444 JACC Vol. 19, No. 2 February 1992:444-9

Electrophysiologic Substrate Associated With Pacing-Induced Failure in Dogs: Potential Value of Programmed Stimulation in Predicting Sudden Death

HUAGUI G. Ll, MB, MSc, DOUGLAS L. JONES, PHD, FACC, RAYMOND YEE, MD, FACC, GEORGE J. KLEIN, MD, FACC London, Ontario, Canada

To investigate the possible mechanisms of sudden death and the 0.001) and the ventricular effective refractory period correlated potential role of electrophysiologic testing in congestive heart inversely with cardiac index (r = -0.49, p < 0.025). failure, this study evaluated the electrophysiologic substrate in a The rest membrane potential of ventricular muscle was less model of heart failure induced by rapid pacing. Seventeen mon· negative in the paced group compared with the control group grel dogs underwent cardiac pacing at 220 to 240 beats/min for 5 (-80.7 ± 2.2 vs. -85.6 ± 2.2 mV, p < 0.05). Intracellularly weeks (paced group) and 11 other dogs served as a sham-operated recorded action potential duration of ventricular muscle was control group. Rapid pacing of the right produced longer in the paced than in the control group (236 ± 9.8 vs. 198.9 clinical and hemodynamic features of congestive heart failure. ± 2.6 ms, p < 0.01, action potential duration at 90% repolariza­ Dogs in the paced group had prolonged cardiac conduction tion). Eight dogs (48%) in the paced group developed ventricular time as reflected by longer epicardial activation time (36.1 ± 2.4 fibrillation after the occurrence of heart failure, whereas no vs. 30.8 ± 0.8 ms, p < 0.05). The ventricular effective refractory control dog had sustained arrhythmia. There was no significant period was significantly prolonged after the development of heart difference in the inducibility of ventricular tachycardia by pro­ failure (141 ± 4 vs. 177 ± 5 ms, p < 0.01, at a basic pacing cycle grammed stimulation between the paced and control groups. length of 300 ms), whereas no significant change was found in the In conclusion, rapid pacing-induced cardiac dysfunction is control group (140 ± 4 vs. 145 ± 4 ms, p =NS). The prolongation associated with electrophysiologic changes that could contribute to of the ventricular effective refractory period correlated with an sudden cardiac death. These changes are not reflected by induc­ increase in left ventricular end-diastolic pressure (r =0.55, p < ibility of sustained ventricular tachycardia. (JAm Coli Cardiol1992;19:444-9)

Complex ventricular arrhythmias are noted in approximately tive heart failure are poorly understood and the cardiac 90% of patients with congestive heart failure (1) and are electrophysiologic abnormalities specifically related to heart closely related to the severity of ventricular dysfunction failure are not clear. (2,3). Patients with severe congestive heart failure have an Tachycardia-induced cardiomyopathy and ventricular increased risk of developing malignant ventricular arrhyth­ dysfunction have been demonstrated in some patients with mias and sudden death (4). However, the efficacy of antiar­ chronic supraventricular tachycardia (8,9). Experimentally, rhythmic drug therapy in preventing sudden death is uncer­ prolonged rapid pacing in the dog has been shown to produce tain (5), whereas -converting enzyme inhibitors clinical features of congestive heart failure, simulating con­ have been shown to reduce ventricular ectopic activity in gestive cardiomyopathy (10-12), and this provides a model patients with severe heart failure (6,7). The mechanisms for studying cardiac electrophysiologic changes and arrhyth­ underlying the genesis of ventricular arrhythmias in conges- mogenesis in congestive heart failure. The objectives of the present study were to determine the electrophysiologic changes associated with the development of heart failure in From the Departments of Physiology and Medicine, University of West­ this model and to explore the value of programmed stimula­ em Ontario and Heart and Circulation Group, Robarts Research Institute, London, Ontario, Canada. This study was supported by the Heart and Stroke tion in assessing susceptibility to sudden cardiac death. Foundation of Ontario. Dr. Li is a Research Fellow and Dr. Klein a Distinguished Research Professor of the Heart and Stroke Foundation of Ontario. Custom pacemakers were provided by Medtronic Canada, Inc., Methods Mississagua, Ontario. Manuscript received April 15, 1991 ; revised manuscript received June 6, Study animals. Twenty-eight mongrel dogs were used; 17 1991, accepted June 28, 1991. in the paced group and 11 in the control group. Baseline Address for reprints: Douglas L. Jones, PhD, Departments of Physiology and Medicine, University of Western Ontario, London, Ontario, Canada N6A physical examination, electrocardiogram (ECG), blood 5Cl. white cell count and determinations of hemoglobin, serum

