- Email: [email protected]
Modulated parasystole as a mechanism of ventricular ectopic activity leading to ventricular fibrillation
Modulated Parasystole a@ a Mechanism of Ventricular Ectopic Activity Leading to Ventricwlar Fibrillation Etienne 0. Robles cje Medina, MD, Mario Delmar, MD, PhD, Serge Sicouri, MD, and Jose Jalife, MD
The electrocardiograms of 2 patients with frequent premature ventiicular complexes characterized by variable coupling intervals and fusions with sinus activations were analyzed according to the modulated parasystole and reflection hypotheses of Moe et al. In adpition, th6 ectopic activiiy was aseociated with couplets, tachycardia and ventricular fibrillation. Departures from the “classic” criteria of parasystole could not be explained satisfactorily if a completely protected (insulated) pacemaker was assumed. In each instance a triphasig response curve could be constructed, suggesting that modulated parasystole was the mechanism common to botF patients. Couplets and runs of ventricular pchycardia were ascribed to single and repetitive reflection, respectively, in the presence of supernoimal excitability of the ectopic pacemaker, the ventricle or both. In these patients, fibrillation probably’resulted from spatial ndquniformity of the ventricular response to the reflected evenf during a phase of vulnerability. This study suggests that modulated parasystole in the presence of supernormal excitability may lead to very severe arrhythmias and trigger ventricular fibrillation. In the clinical settin,+, such patients may be misdiagnosed because of atypical tiatures. (AmJ Cardiol 1989;63:1326-1332)
ecent experimental1-5 and clinical observation& l2 suggest that the activity of a parasystolic pacemaker may be modulated by the electrical influence of events occurring in the surrounding excitable tissue. The cycle length of the pacemaker may be prolonged or abbreviated1-5 by a magnitude that depends upon the timing as well as the strength of the electrotonic current (that is, level of protection) mediating that influence. Modulation of parasystolic pacemaker activity may thus result in episodes of entrainment of the ectopic cycle, with fixed ‘or variable coupling of manifest discharges. 2,3*7~10 In addition, it has been recently suggested that tbe coexistence of parasystole and reflection in the setting of supernormal excitability7 may give rise to fast repetitive discharges that tiere previously thought to be diagnostic of circus movement reentry mechanisms.13-l5 Thus, modulation and reflection may lead to marked departure from .the classic crit& ria16J7 for parasystole. Under these circumstances, parasystole may go unrecognized, and it is possible that the prognostic significance of the arrhythmia per se may be masked in the presence of the so-called atypical manifestations. In this study, we present evidence suggesting that some instances in which premature complexes occur at variable coupling intervals and lead to ventricular fibrillation can be explained on the basis of modulated parasystole and reflection.
R
METHODS Patients: Patient 1 was a 42-year-old man who had
From the Section of Cardiology, Heart-Lung Institute, University Hospital and State University of Utrecht, Utrecht, the Netherlands, and the Department of Pharmacology, State University of New York Health Science Center at Syracuse, Syracuse, New York. This study was supported by a grant from the Wijnand M. Pon Foundation, Leusden, the Netherlands, and by grant RO 129439 from the National Institutes of Health, Bethesda, Maryland. The study was completed during Dr. Jalife’s tenure as an Established Investigator of the American Heart Association. Manuscript received January 3, 1989; revised manuscript received March 10, 1989, and accepted March 12. Address for reprints: Etienne 0. Robles de Medina, MD, HeartLung Institute, Sec$on of Cardiology, University Hospital, 101 Catharynesingel, 3511 GV Utrecht, the Netlierlands.
1326
THE AMERICAN
JOURNAL
OF CARDIOLOGY
VOLUME
63
undergone triple coronary artery bypass grafting at the age of 35 years. He was being treated with a diuretic for mild hypertension but was not receiving antiarrhythmic drugs. An exercise stress test had to be terminated prematurely in the absence of evidence of ischemia because of frequent ventricular ectopy, which was followed by ventricular fibrillation shortly afterwards. The patient was successfully resuscitated and subsequent evaluation revea@l patent grafts and a decreased serum potassium level of 2.7 mM/liter. Correction df the hypokalemia led to complete disappearance of ectopic activity and repeat stress testing was uneventful. Patient 2 was a 72-year-old man with -a recent anteroseptal infarction and episodes of nonsustained ventricular tachycardia. Shortly after admission, continuous recording of an electrocardiographic ‘monitoring
lead during a period of frequent ventricular ectopic activity fortuitously captured the onset of ventricular fibrillation. Use of medication before admission was unknown. Analysis of electrocardiograms: All electrocardiograms were recorded at a paper speed of 25 mm/s. A ventricular origin of the ectopic complexes was postulated, based on their dissociation from atria1 activity and the occurrence of ventricular fusions. Deductive reasoning was applied in differentiating various mechanisms that may lead to variable coupling of ectopic complexes or irregular intervals during parasystole. Neither record could be explained entirely if a regularly firing, completely protected (insulated) pacemaker was assumed; The data were subsequently analyzed on the basis of the modulated parasystole and reflection hypotheses of Moe,2 Jalife, 1,3 Antzelevich4 and their coworkers to determine whether the arrhythmic patterns could be reproduced by assuming variable electrotonic effects and entrainment ratios (that is, the ratio between number of sinus complexes and number of modulated ectopic discharges). Consti-uction of phase-response curves: Phase-response curves were constructed as described by Jalife et a1.7 In patient 1, the mean of the long intervals between consecutive ectopic activations (840 ms) (Figure 1) was used as the intrinsic period. In patient 2, where the parasystolic cycle was not apparent, an “inverse solution” was attempted by initially considering a plausible intrinsic parasystolic period (XX). It was then attempted to fit, empirically, the possible cycle length changes associated with the electrotonic modulation exerted by intervening ventricular responses (R). The analysis started with those interectopic intervals (XRX) that
contained only 1 intervening R complex. These intervals were plotted in terms of percentage of the assumed XX against the respective XR intervals, also expressed as percentage of XX; where 100% XX is equal to the intrinsic parasystolic period. Once the best fit for each of these points was obtained, a “tentative” phase-response curve was constructed. Subsequently, the entire electrocardiogram was analyzed by consecutive measurements of the intervals, and by plotting on the phase-response curve the individual points that corresponded to the concealed or manifest ectopic events.’ Models of electrotonic interaction: In an attempt to explain the overall cellular mechanism of the arrhythmic patterns, armchair models of dynamic electrotonic interaction between the ventricle and the protected pacemaker site were constructed for those instances in which repetitive discharges followed the modulated parasystolic event (Figure 3). These models and their underlying assumptions of (self-)modulation, exit conduction impairment, supernormality and reflection are based on actual experimental data in isolated Purkinje fiber preparations.4>5,18 RESULTS Patient 1: In the example provided by Figure 1, sinus rhythm at a rate of 82 to 96/min with QTc interval of 440 ms is interrupted by frequent ventricular premature complexes, which often occur in groups of 2 or 3 in a row. The latter are composed of both long-short and short-long cycles (Figure 1A). Markedly varying coupling intervals, fusion complexes and long interectopic intervals that are multiples of 400 to 440 ms suggest a “typical” parasystolic tachycardia (136 to 150/min) with exit block.19 At the end of Figure lC, a parasystol-
FlGU,Rg 1. Ekctrecar&ogram of patient 1. With the exception ef a ?-second interval between A and R, the tracing is continirous. The nun$efs in the upper row indiie coupling intervals of ectopic activations, whereas those in the middle indiie intere&tpic intervals. The poditive and negative numbers below each trace indicate the percent change of the intrinsic ectopic cycle length. This was determined from the phase-response curve shown in Figure 2. The uprfgfrf arrows symbolize the position of manifest or conceakd parasystolic activations. The T-crossed bar in A indiites exit block of a self-modulated parasystolii impulse (see also Fire 3). The asterisk represents a premature complex of different origin. Notice close similarity between R-enT complex in C and other parasystolit activations. The electrocardtogram in this and subsequent figures was rewrded at 25 mm/s and the intervals are expressed in ms. F = fusion complex; I = intrinsic parasystolic period (mean 840 ms).
