Determinants of Fast- and Slow-Pathway Conduction in Patients with Dual Atrioventricular Nodal Pathways

Transcription

1 Determinants of Fast- and Slow-Pathway Conduction in Patients with Dual Atrioventricular Nodal Pathways By Delon Wu, Pablo Denes, Ramesh Dhingra. Christopher Wyndham. and Kenneth M. Rosen ABSTRACT Electrophysiological studies were performed in two patients with documented paroxysmal supraventricular tachycardia and dual atrioventricular (AV) nodal pathways as defined by the atrial extra-stimulus technique. Both patients manifested two ranges of A- intervals (AV nodal conduction times) at critical cycle lengths, reflecting fast- and slow-pathway conduction. The occurrence of fast- and slow-pathway conduction at the same cycle length depended on a long fast-pathway effective refractory period relative to the spontaneous or driven cycle length. At critical cycle lengths with fast-pathway conduction, a shift to slow-pathway conduction could be induced by a premature atrial impulse falling within the effective refractory period of the fast pathway. Repetitive retrograde concealed conduction to the fast pathway then maintained antegrade slowpathway conduction. Resumption of fast-pathway conduction was induced with premature atrial impulses falling within the effective refractory periods of both the fast and the slow pathways, allowing recovery of the fast pathway for antegrade conduction. Atrial echoes and AV nodal reentrant paroxysmal supraventricular tachycardia occurred when sufficient slow-pathway delay was achieved to allow recovery of the fast pathway for retrograde conduction. KEY WORDS concealed conduction Wolff-Parkinson-White syndrome refractory periods paroxysmal supraventricular tachycardia is bundle electrograms Recent electrophysiological studies from several laboratories have suggested that the atrioventricular (AV) node can longitudinally dissociate into two pathways with different functional properties (1-11). In man, dual AV nodal pathways have been described in a patient with two P-R intervals and in several patients with AV nodal reentrant paroxysmal supraventricular tachycardia (9-11). In the present investigation, we studied two patients with documented paroxysmal supraventricular tachycardia. In both of these patients, two sets of conduction times (A- intervals) were demonstrated, suggesting dual AV nodal pathways. The determinants of fast- and slow-pathway conduction and their relationship to the echo From the Section of Cardiology, Abraham Lincoln School of Medicine of the University of Illinois College of Medicine, and the West Side Veterans' Administration ospital, Chicago, Illinois This work was supported in part by NI Contract under the Myocardial Infarction Program of the National eart and Lung Institute, by U. S. Public ealth Service Grant L S1 from the National eart and Lung Institute, and by basic institutional support of the West Side Veterans' Administration ospital, Chicago, Illinois. Please address reprint requests to Delon Wu, M.D., Section of Cardiology, University of Illinois ospital, P.O. Box 6998, Chicago, Illinois Received January 13, Accepted for publication March 24, phenomenon and reentrant paroxysmal tachycardia were analyzed. Methods ELECTROPYSIOLOGICAL STUDIES Electrophysiological studies were performed in the postabsorptive, nonsedated state. The research protocol had the prior approval of the University of Illinois Committee of Associates for review of clinical research and investigation involving human beings. Informed consent was obtained. A tripolar electrode catheter was percutaneously introduced into the femoral vein and fluoroscopically positioned across the tricuspid valve for is bundle recording (12). A second quadripolar electrode catheter was percutaneously introduced into the other femoral vein and placed against the lateral wall of the high right atrium for atrial stimulation and recording. Electrocardiographic leads I, II, III, and V, and high right atrial, low right atrial, and is bundle electrograms were simultaneously recorded on a multichannel oscilloscopic recorder (Electronics for Medicine DR-16) at paper speeds of 100 and 200 mm/sec. Recordings were also simultaneously recorded on an eight-channel tape system for further analysis. Stimuli were provided by a programmable digital pulse generator (manufactured by M. Bloom, Philadelphia. Pa.). Stimuli were approximately twice diastolic threshold and 2 msec in duration. The atria were paced at rates slightly faster than the sinus rhythm. Paced rates were increased in 10-beat/min increments until type 1 block proximal to the is bundle was observed. Refractory periods and echo zones were determined with the atrial extra-stimulus technique (13,

