Atrioventricular Nodal Conduction and Refractoriness After Intranodal Collision from Antegrade and Retrograde Impulses

Transcription

1 Atrioventricular Nodal Conduction and Refractoriness After Intranodal Collision from Antegrade and Retrograde Impulses MOHAMMAD SHENASA, M.D., PH.D., STEPHEN DENKER, M.D., REHAN MAHMUD, M.D., MICHAEL LEHMANN, M.D., CAROL J. GILBERT, R.N., AND MASOOD AKHTAR, M.D. Downloaded from by on January 2, 219 SUMMARY Animal studies have suggested that spontaneous or programmed ventricular beats that occur simultaneously with atrial activation may facilitate atrioventricular (AV) nodal conduction during subsequent atrial impulses. However, this possibility has not been systematically studied in the human heart. In the present study the AV nodal conduction during a programmed atrial premature beat (S2) was analyzed. The S2 was delivered after a series of atrial drive beats (SjSj) of constant duration; this was termed stimulation method I. The results were compared with stimulation method II, which was similar to method I except that a single ventricular beat (Vs) was introduced simultaneously with the last Sl. The longest and shortest possible paced atrial cycle lengths (CLs) were scanned during both methods. Twenty-six patients were studied: 14 with a normal PR and normal intraventricular conduction (NIVC), four with first-degree AV nodal block and NIVC, three with a complete left bundle branch block (LBBB) pattern, three with a complete right bundle branch block (RBBB) pattern, and two with an incomplete RBBB pattern. At the same SlS2 intervals, the AV nodal conduction times (S2112 intervals) were consistently shorter with method II than with method I except in three patients, two with complete RBBB and one with complete LBBB. The magnitude of S2H2 shortening with method II was more pronounced at the shorter basic CLs and shorter SlS2 intervals. During method I, the effective refractory period (ERP) of the AV node was measured in 13 patients, eight with NIVC and five with preexisting bundle branch block. With method II, the ERP of the AV node shortened in all but three patients (one with complete RBBB, one with incomplete RBBB and one with complete LBBB pattern), in whom this variable did not change. The findings suggest that intranodal collison from antegrade and retrograde impulses facilitates AV nodal conduction and shortens the ERP. The magnitude of this change is greater at shorter atrial CLs and is probably related to deeper intranodal penetration of a Vs. The shortening in AV nodal conduction and refractoriness is not noted in patients with bundle branch block when retrograde conduction delay or block in the bundle branches coexists with the antegrade counterpart producing delayed or ineffective input of Vs into the AV node. ATRIOVENTRICULAR (AV) nodal conduction and refractoriness in the antegrade direction have been extensively investigated in both animals and man.'-' Although several studies have addressed the effects of programmed ventricular premature stimulation on subsequent AV nodal conduction in animals,9' 11 rarely has such a study been carried out systematically in the human heart. This study was designed to examine the effect of simultaneous AV stimulation on the functional properties of the AV node in man. Patients and Methods Twenty-six patients (12 males and 14 females), ages years (mean 55.6 ± 16.3 years), were studied because of recurrent ventricular arrhythmias, dizziness or syncope. All were in sinus rhythm and none was taking cardioactive medications at the time of the study. Pertinent clinical data are summarized in table 1. Each patient gave informed consent. Electrophysiological studies were performed with the patient in a nonsedated, postabsorptive state. Under local anesthesia, quadripolar electrode catheters were percutane- From the Natalie and Norman Soref and Family Electrophysiology Laboratory, University of Wisconsin-Mount Sinai Medical Center, Milwaukee. Wisconsin. Address for correspondence: Masood Akhtar, M.D., Electrophysiology Laboratory, Mount Sinai Medical Center, P.O. Box 342, Milwaukee, Wisconsin Received August 12, 1982; revision accepted September 24, Circulation 67, No. 3, ously introduced through the femoral and antecubital veins and were positioned under fluoroscopic guidance, in the region of the tricuspid valve to permit recording of the His bundle potential, and in the right atrium and right ventricle for local recording and electrical stimulation using techniques previously described.7 The intracardiac electrograms (filtered at 3-5 Hz) and three surface lead ECGs and time line were simultaneously displayed on a multichannel oscilloscope (Electronics for Medicine VR-16) and recorded on magnetic