THE AMERICAN
JOURNAL
OF CARDIOLOGY
JUNE
1, 1989
1327
VENTRICULAR
PARASYSTOLE
AND
FIBRBLATION
ic complex is followed at an interval of 320 ms by a premature ventricular activation with R-on-T phenomenon that heralds the onset of ventricular fibrillation. The close similarity in configuration between the R-onT complex and parasystolic complexes is striking and suggests a common origin. However, simple dissipation of the exit block during an almost regular parasystolic tachycardia does not provide a satisfactory explanation for the sudden shortening to 320 ms of the ectopic interval of the R-on-T complex. For the analysis of this pattern, we have assumed the presence of a phase of supernormal excitability in both the parasystolic pacemaker and the surrounding tissue beyond the area of “protection.” In addition, postrepolarization refractoriness4,20 and various degrees of cycle length-dependent exit delay4 were postulated, which could give rise to self-modulation of the ectopic pacemaker discharge through the ensuing ventricular response. On the other hand, it was assumed that the ectopic complex marked with an asterisk in Figure 1A in all probability is not a fusion complex because it occurs at a time when neither a sinus nor a coupled parasystolic activation is expected. Figure 2 shows the phase-response curve obtained as an inverse solution for this trace when the mean long ectopic interval of 840 ms was used as the intrinsic pacemaker period (XX, dotted horizontal line). The construction of the curve of modulation was done empirically. As indicated by the upward arrows in Figure 1, as many as 30 interectopic intervals were measured, which resulted in a triphasic response curve (Figure 2) with most points falling on or very near the predicted relation.
FIGURE 3. A, armchair model of the possible dynamic interactions between a protected pacemaker site (lower frace) and the surrounding ventricular tissue (middle trace). The eiectrocardiographic strip shown in the indet was taken from Figure 1A. The arrow pointing downward in the middle trace indicates exit bkwk (see also B, this figure). 6, activation curve derived from the electrocardiogram in Figure 1. Yhe exit conduction time (X-R’) is plotted as a functktn of preceding R-X interval. Symbols represent exit conduction times of corresponding ventricular activations in A and C. C, simulated events as they might have occurred at the celkdar level during the onset of ventricular tachycardia and fibrillatkm in Figure 1C (inset). Format and symbols as in A and B of this figure. R = ventricular response; R’ = ventricular response resuiting from preceding X; X = parasystolii discharge.
FIGURE 2. Triphasic response curve derived from the electrocardiogram shown in Figure 1 (patient 1). lnterectopic intervals (ordinate) plotted as a function of the position of intervening ventrtcular activations (abscissa), expressed as percentage of intri~c cycle (XX). Symbols represent individual XRX hdervals measured in A (closed squares), B (open squares) and C (c/osed friaag/es) of the trace in Figure 1.
1328
THE AMERICAN
JOURNAL
OF CARDIOLOGY
VOLUME
According to the curve in Figure 2, ventricular action potentials produced delays of as much as 25% and accelerations of 62% (see also the positive and negative numbers in Figure l), depending on the timing of the R response. The greatest abbreviations were produced for those instances in which the R response occurred during the supernormal phase of the parasystolic pacemaker. Such instances gave rise to the various runs of ventricular couplets and ultimately triggered fibrillation. This is further shown in Figure 3. In the top traces of Figure 3A and C, we have reproduced the segments of the electrocardiogram showing the development of the actual events. The middle traces are reconstructions of the ventricular responses (R) in the immediate vicinity of the area of protection, and the bottom traces are those representing the action potentials (X) in the pacemaker site. The consecutive changes
63
(downward arrow in Figure 3A) and resulted in a discontinuity in the ventricular activation curve (X-R’ block in Figure 3B). According to this interpretation, the 2 couplets in Figure 3A were the result of supernormal modulation of the parasystolic pacemaker, with exit propagation at relatively long X-R’ intervals (closed triangle and square in Figure 3, A and B). However, in Figure 3C, supernormal capture of the parasystolic pacemaker at an R-X of 220 ms led to a critically coupled premature complex (closed diamond in Figure 3, B and C), which triggered a sequence of repetitive discharges that degenerated into ventricular fibrillation. Hence, in this patient, shortening of the coupling interval of the R-on-T complex may be explained by a sudden abbreviation of exit conduction time due to the presence of a relatively narrow window of cycle lengthdependent supernormal excitability of the tissue in the immediate vicinity of the protection zone. Patient 2: The electrocardiograms presented in Figures 4 and 5 show sinus rhythm with frequent premature ventricular complexes. Some show marked variation of coupling intervals (e.g., 600 and 440 in Figure 4B) and are followed by occasional fusions with the sinus complexes. Long interectopic intervals separating
in parasystolic pacemaker periodicity are those induced electrotonically by the ventricular responses, and the period durations all fit the response curve in Figure 2. Similarly, the resulting ventricular patterns associated with parasystolic modulation and various degrees of exit delay can be derived from the activation curve shown in Figure 3B, in which we have plotted the pacemaker-toventricle exit delay interval (X-R’) for each ventricular response as a function of the immediately preceding RX interval. The various symbols represent the delays incurred during each individual complex. With a “basal” exit delay of 100 ms at R-X intervals of 1750 ms, a fairly close prediction of the events could be derived when we allowed for a discontinuity in the ventricular activation curve4Jo resulting from postrepolarization refractoriness across the exit pathway. For example, the first R-X interval shown (660 ms) gave rise to an X-R’ of 112 ms (open square in Figure 3, A and B), and led to a capture of the parasystolic pacemaker during its repolarization phase and to a subsequent ventricular reactivation with an X-R’ of 200 ms (closed triangle in Figure 3, A and B), thus leading to a ventricular couplet in the electrocardiogram. On the other hand, the following R-X interval (600 ms) was such that the pacemaker discharge failed to propagate to the ventricle
7750
FIGURE 4. Electrocardiigram of patient 2. Note that a, ib and c are parts of a continuous strip. The ectopic activations showing variable coupling intervals are indiited by closed &c/es. The numbers above each trace represent interectopic intervals, those below the trace indiite XR, RX or sinus intervals, as appropriate. F = fusion complex.