2 DUAL ATRIOVENTRICULAR NODAL PATWAYS ). The extra stimulus was coupled to sinus rhythm and decreased in 5-10-msec steps. ELECTROPYSIOLOGICAL DEFINITIONS RA,, A,,,, and V, were, respectively, the high right atrial, the low right atrial, the is bundle, and the ventricular electrograms of sinus beats. RA 2, A 2, 2, and V 2 were, respectively, the high right atrial, the low right atrial, the is bundle, and the ventricular responses to the extra stimulus (S 2 ) or to spontaneous premature atrial beats. Conduction intervals and refractory periods were measured as previously defined (13, 14). A,-A 2,,- 2 and A,-A 2, A 2-2 curves were constructed (9, 10, 13, 14). AV nodal reentrance and echo zones were defined as previously described (10, 11, 15-17). Dual AV nodal pathways were diagnosed when two sets of conduction times and refractory periods were demonstrated during atrial premature stimulation (9-11). Fast-pathway and slow-pathway conduction times and refractory periods were defined as previously described (9-11). Results PATIENT 1 This patient was a 59-year-old woman with documented recurrent paroxysmal supraventricular tachycardia. Electrophysiological studies during sinus rhythm revealed an A- interval of 90 msec (normal msec) and an -V interval of 45 msec (normal msec). Atrial extra stimuli were coupled to sinus rhythm at a cycle length of 830 msec (Fig. 1). As A r A 2 intervals were decreased from 690 to 370 msec,,- 2 intervals shortened from 700 to 470 msec. At an A r A 2 interval of 365 msec, the r 2 interval suddenly increased to 655 msec, reflecting a marked increase in the A 2-2 interval. With A,-A 2 intervals between 365 and 300 msec,,- 2 msec : 600 E O A, A2 (msec) CL»83O msec , A, (msec) FIGURE 1 Atriouentricular conduction curves from patient I, showing dual AV nodal pathways. A:,- 7 responses are plotted against AfA, coupling intervals. B: A 7-7 responses are plotted against A,-At coupling intervals. The fast-pathway effective refractory period was 365 msec. The atrial functional refractory period of 300 msec limited slow-pathway conduction.cl = cycle length. intervals were 655 to 620 msec and A 2-2 intervals were 360 to 440 msec. Atrial echoes and paroxysmal tachycardia were not induced. AV conduction was atrially limited with an atrial functional refractory period of 300 msec. Examination of the A^Aj,,- 2 (Fig. 1A) and A,-A 2, A 2-2 (Fig. IB) curves suggested dual AV nodal pathways with a fastpathway effective refractory period of 365 msec. During constant atrial pacing, intact AV conduction was noted at paced rates up to 140/min. At paced rates between 70 and 100/min, A- intervals were msec (Fig. 2A). At paced rates between 110 and 140/min, A- intervals were either msec (fast-pathway conduction) (Figs. 2B and 3A) or msec (slow-pathway conduction) (Figs. 2C and 3C). The determinants of antegrade AV conduction were studied during sinus rhythm by randomly initiating atrial stimulation at rates of /min. At critical pacing rates, the coupling interval of the first paced beat determined whether that beat and all subsequent beats utilized the fast or the slow pathway for antegrade conduction. At paced rates equal to or less than 100/min, with an initial coupling interval of less than 365 msec, the first paced beat was conducted via the slow pathway (Fig. 2A), but all subsequent beats were conducted via the fast pathway (Fig. 2A). At a paced rate of 110/min, with an initial coupling interval of longer than 365 msec, the first paced beat and all subsequent beats were conducted via the fast pathway (Fig. 2B). At a paced rate of 110/min, with the initial coupling interval equal to or shorter than 365 msec, the first paced beat and all subsequent beats were conducted via the slow pathway (Fig. 2C). At paced rates of 140/min, prolonged periods of both fast-pathway conduction (Fig. 3A) and slowpathway conduction (Fig. 3C) were