- Email: [email protected]
Similar Time-Dependent Recovery Property of Fast and Slow Atrioventricular Nodal Pathways
Similar Time-Dependent Recovery Property of Fast and Slow Atrioventricular Nodal Pathways Ming-Lon Young,
MD, MPH,
Chi-Tai Kuo,
MD,
Vikas Kohli,
MD,
and Grace S. Wolff,
MD
The purpose of this study was to determine whether fast and slow atrioventricular (AV) nodal pathways have the same recovery property. AV nodal recovery property is studied by delivering atrial extrastimuli coupled to atrial beats and plotting nodal coupling intervals against nodal conduction time. In patients with dual pathways the resultant curves will include a fast to fast (F-F) and a fast to slow (F-S) pathway coupled curves. Although fast pathway recovery property can be represented by the former, slow pathway recovery property requires further assessment by studying slow to slow (S-S) pathways coupled curve. In 9 patients with dual pathways F-F, FS, S-F, and S-S curves were obtained by pacing protocols. In 8 patients (control) without dual pathways, F-F curve and atrial extrastimuli coupled to a preceding slowly conducted fast pathway beat (also designated as
S-F curve) were obtained. (1) The S-S curve had a similar time constant as the F-F curve. (2) Although the S-S curve was markedly shifted upward and leftward from the FF curve, the degree of leftward and upward shifts of the S-S curve from the F-F curve were both close to the difference of the basic fast and slow pathway conduction time (a constant). (3) Although the effective refractory period of the fast pathway in dual pathway patients was longer than that of the control patients, the slow pathway effective refractory period when corrected was close to that of fast pathway in control patients. These results suggest that the fast and slow AV nodal pathways have a similar time-dependent recovery property. Q 1997 by Excerpta Medica, Inc. (Am J Cardiol 1997;79:424–430)
trioventricular (AV) nodal recovery property A can be studied by delivering atrial extrastimuli coupled to basic atrial beats and plotting nodal rest-
node, with the lower portion of the node serving as a final common pathway and implying the existence of recovery properties in both fast and slow pathways,8 our results suggest that fast and slow pathway input routes have different conduction time and refractory periods, and converge on the body of the AV node which, in turn, is responsible for the component of time-dependent recovery property.
ing intervals against their resultant nodal conduction time. If the plot shows 2 nonoverlapping ranges of conducting times at the same resting intervals1 – 3 or discontinuous curves1,4 – 6 dual nodal pathways can be recognized. Because basic atrial beats usually conduct through the fast pathway (F), the resultant recovery curves will include an F-F coupling curve and a fast to slow pathway (S) coupling (F-S) curve. Whereas the former represents the true fast pathway recovery curve, the latter does not represent the true slow pathway recovery curve. This was pointed out by Rosen KM et al7 in 1976: ‘‘A truer assessment of slow pathway properties would be provided if the preceding conduction is also via the slow pathway.’’ Therefore, in addition to F-F and F-S curves, we obtained S-F and S-S curves by coupling extrastimuli to a preceding slow pathway conducted beat in patients with dual nodal pathways. We found that although the S-S curve was leftward and upward shifted from the F-F curve, their time-dependent recovery properties were similar. Whereas dual nodal pathways were conventionally interpreted as longitudinal dissociation within the upper portion of the
METHODS Nine patients with demonstrated dual AV nodal pathways in whom dual pathway conduction was not atrial limited formed the study group. Eight patients without dual pathways formed the control group. Informed consent for the electrophysiologic study was obtained from patients or their parents. Studies were performed under either general anesthesia or deep sedation (midazolam and fentanyl) in the postabsorptive state. An electrode catheter was positioned at the His bundle region to record the His bundle electrogram that had recognizable atrial (A), His bundle (H), and ventricular potentials. Study group: EXTRASTIMULI WITH A PRECEDING FAST PATHWAY COUPLING: A progressively shorter
From the Department of Pediatrics, University of Miami, Miami, Florida, and First Division of Cardiology, Department of Medicine, Chang Gung Memorial Hospital, Taipei, Taiwan. This work was supported in part by the Children’s Cardiac Research Foundation of Florida and Grant NSC314-B-182A-016 from the National Science Council of Academia Sinica in Taiwan. Manuscript received July 5, 1996; revised manuscript received and accepted August 29, 1996. Address for reprints: Ming-Lon Young, MD, Pediatric Cardiology, Department of Pediatrics, (R-76), University of Miami, P.O. Box 016960, Miami, Florida 33101.