4710
c
c
2840
'0
c
-
(3x1570)
b FIGURE 5. Electrocardiogram (continusd) of patient 2; b, c and dare continuous. Format as in Figure 4.
(5x15501
2960
,* 1200
^,
d
THE AMERICAN
JOURNAL
OF CARDIOLOGY
JUNE
1, 1989
1329
VENTRICULAR
PARASYSTOLE
AND
FIBRILLATION
these R complexes appear to be multiples of 1,440 to 1,570 ms, which suggests a parasystolic mechanism. However, every time a variably coupled ectopic event is followed by a compensatory pause 2680 ms, the subsequent ectopic complex of similar QRS configuration is more premature, and occurs at a fixed coupling interval with respect to the preceding sinus complex (rule of bigeminy21), which results in shortening of the XRX intervals to 1,120 to 1,240 ms. This series of events was repeated for a number of cycles at a time, thus giving rise to runs of bigeminal rhythm. In some occasions, the closely coupled premature complexes were followed by repetitive discharges, resulting in couplets or in brief runs of nonsustained ventricular tachycardia. Finally, in Figure 5C, an apparently similar sequence of events deteriorated into ventricular fibrillation. With an average sinus cycle length of 600 ms, the best fit for interpreting the arrhythmia pattern in patient 2 was obtained with a triphasic response curve when an intrinsic parasystolic pacemaker period of 1,940 ms was assumed. According to our interpretation (Figure 6), the ventricular discharges caused a maximal prolongation of the ectopic cycle of about 35%, abrupt transition to maximal acceleration of 45% at about 30% (abscissa) and a supernormal phase of excitability at about 15%. The open squares correspond to those ina
DISCUSSION
r
FIGURE 6. Triphasic reeponse curve derived cardiogram of patient 2. Format as in Figure represents the limit of maximal ectopic cycle sible at a given XR interval.
1330
stances in which there was only 1 normal ventricular complex between 2 ectopics. With the assumed intrinsic parasystolic pacemaker period of 1,940 ms, all these complexes must have occurred at about 30% of the ectopic period or longer, that is, during the acceleration phase. The black dots represent the XRX intervals in which at least 1 of the ectopic discharges was concealed; thus, there were >l sinus activation separating 2 manifest ectopic events (see Jalife et al,7 their Figure 3, for an example of sequential measurements in the analysis of electrotonic influence on concealed ectopic discharges). The closed triangles represent those instances in which the ectopic pacemaker was assumed to be captured during its “supernormal” phase. On the basis of this phase response curve, the parasystolic pattern of patient 2 can be readily reproduced. Moreover, in the presence of a phase of supernormal excitability in both the parasystolic pacemaker and surrounding excitable tissues, and with appropriate delay to and from the parasystolic pacemaker, reciprocating activity-indistinguishable from reflection-may occur and sometimes repeat for several cycles, giving rise to couplets and episodes of nonsustained ventricular tachycardia (Figures 4 and 5). On the other hand, it may also lead to sufficient disorganization in the ventricular activation sequence to result in sustained reentrant activity or fibrillation.
THE AMERICAN
JOURNAL
from the electro2. The diagond abbreviation poe-
OF CARDIOLOGY
VOLUME
63
Appraisal of the analysis: In both patients, ventricular parasystole was suggested by the markedly varying coupling intervals of the ectopic complexes, the fusions with activations of sinus origin and the episodes during which a rather fixed common denominator of the interectopic intervals could be identified.16v17 However, in both patients some unusual characteristics were seen. Patient 1 could well represent an example of parasystolic tachycardia at a rate of 136 to 150/min with exit block.19 However, to agree with this interpretation, sudden shortening of the ectopic interval of the R-on-T complex in Figure 1C would have to be attributed to a fortuitous event in which another pacemaker, located close to (but outside) the parasystolic focus, fired during the T wave of the preceding parasystolic complex. On the other hand, in patient 2 (Figures 4 and 5), it seems highly unlikely that switching to bigeminal rhythm with fixed coupling and marked abbreviation of the ectopic intervals was due to dissipation of exit block during a rapid tachycardia with cycle lengths of 240 to 280 ms. First, this would imply an unusually rapid rate of the parasystolic focus, with variation of the intrinsic period of up to f8% from the mean. Second, this interpretation would not explain why the exit block consistently and predictably disappeared, sometimes repetitively, after postectopic intervals 1680 ms. Alternatively, the arrhythmic pattern in patient 2 could result from alternation between typical parasystole with an intrinsic period of 1,440 to 1,570 ms, and manifest and concealed, single and repetitive reentry
within or in the immediate vicinity of the ectopic foCUS.~~-~~ Microreentry due to longitudinal dissociation in the pathway containing the ectopic focus15322may have been enhanced in the setting of the rule of bigeminy.21 According to this line of thought, the interectopic intervals that are longer than a simple multiple of the parasystolic cycle must be explained by concealed reentry,13-15 for example, the interval of 1,880 ms in Figure 5C. This interval may be composed of 1 concealed reentrant cycle of 280 ms plus a parasystolic cycle of 1,600 ms. This, however, also leads to the conclusion that the subsequent manifest ectopic complex which started the run of tachycardia that deteriorated into ventricular fibrillation (Figure 5C) is of parasystolic and not reentrant origin. This is in keeping with the observation that the penultimate RR interval preceding sustainment of the arrhythmia measured only 580 ms, which is significantly shorter than the minimal postectopic interval (680 ms) that heralds fixed coupling of extrasystolic activity. In fact, it was the uncertainty about the reentrant nature of the latter complexes, as well as the assumption of concealed reentrant activity following these events, that led us to consider the possibility of modulated parasystole. Schamroth and Marriott23 have attributed transitions of parasystole to extrasystolic bigeminy to the enhancing effect of Wedensky phenomena. However, their interpretation was subsequently challenged by Moe et a1,24who suggested a modulated parasystole. Unifying hypothesis: In both patients, phase-response curves could be constructed that allowed for accurate prediction of the positions of manifest and concealed ectopic discharges. This required the assumption of phase-related electrotonic effects on the ectopic pacemaker.le5 In addition, frequency-dependent exit conduction delays,4 postrepolarization refractoriness4,20 and supernormal excitability4,5,25 in both the ectopic pacemaker and surrounding ventricular myocardium had to be assumed. The validity of all these assumptions has been demonstrated experimentally4,5,1s and a number of studies have provided additional support for their applicability in the clinical setting.6m12Therefore, our empirically derived models (Figure 3 and Jalife et a1,7 their Figure 3) can be taken to represent close approximations of the possible dynamic interactions between the ventricle and the protected pacemaker site, which may have led to the severe arrhythmias in these patients. Alternatively, one could assume that the modulated parasystolic pacemaker also has the potential of developing triggered activity26 in the form of early afterdepolarizations. This might have triggered secondary or tertiary parasystolic discharges leading to extrasystoles, couplets and tachycardia or, as in patient 1, possibly also torsade de pointes.27 Prognostic significance: To date, only 1 study has described the initiation (and termination) of sustained ventricular tachycardia by a complex of parasystolic origin. The benign prognosis of ventricular parasystole has recently been confirmed by Kuo and Surawicz.14 However, if our interpretation is correct, our patients do
suggest the possibility that at least some instances of premature ventricular ectopic activity that degenerates into ventricular tachycardia or fibrillation may represent the expression of a modulated parasystolic pacemaker that is not recognized because of its atypical behavior. The observation that a modulated parasystole may result in ventricular tachycardia and fibrillation when there is a phase of supernormal excitability in the pacemaker cycle and possibly also in the ventricular site may also shed some light on the benign prognosis of pure or “typical” cases. With little or no modulation, high levels of protection may likewise prevent the early exit of parasystolic impulses. Hence, there will be a reduced chance of meeting sufficient dispersion of refractoriness to initiate a tachycardia or fibrillation. The level of protection of the ectopic pacemaker may thus determine the prognosis of parasystole. Further studies are needed to confirm or reject this hypothesis. Acknowledgment: We thank Justus Anumonwo for reading the manuscript and Wanda Coombs and Gertrude van Eck for excellent secretarial assistance.
