- Email: [email protected]
Quantitative measurement of electrical instability as a function of myocardial infarct size in the dog
Quantitative Measurement of Electrical Instability as a Function of Myocardial Infarct Size in the Dog
BEVERLY A. JONES-COLLINS, MD* RANDOLPH E. PATTERSON, MD, FACC Bethesda,
Mary/and
From the Cardiology Branch, National Heart, Lung, and Blood Institute, National Institutes of Health, Bethesda, Maryland 20205. Manuscript received March 18, 1981; revised manuscript received June 2. 1981, accepted June 11, 1981. Supported by a Fellowship from the Medical Research Council of Canada, Ottawa, Ontario, Canada. Address for reprints: Beverly Jones-Collins, MD, National Institutes of Health, National Heart, Lung, and Blood Institute, Building 10, Room 7515, Bethesda, Maryland 20205. l
858
November 1981
To investigate the relation between electrical instability and myocardial infarct size, 20 foxhounds were studied in the awake state 3 to 5 days afler closed chest coronary occlusion. Programmed right ventricular stimulation was performed with use of an epicardial electrode. After six paced beats at 10 percent greater than control rate, single and then double extrastimuli were introduced, scanning from late diastole to ventricular refractoriness in steps of 10 to 20 ms. Abnormal responses observed after this provocation were repetitive ventricular response, unsustained ventricular tachycardia, sustained ventricular tachycardia and ventricular fibrillation. Scores for electrical instability were determined for each dog, with higher scores assigned for more hazardous tachyarrhythmias (ventricular fibrillation greater than sustained ventricular tachycardia greater than unsustained ventricular tachycardia greater than repetitive ventricular response) and for those provokable later in diastole. An electrical instability index derived from these scores correlated well with infarct size measured with tetrazolium staining (r = 0.94). When scores were given only for the type of abnormal response elicited, excluding the effect of diastolic timing and the number of extrastimuli or vice versa, there was no significant difference in correlation with infarct size (r = 0.85 versus 0.92). Thus the results demonstrate that inducible electrical instability early after infarction is directly related to infarct size. Further, these data demonstrate the usefulness of an electrical instability index derived from the results of programmed right ventricular stimulation in assessing the severity of ischemic damage to the heart.
The predisposing factors for postinfarction ventricular arrhythmias are ill defined, although indirect clinical evidence supports the possibility that infarct size is an important determining factor. Specifically, an association has been reported between the incidence of postinfarction ventricular arrhythmias and each of the following: the incidence of clinically detectable cardiac failure and cardiogenic shock,l maximal serum glutamic oxaloacetic transaminase (SGOT) levels,1,2 serum creatine kinase (CK) estimates of infarct size,3,4 left ventricular ejection fraction and percent akinesia measured with gated cardiac blood pool scintigraphy5 and angiographic estimates of extent of coronary artery disease.6 These results are compatible with the concept that a larger mass of necrotic muscle may predispose to greater heterogeneity of myocardial excitability and refractoriness, promoting arrhythmias and sudden death. However, they are in conflict with the findings of Lawrie et aL7 who reported no correlation between SGOT levels, CK levels or the incidence of clinically detectable heart failure and the occurrence of ventricular fibrillation after infarction. Also, a major problem with these clinical data is the lack of direct measurements of infarct size. (For example, a correlation is not always found between infarct size and enzyme or isoenzyme quantification.) The present study was therefore designed to investigate the relation between myocardial infarction size and electrical instability in dogs 3
The American Journal of CARDIOLOGY
Volume 48
POSTINFARCTION ELECTRICAL INSTABILITY-JONES-COLLINS
to 5 days after acute myocardial infarction. Electrical instability was defined as the ease of induction of ventricular arrhythmias by programmed right ventricular stimulation. Abnormal responses produced during the
electrophysiologic testing protocol described by Josephson et al.8 were given numerical scores. These scores were then used to derive an electrical instability index for each dog that was in turn correlated with infarct size. Methods
Animal Preparation Twenty foxhounds of either sex, weighing 22 to 34 kg, were studied after thoracotomy was performed to implant the following equipment: catheters in the left atrium (by direct puncture) and aorta (by way of the internal mammary artery), a deflated balloon cuff around the circumflex (18dogs) or left anterior descending (2 dogs) coronary artery, a snare around the coronary artery at the site of balloon placement (and in 8 dogs with a circumflex arterial balloon, an additional snare around the left anterior descending artery) and an epicardial right ventricular pacing electrode (bipolar in 19 dogs and unipolar in 1 dog). The ends of these pieces of equipment were tunneled to a subcutaneous pouch behind the dog’s neck. A mean time of 18.5 days was allowed for recovery from surgery. During this recovery period the dogs were trained to lie on a table for study in the conscious state. On return for study, with the aid of morphine sedation (10 to 20 mg intramuscularly or intravenously as needed to a maximum of 60 mg over 4 hours) and local anesthetic (20 to 30 mg of lidoCaine), the subcutaneous pouch was opened and the surgically placed equipment exteriorized. Coronary occlusion: After prophylactic lidocaine treatment (1 mg/kg body weight intravenous bolus injection followed by a 3 mg/min constant infusion) and continuous electrocardiographic, left atrial and aortic pressure monitoring was begun, the coronary arterial balloon was inflated to l/4 volume. At 15 minute intervals another l/4 of the predetermined total balloon volume was injected to produce a gradual occlusion over 45 minutes. At 45 minutes the snare previously positioned at the site of balloon placement was pulled and secured, ensuring total occlusion. Five minutes later the second snare was pulled and secured in the eight dogs that had had two snares implanted. Eight dogs underwent both circumflex and left anterior descending coronary arterial occlusions, whereas 12 underwent a single vessel occlusion (of the left circumflex coronary artery in 10 and of the left anterior descending coronary artery in 2). All surviving dogs were monitored for 3 hours after occlusion before they were returned to their kennel for 3 to 5 days,
Pacing Protocol At the final stage of the study the free ends of the pressure catheters and pacing electrode were again exteriorized. Continuous aortic and left atria1 pressures and a surface electrocardiogram were recorded on a Sanborn 350 series recorder. A Bloom constant current pulse generator was used for pacing, which was performed at twice diastolic threshold or 10 milliamperes (the stimulator’s maximal current output), whichever was less. Each stimulus was 1.5 ms in duration. The dogs were subjected to the following protocol: Pacing (Sr-Si) at a cycle length 10 percent less than the shortest spontaneous cycle length was instituted and continued for six beats, followed by a 4 second pause. The six beat cycle was repeated after each pause. During the pause extrastimuli were introduced as follows: Step 1: An extrastimulus (Sz) was in-
AND PATTERSON
traduced at end-diastole and the coupling interval (Si-S2) was decreased in steps of 20 ms until the refractory period was reached. Step 2: Sr-Ss was then maintained at 50 ms plus the refractory period. An Ss was introduced at twice the Si-Sz interval and the Si-Ss interval was decreased in 20 ms steps until the ventricle was again refractory. Step 3: Finally, Ss was maintained at the refractory interval determined in Step 2 (which ceased to be refractory as Sa was moved) and Sz was moved closer to the T wave by 10 ms decrements until the ventricle was refractory to it once more. Each coupling interval of Si-Sz and Sr-Sz-Ss was repeated three times. Ventricular arrhythmia responses: During the pause after each Sp or Ss-Ss the following responses could be observed: a normal response (that is, a ventricular response to the extrastimulus or extrastimuli, followed by normal sinus rhythm), a repetitive ventricular response (two or three vent&_&r beats following the last extrastimulus before the onset of sinus rhythm), unsustained ventricular tachycardia (four or more ventricular beats lasting less than 4 seconds), SUSmined ventricular tachycardia (lasting 4 seconds or more) and ventricular fibrillation. All responses were observed and recorded from a surface electrocardiogram at a paper speed of 25 mm/s. A response was scored only if it was reproducible. The only responses excluded from analysis were those ventricular beats following the last extrastimulus after a delay greater than the Si-Si interval. These were interpreted as escape beats. The only treatment given for ventricular arrhythmia was a 420 watt second countershock for ventricular fibrillation and pacing conversion by bursts of rapid ventricular stimulation for sustained ventricular tachycardia. If ventricular fibrillation was reproducible at any point in the study, the study was halted.
