificant differences occurred in ejection fractionor regional ventricular function when sequentialexercise radionuclide ventriculograms were per-formed with no intervention. This was the casefor both normals and patients with coronary arterydisease. The parallel changes between radionuclideand haemodynamic data described in this studysupport the findings of Borer et al.20 that changesseen after an intervention are not the result ofintrinsic errors in the method used.The changes in ventricular function, as measured

by total left ventricular ejection fraction, werevariable and were paralleled by changes in contrac-tile segment function in 10 of the 13. The limitationsof comparing results from a single plane geometricmodel with results from a three-dimensionalmeasurement of radioactive counts may contributeto the discrepancies in the other three. It is possiblethat changes detected in the right anterior obliquemay have been offset by opposing changes in the

130



/B


ebruary 1980. Dow

nloaded from



areas of the contractile segment not visualised inthat projection, whereas this will not affect measure-ment of total left ventricular ejection fraction. Thesame problem will apply to comparisons of changesin EFCS and stroke volume index. The variablechanges in left ventricular ejection fraction, EFCS,and stroke volume index that occurred on exerciseare consistent with the spectrum of exerciseresponses described in other studies, in which wideranges in the responses of stroke volume index,cardiac index, and left ventricular stroke work indexoccurred in patients with cardiac failure.2' 22Improvement in ventricular function on exercise

may be brought about by augmented myocardialcontractility under the inotropic stimulus ofexercise and, indeed, the failing ventricle may behypersensitive to circulating catecholamines.2' Con-versely, the increase in peripheral vascular tone andhence ventricular afterload may be poorly toleratedby the failing ventricle because of the rise in systolicwall tension and myocardial oxygen consumption.24Thus, the changes in EFCS, like the changes instroke volume index, may be affected by multipleopposing factors and provide angiographic cor-relates of previously reported variable haemodyna-mic changes.The failure of more than half the patients with

single-vessel disease to improve EFCS on exercise,despite apparently normal residual coronary anat-omy, is worthy of comment. It is possible that inthe chronically dilated and overstretched ventricleexcessive myofibrillar slippage leads to an oftenirreversible depression of contractility25 which isindependent of coronary status.

Ischaemia of the contractile segment would notappear to play an important role in determiningthe response to exercise in view of the goodresponse of those with extensive coronary disease.In addition, no patient experienced angina pectorisduring exercise.

In recent years, dramatic improvements in theperformance of the failing left ventricle have beenshown after the administration of vasodilator drugs.Oral isosorbide dinitrate has been shown to beeffective in causing a substantial reduction in leftventricular filling pressure in patients with conges-tive cardiac failure. 26 27The results of this study indicate that sublingual

isosorbide dinitrate may lead to improvements inglobal and segmental ventricular function onexercise, with significant improvements in totalleft ventricular ejection fraction for the group as awhole, though the changes in EFCS were onlystatistically significant in those with a resting EFCSof greater than 40 per cent. In the other group, thechanges in EFCS paralleled those in left ventricular

ejection fraction and stroke volume index but didnot achieve statistical significance. In four patientsin the study (33%) EFCS failed to improve, however.The detected improvement in exercise EFCS

after isosorbide is consistent with the improvementin exercise haemodynamics that has been previouslyreported in patients with both diffuse impairmentof left ventricular function22 and with left ventricularaneurysms.28 In both these studies, isosorbidedinitrate caused significant improvements in ventri-cular stroke work index and stroke volume index,and significant reductions in left ventricular fillingpressure on exercise.

It has been suggested that the principal action ofvasodilators in general is to reduce left ventricularafterload29 with a concomitant reduction in myo-cardial wall tension. The reduction in filling pressuremay be also partly a result of a fall in preloadproduced by venodilatation.30 The reduced walltension leads to an increased velocity of myocardialfibre shortening.3' In addition, the effects of theinotropic stimulation occurring on exercise may beenhanced by the pharmacological reduction of walltension leading to greater augmentation of cardiacperformance.'2

It is of interest that the more consistent responseto isosorbide dinitrate took place in the patientswith the higher resting EFCS and the improvementwas significant in this group but not in those withan EFCS of less than 40 per cent. A previous studyhas shown that the improvement in exercise cardiacindex was related directly to the resting EFCS.22Nevertheless, improvements in EFCS did occur infour out of seven patients with an EFCS of lessthan 40 per cent. It is also of note that responsesto isosorbide dinitrate could not be entirelypredicted on the basis of the rest to exercise changes,though of the eight who improved after isosorbidedinitrate, six had shown an improvement onexercise.

Again, coronary status did not appear to be ofparamount importance in governing the responseto isosorbide dinitrate, there being a similarspectrum of response in those with single ormultiple vessel disease. The possibility does exist,however, of redistribution of blood by the nitrateto ischaemic areas in the contractile segment,33though the evidence that such alterations in regionalblood flow do occur in man is by no meansconclusive.