REFERENCES 1. Jalife J, Moe GK. Effect of electrotonic potentials on pacemaker activity of canine Purkinje fibers in relation to parasystole. Circ Res 1976;39:802-808. 2. Moe GK, Jalife J, Mueller WJ, Moe B. A mathematical model of parasystole and its application to clinical arrhythmia. Circulation 1977;56;968-979. 3. Jalife J, Moe GK. A biologic model of parasystole. Am J Cardiol1979;43;761772. 4. Antzelevitch C, Jalife J, Moe GK. Characteristics of reflection as a mechanism of re-entrant arrhythmias and its relationship to parasystole. Circulation 1980; 61:182-191. 5. Antzelevitch C, Jalife J, Moe GK. Electrotonic modulation of pacemaker activity: further biological and mathematical observations on the behavior of modulated parasystole. Circulation 1982;66:1225-1232. 6. Furuse A, Matsuo H, Saigusa M. Effects of intervening beats on ectopic cycle length in a patient with ventricular parasystole. Jpn Heart J 1981;22t201-209. 7. Jalife J, Antzelevitch C, Moe GK. The case for modulated parasystole. PACE 1982;91 l-926. 8. Nau GJ, Aldariz AE, Acunzo RS, Halpern MS, Davidenko JM, Elizari MV, Rosenbaum MB. Modulation of parasystolic activity by nonparasystolic beats. Circulation 1982;66:462-469. 9. Oreto G, Satullo G, Luzza F, Consolo F, Schamroth L. Supernormal modulation of ventricular parasystole: the triphasic phase-response curve. Am J Cardiol 1986;57:283-290. 10. Jalife J, Michaels DC, Langendorf R. Modulated parasystole originating in the sinoatrial node. Circulation 1986;74:94.5-954. 11. Tenczer J, Littmann L. Rate-dependent patterns of modulated ventricular parasystole. Am J Cardiol 1986;57:576-581. 12. Tenczer J, Littmann L, Rohla M, Wu DB, Fenyvesi T. A study of modulated ventricular parasystole by programmed stimulation. Am J Cardiol1987;59:846851. 13. Singer DH, Parameswaran R, Drake FT, Meyers SN, Deboer AA. Ventricular parasystole and reentry: clinical-electrophysiological correlations. Am Heart J 1974:88:79-87. 14. Kuo CS, Surawicz B. Coexistence of ventricular parasystole and ventricular couplets: mechanism and clinical significance. Am J Cardiol 1979;33:435-441. 15. Kinoshita S, Kawasaki T. Ventricular parasystole and couplets: the concept of longitudinal dissociation in the microreentry pathway. Am Heart J 1983;105; 1050-l os5. 16. Katz LN, Pick A. Clinical Electrocardiography. Part I. The Arrhythmias. Philadelphia: Lea and Febiger, 1956:185. 17. Scherf D, Schott A. Extrasystoles and Allied Arrhythmias. London William Heineman Medical Books, 1973t269. 18. Antzelevitch C, Bernstein MJ, Feldman HN, Moe GK. Parasystole, x-entry, and tachycardia: a canine preparation of cardiac arrhythmias occurring across inexcitable segments of tissue. Circulation 1983:68:1101-l 115. 19. Scherf D, Borneman C. Parasystole with a rapid ventricular centre. Am Heart J 1961;62:320-331,
THE AMERICAN
JOURNAL
OF CARDIOLOGY
JUNE
1, 1989
1331
VENTRICULAR
PARASYSTOLE
AND FIBRILLATION
20. Davidenko JM, Antzelevitch C. Electrophysiological mechanisms underlying rate-dependent changes in refractoriness in normal and segmentally depressed canine Porkinje fibers. The characteristics of post-repolarization refractoriness. Circ Res 1986;58:257-268. 21. Langendorf R, Pick A, Winternitz M. Mechanisms of intermittent ventricular bigeminy. I. Appearance of ectopic beats dependent upon length of the ventricular cycle, the “rule of bigeminy.” Circulation 1955;l Z:422-430. 22. Kinoshita S, Kato Y, Kawasaki T, Okimori K. Ventricular tachycardia initiated by late-coupled ventricular extrasystoles: the concept of longitudinal dissociation in the microreentry pathway. Am Heart J 1982;103:1090-1095. 23. Schamroth L, Marriott HJL. Intermittent ventricular parasystole with observations on its relationship to extrasystolic bigeminy. Am J Cardiol 1961;7:799809.
1332
THE AMERICAN
JOURNAL
OF CARDIOLOGY
VOLUME
63
24. Moe GK, Jalife J, Mueller WJ. Reciprocation between pacemaker sites: reentrant parasystole? In: Kulbertus HE, ed. Re-Entrant Arrhythmias. Mechanisms and Treatment. Lancaster: MTP Press, 1977:271-280. 25. Spear JF, Moore EN. Supernormal excitability and conduction in the HisPurklnje system of the dog. Circ Res 1974;35:782-792. 26. Wit AL, Cranefield PF, Gadsby DC. Triggered activity. In: Zipes DP, Bailey JC, Alharrar V, eds. The Slow Inward Current and Cardiac Arrhythmias. The Hague: Martinus
Nijhoff,
1980:437-454.