©1992 by the American College of Cardiology 0735-1097/92/$5.00 JACC Vol. 19, No. 2 LI ET AL. 445 February 1992:444-9 ELECTROPHYSIOLOGY IN DOGS WITH HEART FAILURE

electrolytes and creatinine levels were obtained to exclude On completion of the programmed stimulation protocol, any unhealthy animal. The protocol was approved by the the catheters were withdrawn and the right femoral artery University Council on Animal Care of the University of was ligated. A unipolar screw-in pacing lead was inserted Western Ontario and was carried out in accordance with the through the left jugular vein incision and the tip of the lead "Guidelines to the Care and Use of Experimental Animals" was positioned in the right ventricle. The proximal end of the of the Canadian Council on Animal Care. lead was connected to a custom-modified Medtronic 5941 Baseline study. Dogs were tranquillized with an intramus­ programmable generator that was subsequently im­ cular injection of acepromazine (0.3 to 0.4 mg/kg body planted in a subcutaneous pocket in the neck. The pace­ weight) for transportation to the laboratory. They were maker was set at 70 beats/min in demand mode. After the anesthetized with a mixture of 4% halothane, nitrous oxide closure of all incisions, the animals were allowed to recover (2 liters/min), oxygen (lliter/min) and medical air (2 liters/ from anesthesia and returned to their quarters. min) with use of a mask and then were intubated and Twentyjour hours postoperatively, the pacemaker in the ventilated at a rate of 10 to 12 breaths/min and tidal volume paced group was programmed to 240 beats/min and main­ of 12 to 13 mllkg with use of the Hospital 300 Anesthesia tained at that rate for 3 weeks. In accord with previous Ventilator (Hospital Medical Corporation). Halothane in the experience with this animal model, the pacing rate was mixture was subsequently reduced to 0.5% to 1%. A circu­ reduced to 220 beats/min for the next 2 weeks before restudy lating water blanket and controller were used to maintain to produce severe heart failure without increasing mortality. body temperature at 37 ± l°C. Infusion of lactated Ringer's The surface ECG was checked weekly to assure constant solution was maintained during the operation. pacing. The pacemaker in the control dogs remained in A 7F pigtail catheter was inserted through an incision in demand mode at 70 beats/min. Body weight, blood white cell the right femoral artery and placed in the left ventricle under count, hemoglobin, hematocrit and serum creatinine, urea fiuoroscopic guidance. A 6F quadripolar catheter and a and electrolytes were also monitored weekly. tripolar electrode catheter were inserted through an incision Restudy. Five weeks after the initial study, the pace­ in the left jugular vein and placed in the right atrium and right maker was turned off and the dog was transported to the ventricle, respectively. Intracardiac electrograms were fil­ laboratory and anesthetized, with use of the identical proce­ tered at 30 to 500 Hz. Intracardiac electrograms, ECG limb dures used in the baseline study. The ventricular effective leads I and 11 and left ventricular pressure were processed, refractory period and inducibility of ventricular arrhythmia displayed continuously and recorded periodically on photo­ were determined with catheter insertion through the right graphic paper at 50 to 100 mm/s with use of an Electronics jugular vein. Care was taken to position the stimulation For Medicine VR-16 recorder. The endocardium of the right catheter distant from the site of the pacing lead. On comple­ ventricular apex was stimulated at a pulse duration of 2 ms tion of the programmed stimulation protocol, the chest was and twice diastolic threshold with use of a Grass S88 opened through a median sternotomy. The heart was ex­ stimulator (Grass Medical Instruments). A custom-designed posed and the ventricles were wrapped with a nylon sock timer was used to trigger the stimulator. containing 56 bipolar electrode buttons. Epicardial electro­ Ventricular effective refractory period was determined grams were processed by a computerized cardiac mapping