tape (Honeywell model 96). Recordings were reproduced on photographic paper at and 15 mmlsec. Intracardiac electrical stimulation was performed using a digital stimulator. During these studies, the patients were isolated and all electrical equipment was grounded. In all patients, the functional properties of the AV conduction system were studied using two protocols of cardiac stimulation. Method I was the conventional method of antegrade refractory period determination, i.e., introduction of atrial premature beat (S2 or A2) after a series of basic atrial drive beats (S,S, or A,A,). Method II was similar to method I, except that a single right ventricular beat (Vs) was introduced simultaneously with the last beat of the basic drive. The two methods are illustrated in figure 1. Both methods were compared at the longest and shortest possible basic atrial cycle lengths (CLs). The duration of the longest atrial CL was determined by the spontaneous sinus CL. The shortest CL in individual cases was chosen at 651

2 652 CIRCULATION VOL 67, No 3, MARCH 1983 Downloaded from by on January 2, 219 TABLE 1. Clinical and Electrocardiographic Data Pt Age (years) Sex Symptoms and diagnosis Resting ECG* 1 59 F Dizziness, palpitation SR, PR = F Dizziness SR, PR = M IHD, syncope SB, PR = F Syncope SR, PR = F Syncope SB, PR = F Syncope SR, PR = M Syncope SR, PR = M Syncope, IHD, recurrent VT SR, PR = M Syncope, IHD SR, PR = 16 5 F Syncope, sustained VT SR, PR = M Syncope, nonsustained VT SR, PR = F IHD, SSS SR, PR = M IHD, sustained VT SR, PR = F Dizziness, rheumatic mitral valve disease SR, PR = M Dizziness SR, PR = 28, AV block M Dizziness, SSS SR, PR =, AV block F Dizziness, MVP SR, PR = 26, AV block M Palpitation, MVP SB, PR =, 1 AV block F Congestive cardiomyopathy SR, PR = 12, LBBB, VPCs 2 81 F Dizziness, rheumatic mitral valve disease SR, PR = 2, LBBB F Dizziness SR, PR = 2, LBBB F SSS SB, PR = 16, RBBB M IHD, syncope SR, PR = 16, RBBB, VPCs M IHD, recurrent VT SR, PR = 15, RBBB F Dizziness, VT SR, PR = 18, RBBB M Syncope SR, PR = 16, RBBB *Intervals are measured in msec. Abbreviations: AV = atrioventricular; IHD = ischemic heart disease; LBBB = left bundle branch block; MVP = mitral valve prolapse; R = right; SR = sinus rhythm; SB = sinus bradycardia; SSS = sick sinus syndrome; VPC = ventricular premature complex; VT = ventricular tachycardia. which there was 1:1 AV conduction and the programmed ventricular stimulus still produced a ventricular response. Definitions and Measurements A complete set of definitions for antegrade and retrograde conduction and refractory periods has been published. 12 Some definitions pertinent to this report are described as follows: The AH interval was measured from the onset of low atrial deflection to the beginning of the His bundle deflection during sinus beats and was used to estimate the AV nodal conduction time. During atrial pacing, measurements were made from the onset of stimulus artifact (S) as well as from the low atrial electrogram to the His bundle deflection (SH and AH intervals), respectively. Since the exact beginning of low atrial electrogram during paced beats was sometimes difficult to detect, the SH rather than the AH interval was used for comparative analysis unless a stimulus to atrial response latency was noted. The effective refractory period (ERP) of the AV node was the longest A,A2 interval at which A2 failed to depolarize the His bundle. The ERP of the atrium was the longest S,S2 interval at which S did not evoke an atrial response. Statistical Analysis Statistical evaluation was done using a paired t test. All values are mean ± SD. Statistical comparison were calculated only for patients who demonstrated changes at both longest and shortest atrial basic CLs. Results Complete antegrade and retrograde conduction and refractory period data are available in all patients; however, only- data pertinent to this -study are presented here. AV and Intraventricular Conduction During Sinus Rhythm (table 1) PR Intervals Twenty-two patients (nos and 19-26) had normal PR intervals, ranging from 15 to 2 msec