noted. A shift from fast-pathway to slow-pathway conduction occurred via a Wenckebach mechanism of the fast pathway (Fig. 3B). A gradual increase in the fast-pathway A- interval was terminated by prolonged periods of slow-pathway conduction (Fig. 3B). At a paced rate of 130/min, an unexpected shortening of the A- interval during fast-pathway conduction was observed (Fig. 4), which appeared to reflect simultaneous fast- and slow-pathway conduction (from the preceding paced P wave). There was a tendency for beats postulated to be conducted via the slow pathway to show minor notching or plateauing of the QRS complexes in lead II, as opposed to the QRS complexes con-

3 784 WU. DENES. DINGRA. WYNDAM, ROSEN Pacing R = 90/min Coupling interval =365 A = IOO I 01-i 375 W!6oCi IOo\ Pacing R = IIO/min Coupling interval=375 A = I25 At I 830 I 375 I 545 I 545 I 545 Pacing R=IIO/min Coupling interval = 365 A = 360 ^ TV At I I 545 I 545 I. AVN V \v y v\, V' t V 395S 375% 360V- 3«>\ '.A, I ' 1 FIGURE 2 Recordings from patientl, showing dual AV nodal conduction times during constant atrial pacing. Shown are electrocardiographic lead II, the high right atrial electrogram (RA), and the is bundle electrogram (BE). A and are, respectively, low right atrial and is bundle electrograms. Time lines are at 1-second intervals, and paper speed is 100 min/sec in this and subsequent illustrations. Conduction intervals are listed in msec. Electrocardiographic leads I, II, and V, are deleted. In each section, the first two beats are sinus beats with a cycle length of 830 msec and an A- interval of 90 msec. The third and subsequent beats are paced beats. A: Pacing rate (R) was 90/min. The first paced beat had a coupling interval of 365 msec (equal to the fast-pathway effective refractory period) and was conducted via the slow pathway with an A- interval of 375 msec. The second and subsequent paced beats were conducted via the fast pathway with an A- interval of 100 msec. B and C show pacing rates of 110/min. B: Coupling interval of the initial paced beat was 375 msec (fast-pathway effective refractory period), and the beat was conducted via the fast pathway with an A- interval of 180 msec. The subsequent beats were also conducted via the fast pathway with an A- interval of 125 msec. C: Coupling interval of the first paced beat was 365 msec (equal to the fast-pathway effective refractory period), and the beat was conducted via the slow pathway with an A- interval of 395 msec. The second and subsequent paced beats were also conducted via the slow pathway with an A- interval of 360 msec. Proposed mechanisms are present in the ladder diagrams on the right, which depict the atrium (AT), the AV node (AVN) and the is bundle (). Solid lines reflect fast-pathway conduction, and broken lines reflect slow-pathway conduction. Ladder diagrams in all subsequent figures follow this format. ducted via the fast pathway which tended to be of greater amplitude. Examples of this phenomenon are shown in Figures 2A and C, 3A and C, and 4. The phenomenon was not always present, however; exceptions are demonstrated in Figures 2B and 3B. In this patient, the postulated determinants of fast- and slow-pathway conduction are demonstrated in the ladder diagrams on the right of Figure 2. In Figure 2A, at a paced rate of 90/min, the first paced beat encountered the effective refractory period of the fast pathway and conducted via the slow pathway; the paced cycle length was long enough to allow recovery of the fast pathway for antegrade conduction of all of the subsequent beats. In Figure 2B and C, the paced rate was 110/min. In 2B, the first paced beat fell beyond the fast-pathway effective refractory period, so that that beat as well as all subsequent beats were conducted via the fast pathway. In 2C, the first paced beat encountered the effective refractory period of the fast pathway and conducted via the slow pathway. Repetitive retrograde concealed conduction (or possibly repetitive antegrade concealed conduction or a combination of the two) maintained fast-pathway refractoriness, so that all subsequent beats were conducted via the slow pathway. The shift from fast- to slow-pathway conduction at a paced rate of 140/min appeared to reflect type 1 block in the fast pathway (Fig. 3). Once the fast pathway failed, the slow pathway was available for conduction. Repetitive concealed conduction would then maintain slow-pathway conduction. The postulated mechanism for those beats with unexpected shortening of the A- interval at a paced rate of 130/min is shown in the ladder diagram in Figure 4. Pacing-induced lengthening of the conduction time involving both pathways can put these pathways out of phase, so that the Circulation Research, Vol. 36. June 1975