Measurement of Risk Region and Infarct Size At the conclusion of the pacing protocol all dogs were given 10,009 units of heparin intravenously (to keep the coronary arteries patent), then killed and their hearts excised. The region at risk of infarction (area of the left ventricle supplied by the occluded artery or arteries) was stained by intraarterial injection of Evans blue at the site or sites of occlusion. After the right ventricle and atria were excised the heart was sliced from apex through base into 6 to 11 (mean 9) breadloaf slices 0.5 to 1 cm thick, and each was weighed. The border of bluestained region at risk and unstained normal myocardium was marked by a scalpel slash to allow identification of the region at risk after loss of Evans blue in the tetrazolium bath. Infarct staining was done using triphenyl tetrazolium chloride dye.g The slices were photographed and the photographs subjected to video planimetry to obtain the percent by area of infarct, region at risk and normal myocardium for both sides of each slice. By averaging these data for the two sides of each slice, multiplying by the weight of the slice and summing the values obtained, the percent by weight of infarct, region at risk and normal myocardium for the left ventricle was obtained. The region “naturally salvaged” was defined as the area of the region at risk not infarcted, expressed as a percent (by weight) of the left ventricle or region at risk.
Analysis of Data To reduce the many responses of each dog to a manageable number for data analysis the Si-Si cycle length divided by 2 was considered the midpoint of diastole, dividing it into two halves. For steps 1 and 2 of the pacing protocol (decreasing Sr-Sz and Si-Ss cycle lengths), subjects were given credit for the occurrence of a particular abnormal response only once for each half of diastole, whether it was produced by stimu-
November 1991
The American Journal of CARDIOLOGY
Volume 49
959
POSTINFARCTION ELECTRICAL INSTABILITY-JONES-COLLINS
AND PATTERSON
lation at 1 or at each 20 ms interval in that diastolic half. For step 3 of the protocol (decreasing &-Se cycle length in early diastole with Sa fixed at the previously refractory period) all responses were grouped together because the diastolic testing interval was so short. Subjects were again credited with the occurrence of an abnormal response only once, whether it was produced by stimulation at 1 or at each 10 ms testing interval. Thus the many responses of each dog were reduced to a maximum of 20. Each of the four abnormal responses could be scored for its occurrence after provocation during each of the following five stages: (1) Sz testing in the last half of diastole, (2) Sz testing in the first half of diastole, (3) Sa testing in the last half of diastole, (4) Sa testing in the first half of diastole, and (5) decreasing Si-Sz cycle length in early diastole with Ss fixed at the previously iefractory period. Electrical instability indexes: Three numerical scoring systems were developed (Table 1). Each system consists of 20 scores, one for each of the abnormal responses provokable during each of the five stages of the pacing protocol. The scores for responses elicited at each stage were added together to produce one number for each subject, for each electrical instability scoring system. These three numbers are called electrical instability indexes A, B and C. Electrical instability scoring system A scores selectively for the number of extrastimuli (Sz or S&s) required to produce an abnormal response, regardless of its type, and the time in diastole of the stimulus (or stimuli) required. (Any abnormal response was given a score of 20 if produced at stage 1, a score of 15 at stage 2, 10 at stage 3,5 at stage 4 and 1 at stage 5). Scoring system B scores selectively for the type of abnormal responses pro-
15 15 15
duced, regardless of the diastolic timing and the number of stimuli required to produce them. (Ventricular fibrillation provokable at any stage was scored as 15 points, sustained ventricular tachycardia as 10 points, unsustained ventricular tachycardia as 5 points, and a repetitive ventricular response as 1 point.) Scoring system C is a composite of the two systems produced by simple addition. Thus the electrical instability scoring systems variably reflect the type of ventricular arrhythmias produced by stimulation and the number and diastolic timing of stimuli required to produce them. Scoring method: To assess the separate importance of each type of abnormal response in predicting infarct size, a single score (using system A) was given for the first occurrence of each of the four abnormal responses. For example, Dog 6 had no repetitive ventricular response provoked by stimulation during stage 1 but manifested this response with an Sz in the first half of diastole (stage 2). This dog’s score for the first occurrence of repetitive ventricular response was therefore 15 (Table IA), regardless of how many such responses were observed later in the study. Likewise, for unsustained ventricular tachycardia, sustained ventricular tachycardia and ventricular fibrillation, the score (using system A) at the first occurrence of each in each dog was noted and individually correlated with myocardial infarct size. Similarly, to assess the independent contribution of each timing interval/extrastimuli combination in predicting infarct size, five scores were assigned for each dog (using system B), summing all abnormal responses provoked by stimulation during each of the five diastolic timing interval/extrastimuli combinations (stages 1 to 5). For example, Dog 10 had no abnormal responses with stimulation during stage 1 but had a repetitive ventricular response with stimulation during stage 2 and a repetitive ventricular response and sustained ventricular tachycardia during stage 3. The animal was refractory to stimulation during stage 4 but experienced unsustained and sustained ventricular tachycardia and ventricular fibrillation during stage 5. This dog’s scores for each stage are as follows: stage 1 = 0,2 = 1,3 = 11,4 = 0 and 5 = 30 (Table IB). Statistical analysis: Each of the electrical instability indexes and other tabulated scores were correlated with infarct size, size of region at risk and percent of the left ventricle and region at risk naturally salvaged. Infarct size was correlated with percent of the left ventricle and region at risk naturally salvaged. Index A results were correlated with index B results for each dog to provide a comparison of the two systems of scoring. We calculated the best fit linear regression equation by the method of least squares, the standard error of the estimate and the correlation coefficient (r) for these pairs of data.
:: 10
::
Results
z: 20 15 11
35 30 25 20 16
TABLE I Scoring
Systems A, B and C’
Diastolic Timing/Number of Extrastimuli Combination
RVR
VT
sVT
VF
20
20
;: 5 1
:: 5 1
20 15 10 5 1
A. Scoring System A 1. 2. 3. 4. 5.
Ss 2nd half of diastole Ss 1st half of diastole Ss-Ss 2nd half of diastole S&s 1st half of diastole Early Ss with refractory Ss
1. 2. 3. 4. 5.
Sp 2nd half of diastole Ss 1st half of diastole S&s 2nd half of diastole S&s 1st half of diastole Early Ss with refractory Ss
20 15 10 5 1
B. Scoring System B 1
5 :“o
:
:
:
:
C. Scoring System C I. 2. 3. 4. 5.
Sp 2nd half of diastole Ss 1st half of diastole Sp-Ss 2nd half of diastole S&s 1st half of diastole Early Ss with refractory Ss
21 16 11 !