Left ventricular aneurysmectomy may or maynot lead to an improvement in exercise toleranceand left ventricular end-diastolic pressure5; so amethod that could predict the likely results ofsurgical resection in terms of mortality and ofclinical improvement would be welcome.

131



/B


ebruary 1980. Dow

nloaded from


Dymond, Stephens, Stone, Jarritt, Elliott, Britton, Spurrell

One hopes that the pharmacological improve-ments in EFCS, left ventricular ejection fraction,and stroke volume index brought about by thevasodilator would mimic the beneficial haemody-namic effects of reducing ventricular filling pressureby aneurysmectomy, though further prospectiveand postoperative studies are required before apharmacological intervention can be extrapolatedto the surgical situation with any confidence.Contractile reserve, as measured by response toisosorbide dinitrate, does exist even in thosepatients with a low resting EFCS, though apparentlyto a lesser extent than in those with higher EFCS.Conversely, a resting EFCS of 40 per cent or moredoes not always indicate good contractile reserve,since some such patients will have had a badresponse to exercise, and may not achieve restinglevels even after isosorbide dinitrate. Even patientswith one-vessel coronary disease may fall into thiscategory possibly because of irreversibly damagedmuscle in the contractile segment despite normalresidual perfusion. It remains to be seen at whatlevel of response, either good or bad, the surgicalresult becomes unacceptably poor.

In summary, computerised first pass radionuclideventriculography provides a simple and safe way ofassessing segmental and global ventricular functionat rest and exercise. The double hemispheroidmodel is applicable to computer-generated ventri-cular outlines and detected changes in contractilesegment function parallel changes in simultaneouslyrecorded haemodynamics. If the function of thecontractile segment is the prime factor in determin-ing the result of aneurysmectomy, then stressradionuclide ventriculography can provide uniqueangiographic data to aid in selection of patients foraneurysmectomy.The demonstration that isosorbide dinitrate can

improve exercise ventricular function suggests thatthe drug may improve exercise tolerance in somepatients with left ventricular aneurysms. Thistechnique may allow more logical identification ofpatients likely to benefit from long-term nitrateswhen aneurysmectomy is not contemplated orcontraindicated.

References

'Favoloro RG, Effler DB, Groves LK, Westcott RN,Suarez E, Lozada J. Ventricular aneurysm-clinicalexperience. Ann Thorac Surg 1968; 6: 227-45.2Tice DA, Cheng TO, Dolgin M. Surgical treatmentof postmyocardial infarction scars (ventricular an-eurysms). Am Heart J 1970; 80: 282-6.3Kouchoukos NT, Doty DB, Buettner LE, Kirklin JW.Treatment of postinfarction cardiac failure by myo-

cardial excision and revascularization (abstract).Circulation 1971; 43/44 Suppl. II: 108.4Graber JD, Oakley CM, Pickering BN, Goodwin JF,Raphael MJ, Steiner RE. Ventricular aneurysm: anappraisal of diagnosis and surgical treatment. Br HeartJ 1972; 34: 830-8.5Fisher VJ, Alvarez AJ, Shah A, Dolgin M, Tice DA.Left ventricular scars: clinical and haemodynamicresults of excision. Br Heart J 1974; 36: 132-8.6Hamilton GW, Murray JA, Kennedy JW. Quantitativeangiocardiography in ischemic heart disease. Thespectrum of abnormal left ventricular function and therole of abnormally contracting segments. Circulation1972; 45: 1065-80.7Arthur A, Basta L, Kioschos M. Factors influencingprognosis in left ventricular aneurysmectomy (abstract).Circulation 1972; 45/46 Suppl. II: 127.8Watson LE, Dickhaus DW, Martin RH. Left ventri-cular aneurysm: pre-operative hemodynamics, chambervolume and results of aneurysmectomy. Circulation1975; 52: 868-73.9Rigo P, Murray M, Strauss HW, Pitt B. Scinti-photographic evaluation of patients with suspectedleft ventricular aneurysm. Circulation 1974; 50: 985-91.°0Dymond DS, Jarritt PH, Britton KE, Spurrell RAJ.Detection of post-infarction left ventricular aneurysmsby first pass radionuclide ventriculography using amulticrystal gamma camera. Br Heart J 1979; 41:68-78.

"Mullins CB, Leshin SJ, Mierzwiak DS, AlsobrookHD, Mitchell JH. Changes in left ventricular functionproduced by the injection of contrast media. Am HeartJ 1972; 83: 373-81.2Ganz W, Swan HJC. Measurement of blood flow bythermodilution. Am J Cardiol 1972; 29: 241-6.

"Budinger TF, Rollo FD. Physics and instrumentation.Prog Cardiovasc Dis 1977; 20: 19-53.