3 AVN CONDITION AFTER INTRANODAL COLLISION/Shenasa et al. 653 METHOD I METHOD 11 A1 A1 A1 A1 A2 I AVN> I vi V1 V1 V1 Al Al Al Al A2 AVN\ I& V1 V1 v, Vs FIGURE 1. Stimulation protocol. Method I is the conventional method of antegrade refractory period determination, i.e., a series of atrial basic drive beats (A,A1) followed by atrial premature beat (A2). In method II, a simultaneous ventricular beat (Vs) is introduced with the last beat of atrial basic drive (Al). AVN = atrioventricular node. Downloaded from by on January 2, 219 (mean msec). Four patients (nos ) had first-degree AV block with PR intervals of 28,, 26 and msec, respectively. QRS Morphology Eight patients had preexisting bundle branch block (BBB) patterns. Three patients (nos ) had complete left bundle branch block (LBBB). Three (nos ) showed a complete and two (nos. 25 and 26) an incomplete right bundle branch block (RBBB) pattern. AH Intervals The AH interval in the 22 patients with normal PR intervals ranged from 5 to 125 msec (mean msec). The four patients with first-degree AV block had AH intervals of 2, 18, 17 and 155 msec, respectively, indicating AV nodal delay as the cause of prolonged PR interval. HV Intervals The HV interval ranged from 35 to 6 msec (mean 39.6 ± 4.8 msec) for the entire group. The values were not different in patients with or without preexisting BBB. Antegrade AV Nodal Conduction During Methods I and II The atrial CLs were scanned with identical atrial coupling intervals during both methods up to the ERP of the right atrium. AV nodal conduction times (S2H2 intervals) were analyzed at the longest and shortest comparable S1S2 intervals that conducted to the His bundle (table 2). S2H2 Intervals at Comparable S1Sj Intervals (table 2) At the Longest CL The longest atrial CL scanned in this group of patients was 6-12 msec (mean msec). At these CLs, the longest comparable S1S2 intervals were 35-8 msec (mean msec). The corresponding S2H2 interval values were -3 msec (mean ± 26 msec) during method I and 9-29 msec (mean msec) during method II. At the same SIS2, all except patients and 26 showed a smaller S2H2 interval with method II; the shortening in S2H2 was -9 msec (mean msec) (figs. 2A and B, table 2), which represented a decrease of 3-32%. The remaining five cases showed no change in the S2H2 intervals with method II compared with method I. At the shortest comparable S,S2 intervals (range -6 msec, mean msec), the S2H2 intervals were msec (mean msec) during method I and msec (mean msec) using method II. All but four patients (nos and 26) showed a decrease in the S2H2 intervals with method II; the other four showed no change. The range of S2H2 shortening was -14 msec (mean msec), which reflected a change of 4 41 %. None of the patients showed a greater S2H2 interval with method II than with method I at any time. At the Shortest CL The duration of the shortest atrial CLs that were scanned was 4-7 msec (mean msec). At the longest comparable SIS2' the S2H2 values were msec (mean ) with method I, and msec (mean ) with method II. All except patients showed a decrease in the SAH interval with method II (figs. 2C and D). The S2H2 shortening was 5-9 msec (mean 29. ± 19.9 msec), representing a decrease of 2-26% compared with method I. At the shortest comparable S,S2 intervals (range 2-48 msec, msec), the S2H2 intervals were msec (mean 28.6 ± 54.6 msec) with method I and msec (mean msec) with method II. Again, a smaller S2H2 interval was noted with method II in all except four patients (nos ). The S2H2 shortening was msec (mean msec), representing a decrease of 5%. ERP of the AV Node (table 2) The ERP of the AV node was recorded in 13 patients, and in the other 13 patients, the ERP of the atrium exceeded the ERP of the AV node. In patient, the ERP of the AV node was determined only at the longer CL tested, whereas in patients 1 and 23, it could be measured only at the shortest CL tested. The ERP of the AV node in the remaining patients could be determined at both the longest and the shortest CLs tested. Longest Atrial CL The ERP of the AV node was recorded in 11 patients

4 654 CIRCULATION VOL 67. No 3, MARCH 1983 Downloaded from by on January 2, 219 TABLE 2. Electrophysiologic Data Pt SCL AH HV BCL Shortest S2H2 interval at SIH, longest of comparable SIS2 BCL I II A (%) (15%) (13%) (15%) (9%) (%) (%) (18%) (%) (%) (13%) (8%) (22%) (23%) (3%) (I 1%) (26%) (13%) (15%) (%) (1 1%) (8%) (9%) (13%) (32%) (9%) (2%) Longest S2H2 interval at shortest comparable SIS2 I II A (%) (31%) (24%) (18%) (26%) (4%) (2%) (25%) (39%) (11 %) (25%) (4%) (3%) (18%) (26%) (19%) (21%) (4%) (9%) (21%) (14%) (29%) (2%) (2%) (9%) (4%) (15%) (11%) (%) (41%) (4%) ERP-AVN I II A\ <26 < <22 <22 <22 <22 < < < < <28 <28 <28 < <28 <2 >8 >14 <28 <26 <28 <26 <18 <18 <18 < <23 <22 <23 < <26 <26 <22 < <18 <18 <28 < < < 4 44 <3 <3 <3 32 <23 < <26 <26 <22 < <18 <18 <28 < < < < 4 44 <3 <3 <3 32 <23 < > (8%) (8%o) (>22%) (> 14%) (13%) (12%) (%) (14%) (23%) (4%) (>26%) (11%) (3%) (23%) (28%) All electrophysiologic measurements are in milliseconds. Abbreviations: AVN = atrioventricular node; BCL = basic cycle length; ERP = effective refractory period; I = method I; II = method II; SCL = sinus cycle length; A = change. < indicates that the ERP-AV node measured less than the shortest SIS2 available and these values were used to calculate the change and percent shortening.