:: 15 10 6
In scoring system A, the same score is given for any response, regardless of its type, elicited by stimulation at a particular diastolic timing interval and by the number of extrastimuli specified. in scoring system B, the same score is given for a particular abnormal response regardless of the diastolic timing or the number of stimuli required to provoke it. In scoring system C scores A and B are added to give a composite score accounting for the number and diastolic timing of extrastimuli required to produce an abnormal response, and the type of response elicited. RVR = repetitive ventricular response; sVT = sustained ventricular tachycardia; VF = ventricular fibrillation; VT = unsustained ventricular tachycardia. l
880
November 1981
The American Journal of CARDIOLOGY
Four dogs died during or early after occlusion, before electrophysiologic study. Three dogs did not complete the stimulation protocol because of a pacing electrode malfunction. The remaining 13 dogs were studied. The mean left atria1 pressure was 9 f 5 mm Hg and mean aortic pressure 111 f 17 mm Hg during the pacing study. Electrical instability index: This index correlated strongly with infarct size (Fig. 1 and 2), regardless of the scoring scheme used (r = 0.92,0.85 and 0.94 for schemes A, B and C, respectively), Table II shows the infarct size and electrical instability indexes and scores for each dog as well as the correlation coefficients for infarct size versus electrical instability index and each of the other
Volume 48
AND PATTERSON
POSTINFARCTION ELECTRICAL INSTABILITY4ONES-COLLlNS
FIGURE 1. Left, electrical instability (El) index A, derived from programmed right ventricular stimulation data, is correlated with the percent of the left ventricle (LV) infarcted. This index reflects the number of stimuli required to produce an abnormal response (that is, single or double) and their diastolic timing, regardless of response type (Table IA). Right, electrical instability index B is also correlated with the percent of the left ventricle infarcted. This index reflects the types of arrhythmia provoked by stimulation, regardless of the diastolic timing or number of stimuli required to produce them (Table IB). Ml = myocardial infarction; N = number of dogs: r = correlation coefficient; SEE = standard error of the estimate.
SEE=7.7
Ml SIZE (% LV)
tabulated scores. When the scoring systems were repeated using the same format but the lowest possible scores at each interval (1,2,3,4 and 5 instead of 1,5,10,15 and 20) to evaluate the effect of reduced dispersion of data, the correlation coefficients obtained were unchanged. The results were also unchanged when the dog with the unipolar pacing wire (Dog 9) was excluded from analysis. Ventricular tachycardia: The type of arrhythmia that correlated best with myocardial infarct size (when each response was scored only on its first occurrence, using scoring system A) was sustained ventricular tachycardia (r = 0.90) (Fig. 3). No dog with an infarct size less than 20 percent of the left ventricle exhibited this arrhythmia. Responses at each of the diastolic timing interval/extrastimuli combinations: When these responses were individually correlated with infarct size (using scoring system B), the best correlation was found with stimulation by a single extrastimulus (Sz) in the first half of diastole (r = 0.83, Table II). However, the poorer correlation at seemingly more vulnerable diastolic positions is partially due to exclusion of dogs experiencing ventricular fibrillation early in the protocol, from later analysis (46 percent of the dogs manifested ventricular fibrillation during the protocol, 31 percent when Sz was introduced into the first half of diastole). Size of risk region: No correlation existed between the size of the region at risk and the electrical instability index (r = 0.08 using index C). There was a strong negative correlation between the percent of the left ventricle or region at risk naturally salvaged and the electrical instability index (r = -0.73 and -0.88, respectiveiy, using index C). As expected, negative correlations were also observed for percent salvage of left ventricle versus percent of the left ventricle infarcted (r = -0.60) and percent salvage of risk region versus percent of the left ventricle infarcted (r = -0.80). Discussion The results of this investigation provide unique direct evidence that the degree of electrical instability corre-
Ml SIZE (% LVI
lates strongly with infarct size 3 to 5 days after acute myocardial infarction in the dog. These data support previous clinical observations that patients with larger CK curve estimates of infarct size, higher SGOT levels, poorer global left ventricular function and angiographic evidence of more extensive coronary artery diseasel-a demonstrate more frequent and complex arrhythmias after infarction. The effects of anesthesia and surgical trauma were minimized by studying dogs in the awake state after recovery from thoracotomy. No drugs were given at the time of the pacing study other than morphine for sedation and lidocaine given subcutaneously as a local anesthetic agent 30 to 60 minutes before the pacing
y=2.lx+5.7 P.34 n=13 SEE=11.2
& 0’10
I 20
30
40
50
Ml SIZE (% LV 1 FIGURE 2. Electrical instability (El) index C. derived from programmed right ventricular stimulation data is correlated with the percent of the left ventricle (LV) infarcted. This index reflects the types of arrhythmia produced by stimulation, as well as the number and diastolic timing of stimuli required to produce them (Table C). Abbreviations as before.
November 1981
The American Journal ol CARDIOLOGY
Volume 48
861
POSTINFARCTION ELECTRICAL INSTABILITY-JONES-COLLINS AND PATTERSON
TABLE
II
Electrical
Instability
Indexes
and Score Results Score (System B) at Each Diastolic Timing/Extrastimuli Combination
Electrical Instability Index Dog
% LV Infarcted
B
A
Score (System A) at First Occurrence RVR
C
VT
sVT
S2 2nd ‘I2 of Diastole
VF
s2-s3
s2
1st ‘12 of Diastole
s2-s3
2nd ‘1s of Diastole
1st ‘Is of Diastole
Early Sp With Refractory S3
1 :
4 0”
z
4 5
6 7 : :‘: :z r* =
:;
11 :
:
:
10 :
:
:
:0
0 :
0 :
:
:
:1
1
1;
19 :“6
0
:0
:
1
:
:
:,
:1
151
: 15 ::
z;
:: 1:
:
: 1:
: 15 :
:
; 1:
:
:
:
1 15 :
:
:
z
::
::
15 :z
1:
1:
0
3:
1:0
:
300
z 0.;:
;: 0.85
:: 0.94
15 0.380
: 0.33
0
;: 0.83
: 0.13
:; :z
;: 3’: ::
2:
:: 15 1: 15 0.9’0”
15 0.::
0.::
: -0.25
: -0.003
Correlation between percent of the left ventricle infarcted and each measurement of electrical instability. LV = left ventricle; RVR = repetitive ventricular response; S2 and S3 = first and second premature stimuli; sVT = sustained ventricular tachycardia; VF = ventricular fibrillation: VT = unsustained ventricular tachycardia. l
study. This small amount of locally injected lidocaine would not be expected to affect the results of electrophysiologic testing.lO Mechanism of observed ventricular arrhythmias and correlation with infarct size: No attempt was made to determine the mechanism of the observed arrhythmias. Bundle branch reentry-the phenomenon of retrograde, then anterograde conduction of an extrastimulus through the bundle branches to reexcite the ventricles-cannot be excluded as a mechanism for some of the responses observed because no endocardial recordings were obtained. However, excluding this mechanism as a cause of the reported observations would not be appropriate, because it may have a role in the initiation of postinfarction arrhythmias.li Also, the lack of multiple recordings made it impossible to determine the site of origin of the induced arrhythmias. However, the observed strong correlation between infarct size and ventricular arrhythmias, regardless of origin, is of primary interest here. If myocardial infarction causes some change in noninfarcted tissue leading to induction of arrhythmia, and the reported
l /
J. .
g
I=.90
/ 5
y= 5x-2.5
IF13 SEE=3.5
::
O-?ElrGr
40
MI SIZE I% LV)
FIGURE 3. Electrical instability score (using scoring system A) at the first occurrence of sustained ventricular tachycardia @VT) is correlated with the percent of the left ventricle (LV) infarcted. No dog with an infarct size of less than 20 percent of the left ventricle experienced this arrhythmia. Abbreviations as before.