27. Brachmann J, Scherlag BJ, Rosenshtraukh LV, Lazzara R. Bradycardiadependent triggered activity: relevance to drug induced multiform ventricular tachycardia. Circulation 1983;68:846-856. 26. Jacobsen LB. Spontaneous ventricular parasystole initiating ventricular tachycardia. J Electrocardiol 1973,6:63-70.
The electrocardiograms of 2 patients with frequent premature ventiicular complexes characterized by variable coupling intervals and fusions with sinus activations were analyzed according to the modulated parasystole and reflection hypotheses of Moe et al. In adpition, th6 ectopic activiiy was aseociated with couplets, tachycardia and ventricular fibrillation. Departures from the “classic” criteria of parasystole could not be explained satisfactorily if a completely protected (insulated) pacemaker was assumed. In each instance a triphasig response curve could be constructed, suggesting that modulated parasystole was the mechanism common to botF patients. Couplets and runs of ventricular pchycardia were ascribed to single and repetitive reflection, respectively, in the presence of supernoimal excitability of the ectopic pacemaker, the ventricle or both. In these patients, fibrillation probably’resulted from spatial ndquniformity of the ventricular response to the reflected evenf during a phase of vulnerability. This study suggests that modulated parasystole in the presence of supernormal excitability may lead to very severe arrhythmias and trigger ventricular fibrillation. In the clinical settin,+, such patients may be misdiagnosed because of atypical tiatures. (AmJ Cardiol 1989;63:1326-1332)
ecent experimental1-5 and clinical observation& l2 suggest that the activity of a parasystolic pacemaker may be modulated by the electrical influence of events occurring in the surrounding excitable tissue. The cycle length of the pacemaker may be prolonged or abbreviated1-5 by a magnitude that depends upon the timing as well as the strength of the electrotonic current (that is, level of protection) mediating that influence. Modulation of parasystolic pacemaker activity may thus result in episodes of entrainment of the ectopic cycle, with fixed ‘or variable coupling of manifest discharges. 2,3*7~10 In addition, it has been recently suggested that tbe coexistence of parasystole and reflection in the setting of supernormal excitability7 may give rise to fast repetitive discharges that tiere previously thought to be diagnostic of circus movement reentry mechanisms.13-l5 Thus, modulation and reflection may lead to marked departure from .the classic crit& ria16J7 for parasystole. Under these circumstances, parasystole may go unrecognized, and it is possible that the prognostic significance of the arrhythmia per se may be masked in the presence of the so-called atypical manifestations. In this study, we present evidence suggesting that some instances in which premature complexes occur at variable coupling intervals and lead to ventricular fibrillation can be explained on the basis of modulated parasystole and reflection.
R
METHODS Patients: Patient 1 was a 42-year-old man who had
From the Section of Cardiology, Heart-Lung Institute, University Hospital and State University of Utrecht, Utrecht, the Netherlands, and the Department of Pharmacology, State University of New York Health Science Center at Syracuse, Syracuse, New York. This study was supported by a grant from the Wijnand M. Pon Foundation, Leusden, the Netherlands, and by grant RO 129439 from the National Institutes of Health, Bethesda, Maryland. The study was completed during Dr. Jalife’s tenure as an Established Investigator of the American Heart Association. Manuscript received January 3, 1989; revised manuscript received March 10, 1989, and accepted March 12. Address for reprints: Etienne 0. Robles de Medina, MD, HeartLung Institute, Sec$on of Cardiology, University Hospital, 101 Catharynesingel, 3511 GV Utrecht, the Netlierlands.
1326
THE AMERICAN
JOURNAL
OF CARDIOLOGY
VOLUME
63
undergone triple coronary artery bypass grafting at the age of 35 years. He was being treated with a diuretic for mild hypertension but was not receiving antiarrhythmic drugs. An exercise stress test had to be terminated prematurely in the absence of evidence of ischemia because of frequent ventricular ectopy, which was followed by ventricular fibrillation shortly afterwards. The patient was successfully resuscitated and subsequent evaluation revea@l patent grafts and a decreased serum potassium level of 2.7 mM/liter. Correction df the hypokalemia led to complete disappearance of ectopic activity and repeat stress testing was uneventful. Patient 2 was a 72-year-old man with -a recent anteroseptal infarction and episodes of nonsustained ventricular tachycardia. Shortly after admission, continuous recording of an electrocardiographic ‘monitoring
lead during a period of frequent ventricular ectopic activity fortuitously captured the onset of ventricular fibrillation. Use of medication before admission was unknown. Analysis of electrocardiograms: All electrocardiograms were recorded at a paper speed of 25 mm/s. A ventricular origin of the ectopic complexes was postulated, based on their dissociation from atria1 activity and the occurrence of ventricular fusions. Deductive reasoning was applied in differentiating various mechanisms that may lead to variable coupling of ectopic complexes or irregular intervals during parasystole. Neither record could be explained entirely if a regularly firing, completely protected (insulated) pacemaker was assumed; The data were subsequently analyzed on the basis of the modulated parasystole and reflection hypotheses of Moe,2 Jalife, 1,3 Antzelevich4 and their coworkers to determine whether the arrhythmic patterns could be reproduced by assuming variable electrotonic effects and entrainment ratios (that is, the ratio between number of sinus complexes and number of modulated ectopic discharges). Consti-uction of phase-response curves: Phase-response curves were constructed as described by Jalife et a1.7 In patient 1, the mean of the long intervals between consecutive ectopic activations (840 ms) (Figure 1) was used as the intrinsic period. In patient 2, where the parasystolic cycle was not apparent, an “inverse solution” was attempted by initially considering a plausible intrinsic parasystolic period (XX). It was then attempted to fit, empirically, the possible cycle length changes associated with the electrotonic modulation exerted by intervening ventricular responses (R). The analysis started with those interectopic intervals (XRX) that
contained only 1 intervening R complex. These intervals were plotted in terms of percentage of the assumed XX against the respective XR intervals, also expressed as percentage of XX; where 100% XX is equal to the intrinsic parasystolic period. Once the best fit for each of these points was obtained, a “tentative” phase-response curve was constructed. Subsequently, the entire electrocardiogram was analyzed by consecutive measurements of the intervals, and by plotting on the phase-response curve the individual points that corresponded to the concealed or manifest ectopic events.’ Models of electrotonic interaction: In an attempt to explain the overall cellular mechanism of the arrhythmic patterns, armchair models of dynamic electrotonic interaction between the ventricle and the protected pacemaker site were constructed for those instances in which repetitive discharges followed the modulated parasystolic event (Figure 3). These models and their underlying assumptions of (self-)modulation, exit conduction impairment, supernormality and reflection are based on actual experimental data in isolated Purkinje fiber preparations.4>5,18 RESULTS Patient 1: In the example provided by Figure 1, sinus rhythm at a rate of 82 to 96/min with QTc interval of 440 ms is interrupted by frequent ventricular premature complexes, which often occur in groups of 2 or 3 in a row. The latter are composed of both long-short and short-long cycles (Figure 1A). Markedly varying coupling intervals, fusion complexes and long interectopic intervals that are multiples of 400 to 440 ms suggest a “typical” parasystolic tachycardia (136 to 150/min) with exit block.19 At the end of Figure lC, a parasystol-
FlGU,Rg 1. Ekctrecar&ogram of patient 1. With the exception ef a ?-second interval between A and R, the tracing is continirous. The nun$efs in the upper row indiie coupling intervals of ectopic activations, whereas those in the middle indiie intere&tpic intervals. The poditive and negative numbers below each trace indicate the percent change of the intrinsic ectopic cycle length. This was determined from the phase-response curve shown in Figure 2. The uprfgfrf arrows symbolize the position of manifest or conceakd parasystolic activations. The T-crossed bar in A indiites exit block of a self-modulated parasystolii impulse (see also Fire 3). The asterisk represents a premature complex of different origin. Notice close similarity between R-enT complex in C and other parasystolit activations. The electrocardtogram in this and subsequent figures was rewrded at 25 mm/s and the intervals are expressed in ms. F = fusion complex; I = intrinsic parasystolic period (mean 840 ms).