5 AVN CONDITION AFTER INTRANODAL COLLISION/Shenasa et al. 655 A C 1* J I I I B D Downloaded from by on January 2, 219 FIGURE 2. Comparison ofmethods I and II in patient 5. Tracings in this and subsequentfiguresfrom top to bottom are surface ECG lead, high right atrial (HRA), His bundle (HB) electrograms and time line (T). At the longer basic cycle length (SIS,) of 7 msec (panels A and B) and an SIS2 of4 msec, the S2H2 is 2 msec with method I (A), while with method II (B), the S2H2 is 195 msec. At the shortest basic cycle length of5 msec (C and D) and an S,S2 of 4 msec, the S2H2 interval is 245 msec with method I (C) and 2 msec with method II (D). The magnitude of shortening in S2H2 with method II is greater at shorter cycle lengths (C and D) than at longer cycle lengths (A and B). All measurements are in msec; pertinent deflections and intervals are labeled. VS = ventricular beat introduced simultaneously with the last A,. HRA A BCL-7 11 '5 V, I-I. i L-L -L-t-" -1 " 1-t-t t I -L-L- 1- " t A I L I L-1 --L-A--I- -L-I-A L-L. L. 1-- LA B I iii6'1 -A- * C I I.I I I II.I,~~~~~~~~~~-- ' v C D A i I 1 FIGURE 3. Comparison of methods I and II in 9 patients with first-degree atrioventricular block. Tracings are arranged as in figure 2. The basic atrial drive in all panels in 7 msec. The effective refractory period of the atrioventricular node (ERP-AVN) during method I and panels C and D shows ERP-AVN during method II. With method I, the ERP-AVN is 6 msec (B) and 46 msec during method HI, a 23% decrease.

6 656 CIRCULATION VOL 67, No 3, MARCH 1983 TABLE 3. Mean Values and Statistical Analysis of Electrophysiologic Measurements At longest atrial cycle length At shortest atrial cycle length Shortest S2H2 Longest S2H2 Shortest S2H2 Longest S2H2 at longest at shortest at longest at shortest comparable S1H2 comparable S1H2 comparable S1H2 comparable SIH2 S2H2 intervals (n = 21) Method I ± ± ±5.8 Method I ± A S2H2 with method Il ± ± p* <.5 <.5 <.5 <.5 At longest atrial cycle length At shortest atrial cycle length ERP-AVN (n = 7) Method I ± ± 3.2 Method II 381.4± ±112.2 A ERP-AVN with method II 61.4 ± ± 57.3 p <.5 <.5 All values pertaining to electrophysiologic data are in msec. *Comparison is made only in those patients who demonstrated changes with method II compared with method I at both the longest and shortest CL. Downloaded from by on January 2, 219 (nos. 5, 8,, 15-19, 21, 25 and 26). In nine, it was shorter with method II than with method I (fig. 3). The degree of shortening ranged from to > msec,* representing a percent decrease of 3% to > 26%. In patients 21 and 26, the ERP of the AV node was the same with either method. Shortest Atrial CL The ERP of the AV node was recorded in 12 patients (nos. 1, 5, 8, 15-19, 21, 23, 25 and 26). The ERP of the AV node with method II did not change in three patients (nos. 21, 23 and 26) and decreased in nine (by 2 to > 14 msec, i.e., a shortening of 5% to > 41% (fig. 3). Effect of First-degree AV Block In the four patients with first-degree AV nodal block (nos ), the ERP of the AV node was reached at longer SIS2 than in the patients with a normal PR interval. The relative shortening in AV nodal conduction time and ERP of the AV node using method II in these patients was not significantly different from that in patients with normal PR intervals. However, at shorter CLs, which were associated with longer S1H, intervals (table 2), the degree of S2H2 shortening with method II was greater at longer as well as shorter S1S2 intervals in most patients. Similarly, the decrease in the ERP of the AV node with method II was greater at shorter CLs than at longer CLs (table 3). (Table 3 lists the values only of patients in whom the changes were noted at both CLs.) Effect of Bundle Branch Block Two of the three patients with LBBB (nos. 19 and 2) showed results similar to patients with narrow QRS, i.e., S2H2 and ERP of the AV node became *The " >" sign indicates that the ERP of the AV node shortened by an amount greater than the shortest available SIS2 intervals. shorter using method II. In the third patient (no. 21), there was no change in either variable with method II (fig. 4). This patient also demonstrated intermittent antegrade block in the His-Purkinje system and retrograde LBBB as well as RBBB during ventricular premature stimulation (i.e., no H2 followed V2 of any V1V2). On the