882
November 1981
The American Journal of CARDIOLOGY
responses were demonstrated to originate outside the infarct zone, they are not any less important. The protocol was carried out at only one cycle length and stimulation site because the purpose of the experiment was to define electrical instability in a general sense, not to stimulate in a multitude of ways until a previously demonstrated specific arrhythmia was reproduced. The ease of induction of ventricular arrhythmias by the experimental protocol, using one cycle length and one stimulation site per dog was tested and proved to be a good indicator of severity of disease. The right ventricle, remote from the septum, was selected as the site for pacing electrode placement to be sure it would not overlie any infarcted tissue at the time of the pacing study. Quantitating electrical instability-validity of method: The problem of quantitating electrical instability has concerned many previous investigators.12 Arrhythmias induced in the laboratory are being used to investigate and evaluate patients’ and experimental subjects’ predisposition to spontaneously occurring arrhythmias. Various methods of stimulation have been used and a variety of responses have been observed.ls17 In this study severity of disease (that is, infarct size) could be measured directly and was shown to correlate well with electrical instability indexes. These correlations support the validity of the method of stimulation and interpretation used. The scoring systems focus on two principles: First, it has been shown that stimulation early in diastole, near the T wave (“vulnerable period”), is able to produce ventricular arrhythmias in normal dogs, whereas stimulation elsewhere in diastole is not.ls Therefore, for score A lower numbers were assigned to responses to stimulation early in diastole, when vulnerability is greater, and higher numbers to abnormal responses to stimulation in late diastole, when propensity for arrhythmia development should be low, despite provo-
Volume 48
POSTINFARCTION ELECTRICAL INSTABILITY-JONES-COLLINS
cation. Second, it has been shown that repetitive ventricular responses and ventricular tachycardia, as indexes of ventricular vulnerability, occur reproducibly in response to lower amplitude stimuli than does ventricular fibrillation when elicited at the same diastolic position. lg Therefore, the repetitive ventricular response, unsustaided ventricular tachycardia, sustained ventricular tachycardia and ventricular fibrillation were considered to indicate increasing severity of electrical instability in that order when produced by an extrastimulus of fixed amplitude and at the same time in diastole. Hence, for score B, progressively higher numbers were assigned to progressively more hazardous and complex arrhythmias. Correlation with infarct size: The correlation between infarct size and electrical instability index A (r = 0.92) was not significantly different from that between infarct size and electrical instability index B (r = 0.85). Moreover, each dog’s index A was well correlated with the same dog’s index B (r = 0.78). Electrical instability index C was designed ta account for diastolic time, number of extrastimuli required and type of arrhythmic response together. This composite index also correlated well with infarct size (r = 0.94) but did not significantly improve the correlation of either index alone. Of particular note is the finding that the electrical instability index was unrelated to size of region at risk, despite such a strong correlation with infarct size. This
AND PATTERSON
indicates that the crucial factor in determining electrical instability 3 to 5 days after infarction may be mass of necrotic myocardium rather than mass of myocardium initially at risk of infarction. A strong negative correlation between percent region at risk salvaged (at risk but not infarcted) and the electrical instability index was demonstrated. We considered this result primarily due to the strong negative correlation between infarct size and region at risk salvaged. Although the two dogs without infarction may not have actually experienced total coronary occlusion, they serve to illustrate the results of this electrophysiologic protocol in dogs without infarcts. Not one abnormal response was produced in either dog. Implications: These data provide direct evidence that induced electrical instability is strongly correlated with infarct size. Furthermore, the scoring system introduced here provides a unique quantitative method for deriving an electrical instability index from programmed right ventricular stimulation data. This derived index may eventually prove to be a very useful investigative and perhaps clinical tool.
Acknowledgment We thank William Parker for technical assistance, Sharla Goldstein for preparing the manuscript and Stephen E. Epstein, MD for editing it,
References 1. Morgensen L. A clinical study of cardiac arrhythmias in 421 consecutive patients with acute myocardial infarction treated in a coronary care unit. Acta MedStand 1971;513:10-29. 2. Chapman BL. Relation of cardiac complications to SGOT level in acute myocardial infarction. Br Heart J 1972;34:890-6. 3. Roberts R, Husaln A, Ambos HD, Oliver GC, Cox JR, Sobel BE. Relation between infarct size and ventricular arrhythmia. Br Heart J 1975;37:1169-75. 4. Geftman EM, Ehsanl AA, Campbell MK, Schechtman K, Roberls R, Sobel BE. The influence of location and extent of myocardial infarction on long term dysrhythmia and mortality. Circulation 1979;60:805-14. 5. Schulze RA, Rouleau J, Rlgo P, Sowers S, Strauss HW, Pltl B. Ventricular arrhythmias in the late hospital phase of acute myocardial infarction: relation to left ventricular function detected by gated cardiac blood pool scanning. Circulation 1975:52:100611. 6. Schulze RA, Humphrles JO, Grffflth LS, et al. Lett ventricular and coronary angiographic anatomy. Relationship lo ventricular irritability in the late hospital phase of acute myocardial infarction. Circulation 1977;55:839-43. 7. Lawrle DM, Hlgglns MR, Goodman MJ, Ollver MF, Jullan DG, Donald KW. Ventricular fibrillation complicating acute myocardial infarction. Lancet 1968;2:523-8. 6. Josephson ME, Horowitz LN, Farshldl A, Kastor JA. Recurrent sustained ventricular tachycardia. I. Mechanisms. Circulation 1978571431-40. 9. Lle JT, Palrofero PC, Holtey KE, Tltus JL. Macroscopic enzymemapping verification of large, homogeneous, experimental myocardial infarcts of predictable size and location in dogs. J Thorac
Cardiovasc Surg 1975;69:599-605. 10. Nattel S, Rlnkenberger RL, Lehrman LL, Zl~es DP. Therapeutic blood lidocaine concentrations after local anesthesia for cardiac electrophysiologic studies. N Engl J h4ed 1979;301:418-20. 11. Foster JR, Simpson RJ. Initiation of ventricular tachycardia by reentry within the bundle branches. Am J Cardiol 1980;45:895900. 12. Kent KM. Assessing the electrical stability of the heart in human beings. Am J Cardiol 1980;45:1305-7. 13. Han J. Ventricular vulnerability during acute coronary occlusion. Am J Cardiol 1989;24:857-64. 14. Jenzer H, Lohrbauer L, Lown 5. Response to single threshold stimuli following acute myocardial infarction. Proc Sot Exp Biol Med 1972;141:606-8. 15. Wellens HJ, Schullenburg RM, Dorrer D. Electrical stimulation of the heart in patients with ventricular tachycardia. Circulation 1972;46:216-26. 16. Denes P, Wu D, Dhlngra RC, et al. Electrophysiological studies in patients with chronic recurrent ventricular tachycardia. Circulation 1976;54:229-38. 17. Flsher JD, Cohen HL, Mehra R, Altschuler H, Escher DJW, Furman S. Cardiac pacing and pacemakers. II. Serial electrophysiologic-pharmacologic testing for control of recurrent tachyarrhythmias. Am Heart J 1977;93:658-68. 18. Wlggers CJ, Wegrla R. Ventricular fibrillation due to single localized induction and condenser shocks applied during the vulnerable phase of ventricular systole. Am J Physiol 1940;128:500-5. 19. Matta RJ, Verrler RL, Lown 6. Repetitive extrasystole as an index of vulnerability to ventricular fibrillation. Am J Physiol 1978;230: 1469-73.