THE AMERICAN
JOURNAL
OF CARDIOLOGY
JUNE
1, 1989
1327
VENTRICULAR
PARASYSTOLE
AND
FIBRBLATION
ic complex is followed at an interval of 320 ms by a premature ventricular activation with R-on-T phenomenon that heralds the onset of ventricular fibrillation. The close similarity in configuration between the R-onT complex and parasystolic complexes is striking and suggests a common origin. However, simple dissipation of the exit block during an almost regular parasystolic tachycardia does not provide a satisfactory explanation for the sudden shortening to 320 ms of the ectopic interval of the R-on-T complex. For the analysis of this pattern, we have assumed the presence of a phase of supernormal excitability in both the parasystolic pacemaker and the surrounding tissue beyond the area of “protection.” In addition, postrepolarization refractoriness4,20 and various degrees of cycle length-dependent exit delay4 were postulated, which could give rise to self-modulation of the ectopic pacemaker discharge through the ensuing ventricular response. On the other hand, it was assumed that the ectopic complex marked with an asterisk in Figure 1A in all probability is not a fusion complex because it occurs at a time when neither a sinus nor a coupled parasystolic activation is expected. Figure 2 shows the phase-response curve obtained as an inverse solution for this trace when the mean long ectopic interval of 840 ms was used as the intrinsic pacemaker period (XX, dotted horizontal line). The construction of the curve of modulation was done empirically. As indicated by the upward arrows in Figure 1, as many as 30 interectopic intervals were measured, which resulted in a triphasic response curve (Figure 2) with most points falling on or very near the predicted relation.
FIGURE 3. A, armchair model of the possible dynamic interactions between a protected pacemaker site (lower frace) and the surrounding ventricular tissue (middle trace). The eiectrocardiographic strip shown in the indet was taken from Figure 1A. The arrow pointing downward in the middle trace indicates exit bkwk (see also B, this figure). 6, activation curve derived from the electrocardiogram in Figure 1. Yhe exit conduction time (X-R’) is plotted as a functktn of preceding R-X interval. Symbols represent exit conduction times of corresponding ventricular activations in A and C. C, simulated events as they might have occurred at the celkdar level during the onset of ventricular tachycardia and fibrillatkm in Figure 1C (inset). Format and symbols as in A and B of this figure. R = ventricular response; R’ = ventricular response resuiting from preceding X; X = parasystolii discharge.
FIGURE 2. Triphasic response curve derived from the electrocardiogram shown in Figure 1 (patient 1). lnterectopic intervals (ordinate) plotted as a function of the position of intervening ventrtcular activations (abscissa), expressed as percentage of intri~c cycle (XX). Symbols represent individual XRX hdervals measured in A (closed squares), B (open squares) and C (c/osed friaag/es) of the trace in Figure 1.
1328
THE AMERICAN
JOURNAL
OF CARDIOLOGY
VOLUME
According to the curve in Figure 2, ventricular action potentials produced delays of as much as 25% and accelerations of 62% (see also the positive and negative numbers in Figure l), depending on the timing of the R response. The greatest abbreviations were produced for those instances in which the R response occurred during the supernormal phase of the parasystolic pacemaker. Such instances gave rise to the various runs of ventricular couplets and ultimately triggered fibrillation. This is further shown in Figure 3. In the top traces of Figure 3A and C, we have reproduced the segments of the electrocardiogram showing the development of the actual events. The middle traces are reconstructions of the ventricular responses (R) in the immediate vicinity of the area of protection, and the bottom traces are those representing the action potentials (X) in the pacemaker site. The consecutive changes
63
(downward arrow in Figure 3A) and resulted in a discontinuity in the ventricular activation curve (X-R’ block in Figure 3B). According to this interpretation, the 2 couplets in Figure 3A were the result of supernormal modulation of the parasystolic pacemaker, with exit propagation at relatively long X-R’ intervals (closed triangle and square in Figure 3, A and B). However, in Figure 3C, supernormal capture of the parasystolic pacemaker at an R-X of 220 ms led to a critically coupled premature complex (closed diamond in Figure 3, B and C), which triggered a sequence of repetitive discharges that degenerated into ventricular fibrillation. Hence, in this patient, shortening of the coupling interval of the R-on-T complex may be explained by a sudden abbreviation of exit conduction time due to the presence of a relatively narrow window of cycle lengthdependent supernormal excitability of the tissue in the immediate vicinity of the protection zone. Patient 2: The electrocardiograms presented in Figures 4 and 5 show sinus rhythm with frequent premature ventricular complexes. Some show marked variation of coupling intervals (e.g., 600 and 440 in Figure 4B) and are followed by occasional fusions with the sinus complexes. Long interectopic intervals separating
in parasystolic pacemaker periodicity are those induced electrotonically by the ventricular responses, and the period durations all fit the response curve in Figure 2. Similarly, the resulting ventricular patterns associated with parasystolic modulation and various degrees of exit delay can be derived from the activation curve shown in Figure 3B, in which we have plotted the pacemaker-toventricle exit delay interval (X-R’) for each ventricular response as a function of the immediately preceding RX interval. The various symbols represent the delays incurred during each individual complex. With a “basal” exit delay of 100 ms at R-X intervals of 1750 ms, a fairly close prediction of the events could be derived when we allowed for a discontinuity in the ventricular activation curve4Jo resulting from postrepolarization refractoriness across the exit pathway. For example, the first R-X interval shown (660 ms) gave rise to an X-R’ of 112 ms (open square in Figure 3, A and B), and led to a capture of the parasystolic pacemaker during its repolarization phase and to a subsequent ventricular reactivation with an X-R’ of 200 ms (closed triangle in Figure 3, A and B), thus leading to a ventricular couplet in the electrocardiogram. On the other hand, the following R-X interval (600 ms) was such that the pacemaker discharge failed to propagate to the ventricle
7750
FIGURE 4. Electrocardiigram of patient 2. Note that a, ib and c are parts of a continuous strip. The ectopic activations showing variable coupling intervals are indiited by closed &c/es. The numbers above each trace represent interectopic intervals, those below the trace indiite XR, RX or sinus intervals, as appropriate. F = fusion complex.