424
pacing impulse (with corresponding A2 and H2) coupled to 8 atrial paced beats (with stable A1H1 intervals) was delivered to the right atrium (until atrial tissue became refractory). Both fast and slow pathway recovery curves with a preceding fast pathway coupling (F-F and F-S curves) were obtained. EXTRASTIMULI WITH A PRECEDING SLOW PATHWAY COUPLING: The protocol was similar except that the
basic beats were conducted to the ventricle via the slow nodal pathway. If the immediate preceding conducted beat to the extrastimuli was conducted
Q1997 by Excerpta Medica, Inc. All rights reserved.
/ 2w19 0766 Mp
424
Thursday Jan 23 04:47 PM
0002-9149/97/$17.00 PII S0002-9149(96)00780-1
EL–AJC (v. 79, no. 4 ’97)
0766
through the fast pathway (this occurred in most cases), an extra atrial stimulus was added with a short coupling interval to force conduction to the ventricle via the slow pathway without causing reentrant beat(s). Thus, both fast and slow pathway recovery curves with a preceding slow pathway coupling (SF and S-S curves) were obtained. Evidences of dual nodal pathways were provided by: (1) 2 distinctively different ranges of A2H2 at equivalent H1A21 – 3; (2) the pacing-induced AV nodal echoes or tachycardia showed anterograde slow-retrograde fast-type nodal reentry4,5; or, (3) a sudden prolongation of A2H2 of ¢50 ms in a 10-ms decrement of H1A2.1,4,6,7 In patients with AV nodal reentrant tachycardia, direct proof of a slow pathway was provided by a later successful slow pathway ablation performed by an anatomically guided approach.9 Control group: In control patients, similar protocols were applied except that an extra beat of a slowly conducted fast pathway impulse was introduced after the last beat of basic train with a short coupling interval that was close to the fast pathway effective refractory period. Thus, fast pathway curves with a preceding fast pathway (F-F curve) or a preceding slowly conducted fast pathway (also designated as S-F curve) were obtained. Construction of nodal recovery curves (HA method):
With the following reasoning, we chose HA as the index of nodal resting period to construct nodal recovery: Mobitz in 192410 and Lewis and Master11 first noticed that PR intervals depend on the recovery time measured from the preceding RP intervals. Because P (or, equivalently A) is the earliest input to the node while R (or H) is the latest output from the AV node, PP intervals do not take into account the conduction time change of the preceding beat, Mobitz proposed that an intermediate point between P and R should be used as standard for measurement of the recovery period. Whereas AA would overestimate and HA would underestimate the true nodal recovery period, the degree of overestimation in the former could be varied depending on the degree of prolongation of the preceding AH interval (especially when dealing with the long AH of a slow pathway conducted beat). As stated by Billete in 1976,12 the underestimation of the nodal resting period by HA method is due to: ‘‘The HA interval preceding a given atrial beat includes neither the time elapsed in activation of NH cells before the occurrence of the H complex in the preceding beat, nor the time elapsed after the A complex for the activation of late AN cells of the node in the considered beat. These 2 values should be added, respectively, at the left and right ends of the HA interval in order to obtain the exact value of the phase of excitability at which the impulse entered the delay-providing part of the node.’’ Because the sum of ‘‘the time elapsed after A for the activation of AN cells’’ and ‘‘the time elapsed in activation of NH cells before the occurrence of H’’ is equal to basic AH after an infinite long recovery time,13 – 16 use of the HA method to study nodal recovery property could underestimate
the recovery period by a constant of the basic AH. Thus, for the reason of simplicity we chose HA as the index of nodal resting period. After plotting A2H 2 against H1A2, every (H1A2,A2H2) point was classified into 4 groups: FF, F-S, S-F, and S-S couplings. In each group the individual recovery curve was fitted to the equation12,17: AH Å AHmin / exp (a 0 HA/t), for HA ¢ u, where AH (a simplified notation for A2H2) is the AH at any given HA (a simplified notation for H1A2). AHmin is the AH after an infinite recovery time (or the shortest observed A2H2). The shortest AH obtainable at the longest cycle length was taken as the fast pathway AHmin (AHmin (fast)). The shortest recognizable slow pathway conducted AH was taken as the slow pathway AHmin (AHmin(slow)). a is a positive constant, t is the recovery time constant, and u is the minimum H1A2 at which a conducted A2H2 is registered (e.g., nodal effective refractory period). The curve-fitting process was performed using a commercially available software (SigmaPlott, Jandel Scientific, San Rafael, California). As stated above, using the HA method to study the nodal recovery property could underestimate the recovery period of the fast pathway by a constant of AHmin(fast) and the slow pathway by a constant of AHmin(slow). Therefore, the difference of underestimation of true nodal coupling for recovery curves between fast and slow pathways could be equal to a d, where d Å AHmin(slow) 0 AHmin(fast). To compare fast and slow pathway recovery properties, the S-S curve was translocated by this d rightward (HA/d,AH) and rightward and downward (HA/d, AH-d) to form corrected S-S curves (S-Sc curves)(this S-Sc curve will have the same time constant as the original S-S curve). In using HA as a gauge to compare the effective refractory period of fast and slow pathways, the different degree of underestimation of true nodal coupling between fast and slow pathway should also be considered. Thus, the same observed d was used to correct the slow pathway effective refractory period (uS-Sc Å uS-S / d) to be compared with that of the fast pathway (uF-F). Five to 95% confidence intervals of the F-F curve were calculated by a commercially available software (SigmaStatt, Jandel Scientific). Mean { SD were obtained and paired Student’s t tests were applied when appropriate. A p value õ0.05 was considered significant.