November 1981
The American Journal of CARDIDLDGY
Volume 48
863
BEVERLY A. JONES-COLLINS, MD* RANDOLPH E. PATTERSON, MD, FACC Bethesda,
Mary/and
From the Cardiology Branch, National Heart, Lung, and Blood Institute, National Institutes of Health, Bethesda, Maryland 20205. Manuscript received March 18, 1981; revised manuscript received June 2. 1981, accepted June 11, 1981. Supported by a Fellowship from the Medical Research Council of Canada, Ottawa, Ontario, Canada. Address for reprints: Beverly Jones-Collins, MD, National Institutes of Health, National Heart, Lung, and Blood Institute, Building 10, Room 7515, Bethesda, Maryland 20205. l
858
November 1981
To investigate the relation between electrical instability and myocardial infarct size, 20 foxhounds were studied in the awake state 3 to 5 days afler closed chest coronary occlusion. Programmed right ventricular stimulation was performed with use of an epicardial electrode. After six paced beats at 10 percent greater than control rate, single and then double extrastimuli were introduced, scanning from late diastole to ventricular refractoriness in steps of 10 to 20 ms. Abnormal responses observed after this provocation were repetitive ventricular response, unsustained ventricular tachycardia, sustained ventricular tachycardia and ventricular fibrillation. Scores for electrical instability were determined for each dog, with higher scores assigned for more hazardous tachyarrhythmias (ventricular fibrillation greater than sustained ventricular tachycardia greater than unsustained ventricular tachycardia greater than repetitive ventricular response) and for those provokable later in diastole. An electrical instability index derived from these scores correlated well with infarct size measured with tetrazolium staining (r = 0.94). When scores were given only for the type of abnormal response elicited, excluding the effect of diastolic timing and the number of extrastimuli or vice versa, there was no significant difference in correlation with infarct size (r = 0.85 versus 0.92). Thus the results demonstrate that inducible electrical instability early after infarction is directly related to infarct size. Further, these data demonstrate the usefulness of an electrical instability index derived from the results of programmed right ventricular stimulation in assessing the severity of ischemic damage to the heart.
The predisposing factors for postinfarction ventricular arrhythmias are ill defined, although indirect clinical evidence supports the possibility that infarct size is an important determining factor. Specifically, an association has been reported between the incidence of postinfarction ventricular arrhythmias and each of the following: the incidence of clinically detectable cardiac failure and cardiogenic shock,l maximal serum glutamic oxaloacetic transaminase (SGOT) levels,1,2 serum creatine kinase (CK) estimates of infarct size,3,4 left ventricular ejection fraction and percent akinesia measured with gated cardiac blood pool scintigraphy5 and angiographic estimates of extent of coronary artery disease.6 These results are compatible with the concept that a larger mass of necrotic muscle may predispose to greater heterogeneity of myocardial excitability and refractoriness, promoting arrhythmias and sudden death. However, they are in conflict with the findings of Lawrie et aL7 who reported no correlation between SGOT levels, CK levels or the incidence of clinically detectable heart failure and the occurrence of ventricular fibrillation after infarction. Also, a major problem with these clinical data is the lack of direct measurements of infarct size. (For example, a correlation is not always found between infarct size and enzyme or isoenzyme quantification.) The present study was therefore designed to investigate the relation between myocardial infarction size and electrical instability in dogs 3
The American Journal of CARDIOLOGY
Volume 48
POSTINFARCTION ELECTRICAL INSTABILITY-JONES-COLLINS
to 5 days after acute myocardial infarction. Electrical instability was defined as the ease of induction of ventricular arrhythmias by programmed right ventricular stimulation. Abnormal responses produced during the
electrophysiologic testing protocol described by Josephson et al.8 were given numerical scores. These scores were then used to derive an electrical instability index for each dog that was in turn correlated with infarct size. Methods
Animal Preparation Twenty foxhounds of either sex, weighing 22 to 34 kg, were studied after thoracotomy was performed to implant the following equipment: catheters in the left atrium (by direct puncture) and aorta (by way of the internal mammary artery), a deflated balloon cuff around the circumflex (18dogs) or left anterior descending (2 dogs) coronary artery, a snare around the coronary artery at the site of balloon placement (and in 8 dogs with a circumflex arterial balloon, an additional snare around the left anterior descending artery) and an epicardial right ventricular pacing electrode (bipolar in 19 dogs and unipolar in 1 dog). The ends of these pieces of equipment were tunneled to a subcutaneous pouch behind the dog’s neck. A mean time of 18.5 days was allowed for recovery from surgery. During this recovery period the dogs were trained to lie on a table for study in the conscious state. On return for study, with the aid of morphine sedation (10 to 20 mg intramuscularly or intravenously as needed to a maximum of 60 mg over 4 hours) and local anesthetic (20 to 30 mg of lidoCaine), the subcutaneous pouch was opened and the surgically placed equipment exteriorized. Coronary occlusion: After prophylactic lidocaine treatment (1 mg/kg body weight intravenous bolus injection followed by a 3 mg/min constant infusion) and continuous electrocardiographic, left atrial and aortic pressure monitoring was begun, the coronary arterial balloon was inflated to l/4 volume. At 15 minute intervals another l/4 of the predetermined total balloon volume was injected to produce a gradual occlusion over 45 minutes. At 45 minutes the snare previously positioned at the site of balloon placement was pulled and secured, ensuring total occlusion. Five minutes later the second snare was pulled and secured in the eight dogs that had had two snares implanted. Eight dogs underwent both circumflex and left anterior descending coronary arterial occlusions, whereas 12 underwent a single vessel occlusion (of the left circumflex coronary artery in 10 and of the left anterior descending coronary artery in 2). All surviving dogs were monitored for 3 hours after occlusion before they were returned to their kennel for 3 to 5 days,
Pacing Protocol At the final stage of the study the free ends of the pressure catheters and pacing electrode were again exteriorized. Continuous aortic and left atria1 pressures and a surface electrocardiogram were recorded on a Sanborn 350 series recorder. A Bloom constant current pulse generator was used for pacing, which was performed at twice diastolic threshold or 10 milliamperes (the stimulator’s maximal current output), whichever was less. Each stimulus was 1.5 ms in duration. The dogs were subjected to the following protocol: Pacing (Sr-Si) at a cycle length 10 percent less than the shortest spontaneous cycle length was instituted and continued for six beats, followed by a 4 second pause. The six beat cycle was repeated after each pause. During the pause extrastimuli were introduced as follows: Step 1: An extrastimulus (Sz) was in-
AND PATTERSON
traduced at end-diastole and the coupling interval (Si-S2) was decreased in steps of 20 ms until the refractory period was reached. Step 2: Sr-Ss was then maintained at 50 ms plus the refractory period. An Ss was introduced at twice the Si-Sz interval and the Si-Ss interval was decreased in 20 ms steps until the ventricle was again refractory. Step 3: Finally, Ss was maintained at the refractory interval determined in Step 2 (which ceased to be refractory as Sa was moved) and Sz was moved closer to the T wave by 10 ms decrements until the ventricle was refractory to it once more. Each coupling interval of Si-Sz and Sr-Sz-Ss was repeated three times. Ventricular arrhythmia responses: During the pause after each Sp or Ss-Ss the following responses could be observed: a normal response (that is, a ventricular response to the extrastimulus or extrastimuli, followed by normal sinus rhythm), a repetitive ventricular response (two or three vent&_&r beats following the last extrastimulus before the onset of sinus rhythm), unsustained ventricular tachycardia (four or more ventricular beats lasting less than 4 seconds), SUSmined ventricular tachycardia (lasting 4 seconds or more) and ventricular fibrillation. All responses were observed and recorded from a surface electrocardiogram at a paper speed of 25 mm/s. A response was scored only if it was reproducible. The only responses excluded from analysis were those ventricular beats following the last extrastimulus after a delay greater than the Si-Si interval. These were interpreted as escape beats. The only treatment given for ventricular arrhythmia was a 420 watt second countershock for ventricular fibrillation and pacing conversion by bursts of rapid ventricular stimulation for sustained ventricular tachycardia. If ventricular fibrillation was reproducible at any point in the study, the study was halted.