with use of a driving train (S 1) of eight beats at a cycle length system (CMS 1000, Biomedical Instrumentation). Isochronal of 300 and 250 ms, followed by an extrastimulus (S 2) that was activation maps were generated and the total activation time decremented in 10-ms intervals. Responses were monitored was calculated. In nine dogs (six in the paced and three in the on a Tektronix 5513 storage oscilloscope. Ventricular effec­ control group) programmed stimulation was also under­ tive refractory period was defined as the longest S1S2 inter­ taken, stimulating the epicardium of the right ventricular val at which S2 failed to elicit ventricular activation. outflow tract, apex and left ventricular free wall to evaluate Inducibility of ventricular arrhythmia was determined dispersion of refractoriness and inducibility of ventricular using programmed stimulation at basic pacing cycle lengths arrhythmia. of 400, 300 and 250 ms and up to three extrastimuli. The On completion of the restudy, the heart was removed and second extrastimulus (S 3) was started with the S1S2 interval examined for size, ventricular weight and gross abnormali­ fixed at 30 ms longer than the ventricular effective refractory ties. The pericardium, lungs and abdominal organs were also period. The S2S3 interval was set at 80% of basic pacing examined. Papillary muscles were removed for cellular cycle length, again with use of 10-ms scanning decrements electrophysiologic study. until S3 failed to elicit ventricular activation. If double measurement. In 10 dogs (5 in the paced extrastimulation failed to induce arrhythmia, a third extra­ and 5 in the control group), in addition to the electrophysi­ stimulus (S4) was added with a similar scanning procedure ologic study, a Swan-Ganz catheter was inserted through the (that is, the s2s3 interval fixed at 30 ms longer than the right femoral vein into a . Triplicate mea­ longest S2S3 interval at which S3 failed to capture). Nonsus­ surements of cardiac output were obtained with use of 5 ml tained ventricular tachycardia was defined as tachycardia of iced dextrose injections and a 9520A Thermodilution lasting between three consecutive beats to 30 s and self­ Cardiac Output Computer (American Edwards Laborato­ terminating. ries) during sinus rhythm and right atrial pacing at a cycle 446 LI ET AL. JACC Vol. 19, No. 2 ELECTROPHYSIOLOGY IN DOGS WITH HEART FAILURE February 1992:444-9

Table 1. Selected Hemodynamic Observations at Baseline and Restudy in the Paced and Control Groups Cardiac Index Cardiac Output (liters/min) (liters/kg per min) LVSP(mmHg) LVDP(mm Hg) Baseline Restudy Baseline Restudy Baseline Re study Baseline Re study Control group 3.81 ± 0.34 3.76 ± 0.20 0.18 ± 0.02 0.18 ± 0.01 104.8 ± 3.4 101.1 ± 3.5 2.8 ± 0.8 4.7 ± 0.9 No. 5 5 5 5 11 11 11 11 Paced group 3.46 ± 0.34 2.11 ± 0.38 0.18 ± 0.01 0.11 ± 0.03* 102.1 ± 3.8 72.9 ± 3.4* 3.5 ± 0.8 12.5 ± 1.6* No. 5 5 5 5 16 16 16 16

*p < 0.01 compared with baseline. LVDP and LVSP = left ventricular diastolic and systolic pressure, respectively. Cardiac output and cardiac index were measured during right atrial pacing at a cycle length of 300 ms. length of 300 ms. Cardiac output was measured during both failure in the rapidly paced group but not in the control the initial study and the restudy in all 10 dogs. group. Twelve (71%) of the 17 dogs in the paced group were Microelectrode study. In 19 dogs (10 in the paced and 9 in lethargic before re study, whereas no behavior change was the control group) at the end of in vivo study, a papillary noted in the control dogs. Cardiac output and cardiac index muscle was dissected from the beating heart and immersed were significantly decreased at restudy in the paced group, in cold Tyrode's solution. The preparation was mounted in a but there was no significant change in the control group. Left tissue bath and perfused with a 95% oxygen, 5% carbon ventricular systolic pressure was significantly decreased and dioxide balanced Tyrode's solution at a flow rate of the diastolic pressure was significantly increased after the 12 ml/min. The temperature of the solution was occurrence of heart failure in the paced group, whereas no monitored by a thermistor probe in the bath and maintained significant change was found in the control group. Although at 37 ± 1oc by a heating pump. The composition of the the heart was visibly dilated in the paced group, no signifi­ Tyrode's solution at pH 7.38 to 7.44 was as follows (mM): cant difference in ventricular weight