4 DUAL ATRIOVENTRICULAR NODAL PATWAYS 785 Pacing R = l40/min A = I85 JL RA- BE Pacing R = l40/min. u l l k U LJ la A i ja ^ I. I' IP \l, L I ' I I II, I A=420 \ 11 R, fri i\ FIGURE 3 ^WA^^b; Recordings from patient 1, showing two A- intervals at a paced heart rate of 140/min. A: A- interval was 185 msec, consistent with fast-pathway conduction. B: Progressive lengthening of the A- interval from 100 to 190 msec was followed by a sudden jump of the A- interval to 400 msec, suggesting type 1 block in the fast pathway with shift of conduction to the slow pathway. C: A- interval 420 msec, consistent with slow-pathway conduction. Abbreviations are the same as they are in Figure 2. ventricles can respond twice to a single atrial impulse. The fourth QRS complex in Figure 4 was the first ventricular response to the fourth atrial paced beat conducted via the fast pathway. The fifth QRS complex, which shows minor notching, is At Pacing R= 30/min. postulated to be the second ventricular response to the fourth atrial paced beat conducted via the slow pathway. The fifth atrial paced beat was blocked in both pathways. Superimposition of this and the fifth QRS complex resulted in an apparent short- FIGURE 4 Recordings from patient 1, showing pseudoshortening of the A- interval and shift of conduction from the fast to the slow pathway. The paced heart rate was 130/min. The fifth paced beat had an unexpected shortening of the A- interval {65 msec), which occurred during a pacing-induced Wenckebach sequence. The sixth paced beat was blocked in the AV node. The seventh and subsequent paced beats showed a progressive lengthening of the A- intervals, followed by a sudden jump of the A- interval to 385 msec; the interval then stabilized, suggesting type 1 block in the fast pathway with shifting of conduction to the slow pathway. Interpretations are shown in the ladder diagram. See text for further discussion. Abbreviations are the same as they are in Figure 2.

5 786 WU. DENES. DINGRA. WYNDAM. ROSEN ening of the A- interval. The sixth to ninth atrial paced beats were conducted via the fast pathway with type 1 block. The tenth and subsequent beats were conducted via the slow pathway. PATIENT 2 This patient was a 61-year-old man with arteriosclerotic heart disease and documented recurrent paroxysmal supraventricular tachycardia. Electrocardiograms taken over 2 years revealed two ranges of P-R intervals (0.18 and 0.28 seconds, respectively) without intermediate values. Electrophysiological studies during sinus rhythm at a cycle length of 520 msec revealed A- intervals of either 160 msec (fast pathway) or 260 msec (slow pathway), suggesting dual AV nodal pathways (Figs. 5 and 6). The response to coupled stimulation was studied by analyzing the responses to both spontaneous atrial premature contractions and test extra stimuli. During sinus rhythm with fast-pathway conduction (Fig. 5), extra stimuli and premature atrial contractions, as well as subsequent sinus beats, were conducted via the fast pathway at coupling intervals of 500 to 410 msec (with the A 2-2 interval ranging from 160 to 205 msec) (Fig. 5A). Extra stimuli and premature atrial contractions and all subsequent sinus beats were conducted via the slow pathway at coupling intervals of 400 to 375 msec (with the A 2-2 interval ranging from 280 to 310 msec) (Fig. 5B). AV nodal reentrant atrial echoes with paroxysmal supraventricular tachycardia occurred at coupling intervals of 365 to 300 msec, with a critical A 2-2 interval (slow pathway) of 320 msec or greater (Fig. 5C). The AV nodal effective refractory period of 290 msec limited AV conduction (Fig. 5D). During sinus rhythm with slow-pathway conduction (Fig. 6), extra stimuli and premature atrial contractions were conducted via the slow pathway (A 2-2 interval of 260 to 340 msec) at all coupling intervals (Fig. 6A and