Measurement of Risk Region and Infarct Size At the conclusion of the pacing protocol all dogs were given 10,009 units of heparin intravenously (to keep the coronary arteries patent), then killed and their hearts excised. The region at risk of infarction (area of the left ventricle supplied by the occluded artery or arteries) was stained by intraarterial injection of Evans blue at the site or sites of occlusion. After the right ventricle and atria were excised the heart was sliced from apex through base into 6 to 11 (mean 9) breadloaf slices 0.5 to 1 cm thick, and each was weighed. The border of bluestained region at risk and unstained normal myocardium was marked by a scalpel slash to allow identification of the region at risk after loss of Evans blue in the tetrazolium bath. Infarct staining was done using triphenyl tetrazolium chloride dye.g The slices were photographed and the photographs subjected to video planimetry to obtain the percent by area of infarct, region at risk and normal myocardium for both sides of each slice. By averaging these data for the two sides of each slice, multiplying by the weight of the slice and summing the values obtained, the percent by weight of infarct, region at risk and normal myocardium for the left ventricle was obtained. The region “naturally salvaged” was defined as the area of the region at risk not infarcted, expressed as a percent (by weight) of the left ventricle or region at risk.
Analysis of Data To reduce the many responses of each dog to a manageable number for data analysis the Si-Si cycle length divided by 2 was considered the midpoint of diastole, dividing it into two halves. For steps 1 and 2 of the pacing protocol (decreasing Sr-Sz and Si-Ss cycle lengths), subjects were given credit for the occurrence of a particular abnormal response only once for each half of diastole, whether it was produced by stimu-
November 1991
The American Journal of CARDIOLOGY
Volume 49
959
POSTINFARCTION ELECTRICAL INSTABILITY-JONES-COLLINS
AND PATTERSON
lation at 1 or at each 20 ms interval in that diastolic half. For step 3 of the protocol (decreasing &-Se cycle length in early diastole with Sa fixed at the previously refractory period) all responses were grouped together because the diastolic testing interval was so short. Subjects were again credited with the occurrence of an abnormal response only once, whether it was produced by stimulation at 1 or at each 10 ms testing interval. Thus the many responses of each dog were reduced to a maximum of 20. Each of the four abnormal responses could be scored for its occurrence after provocation during each of the following five stages: (1) Sz testing in the last half of diastole, (2) Sz testing in the first half of diastole, (3) Sa testing in the last half of diastole, (4) Sa testing in the first half of diastole, and (5) decreasing Si-Sz cycle length in early diastole with Ss fixed at the previously iefractory period. Electrical instability indexes: Three numerical scoring systems were developed (Table 1). Each system consists of 20 scores, one for each of the abnormal responses provokable during each of the five stages of the pacing protocol. The scores for responses elicited at each stage were added together to produce one number for each subject, for each electrical instability scoring system. These three numbers are called electrical instability indexes A, B and C. Electrical instability scoring system A scores selectively for the number of extrastimuli (Sz or S&s) required to produce an abnormal response, regardless of its type, and the time in diastole of the stimulus (or stimuli) required. (Any abnormal response was given a score of 20 if produced at stage 1, a score of 15 at stage 2, 10 at stage 3,5 at stage 4 and 1 at stage 5). Scoring system B scores selectively for the type of abnormal responses pro-
15 15 15
duced, regardless of the diastolic timing and the number of stimuli required to produce them. (Ventricular fibrillation provokable at any stage was scored as 15 points, sustained ventricular tachycardia as 10 points, unsustained ventricular tachycardia as 5 points, and a repetitive ventricular response as 1 point.) Scoring system C is a composite of the two systems produced by simple addition. Thus the electrical instability scoring systems variably reflect the type of ventricular arrhythmias produced by stimulation and the number and diastolic timing of stimuli required to produce them. Scoring method: To assess the separate importance of each type of abnormal response in predicting infarct size, a single score (using system A) was given for the first occurrence of each of the four abnormal responses. For example, Dog 6 had no repetitive ventricular response provoked by stimulation during stage 1 but manifested this response with an Sz in the first half of diastole (stage 2). This dog’s score for the first occurrence of repetitive ventricular response was therefore 15 (Table IA), regardless of how many such responses were observed later in the study. Likewise, for unsustained ventricular tachycardia, sustained ventricular tachycardia and ventricular fibrillation, the score (using system A) at the first occurrence of each in each dog was noted and individually correlated with myocardial infarct size. Similarly, to assess the independent contribution of each timing interval/extrastimuli combination in predicting infarct size, five scores were assigned for each dog (using system B), summing all abnormal responses provoked by stimulation during each of the five diastolic timing interval/extrastimuli combinations (stages 1 to 5). For example, Dog 10 had no abnormal responses with stimulation during stage 1 but had a repetitive ventricular response with stimulation during stage 2 and a repetitive ventricular response and sustained ventricular tachycardia during stage 3. The animal was refractory to stimulation during stage 4 but experienced unsustained and sustained ventricular tachycardia and ventricular fibrillation during stage 5. This dog’s scores for each stage are as follows: stage 1 = 0,2 = 1,3 = 11,4 = 0 and 5 = 30 (Table IB). Statistical analysis: Each of the electrical instability indexes and other tabulated scores were correlated with infarct size, size of region at risk and percent of the left ventricle and region at risk naturally salvaged. Infarct size was correlated with percent of the left ventricle and region at risk naturally salvaged. Index A results were correlated with index B results for each dog to provide a comparison of the two systems of scoring. We calculated the best fit linear regression equation by the method of least squares, the standard error of the estimate and the correlation coefficient (r) for these pairs of data.
:: 10
::
Results
z: 20 15 11
35 30 25 20 16
TABLE I Scoring
Systems A, B and C’
Diastolic Timing/Number of Extrastimuli Combination
RVR
VT
sVT
VF
20
20
;: 5 1
:: 5 1
20 15 10 5 1
A. Scoring System A 1. 2. 3. 4. 5.
Ss 2nd half of diastole Ss 1st half of diastole Ss-Ss 2nd half of diastole S&s 1st half of diastole Early Ss with refractory Ss
1. 2. 3. 4. 5.
Sp 2nd half of diastole Ss 1st half of diastole S&s 2nd half of diastole S&s 1st half of diastole Early Ss with refractory Ss
20 15 10 5 1
B. Scoring System B 1
5 :“o
:
:
:
:
C. Scoring System C I. 2. 3. 4. 5.
Sp 2nd half of diastole Ss 1st half of diastole Sp-Ss 2nd half of diastole S&s 1st half of diastole Early Ss with refractory Ss
21 16 11 !