RESULTS Table I lists the demographic features and parameter data of nodal recovery curves in our study and in control patients. Of the 6 patients with nodal reentrant tachycardia, the slow pathway was successfully ablated in 5 and the fast pathway in 1. In Figure 1 (case 1, upper panel), the shortest conducted beat of the fast pathway was 85 ms (AHmin(fast)). When HA was shortened from 200 ms (u of the fast pathway with a F-F coupling, uF-F) to 190 ms, there was a sudden jump in AH. The last
ARRHYTHMIAS AND CONDUCTION DISTURBANCES/AV NODAL PATHWAYS
/ 2w19 0766 Mp
425
Thursday Jan 23 04:47 PM
EL–AJC (v. 79, no. 4 ’97)
0766
425
426
THE AMERICAN JOURNAL OF CARDIOLOGYT
/ 2w19 0766 Mp
426
VOL. 79
Thursday Jan 23 04:47 PM
FEBRUARY 15, 1997
EL–AJC (v. 79, no. 4 ’97)
0766
9M 12F 15F 10F 9M 5F 14M 14F
JET WPW WPW WPW WPW WPW SD NCS
Slow area 4 attempts Slow area 2 attempts Slow area 2 attempts Slow area 1 attempt Not applicable Not applicable
Ablation
460 700 680 600 620 470 920 650 AP AP AP AP AP
ablation ablation ablation ablation ablation
u
a
t
F-S Curve
232
170
210 280
200 330 240 190 240 230
8.64
8.05
6.26 6.70 7.46 7.60 6.52 10.40
40
58
123 126 56 55 81 29
u
a
6.75
5.60
7.07 6.38
7.17 7.18 4.78 7.64 8.14 6.77
6.66
Mean 81 {SD {13
108
81 95 121 48 118 92 169 140 170
130 150 130 190 170 110 170 310
{1.04 {37 {62
6.60 6.50 5.70 8.76 6.26 5.54 6.38 7.55
70 80 100 70 80 70 100 75
108
122
114 118
103 103 205 69 63 77
137
5.90
5.58
5.90 4.63
4.87 5.46 4.80 4.77 6.39 10.67
a
100
108
64 176
115 161 83 139 33 20
t
151
130 180 100 120 170 20 120 370
187
197
310 110
170 190 140 105 220 245
uc
{75 {66
20
50
80 0100
30 20 040 90 120
070
u
S-S Curve
{40 {1.89 {55
211
130
180 230
220 270 210 200 250 210
u
{0.85 {43 {101
5.98
5.48 6.14 5.05 6.54 5.56 5.30 6.07 7.69
143 103 186 80 178 86 180 138
{31 {1.03 {42
160
135
180
100 165 180 160 200 160
t
S-F Curve
Control Group (patients without dual AV nodal pathways)
71
88
76
64 123
95 44 113 94 119 63
{36
7.63
6.95
8.63 6.81
7.37 11.75 6.42 6.89 5.96 7.85
7.71
249
t
F-F Curve
Study Group (patients with dual AV nodal pathways)
a
{51 {1.73 {28 {49 {1.35
Mean 82 {SD {22
300 290
325 250 180 230 260 180
229
70 80
85 90 60 85 130 55
Slow
AHmin Fast
82
578 Fast area 1 attempt
730 AP ablation 950 Slow area 1 attempt
880 960 660 650 800 580
SCL
Atrioventricular Nodal Recovery Curve Before Ablation
181
AHmin Slow
85 80 60 70
AHmin Fast
AP Å accessory pathway; AVNRT Å atrioventricular nodal reentrant tachycardia; JET Å junctional ectopic tachycardia; NCS Å neurocardiac syncope; SCL Å sinus cycle length (ms); SD Å aborted sudden death; WPW Å Wolff-Parkinson-White syndrome; uc Å corrected u value.