B). AV nodal reentrant atrial echoes with paroxysmal supraventricular tachycardia occurred at A,-A 2 coupling intervals of 365 to 300 msec, with the A 2-2 interval equal to or greater than 320 msec (Fig. 6B). A shift to fastpathway conduction occurred following extra stimuli which were blocked in the AV node (coupling intervals of 290 to 260 msec) (Fig. 6C). The atrial effective refractory period was 250 msec. A shift of AV conduction from the slow to the fast pathway also occurred following spontaneous ventricular premature beats (Fig. 6D). The postulated mechanisms of shifting conduction and the relationship to paroxysmal supraventricular tachycardia are presented in the ladder diagrams in Figure 7. Figure 7A-C reflects fastpathway conduction. In 7A, the extra stimulus and all subsequent sinus beats were conducted via the fast pathway. In 7B, the extra stimulus encountered the effective refractory period of the fast pathway (400 msec) and conducted via the slow pathway. Repetitive retrograde concealed conduction kept the fast pathway refractory, and, therefore, antegrade conduction shifted to the slow pathway. In 7C, the extra stimulus achieved a critical delay (A 2-2 interval of 320 msec or greater) in the slow pathway, allowing recovery of the fast pathway for total retrograde conduction; therefore, paroxysmal supraventricular tachycardia occurred. Figure 7D-E reflects slow-pathway conduction. In 7D, the extra stimulus achieved a critical A 2-2 delay of 320 msec with occurrence of atrial echoes and paroxysmal supraventricular tachycardia, as in 7C. In 7E, the extra stimulus encountered the effective refractory period of the slow pathway (290 msec) and was blocked. Retrograde concealed conduction to the fast pathway therefore could not occur, and conduction shifted back to the fast pathway. In 7F, the premature ventricular beat interfered with the slow pathway (retrograde concealed conduction), allowing the fast pathway to recover. Therefore, conduction resumed in the fast pathway. Discussion Electrophysiological studies in both animal and human hearts suggest that the AV node can undergo longitudinal dissociation into two pathways with different functional properties (1-11). Rosen et al. (9), utilizing is bundle recording and the atrial extra stimulus technique in a patient with two PR and A- intervals, have reported discontinuity in the A,-A 2, r 2 curve, suggesting dual AV nodal pathways. Denes et al. (10) have demonstrated similar curves in two patients with AV nodal reentrant paroxysmal tachycardia. Wu et al. (11) have demonstrated conduction curves suggesting dual pathways in 7 of 12 patients with AV nodal reentrant paroxysmal tachycardia. The present study suggests that two ranges of P-R intervals during sinus rhythm or two ranges of A- intervals at critical paced atrial rates can occur in patients with AV nodal reentrant paroxysmal supraventricular tachycardia. Presumably, dual conduction times and AV nodal reentrant paroxysmal supraventricular tachycardia are both manifestations of dual AV nodal pathways. The occurrence of two P-R or A- intervals

6 DUAL ATRIOVENTRICULAR NODAL PATWAYS 787 'ia B A, A 2 =400, 2 =520 FIGURE 5 Recordings from patient 2, showing a shift of conduction to the slow pathway and an induction of paroxysmal supraventricular tachycardia during fast-pathway conduction. Shown are electrocardiographic leads III, V,, RA, and BE. A, and, are atrial and is bundle electrograms of the last sinus beat. A, and, are atrial and is bundle responses to a test stimulus (S,) or to a spontaneous premature atrial beat (PAB). E represents echo. A r A t,,- t, and A- intervals (in msec) are listed. The sinus cycle length was 520 msec, and the A- interval was 160 msec (fast pathway). A: A,-A t interval was 430 msec. A 7 -, interval was 200 msec. Subsequent beats were conducted via the fast pathway with an A- interval of 160 msec. B: A,-A, interval was 400 msec, A,- t interval was 280 msec. Subsequent beats were conducted via the slow pathway with an A- interval of 260 msec. C: A,-A, interval was 320 msec. A,- 2 interval was 325 msec. Echoes and paroxysmal supraventricular tachycardia occurred. D: A,-Aj interval was 290 msec. A, was blocked in the A V node. Subsequent beats were conducted via the fast pathway with an A- interval of 160 msec.