was found between the sodium chloride 129, sodium bicarbonate 20, sodium phos­ paced and control groups (6.6 ± 0.2 vs. 6.4 ± 0.2 g/kg body phate, monobasic 0.9, magnesium sulfate 0.5, potassium weight, p = NS). Local fibrosis of the endocardium at the chloride 4, calcium chloride 2.5 and glucose 5.5. site of the permanent pacing lead was observed in both Glass microelectrodes with tip resistance of 15 to 20 groups. In the paced group, 13 dogs (77%) had one or more megohms were used for transmembrane potential recording. of the following abnormal findings at postmortem examina­ A silver-silver chloride wire in a glass tube filled with an tion: ascites in 8 (47%), pericardial effusion in 10 (59%) and agar-3M potassium chloride mixture was positioned in the multifocallung infarction in 3 (18%). No abnormalities were bath as a reference electrode. The tissue was stimulated at noted in the control group. No significant change in body 1 Hz through a bipolar silver wire electrode with use of a weight was noted in either the paced group (20.1 ± 0.8 kg at WPI 1381 pulse generator (World Precision Instruments, baseline study vs. 20.2 ± 0.9 kg at re study, p > 0.05) or the Inc.) connected to a WPI 1850A stimulus isolator. Trans­ control group (21.1 ± 1.2 kg at baseline study vs. 21.4 ± membrane potentials were processed with use of a WPI FD 1.1 kg at re study, p > 0.05), but there was evident loss of 223 electrometer and recorded on a Gould 2400s chart recorder muscle mass in the paced dogs. at paper speeds of 50 to 100 mrnls. Transmembrane potential Serum electrolyte disturbances were not evident in either was also digitalized and analyzed with use of a custom­ group. Hematocrit was significantly decreased in the paced designed computer system. Rest membrane potential, action group (40.3% vs. 46.3%, p < 0.05), whereas no significant potential amplitude and action potential duration at 50% and change was found in the control group (44.5% vs. 44.3%, p > 90% repolarization were derived from the computer. 0.05). Statistical analysis. Values were expressed as mean val­ Ventricular fibrillation. This arrhythmia occurred in eight ues ± SEM. The unpaired t test was used to compare mean dogs (47%) in the paced group. In one dog it occurred values between the control and paced groups. The difference spontaneously before restudy with ECG monitoring. In the between baseline study and restudy in each group was other seven it occurred during restudy: initiated by sponta­ analyzed with the paired t test. The relation between hemo­ neous premature ventricular complexes without preceding dynamic and electrophysiologic variables was evaluated by evidence of ischemia or hypotension in four, preceded by linear regression analysis. A p value $0.05 was considered bradycardia of sudden onset in two and induced by triple significant. extrastimuli in one. High energy shocks were usually re­ quired to terminate ventricular fibrillation in these dogs. Five dogs developed electromechanical dissociation after defibril­ Results lation and could not be resuscitated and thus failed to Indexes of heart failure (Table 1). One dog in the paced complete the electrophysiologic re study. group died of spontaneous ventricular fibrillation. The others Ventricular effective refractory period. This was signifi­ survived until restudy. There were multiple indexes of heart cantly prolonged after development of heart failure in the JACC Vol. 19, No. 2 L1 ET AL. 447 February 1992:444-9 ELECTROPHYSIOLOGY IN DOGS WITH HEART FAILURE

Jll) • t • - f-- r- f=== 110 t::::::::~ f=== Cc: f-- -r f=: - .-f=== f-- !=== f-- = f-- DJ t:- = f-- f-- f--- Ill f-- f-- 1--- 60 f-- f-- t::= F= t-- 10 1--- r----= f-- 1--- = t-- control paced control paced XX t-- 0 Cycle Length = 300 ms Cycle Length = 250 ms RV(endo) RV(epi) RVO'l'(epi) LVFW(epi) Figure 1. Bar graph of the ventricular effective refractory period at Figure 2. Bar graph of the ventricular effective refractory period initial study and restudy in 11 dogs in each of the control and paced measured at restudy from different sites in six dogs in the paced groups. The ventricular effective refractory period was shorter with group and in three dogs in the control group. *p < 0.01 and tp < 0.05 a basic pacing cycle length of 250 ms (right four bars) than with a compared with the control group. endo = endocardium; epi = cycle length of 300 ms (left four bars) for both paced and control epicardium; LVFW = left ventricular free wall; RV = right ventri­ dogs. Dogs