other hand, in two of three patients with a complete RBBB pattern (nos. 22 and 23), the S2H2 interval with method II did not change at either the longest or the shortest atrial CL, even though retrograde conduction through the left bundle branch was intact in these two patients. In case 24, a 15-msec shortening in S2H2 was noted with method II at the longer CL. In patient 25, who had an incomplete RBBB pattern, the decrease in S2H2 with method IL was similar to that in patients with narrow QRS complexes, whereas patient 26 showed no change at longer but a 4-msec decrease in S2H2 at shorter CLs. Effect of Vs Alone Omission of last Si resulted in retrograde atrial capture from Vs in all the 13 patients who had intact ventriculoatrial conduction such that the S2 was frequently within the atrial ERP. However, the effect of Vs alone was tested in seven of the remaining 13 patients who demonstrated retrograde AV nodal block with isolated Vs. Compared with both methods I and II, Vs alone produced longer S2H2 delays (fig. 5) in all seven. Effect of Simultaneous Pacing on the Atrial ERP Atrial ERP was the same with both methods. Discussion Our results strongly suggest that introduction of a Vs simultaneously with the last A1 produces shortening of AV nodal conduction and the ERP during A2. Before addressing the underlying electrophysiologic mecha-

7 AVN CONDITION AFTER INTRANODAL COLLISION/Shenasa et al. 657 A 2 - _ vi C a Er, -% A l HRA S1½1.55.6OS... _11- -IAI -I.- K. I I V - I Al VA1VA2 V HBBji ~1 11 I 11 L II B D Downloaded from by on January 2, 219. FIGURE 4. Comparison of methods I and II in a patient with left bundle branch block. Panels are arranged as infigure 2. The basic atrial drive is 55 msec in allfour panels. Panels A and B show method I and panels C and D method 11. The S,H, in allfour panels is constant at 2 msec. After introduction ofa single ventricular beat (Vs), the S1H, still measures 2 msec, indicating that the Vs did not reach the His, and consequently, it did not affect the subsequent S212 interval (C and D). Similarly, no change in the effective refractory period of the atrioventricular node (ERP-AVN) is noted after method HI compared with method I (B and D). nisms, it is important to examine whether retrograde AV nodal penetration of Vs with or without its subsequent intranodal collision with the atrial impulse caused the phenomena. In patients with normal intraventricular conduction, paced right ventricular impulses are expected to reach the AV node through the RBB. If the retrograde impulse can activate the His bundle before its anticipated antegrade depolarization, then intranodal penetration of Vs should occur. iiiiini 1111j" Since the retrograde depolarization after Vs was not identifiable in most cases, the Vs- H interval could not be directly compared to the antegrade S,H,. However, the range of antegrade S Hj intervals noted in this series at the longer CL exceeded the usual retrograde His-Purkinje system conduction time during constant CL pacing from the right ventricle.7 It has also been reported that even closely coupled Vs during atrial drive generally are not associated with significant retrograde delays in the His-Purkinje system.'4 Therefore, one could assume that during this study, the Vs did reach the AV node in patients with normal intraventricular conduction and intact retrograde conduction along the RBB. Intranodal penetration of Vs would be even more likely at shorter paced atrial CL, since the latter are associated with longer antegrade S,H, intervals and shorter refractoriness of the His-Purkinje system. That the results were related to intranodal penetration of Vs is also supported by the observation that two of three patients with antegrade complete RBBB pattern and one patient with intermittent infranodal AV block showed no change in the antegrade AV nodal conduction with method II. The results in these three may have been caused by coexisting retrograde conduction delay or RBBB such that the His activation occurred via the A, rather than the Vs impulse."5 Similarly, relatively intact retrograde RBB conduction in two of three patients with LBBB and one of two patients with incomplete RBBB can also explain why these cases showed shortening of AV nodal conduction and refractoriness similar to those with normal interventricular conduction.'5 A greater magnitude