4710
c
c
2840
'0
c
-
(3x1570)
b FIGURE 5. Electrocardiogram (continusd) of patient 2; b, c and dare continuous. Format as in Figure 4.
(5x15501
2960
,* 1200
^,
d
THE AMERICAN
JOURNAL
OF CARDIOLOGY
JUNE
1, 1989
1329
VENTRICULAR
PARASYSTOLE
AND
FIBRILLATION
these R complexes appear to be multiples of 1,440 to 1,570 ms, which suggests a parasystolic mechanism. However, every time a variably coupled ectopic event is followed by a compensatory pause 2680 ms, the subsequent ectopic complex of similar QRS configuration is more premature, and occurs at a fixed coupling interval with respect to the preceding sinus complex (rule of bigeminy21), which results in shortening of the XRX intervals to 1,120 to 1,240 ms. This series of events was repeated for a number of cycles at a time, thus giving rise to runs of bigeminal rhythm. In some occasions, the closely coupled premature complexes were followed by repetitive discharges, resulting in couplets or in brief runs of nonsustained ventricular tachycardia. Finally, in Figure 5C, an apparently similar sequence of events deteriorated into ventricular fibrillation. With an average sinus cycle length of 600 ms, the best fit for interpreting the arrhythmia pattern in patient 2 was obtained with a triphasic response curve when an intrinsic parasystolic pacemaker period of 1,940 ms was assumed. According to our interpretation (Figure 6), the ventricular discharges caused a maximal prolongation of the ectopic cycle of about 35%, abrupt transition to maximal acceleration of 45% at about 30% (abscissa) and a supernormal phase of excitability at about 15%. The open squares correspond to those ina
DISCUSSION
r
FIGURE 6. Triphasic reeponse curve derived cardiogram of patient 2. Format as in Figure represents the limit of maximal ectopic cycle sible at a given XR interval.
1330
stances in which there was only 1 normal ventricular complex between 2 ectopics. With the assumed intrinsic parasystolic pacemaker period of 1,940 ms, all these complexes must have occurred at about 30% of the ectopic period or longer, that is, during the acceleration phase. The black dots represent the XRX intervals in which at least 1 of the ectopic discharges was concealed; thus, there were >l sinus activation separating 2 manifest ectopic events (see Jalife et al,7 their Figure 3, for an example of sequential measurements in the analysis of electrotonic influence on concealed ectopic discharges). The closed triangles represent those instances in which the ectopic pacemaker was assumed to be captured during its “supernormal” phase. On the basis of this phase response curve, the parasystolic pattern of patient 2 can be readily reproduced. Moreover, in the presence of a phase of supernormal excitability in both the parasystolic pacemaker and surrounding excitable tissues, and with appropriate delay to and from the parasystolic pacemaker, reciprocating activity-indistinguishable from reflection-may occur and sometimes repeat for several cycles, giving rise to couplets and episodes of nonsustained ventricular tachycardia (Figures 4 and 5). On the other hand, it may also lead to sufficient disorganization in the ventricular activation sequence to result in sustained reentrant activity or fibrillation.
THE AMERICAN
JOURNAL
from the electro2. The diagond abbreviation poe-
OF CARDIOLOGY
VOLUME
63
Appraisal of the analysis: In both patients, ventricular parasystole was suggested by the markedly varying coupling intervals of the ectopic complexes, the fusions with activations of sinus origin and the episodes during which a rather fixed common denominator of the interectopic intervals could be identified.16v17 However, in both patients some unusual characteristics were seen. Patient 1 could well represent an example of parasystolic tachycardia at a rate of 136 to 150/min with exit block.19 However, to agree with this interpretation, sudden shortening of the ectopic interval of the R-on-T complex in Figure 1C would have to be attributed to a fortuitous event in which another pacemaker, located close to (but outside) the parasystolic focus, fired during the T wave of the preceding parasystolic complex. On the other hand, in patient 2 (Figures 4 and 5), it seems highly unlikely that switching to bigeminal rhythm with fixed coupling and marked abbreviation of the ectopic intervals was due to dissipation of exit block during a rapid tachycardia with cycle lengths of 240 to 280 ms. First, this would imply an unusually rapid rate of the parasystolic focus, with variation of the intrinsic period of up to f8% from the mean. Second, this interpretation would not explain why the exit block consistently and predictably disappeared, sometimes repetitively, after postectopic intervals 1680 ms. Alternatively, the arrhythmic pattern in patient 2 could result from alternation between typical parasystole with an intrinsic period of 1,440 to 1,570 ms, and manifest and concealed, single and repetitive reentry
within or in the immediate vicinity of the ectopic foCUS.~~-~~ Microreentry due to longitudinal dissociation in the pathway containing the ectopic focus15322may have been enhanced in the setting of the rule of bigeminy.21 According to this line of thought, the interectopic intervals that are longer than a simple multiple of the parasystolic cycle must be explained by concealed reentry,13-15 for example, the interval of 1,880 ms in Figure 5C. This interval may be composed of 1 concealed reentrant cycle of 280 ms plus a parasystolic cycle of 1,600 ms. This, however, also leads to the conclusion that the subsequent manifest ectopic complex which started the run of tachycardia that deteriorated into ventricular fibrillation (Figure 5C) is of parasystolic and not reentrant origin. This is in keeping with the observation that the penultimate RR interval preceding sustainment of the arrhythmia measured only 580 ms, which is significantly shorter than the minimal postectopic interval (680 ms) that heralds fixed coupling of extrasystolic activity. In fact, it was the uncertainty about the reentrant nature of the latter complexes, as well as the assumption of concealed reentrant activity following these events, that led us to consider the possibility of modulated parasystole. Schamroth and Marriott23 have attributed transitions of parasystole to extrasystolic bigeminy to the enhancing effect of Wedensky phenomena. However, their interpretation was subsequently challenged by Moe et a1,24who suggested a modulated parasystole. Unifying hypothesis: In both patients, phase-response curves could be constructed that allowed for accurate prediction of the positions of manifest and concealed ectopic discharges. This required the assumption of phase-related electrotonic effects on the ectopic pacemaker.le5 In addition, frequency-dependent exit conduction delays,4 postrepolarization refractoriness4,20 and supernormal excitability4,5,25 in both the ectopic pacemaker and surrounding ventricular myocardium had to be assumed. The validity of all these assumptions has been demonstrated experimentally4,5,1s and a number of studies have provided additional support for their applicability in the clinical setting.6m12Therefore, our empirically derived models (Figure 3 and Jalife et a1,7 their Figure 3) can be taken to represent close approximations of the possible dynamic interactions between the ventricle and the protected pacemaker site, which may have led to the severe arrhythmias in these patients. Alternatively, one could assume that the modulated parasystolic pacemaker also has the potential of developing triggered activity26 in the form of early afterdepolarizations. This might have triggered secondary or tertiary parasystolic discharges leading to extrasystoles, couplets and tachycardia or, as in patient 1, possibly also torsade de pointes.27 Prognostic significance: To date, only 1 study has described the initiation (and termination) of sustained ventricular tachycardia by a complex of parasystolic origin. The benign prognosis of ventricular parasystole has recently been confirmed by Kuo and Surawicz.14 However, if our interpretation is correct, our patients do
suggest the possibility that at least some instances of premature ventricular ectopic activity that degenerates into ventricular tachycardia or fibrillation may represent the expression of a modulated parasystolic pacemaker that is not recognized because of its atypical behavior. The observation that a modulated parasystole may result in ventricular tachycardia and fibrillation when there is a phase of supernormal excitability in the pacemaker cycle and possibly also in the ventricular site may also shed some light on the benign prognosis of pure or “typical” cases. With little or no modulation, high levels of protection may likewise prevent the early exit of parasystolic impulses. Hence, there will be a reduced chance of meeting sufficient dispersion of refractoriness to initiate a tachycardia or fibrillation. The level of protection of the ectopic pacemaker may thus determine the prognosis of parasystole. Further studies are needed to confirm or reject this hypothesis. Acknowledgment: We thank Justus Anumonwo for reading the manuscript and Wanda Coombs and Gertrude van Eck for excellent secretarial assistance.