in the paced group whose ventricular effective refractory cle; RVOT = right ventricular outflow tract. Cross-hatched bars = period was not obtained at restudy because of death from ventric­ control group; hatched bars = paced group. ular fibrillation were not included. *p < 0.01 compared with data from the initial study. Hatched bars= baseline; cross-hatched bars= restudy. Epicardial activation time. Epicardial activation patterns in sinus rhythm were similar in both groups, with the earliest paced group, whereas there was no significant change in the epicardial breakthrough at the right ventricular anterior wall control group (Fig. 1). Ventricular effective refractory period near the interventricular septum. However, the total epicar­ measured at different ventricular sites during restudy was dial activation time was longer in the paced group than in the consistently longer in the paced group than in the control control group (36.4 ± 2.4 vs. 30.8 ± 0.8 ms, p < 0.05). group (Fig. 2). Dispersion of ventricular refractoriness was Conduction slowing was evenly distributed as demonstrated defined as the difference between the longest and the short­ by uniformity of interisochrone distance. est ventricular effective refractory period measured at dif­ Correlation with . Left ventricular dia­ ferent sites in the same heart and was similar in both groups stolic pressure was significantly correlated with ventricular (21 ± 5 .1 ms in the paced group vs. 28 ± 11.5 ms of the effective refractory period (r = 0.52, p < 0.01). The increase control group, p > 0.05). in left ventricular diastolic pressure was positively corre­ Ventricular tachycardia. Single and double extrastimuli lated with the prolongation of ventricular effective refractory did not induce ventricular tachycardia at initial study or period (r = 0.55, p < 0.001). There was also an inverse restudy in either group. Triple extrastimuli of the right correlation between cardiac index and ventricular effective ventricular endocardium induced nonsustained ventricular refractory period (r = -0.49, p < 0.025) but no correlation tachycardia once each in Dogs 10 and 12 in the paced group between cardiac output and ventricular effective refractory (four to five beats) and in Dog 10 in the control group (three period (r = 0.13, p > 0.05). beats) at the initial study. On restudy, nonsustained ventric­ Microelectrode findings (Table 2). Rest membrane poten­ ular tachycardia was induced repeatedly in Dogs 6, 9 and 15 tial was decreased (less negative) and the action potential in the paced group (four to nine beats) but only once each in duration was significantly longer in the paced group than in the two dogs (Dogs 7 and 11; four beats) in the control group. the control group. Action potential duration at 90% repolar- Triple extrastimuli at multiple epicardial sites induced non­ sustained ventricular tachycardia repeatedly in two of six dogs in the paced group (Dogs 8 and 9; four to eight beats) Table 2. Transmembrane Potentials During Microlectrode Studies but only once each in two of the three control dogs (Control Rest Pot Amplitude APD50 APD90 Dogs 6 and 7; 3 and 12 beats). In the three dogs resuscitated (mY) (mY) (ms) (ms) from ventricular fibrillation, findings differed from those of Control group -85.6 ± 2.2 102.3 ± 2.0 157.9 ± 4.8 198.9 ± 2.6 the control group but there were no differences between Paced group -80.7 ± 2.2 101.3 ± 2.2 182.9 ± 12.0 236.0 ± 9.8 values for ventricular effective refractory period, dispersion p value <0.05 >0.05 <0.05 <0.01 of ventricular refractoriness and inducibility of ventricular APD50 and APD90 = action potential duration at 50% and 90% repolar­ tachycardia and those of other dogs in the paced group. ization, respectively; Pot = potential. 448 Ll ET AL. JACC Vol. 19, No. 2 ELECTROPHYSIOLOGY IN DOGS WITH HEART FAILURE February 1992:444-9 ization was significantly correlated with ventricular effective ventricular fibrillation with low inducibility of ventricular refractory period (r = 0.62, p < 0.01). arrhythmia by programmed stimulation in the paced group suggests that the initiating mechanism of ventricular fibrilla­ tion may be other than a fixed reentry substrate. The Discussion possibility that ventricular fibrillation is initiated by mecha­ Chronic supraventricular tachycardias have been re­ nisms other than reentry was suggested in an ischemia­ ported (8,9) to be able to induce ventricular dysfunction in reperfusion model (15) and supported by the low inducibility