7 788 A CL=52O A. A 2 =385 i 2 =425 WU. DENES. DINGRA. WYNDAM. ROSEN BE-v 26b, 300 ', 260 V i-i A, A, =365,, =425 \,'> '260 \- 260 \ D FIGURE 6 Recordings from patient 2, showing slow-pathway conduction and paroxysmal supraventricular tachycardia. Sinus cycle length was 520 msec, and the A- interval was 260 msec (slow pathway). A: A,-A, interval was 385 msec. A,-, interval was 330 msec. Echoes and paroxysmal supraventricular tachycardia were not seen. Subsequent beats were conducted via the slow pathway with ana- interval of 260 msec. B: A^-A, interval was 365 msec. A t - t interval was 320 msec. Echoes and paroxysmal supraventricular tachycardia were induced. C: A,-A 2 interval was 290 msec. A 2 was blocked in the A V node. Subsequent beats were conducted via the fast pathway with an A- interval of 160 msec. D: Sinus beats following a premature ventricular beat (PVC) were conducted via the fast pathway with an A- interval of 160 msec. Abbreviations are the same as they are in Figure 5.

8 DUAL ATRIOVENTRICULAR NODAL PATWAYS 789 B *l At 520 I AVN A A2 At 520 / 400/ 560 I 520 I 520~l II 260\ 260p A A 2 At I 520 At FIGURE 7 AVN I 520 I I: 260'i 26' '\- ZB rrpvc Ladder diagrams for Figures 5 and 6, showing determinants of fast and slow pathway conduction and the relationships to induction paroxysmal suprauentricular tachycardia. See text for discussion. Abbreviations are the same as they are in Figure 2. of depends on a long fast-pathway effective refractory period relative to cycle length. At critical heart rates, a premature impulse can encounter the effective refractory period of the fast pathway and conduct via the slow pathway. Subsequent repetitive retrograde concealed conduction (and possibly antegrade concealed conduction) to the fast pathway can keep the fast pathway refractory for subsequent antegrade conduction and maintain slow-pathway conduction. With slower rates, the fast pathway can recover for subsequent antegrade conduction despite concealed conduction. Thus, the shift to persistent slow-pathway conduction can be induced by a critically timed premature atrial beat only at critical heart rates. The difference between the cycle length of the critical heart rate and the effective refractory period of the fast pathway could reflect the conduction time of the retrograde concealed impulse to the fast pathway. A shift of conduction to the slow pathway can occur at a heart rate which induces type 1 block in the fast pathway. A shift of conduction from the slow to the fast pathway occurs when the premature atrial impulse is blocked in both pathways, allowing recovery of the fast pathway for subsequent antegrade conduction. The occurrence of echoes and paroxysmal supraventricular tachycardia due to AV nodal reentrance depends on a slow conduction delay sufficient to allow recovery of the fast pathway for retrograde conduction (10, 11, 17). Echoes and paroxysmal supraventricular tachycardia were always induced in patient 2 at coupling intervals which achieved a slow pathway A 2-2 interval of 320 msec or more. Inability to induce atrial echoes and paroxysmal supraventricular tachycardia in patient 1 probably reflected increased retrograde refractoriness of the fast pathway on the day of study. A dual AV nodal pathway can result in double ventricular responses to a single P wave if the slow-pathway conduction time (relative to the fast-pathway conduction time) is sufficient for recovery of the distal is-purkinje system and ventricle for reexcitation. Bailey et al. (18), working with depressed canine Purkinje fibers, have suggested that a single impulse can excite distal tissue twice due to longitudinal dissociation of conduction. In man, double ventricular responses to a single P wave have recently been demonstrated by Puech and Grolleau (19) in a patient with preexcitation, in whom single atrial extra stimuli are conducted via both anomalous (first QRS) and normal (second QRS) pathways. In patient 1, double ventricular responses occurred during pacing-induced type 1 block in both pathways. Superimposition of the second ventricular response with the subsequent atrial paced beat resulted in pseudoshortening of the A- interval. The mechanism of minor notching of the QRS complex without a change in the -V interval during antegrade slow-pathway conduction in patient 1 is unclear. In other patients with dual AV nodal pathways reported on previously, the ventricles responded with similar QRS complexes whether the fast or the slow pathway was utilized