:: 15 10 6
In scoring system A, the same score is given for any response, regardless of its type, elicited by stimulation at a particular diastolic timing interval and by the number of extrastimuli specified. in scoring system B, the same score is given for a particular abnormal response regardless of the diastolic timing or the number of stimuli required to provoke it. In scoring system C scores A and B are added to give a composite score accounting for the number and diastolic timing of extrastimuli required to produce an abnormal response, and the type of response elicited. RVR = repetitive ventricular response; sVT = sustained ventricular tachycardia; VF = ventricular fibrillation; VT = unsustained ventricular tachycardia. l
880
November 1981
The American Journal of CARDIOLOGY
Four dogs died during or early after occlusion, before electrophysiologic study. Three dogs did not complete the stimulation protocol because of a pacing electrode malfunction. The remaining 13 dogs were studied. The mean left atria1 pressure was 9 f 5 mm Hg and mean aortic pressure 111 f 17 mm Hg during the pacing study. Electrical instability index: This index correlated strongly with infarct size (Fig. 1 and 2), regardless of the scoring scheme used (r = 0.92,0.85 and 0.94 for schemes A, B and C, respectively), Table II shows the infarct size and electrical instability indexes and scores for each dog as well as the correlation coefficients for infarct size versus electrical instability index and each of the other
Volume 48
AND PATTERSON
POSTINFARCTION ELECTRICAL INSTABILITY4ONES-COLLlNS
FIGURE 1. Left, electrical instability (El) index A, derived from programmed right ventricular stimulation data, is correlated with the percent of the left ventricle (LV) infarcted. This index reflects the number of stimuli required to produce an abnormal response (that is, single or double) and their diastolic timing, regardless of response type (Table IA). Right, electrical instability index B is also correlated with the percent of the left ventricle infarcted. This index reflects the types of arrhythmia provoked by stimulation, regardless of the diastolic timing or number of stimuli required to produce them (Table IB). Ml = myocardial infarction; N = number of dogs: r = correlation coefficient; SEE = standard error of the estimate.
SEE=7.7
Ml SIZE (% LV)
tabulated scores. When the scoring systems were repeated using the same format but the lowest possible scores at each interval (1,2,3,4 and 5 instead of 1,5,10,15 and 20) to evaluate the effect of reduced dispersion of data, the correlation coefficients obtained were unchanged. The results were also unchanged when the dog with the unipolar pacing wire (Dog 9) was excluded from analysis. Ventricular tachycardia: The type of arrhythmia that correlated best with myocardial infarct size (when each response was scored only on its first occurrence, using scoring system A) was sustained ventricular tachycardia (r = 0.90) (Fig. 3). No dog with an infarct size less than 20 percent of the left ventricle exhibited this arrhythmia. Responses at each of the diastolic timing interval/extrastimuli combinations: When these responses were individually correlated with infarct size (using scoring system B), the best correlation was found with stimulation by a single extrastimulus (Sz) in the first half of diastole (r = 0.83, Table II). However, the poorer correlation at seemingly more vulnerable diastolic positions is partially due to exclusion of dogs experiencing ventricular fibrillation early in the protocol, from later analysis (46 percent of the dogs manifested ventricular fibrillation during the protocol, 31 percent when Sz was introduced into the first half of diastole). Size of risk region: No correlation existed between the size of the region at risk and the electrical instability index (r = 0.08 using index C). There was a strong negative correlation between the percent of the left ventricle or region at risk naturally salvaged and the electrical instability index (r = -0.73 and -0.88, respectiveiy, using index C). As expected, negative correlations were also observed for percent salvage of left ventricle versus percent of the left ventricle infarcted (r = -0.60) and percent salvage of risk region versus percent of the left ventricle infarcted (r = -0.80). Discussion The results of this investigation provide unique direct evidence that the degree of electrical instability corre-
Ml SIZE (% LVI
lates strongly with infarct size 3 to 5 days after acute myocardial infarction in the dog. These data support previous clinical observations that patients with larger CK curve estimates of infarct size, higher SGOT levels, poorer global left ventricular function and angiographic evidence of more extensive coronary artery diseasel-a demonstrate more frequent and complex arrhythmias after infarction. The effects of anesthesia and surgical trauma were minimized by studying dogs in the awake state after recovery from thoracotomy. No drugs were given at the time of the pacing study other than morphine for sedation and lidocaine given subcutaneously as a local anesthetic agent 30 to 60 minutes before the pacing
y=2.lx+5.7 P.34 n=13 SEE=11.2
& 0’10
I 20
30
40
50
Ml SIZE (% LV 1 FIGURE 2. Electrical instability (El) index C. derived from programmed right ventricular stimulation data is correlated with the percent of the left ventricle (LV) infarcted. This index reflects the types of arrhythmia produced by stimulation, as well as the number and diastolic timing of stimuli required to produce them (Table C). Abbreviations as before.
November 1981
The American Journal ol CARDIOLOGY
Volume 48
861
POSTINFARCTION ELECTRICAL INSTABILITY-JONES-COLLINS AND PATTERSON
TABLE
II
Electrical
Instability
Indexes
and Score Results Score (System B) at Each Diastolic Timing/Extrastimuli Combination
Electrical Instability Index Dog
% LV Infarcted
B
A
Score (System A) at First Occurrence RVR
C
VT
sVT
S2 2nd ‘I2 of Diastole
VF
s2-s3
s2
1st ‘12 of Diastole
s2-s3
2nd ‘1s of Diastole
1st ‘Is of Diastole
Early Sp With Refractory S3
1 :
4 0”
z
4 5
6 7 : :‘: :z r* =
:;
11 :
:
:
10 :
:
:
:0
0 :
0 :
:
:
:1
1
1;
19 :“6
0
:0
:
1
:
:
:,
:1
151
: 15 ::
z;
:: 1:
:
: 1:
: 15 :
:
; 1:
:
:
:
1 15 :
:
:
z
::
::
15 :z
1:
1:
0
3:
1:0
:
300
z 0.;:
;: 0.85
:: 0.94
15 0.380
: 0.33
0
;: 0.83
: 0.13
:; :z
;: 3’: ::
2:
:: 15 1: 15 0.9’0”
15 0.::
0.::
: -0.25
: -0.003
Correlation between percent of the left ventricle infarcted and each measurement of electrical instability. LV = left ventricle; RVR = repetitive ventricular response; S2 and S3 = first and second premature stimuli; sVT = sustained ventricular tachycardia; VF = ventricular fibrillation: VT = unsustained ventricular tachycardia. l
study. This small amount of locally injected lidocaine would not be expected to affect the results of electrophysiologic testing.lO Mechanism of observed ventricular arrhythmias and correlation with infarct size: No attempt was made to determine the mechanism of the observed arrhythmias. Bundle branch reentry-the phenomenon of retrograde, then anterograde conduction of an extrastimulus through the bundle branches to reexcite the ventricles-cannot be excluded as a mechanism for some of the responses observed because no endocardial recordings were obtained. However, excluding this mechanism as a cause of the reported observations would not be appropriate, because it may have a role in the initiation of postinfarction arrhythmias.li Also, the lack of multiple recordings made it impossible to determine the site of origin of the induced arrhythmias. However, the observed strong correlation between infarct size and ventricular arrhythmias, regardless of origin, is of primary interest here. If myocardial infarction causes some change in noninfarcted tissue leading to induction of arrhythmia, and the reported
l /
J. .
g
I=.90
/ 5
y= 5x-2.5
IF13 SEE=3.5
::
O-?ElrGr
40
MI SIZE I% LV)
FIGURE 3. Electrical instability score (using scoring system A) at the first occurrence of sustained ventricular tachycardia @VT) is correlated with the percent of the left ventricle (LV) infarcted. No dog with an infarct size of less than 20 percent of the left ventricle experienced this arrhythmia. Abbreviations as before.