patients. Similarly, in the present experimental study, the of ventricular tachycardia by programmed stimulation in clinical features of congestive heart failure were produced in patients with cardiomyopathy who had a high prevalence of dogs after rapid pacing of the right ventricle for 5 weeks as spontaneous ventricular arrhythmias (14). manifested by lethargic behavior, decreased cardiac output, Transmembrane potential changes. Depletion of high en­ increased ventricular diastolic pressure and fluid retention. ergy phosphate has been demonstrated in heart failure Despite visible dilation, total ventricular weight was not induced by rapid pacing (10). Shortage of adenosine triphos­ statistically different from that in the control group, indicat­ phate (A TP) may cause dysfunction of the electrogenic ing thinning of the ventricular wall in this model. sodium-potassium pump. Thus, the contribution of the elec­ Prolongation of ventricular effective refractory period. trogenic sodium-potassium pump to rest membrane potential Significant prolongation of the ventricular effective refrac­ may be decreased, which in turn may produce partial depo­ tory period and action potential duration was found after larization of the rest membrane potential. The decreased rest development of heart failure in the present study. Slowing of membrane potential can lead to increased automaticity and the inactivation of a calcium-activated inward current and a thus constitute a substrate of arrhythmia. Impulse conduc­ decrease in outward repolarizing current have been pro­ tion velocity is also affected by rest membrane potential. posed to explain the action potential prolongation produced Thus, the decreased rest membrane potential may contribute by myocardial hypertrophy (13) and this might also have to conduction slowing as shown by increased total duration happened in the dogs with rapid pacing-induced heart fail­ of epicardial activation. Disorganization of myocardial fiber ure. By placing the stimulation catheter distant from the site orientation and collagen arrangement may also contribute to of the pacing lead, measurement of ventricular effective conduction slowing (16). Increased conduction distance refractory period in the area of fibrosis caused by the pacing caused by the dilation of the heart may prolong total activa­ lead was avoided. Measurements at multiple epicardial sites tion time. also revealed a consistently longer ventricular effective Implications. The implications of the prolongation of the refractory period in the paced group than in the control ventricular effective refractory period and action potential group. Furthermore, action potential duration of papillary duration deserve further investigation. Agents that prolong muscle that was not affected by pacing lead-induced local the effective refractory period and action potential duration fibrosis was also prolonged consistently. Therefore, the have been shown to induce early triggered activity experi­ prolongation of the ventricular refractory period is a global mentally and torsade de pointes clinically (17-20). Recently, abnormality associated with cardiac dysfunction. prolongation of action potential duration caused by myocar­ The inverse correlation of the ventricular effective refrac­ dial hypertrophy has been reported (21) to predispose the tory period with cardiac index suggests that the degree of heart to early triggered activity and ventricular tachycardia. electrophysiologic abnormality is related to the severity of Whether prolongation of effective refractory period and cardiac dysfunction. Prolongation of the ventricular effective action potential duration in the present study can produce refractory period also correlated with an increase in left early triggered activity and contribute to the high incidence ventricular end-diastolic pressure, indicating that the loading of ventricular fibrillation deserves further study. condition and the increased stretch of ventricular wall may be factors causing the electrophysiologic change. We thank G. Kim Wood, AHT for excellent technical assistance. Inducibility of ventricular tachycardia. Despite severe dilation of the heart in all dogs in the paced group, sustained ventricular tachycardia was not induced in any surviving animal; it may have been inducible in the nonsurvivors, but References this could not be assessed. In contrast, ventricular fibrilla­ I. Francis GS. Development of arrhythmias in the patient with congestive tion