9 790 WU. DENES. DINGRA. WYNDAM. ROSEN for antegrade conduction (10-12). Sherf and James (20) have postulated longitudinal dissociation and preferential conduction in the is-purkinje system, dependent on atrial input. The QRS complex during slow-pathway conduction may be explained by this concept. In conclusion, the occurrence of both fast- and slow-pathway conduction at identical cycle lengths depends on a long fast-pathway effective refractory period relative to cycle length. Repetitive retrograde concealed conduction to the fast pathway is responsible for the maintenance of slow-pathway conduction. Atrial echoes and paroxysmal supraventricular tachycardia occur when sufficient slowpathway conduction delay is achieved, allowing recovery of the fast pathway for retrograde conduction. Shift from the fast to the slow pathway can result in changes in ventricular activation by an unknown mechanism. Acknowledgment The authors wish to express their appreciation for the secretarial help of Ms. Valerie Woods and Ms. Therese Calderon. References 1. MOE GK, PRESTON JB, BURLINGTON : Physiologic evidence for a dual AV transmission system. Circ Res 4: , MENDEZ C, AN J, GARCIA DE JALON PD, MOE GK: Some characteristics of ventricular echoes. Circ Res 16: , MENDEZ C, MOE GK: Demonstration of a dual AV nodal conduction system in the isolated rabbit heart. Circ Res 19: , ROSENBLUT A, RUBIO R: Ventricular echoes. Am J Physiol 195:53-60, KISTIN AD: Multiple pathways of conduction and reciprocal rhythm with interpolated ventricular premature systoles. Am eart J 65: , WATANABE Y, DREIFUS LS: Inhomogeneous conduction in the AV node: A model for reentry. Am eart J 70: , SCUILENBURC RM, DURRER D: Ventricular echo beats in the human heart elicited by induced ventricular premature beats. Circulation 40: , JANSE MJ, VAN CAPELLE GJL, FREUD GE, DURRER D: Circus movement within the AV node as a basis for supraventricular tachycardia as shown by multiple microelectrode recording in the isolated rabbit heart. Circ Res 28: , ROSEN KM, META A, MILLER RA: Demonstration of dual atrioventricular nodal pathways in man. Am J Cardiol 33: , DENES P, WU D, DINGRA RC, CUQUIMIA R, ROSEN KM: Demonstration of dual A-V nodal pathway in patients with paroxysmal supraventricular tachycardia. Circulation 48: , Wu D, DENES P, DINGRA RC, KAN A, ROSEN KM: Effects of propranolol on induction of A-V nodal reentrant paroxysmal tachycardia. Circulation 50: , SCERLAG BJ, LAU S, ELFANT R, STEIN E, BERKOWITZ WD, DAMATO AN: Catheter technique for recording is bundle activity in man. Circulation 39:13-18, WIT AL, WEISS MB, BERKOWITZ WD, ROSEN KM, STEINER C, DAMATO AN: Patterns of atrioventricular conduction in the human heart. Circ Res 27: , DENES P, WU D, DINGRA R, PIETRAS RJ, ROSEN KM: Effects of cycle length on cardiac refractory periods in man. Circulation 49:32-41, BIGGER JT JR, GOLDREYER BN: Mechanism of supraventricular tachycardia. Circulation 42: , GOLDREYER BN, BIGGER JT JR: Site of reentry in paroxysmal supraventricular tachycardia in man. Circulation 43:15-26, GOLDREYER BN, DAMATO AN: Essential role of atrioventricular conduction delay in the initiation of paroxysmal supraventricular tachycardia. Circulation 43: , BAILEY JC, ANDERSON GJ, FISC C: Digitalis-induced longitudinal dissociation in canine cardiac Purkinje fibers. Am J Cardiol 32: , PUEC P, GROLLEAU R: L'activite du faisceau de is normale et pathologique. Sandoz Editions, 1972, pp SERF L, JAMES TN: New electrocardiographic concept: Synchronized sinoventricular conduction. Dis Chest 55: , 1969 Circulation Research, Vol. 36. June 1975