REFERENCES 1. Jalife J, Moe GK. Effect of electrotonic potentials on pacemaker activity of canine Purkinje fibers in relation to parasystole. Circ Res 1976;39:802-808. 2. Moe GK, Jalife J, Mueller WJ, Moe B. A mathematical model of parasystole and its application to clinical arrhythmia. Circulation 1977;56;968-979. 3. Jalife J, Moe GK. A biologic model of parasystole. Am J Cardiol1979;43;761772. 4. Antzelevitch C, Jalife J, Moe GK. Characteristics of reflection as a mechanism of re-entrant arrhythmias and its relationship to parasystole. Circulation 1980; 61:182-191. 5. Antzelevitch C, Jalife J, Moe GK. Electrotonic modulation of pacemaker activity: further biological and mathematical observations on the behavior of modulated parasystole. Circulation 1982;66:1225-1232. 6. Furuse A, Matsuo H, Saigusa M. Effects of intervening beats on ectopic cycle length in a patient with ventricular parasystole. Jpn Heart J 1981;22t201-209. 7. Jalife J, Antzelevitch C, Moe GK. The case for modulated parasystole. PACE 1982;91 l-926. 8. Nau GJ, Aldariz AE, Acunzo RS, Halpern MS, Davidenko JM, Elizari MV, Rosenbaum MB. Modulation of parasystolic activity by nonparasystolic beats. Circulation 1982;66:462-469. 9. Oreto G, Satullo G, Luzza F, Consolo F, Schamroth L. Supernormal modulation of ventricular parasystole: the triphasic phase-response curve. Am J Cardiol 1986;57:283-290. 10. Jalife J, Michaels DC, Langendorf R. Modulated parasystole originating in the sinoatrial node. Circulation 1986;74:94.5-954. 11. Tenczer J, Littmann L. Rate-dependent patterns of modulated ventricular parasystole. Am J Cardiol 1986;57:576-581. 12. Tenczer J, Littmann L, Rohla M, Wu DB, Fenyvesi T. A study of modulated ventricular parasystole by programmed stimulation. Am J Cardiol1987;59:846851. 13. Singer DH, Parameswaran R, Drake FT, Meyers SN, Deboer AA. Ventricular parasystole and reentry: clinical-electrophysiological correlations. Am Heart J 1974:88:79-87. 14. Kuo CS, Surawicz B. Coexistence of ventricular parasystole and ventricular couplets: mechanism and clinical significance. Am J Cardiol 1979;33:435-441. 15. Kinoshita S, Kawasaki T. Ventricular parasystole and couplets: the concept of longitudinal dissociation in the microreentry pathway. Am Heart J 1983;105; 1050-l os5. 16. Katz LN, Pick A. Clinical Electrocardiography. Part I. The Arrhythmias. Philadelphia: Lea and Febiger, 1956:185. 17. Scherf D, Schott A. Extrasystoles and Allied Arrhythmias. London William Heineman Medical Books, 1973t269. 18. Antzelevitch C, Bernstein MJ, Feldman HN, Moe GK. Parasystole, x-entry, and tachycardia: a canine preparation of cardiac arrhythmias occurring across inexcitable segments of tissue. Circulation 1983:68:1101-l 115. 19. Scherf D, Borneman C. Parasystole with a rapid ventricular centre. Am Heart J 1961;62:320-331,
THE AMERICAN
JOURNAL
OF CARDIOLOGY
JUNE
1, 1989
1331
VENTRICULAR
PARASYSTOLE
AND FIBRILLATION
20. Davidenko JM, Antzelevitch C. Electrophysiological mechanisms underlying rate-dependent changes in refractoriness in normal and segmentally depressed canine Porkinje fibers. The characteristics of post-repolarization refractoriness. Circ Res 1986;58:257-268. 21. Langendorf R, Pick A, Winternitz M. Mechanisms of intermittent ventricular bigeminy. I. Appearance of ectopic beats dependent upon length of the ventricular cycle, the “rule of bigeminy.” Circulation 1955;l Z:422-430. 22. Kinoshita S, Kato Y, Kawasaki T, Okimori K. Ventricular tachycardia initiated by late-coupled ventricular extrasystoles: the concept of longitudinal dissociation in the microreentry pathway. Am Heart J 1982;103:1090-1095. 23. Schamroth L, Marriott HJL. Intermittent ventricular parasystole with observations on its relationship to extrasystolic bigeminy. Am J Cardiol 1961;7:799809.
1332
THE AMERICAN
JOURNAL
OF CARDIOLOGY
VOLUME
63
24. Moe GK, Jalife J, Mueller WJ. Reciprocation between pacemaker sites: reentrant parasystole? In: Kulbertus HE, ed. Re-Entrant Arrhythmias. Mechanisms and Treatment. Lancaster: MTP Press, 1977:271-280. 25. Spear JF, Moore EN. Supernormal excitability and conduction in the HisPurklnje system of the dog. Circ Res 1974;35:782-792. 26. Wit AL, Cranefield PF, Gadsby DC. Triggered activity. In: Zipes DP, Bailey JC, Alharrar V, eds. The Slow Inward Current and Cardiac Arrhythmias. The Hague: Martinus
Nijhoff,
1980:437-454.
27. Brachmann J, Scherlag BJ, Rosenshtraukh LV, Lazzara R. Bradycardiadependent triggered activity: relevance to drug induced multiform ventricular tachycardia. Circulation 1983;68:846-856. 26. Jacobsen LB. Spontaneous ventricular parasystole initiating ventricular tachycardia. J Electrocardiol 1973,6:63-70.