of change in AV nodal conduction and refractoriness noted at shorter rather than longer CLs can be best explained by deeper AV nodal penetration of Vs at shorter CL. The latter is not unexpected and is related to an increase in the antegrade S,H at shorter CLs without a concomitant increase (or even decrease) in the retrograde conduction time within the His-Purkinje system during Vs. Shortening of atrial CL per se does not produce an abbreviation of AV nodal refractoriness. In fact, an inverse relation between the basic atrial CL and AV nodal refractoriness is frequently noted.7'8 One might argue that the shortening of AV nodal conduction and refractoriness noted in these cases may be related to sympathetic stimulation with Vs. Similar-

8 658 CIRCULATION ~~~~~~~~~~~VOL 67, No 3, MARCH 1983 A 2 T I II 'I I III it U it It III a liiw B Downloaded from by on January 2, I....I I.. I.. I...1 -I I.. I.. I.. C~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~ J ' FIGURE 5. At a basic atrial cycle length of 55 msec, scanning with methods I (A) and II (B) are shown. (C) Introduction of single ventricular beat (Vs) at the same time as in panel B, while omitting the last S]. The VS-S2 interval in panels B and C are identical. The S2H2 interval measures 28 during metho-d I (A) and 26 msec with method 11(B). However, with omission of last SI (Al) in panel C, the S2 is followed by S2H2 of3 msec, which is even longer than that in panel A. This suggests that occurrence of last A1 simultaneously with Vs shortens atrioventricular nodal conduction during the subsequent S2.

9 AVN CONDITION AFTER INTRANODAL COLLISION/Shenasa et al. 659 Downloaded from by on January 2, 219 ly, a greater change in AV nodal conduction and refractoriness could be due to additional sympathetic stimulation secondary to faster heart rates. However, during cardiac pacing, sympathetic stimulation is generally reflex-mediated and the response is not immediate.'6 Atrial pacing at shorter CLs did not shorten the AV nodal conduction measurements during this study; in fact, the S,H,, S2H2 at comparable S152 intervals were frequently longer at shorter as compared with longer CLs during method I. It seems logical to conclude, therefore, that introduction of Vs did indeed produce the shortening of AV nodal refractoriness during this study. Another relevant issue is whether retrograde penetration of Vs per se produced the change or whether concomitant AV nodal activation from atrial impulse was necessary to shorten refractoriness of the AV node. An isolated ventricular premature beat, while failing to conduct to the atria, may result in prolongation of AV nodal conduction or block of subsequent sinus impulses within the AV node (retrograde concealed conduction in the AV node). This hypothesis was tested in seven of 13 patients with no ventriculoatrial conduction (retrograde AV nodal block) and Vs alone produced longer AV nodal conduction during S2 (fig. 5) compared with either method I or II. These findings suggest that simultaneous activation of the AV node from both atrial and ventricular impulses contributed to the shortening of AV nodal conduction and refractoriness. Programming of VS alone was not performed in many of the patients in this series, and no final conclusions can be drawn in this regard. Electrophysiologic Mechanisms Although present data suggest the simultaneous AV nodal penetration from atrial and ventricular impulses produced the noted changes, these results do not provide direct insight into the underlying cellular mechanisms for the observed phenomena. Several explanations, however, could account for these results. In the isolated rabbit AV node, Zipes et al.17 demonstrated that the dv/dt and amplitude of action potential of cells within the N region was larger during an intranodal collision of impulses originating in the atrium and His bundle than during atrial or His pacing alone. This effect, which was attributed to summation of the two impulses and the resultant change in the characteristics of action potential, could produce facilitation of local recovery and consequently conduction of subsequent impulse through the N region of the AV node. In the present study, although the precise level of intranodal collision could not be controlled, a deeper penetration by retrograde impulse was expected at shorter CLs, resulting in a higher level of intranodal collision than at longer CLs. Since greater shortening in AV nodal conduction occurred at shorter CL, Vs may have more frequently reached the area of greater physiologic delay at the shorter CLs. An alternative explanation might be that simultaneous activation of AV node from atrial and ventricular impulses shortened the total time needed to depolarize the AV node and consequently shortened the recovery time. Facilitation in the recovery of excitability of the area penetrated by Vs as a result of its earlier depolarization (as compared to A1 alone), could enhance AV nodal conduction and shorten refractoriness." 