1Kitamura S, Kay JH, Krohn BG, Magidson O,Dunne EF. Geometric and functional abnormalitiesof the left ventricle with a chronic localized non-contractile area. Am J Cardiol 1973; 31: 701-7.

"Marshall RC, Berger HJ, Costin JC, et al. Assessmentof cardiac performance with quantitative radionuclideangiography. Sequential left ventricular ejection frac-tion, normalized left ventricular ejection rate andregional wall motion. Circulation 1977; 56: 820-9.

"6Key JA, Aldridge HE, MacGregor DC. The selectionof patients for resection of left ventricular aneurysms.Y Thorac Cardiovasc Surg 1968; 56: 477-83."Kluge TH, Ullal SR, Hill JD, Kerth WJ, Gerbode F.Dyskinesia and aneurysm of the left ventricle. Surgicalexperience in 36 patients. J Cardiovasc Surg 1971; 12:273-80.'Lee, DCS, Johnson RA, Boucher CA, Wexler LF,McEnany MT. Angiographic predictors of survivalfollowing left ventricular aneurysmectomy. Circulation1977; 55/56 Suppl. II: 12-18.

'9Feild BL, Russell RO Jr, Dowling JT, Rackley CE.Regional left ventricular performance in the yearfollowing myocardial infarction. Circulation 1972; 46:679-89.

132



/B


ebruary 1980. Dow

nloaded from



20Borer JS, Bacharach SL, Green MV, Kent KM,Johnston GS, Epstein SE. Effect of nitroglycerin onexercise-induced abnormalities of left ventricularregional function and ejection fraction in coronaryartery disease. Assessment by radionuclide cine-angiography in symptomatic and asymptomaticpatients. Circulation 1978; 57: 314-20.

21Ross J Jr, Gault JH, Mason DT, Linhart JW, Braun-wald E. Left ventricular performance during muscularexercise in patients with and without cardiac dysfun-tion. Circulation 1966; 34: 597-608.

22Stephens JD, Camm AJ, Spurrell RAJ. Haemodynamiceffects of isosorbide dinitrate in patients with congestivecardiac failure at rest and during submaximal supineexercise. Arch Mal Coeur 1978; 71: 241-50.

2Spann JE Jr, Buccino RA, Sonnenblick E, BraunwaldE. Contractile state of cardiac muscle obtained fromcats with experimentally produced ventricular hyper-trophy and heart failure. Circ Res 1967; 21: 341-54.

24Braunwald E. Control of myocardial oxygen consump-tion: physiological and clinical considerations. Am JCardiol 1971; 27: 416-32.oMason DT. Regulation of cardiac performance inclinical heart disease. Interactions between contractilestate mechanical abnormalities and ventricular com-pensatory mechanisms. Am J Cardiol 1973; 32: 437-48.6Franciosa JA, Mikulic E, Cohn JN, Jose E, Fabie A.Hemodynamic effects of orally administered isosorbidedinitrate in patients with congestive cardiac failure.Circulation 1974; 50: 1020-4.

27Williams DO, Bommer WJ, Miller RR, AmsterdamEA, Mason DT. Hemodynamic assessment of oralperipheral vasodilator therapy in chronic congestiveheart failure. Prolonged effectiveness of isosorbidedinitrate. Am 7 Cardiol 1977; 39: 84-90.

28Stephens JD, Dymond DS, Spurrell RAJ. Effects ofafterload reduction in exercise left ventricular per-formance in patients with left ventricular aneurysm.Trans Eur Soc Cardiol 1978; 1: 106.

29Cohn JN. Vasodilator therapy for heart failure. Theinfluence of impedance on left ventricular performance.Circulation 1973; 48: 5-8.

30Mason DT, Braunwald E. Effects of nitroglycerin andamyl nitrite on arteriolar and venous tone in the humanforearm. Circulation 1965; 32: 755-66.

3'Sonnenblick E, Parmley WW, Urschel CW. Thecontractile state of the heart as expressed by force-velocity relations. Am J Cardiol 1969; 23: 488-503.

32Miller RR, Awan NA, Joye JA, et al. Combineddopamine and nitroprusside therapy in congestiveheart failure. Greater augmentation of cardiac per-formance by addition of inotropic stimulation toafterload reduction. Circulation 1977; 55: 881-4.

33Horwitz LD, Gorlin R, Taylor WJ, Kemp HG.Effect of nitroglycerine on regional myocardial bloodflow in coronary artery disease. J Clin Invest 1971;50: 1578-84.

Requests for reprints to Dr D S Dymond, Depart-ment of Cardiology, St. Bartholomew's Hospital,London EClA 7BE.

133



/B


ebruary 1980. Dow

nloaded from


quantitative radionuclide haemodynamic rest exercise*

Documents