was common. There was no statistically significant heart failure: pathophysiology, prevalence, and prognosis. Am J Cardiol difference in the inducibility of nonsustained ventricular 1986;57:3B-7B. tachycardia between the initial study and restudy or between 2. Bigger JT Jr, Fleiss JL, Kleiger R, Miller JP, Rolnitzky LM and the Multicenter Postinfarction Group. The relationship among ventricular the paced and control groups. This finding suggests that the arrhythmias, left ventricular dysfunction and mortality in the 2 years after susceptibility to ventricular fibrillation cannot be predicted myocardial infarction. Circulation 1984;69:250-8. by programmed stimulation in congestive heart failure, an 3. Schultze RA, Roleau J, Rigo P, Bowers S, Strauss HW, Pitt B. Ventric­ ular arrhythmias in the late hospital phase of acute myocardial infarction: observation consistent with the results of programmed stim­ relation of left ventricular function detected by gated cardiac blood pool ulation in dilated cardiomyopathy (14). The high incidence of scanning. Circulation 1975;52:1006-11. JACC Vol. 19, No. 2 LI ET AL. 449 February 1992:444-9 ELECTROPHYSIOLOGY IN DOGS WITH HEART FAILURE

4. Packer M. Sudden unexpected death in patients with congestive heart 14. Kulick DL, Bhandari AK, Hong R, Petersen R, Leon C, Rahimtoola SH. failure: a second frontier. Circulation 1985;72:681-5. Effect of acute hemodynamic decompensation on electrical inducibility of 5. Podrid PJ, Wilson JS. Should asymptomatic ventricular arrhythmia in ventricular arrhythmias in patients with dilated cardiomyopathy and patients with congestive heart failure be treated? An antagonist's view. complex nonsustained ventricular arrhythmias. Am Heart J 1990;119: Am J Cardiol1990;66:451-7. 878-83. 6. Captopril-Digoxin Multicenter Research Group. Comparative effects of 15. ldeker RE, Klein GJ, Harrison L, et al. The transition to ventricular captopril and digoxin in patients with mild to moderate heart failure. fibrillation induced by reperfusion after acute ischemia in the dog: a JAMA 1988;259:539-44. period of organized epicardial activation. Circulation 1981;63:1371-9. 7. Cleland JGF, Dargie HJ, Hodsman GP, et al. Captopril in heart failure: a double blind controlled trial. Br Heart J 1984;52:530-5. 16. Weber KT, Pick R, Silver MA, et al. Fibrillar collagen and remodelling of dilated canine left ventricle. Circulation 1990;82: 1387-401. 8. Gillette PC, Wampler DG, Garson A Jr, Zinner A, Ott D, Cooley D. Treatment of atrial automatic tachycardia by ablation procedures. J Am 17. Brachmann J, Scherlag BJ, Rosenshtraukh LV, Lazzara R. Bradycardia­ Coli Cardioll985;6:405-9. dependent triggered activity: relevance to drug-induced multiform ven­ 9. Packer DL, Bardy GH, Worley SJ, et al. Tachycardia-induced cardiomy­ tricular tachycardia. Circulation 1983;68:846-56. opathy: a reversible form of left ventricular dysfunction. Am J Cardiol 18. Roden DM, Hoffman BF. Action potential prolongation and induction of 1986 ;57: 563-70. abnormal automaticity by low quinidine concentrations in canine Purkinje 10. Coleman HN, Taylor RR, Pool PE, et al. Congestive heart failure fibers: relationship to potassium and cycle length. Circ Res 1985 ;56:857- following chronic tachycardia. Am Heart J 1971;81:790-8. 67. 11. Armstrong PW, Stopps TP, Ford SE, de Bold AJ. Rapid ventricular 19. Jackman WM, Friday KJ, Anderson JL, Aliot EM, Clark M, Lazzara R. pacing in the dog: pathophysiologic studies of heart failure. Circulation The long QT syndromes: a critical review, new clinical observations and 1987;74:1075-84. a unifying hypothesis. Prog Cardiovasc Dis 1988;31:115-72. 12. Wilson JR, Douglas P, Hichey WF, et al. Experimental congestive heart failure produced by rapid ventricular pacing in the dog: cardiac effects. 20. Cranefield PF, Aronson RS. Torsade de pointes and other pause-induced Circulation 1987;75:857-67. ventricular tachyarrhythmias: the short-long -short sequence and afterde­ 13. TenEick RE, Houser SR, Basset AL. Cardiac hypertrophy and altered polarizations. PACE 1988;11:670-8. cellular electrical activity of the myocardium: possible electrophysiolog­ 21. Ben-David J, Ayers GM, Pride HP, Zipes DP. Left ventricular hypertro­ ical basis for changes. In: Sperelakis N, ed. phy predisposes to development of early afterdepolarizations that cause Physiology and Pathophysiology of the Heart. 2nd ed. Boston: Kluwer ventricular tachycardia in dogs (abstr). Circulation 1990;82(suppliii):III- Academic. 1989:573-94. 99.