18 Clinical Implications Previous studies on animal models have shown that retrograde excitation simultaneous with, or before an atrial impulse can facilitate subsequent AV nodal conduction, and this phenomenon could explain some forms of so called supernormal conduction in the AV node The present study demonstrates the occurrence of enhanced AV nodal conduction and shortening of AV nodal ERP after Vs in the human heart as well, probably by similar mechanisms. This study also extends our previous observations that shortening of AV nodal refractoriness is partly responsible for the prevention of AV nodal reentry tachycardia induction with A2 when simultaneous AV pacing is used during the basic drive as opposed to atrial pacing alone.'2 Another clinically relevant aspect of this study is the observation that in patients with retrograde RBBB (with or without antegrade RBBB), isolated ventricular premature complexes may have difficulty in penetrating the AV node. Underdrive ventricular pacing in such cases, therefore, may be less effective in terminating AV nodal reentrant tachycardia, a method previously used for termination of this arrhythmia. 19 Acknowledgment We gratefully acknowledge Ray Grenier, Ann Edwards and Brian Miller for their assistance. References 1. Hoffman BF: Physiology of atrioventricular transmission. Circulation 24: 56, Hoffman BF, Moore EN, Stuckey JH, Cranefield PF: Functional properties of the atrioventricular system. Circ Res 13: 38, Damato AN, Lau SH, Patton RD, Steiner C, Berkowitz WD: Study of atrioventricular conduction in man using premature atrial stimulation and His bundle recordings. Circulation 4: 61, Wit AL, Weiss MB, Berkowitz WD, Rosen KM, Steiner C, Damato AN: Patterns of atrioventricular conduction in the human heart. Circ Res 27: 345, Van Capelle, Du Perron JC, Durrer D: Atrioventricular conduction in isolated rat heart. Am J Physiol 221: 284, Ferrier GR, Dresel PE: Relationship of functional refractory period to conduction in the atrioventricular node. Circ Res 35: 24, Akhtar M, Damato AN, Batsford WO, Ruskin JN, Ogunkelu JG: A comparative analysis of anterograde and retrograde conduction patterns in man. Circulation 52: 776, Denes P, Wu D, Dhingra R, Peitras RJ, Rosen KM: The effects of cycle length on cardiac refractory periods in man. Circulation 49: 32, Moore EN: Microelectrode studies on retrograde concealment of multiple premature responses. Circ Res 2: 88, Akhtar M, Damato AN, Ruskin JN, Batsford WP, Reddy CP, Ticzone AR, Dhatt MS, Gomes JAC, Calon AH: Anterograde and retrograde conduction characteristics in three patterns of paroxysmal atrioventricular junctional reentrant tachycardia. Am Heart J 95: 22, Moore EN, Spear JF: Experimental studies on the facilitation of AV conduction by ectopic beats in dogs and rabbits. Circ Res 29: 29, Akhtar M, Gilbert CJ, Al-Nouri M, Schmidt DH: Electrophysiolo-

10 66 CIRCULATION VOL 67, No 3, MARCH 1983 gic mechanisms for modification and abolition of atrioventricular junctional tachycardia with simultaneous and sequential atrial and ventricular pacing. Circulation 6: 1443, Akhtar M, Gilbert CJ, Wolf FG, Schmidt DH: Retrograde conduction in the His-Purkinje system. Analysis of the route of impulse propagation using His and right bundle branch recordings. Circulation 59: 1252, Fuchs M, Akhtar M, Gilbert C, Kumar L: Incidence of macro reentry in the His Purkinje system following ventricular extrastimuli delivered during regular atrial and ventricular paced rhythms. (abstr) Clin Res 27: 167A, Saleem T, Akhtar M, Gilbert C: Bidirectional block: a common finding in patients with anterograde bundle branch block. (abstr) Circulation 62 (suppl III): , Levy MN, Martin PJ: Neurol control of the