882
November 1981
The American Journal of CARDIOLOGY
responses were demonstrated to originate outside the infarct zone, they are not any less important. The protocol was carried out at only one cycle length and stimulation site because the purpose of the experiment was to define electrical instability in a general sense, not to stimulate in a multitude of ways until a previously demonstrated specific arrhythmia was reproduced. The ease of induction of ventricular arrhythmias by the experimental protocol, using one cycle length and one stimulation site per dog was tested and proved to be a good indicator of severity of disease. The right ventricle, remote from the septum, was selected as the site for pacing electrode placement to be sure it would not overlie any infarcted tissue at the time of the pacing study. Quantitating electrical instability-validity of method: The problem of quantitating electrical instability has concerned many previous investigators.12 Arrhythmias induced in the laboratory are being used to investigate and evaluate patients’ and experimental subjects’ predisposition to spontaneously occurring arrhythmias. Various methods of stimulation have been used and a variety of responses have been observed.ls17 In this study severity of disease (that is, infarct size) could be measured directly and was shown to correlate well with electrical instability indexes. These correlations support the validity of the method of stimulation and interpretation used. The scoring systems focus on two principles: First, it has been shown that stimulation early in diastole, near the T wave (“vulnerable period”), is able to produce ventricular arrhythmias in normal dogs, whereas stimulation elsewhere in diastole is not.ls Therefore, for score A lower numbers were assigned to responses to stimulation early in diastole, when vulnerability is greater, and higher numbers to abnormal responses to stimulation in late diastole, when propensity for arrhythmia development should be low, despite provo-
Volume 48
POSTINFARCTION ELECTRICAL INSTABILITY-JONES-COLLINS
cation. Second, it has been shown that repetitive ventricular responses and ventricular tachycardia, as indexes of ventricular vulnerability, occur reproducibly in response to lower amplitude stimuli than does ventricular fibrillation when elicited at the same diastolic position. lg Therefore, the repetitive ventricular response, unsustaided ventricular tachycardia, sustained ventricular tachycardia and ventricular fibrillation were considered to indicate increasing severity of electrical instability in that order when produced by an extrastimulus of fixed amplitude and at the same time in diastole. Hence, for score B, progressively higher numbers were assigned to progressively more hazardous and complex arrhythmias. Correlation with infarct size: The correlation between infarct size and electrical instability index A (r = 0.92) was not significantly different from that between infarct size and electrical instability index B (r = 0.85). Moreover, each dog’s index A was well correlated with the same dog’s index B (r = 0.78). Electrical instability index C was designed ta account for diastolic time, number of extrastimuli required and type of arrhythmic response together. This composite index also correlated well with infarct size (r = 0.94) but did not significantly improve the correlation of either index alone. Of particular note is the finding that the electrical instability index was unrelated to size of region at risk, despite such a strong correlation with infarct size. This
AND PATTERSON
indicates that the crucial factor in determining electrical instability 3 to 5 days after infarction may be mass of necrotic myocardium rather than mass of myocardium initially at risk of infarction. A strong negative correlation between percent region at risk salvaged (at risk but not infarcted) and the electrical instability index was demonstrated. We considered this result primarily due to the strong negative correlation between infarct size and region at risk salvaged. Although the two dogs without infarction may not have actually experienced total coronary occlusion, they serve to illustrate the results of this electrophysiologic protocol in dogs without infarcts. Not one abnormal response was produced in either dog. Implications: These data provide direct evidence that induced electrical instability is strongly correlated with infarct size. Furthermore, the scoring system introduced here provides a unique quantitative method for deriving an electrical instability index from programmed right ventricular stimulation data. This derived index may eventually prove to be a very useful investigative and perhaps clinical tool.
Acknowledgment We thank William Parker for technical assistance, Sharla Goldstein for preparing the manuscript and Stephen E. Epstein, MD for editing it,
References 1. Morgensen L. A clinical study of cardiac arrhythmias in 421 consecutive patients with acute myocardial infarction treated in a coronary care unit. Acta MedStand 1971;513:10-29. 2. Chapman BL. Relation of cardiac complications to SGOT level in acute myocardial infarction. Br Heart J 1972;34:890-6. 3. Roberts R, Husaln A, Ambos HD, Oliver GC, Cox JR, Sobel BE. Relation between infarct size and ventricular arrhythmia. Br Heart J 1975;37:1169-75. 4. Geftman EM, Ehsanl AA, Campbell MK, Schechtman K, Roberls R, Sobel BE. The influence of location and extent of myocardial infarction on long term dysrhythmia and mortality. Circulation 1979;60:805-14. 5. Schulze RA, Rouleau J, Rlgo P, Sowers S, Strauss HW, Pltl B. Ventricular arrhythmias in the late hospital phase of acute myocardial infarction: relation to left ventricular function detected by gated cardiac blood pool scanning. Circulation 1975:52:100611. 6. Schulze RA, Humphrles JO, Grffflth LS, et al. Lett ventricular and coronary angiographic anatomy. Relationship lo ventricular irritability in the late hospital phase of acute myocardial infarction. Circulation 1977;55:839-43. 7. Lawrle DM, Hlgglns MR, Goodman MJ, Ollver MF, Jullan DG, Donald KW. Ventricular fibrillation complicating acute myocardial infarction. Lancet 1968;2:523-8. 6. Josephson ME, Horowitz LN, Farshldl A, Kastor JA. Recurrent sustained ventricular tachycardia. I. Mechanisms. Circulation 1978571431-40. 9. Lle JT, Palrofero PC, Holtey KE, Tltus JL. Macroscopic enzymemapping verification of large, homogeneous, experimental myocardial infarcts of predictable size and location in dogs. J Thorac
Cardiovasc Surg 1975;69:599-605. 10. Nattel S, Rlnkenberger RL, Lehrman LL, Zl~es DP. Therapeutic blood lidocaine concentrations after local anesthesia for cardiac electrophysiologic studies. N Engl J h4ed 1979;301:418-20. 11. Foster JR, Simpson RJ. Initiation of ventricular tachycardia by reentry within the bundle branches. Am J Cardiol 1980;45:895900. 12. Kent KM. Assessing the electrical stability of the heart in human beings. Am J Cardiol 1980;45:1305-7. 13. Han J. Ventricular vulnerability during acute coronary occlusion. Am J Cardiol 1989;24:857-64. 14. Jenzer H, Lohrbauer L, Lown 5. Response to single threshold stimuli following acute myocardial infarction. Proc Sot Exp Biol Med 1972;141:606-8. 15. Wellens HJ, Schullenburg RM, Dorrer D. Electrical stimulation of the heart in patients with ventricular tachycardia. Circulation 1972;46:216-26. 16. Denes P, Wu D, Dhlngra RC, et al. Electrophysiological studies in patients with chronic recurrent ventricular tachycardia. Circulation 1976;54:229-38. 17. Flsher JD, Cohen HL, Mehra R, Altschuler H, Escher DJW, Furman S. Cardiac pacing and pacemakers. II. Serial electrophysiologic-pharmacologic testing for control of recurrent tachyarrhythmias. Am Heart J 1977;93:658-68. 18. Wlggers CJ, Wegrla R. Ventricular fibrillation due to single localized induction and condenser shocks applied during the vulnerable phase of ventricular systole. Am J Physiol 1940;128:500-5. 19. Matta RJ, Verrler RL, Lown 6. Repetitive extrasystole as an index of vulnerability to ventricular fibrillation. Am J Physiol 1978;230: 1469-73.
November 1981
The American Journal of CARDIDLDGY
Volume 48
863