heart. In Handbook of Physiology, Section 2: The Cardiovascular System, vol 1, The Heart, edited by Berne RM, Sperelakis N, Geiger SR. Bethesda, MD, American Physiological Society, 1979, pp Zipes DP, Mendez C, Moe GK: Evidence for summation and voltage dependency in rabbit atrioventricular nodal fibers. Circ Res 32: 17, Moe GK, Childers RW, Merideth J: An appraisal of "supernormal" AV conduction. Circulation 38: 5, Curry PVL, Rowland E, Krikler DM: Dual-demand pacing for refractory atrioventricular re-entry tachycardia. PACE 2: 137, 1979 Comparison of Total Body Surface Map Depolarization Patterns of Left Bundle Branch Block and Normal Axis with Left Bundle Branch Block and Left-axis Deviation GURBACHAN S. SOHI, M.D., NANCY C. FLOWERS, M.D., LEO G. HORAN, M.D., M.R. SRIDHARAN, M.D., AND JENNIFER C. JOHNSON, M.D. Downloaded from by on January 2, 219 SUMMARY Total body surface maps from 15 subjects with left bundle branch block and normal axis (LBBB-NA) and subjects with left bundle branch block and left axis (LBBB-LA) were analyzed and compared with maps from normal subjects. In 19 of the 25 subjects with LBBB, the timing of early upper sternal positivity was similar to that of normal subjects, indicative of timely but oppositely directed septal activation. The right ventricular breakthrough was normally located in all, but was earlier after the onset of QRS than expected in some. The initial portion of the positivity produced by left ventricular activation was located in the upper anterior chest in both LBBB-NA and LBBB-LA, but its onset was generally delayed compared with that in normal subjects, presumably because of the time taken by the right-to-left septal activation. Also, the total duration of this positivity was longer than in normal subjects and extended considerably beyond 9 msec, indicating prolonged activation of the anterior free wall of the left ventricle. In LBBB-NA, this upper anterior positivity remained anterior throughout depolarization, but in LBBB-LA it moved toward the left shoulder and the left upper back, presumably due to the posterior orientation of the terminal portion of depolarization. This terminal orientation in patients with LBBB-LA was thought to be due to the additional delay in the activation of the anterobasal portion of the left ventricle caused by selective involvement of the left anterior fascicle. THE ALTERATIONS in the ventricular depolarization produced by left bundle branch block (LBBB) have been basically understood for a long time. 1 Recently, additional knowledge has been gained from studies based on recordings of the epicardial and endocardial spread of excitation. -7 In this report, we describe the depolarization process in LBBB as deduced from total body surface mapping, as well as the distinguishing features of LBBB with normal axis (LBBB- NA) and with left-axis deviation (LBBB-LA). Methods Total body surface maps were recorded in 25 subjects in whom an electrocardiographic diagnosis of left From the Division of Cardiology, Department of Medicine, University of Louisville, and Veterans Administration Hospital, Louisville, Kentucky. Address for correspondence: G. S. Sohi, M.D., Division of Cardiology, 55 South Jackson Street, Louisville, Kentucky 422. Received April 22, 1982; revision accepted September 15, Circulation 67, No. 3, bundle branch block was made from the presence of a total QRS duration of. 12 second or more, absence of a Q or an S in leads I, avl, V5 and V6, and the presence of an RR' pattern in V5 and V6.I Fifteen subjects had a normal frontal QRS axis of + 9 to - 29 (LBBB- NA) and a left axis of -3 or less (LBBB-LA). The LBBB-NA subjects were years old (mean 67 years) and 11 were males. All the -subjects in this group had clinical evidence of left ventricular dysfunction and five had cardiac catheterization evidence of severe three-vessel coronary artery disease. The subjects in LBBB-LA group were years old (mean 62 years); five were males and five females. Similarly, all the subjects in this group had clinical evidence of left ventricular dysfunction, and one also had hypertrophic obstructive cardiomyopathy. In addition to the clinical data, chest x-rays, cardiac catheterization data in six, ECGs and vectorcardiograms in most, total body surface maps from 14 sites were recorded. No intracardiac electrophysiologic studies were performed.