10 11 12 13 14 15 16 17
AVNRT
9 65F
AVNRT AVNRT AVNRT AVNRT JET NCS
WPW AVNRT
14F 17M 13F 31F 10M 15M
Diagnosis
7 14M 8 55F
1 2 3 4 5 6
Age (yr) No. & Sex
TABLE I Demographic Data and Atrioventricular Nodal Recovery Curve Parameters
a 5.73
74
t
u
30
u
110 220 230 130
S-S Curve
102 110 128 100
t
F-F Curve
6.65 6.55 5.83 6.22
a
After Ablation
FIGURE 1. Atrioventricular nodal recovery curves for case 1. Upper panel, individual (HA,AH) points of different coupling types and F-F, F-S, S-F, S-S, and postablation (p.a.) F-F recovery curves. Lower panel, corrected S-S (S-Sc) curve and F-F curve 5% to 95% confidence interval (C.I.) together with the above curves (postablation F-F curve is not shown). BCL Å basic driven cycle lengths (see text for discussion).
conducted beat of the slow pathway had an HA of 100 ms (uF-S). When the preceding conducted beat was via the slow pathway and the HA was shortened from 220 ms (uS-F) to 210 ms, there was a sudden AH jump from the fast pathway conducted beat to the slow pathway conducted beat of 325 ms (AHmin(slow)). The last slow pathway conducted beat had an HA of 070 ms (uS-S). Although F-F and SF curves were found to coincide (Figure 1, lower panel), the S-S curve was shifted leftward from the F-S curve. When the S-S curve was rightward translocated by the d (240 Å AHmin(slow) 325 0 AHmin(fast) 85), the S-Sc curve became rightward to the F-S curve. When the S-S curve was rightward and downward translocated by the d (e.g., [H1A2/240,A2H20240]), the S-Sc curve became superimposed on the F-F curve and was within its 95% confidence intervals. The uS-Sc (170 ms Å uS-S of 070 ms / d value of 240 ms) was shorter than the uF-F (200 ms).
Figure 2 shows nodal recovery curves of case 1 to 8 (in case 7, only a few beats of slow pathway conducted beats of F-S coupling were induced; thus, the F-S curve was not provided). In each case, although the F-F and S-F curves were closely associated, the S-S curve was shifted leftward from the F-S curve. When the S-S curve was translocated rightward by their individual d, the S-Sc curve became either rightward (cases 1 and 6) or close to (cases 2 to 5 and 8) the F-S curve. When the S-S curve was translocated rightward and downward by their individual d, the SSc curve became close to or superimposed to the FF curves (except in case 3 in which the S-Sc curve was shifted leftward of the F-F curve and was closely approximated to the S-F curve) and were within their 5% to 95% confidence intervals. In case 9 (Figure 3, upper panel) a single ablation attempt to eliminate the slow pathway resulted in fast pathway ablation. The postablation slow pathway curve (S-Spostablation curve) was obtained by atrial extrastimulation coupled to 2 basic driven cycle lengths (all A1H1 were conducted via the slow pathway). Before the fast pathway ablation, the minimum slow pathway conducted AH (AHmin(slow)preablation of 229 ms) was obscured by the fast pathway uF-F or uS-F. The last slow pathway conducted beat had an HA of 50 ms (uS-S(preablation)). When the S-S curve was translocated rightward by the d (147 Å 229 0 AHmin(fast) 82), the S-Sc curve became rightward to the F-S curve (Figure 3, lower panel). When the S-S curve was translocated rightward and downward by the d, the S-Sc curve became closely associated with the F-F curve within its 5% to 95% confidence interval. When the AHmin(slow) that was observed after fast pathway ablation (AHmin(slow)postablation of 181 ms) was taken as AHmin(slow), a d value of 99 ms was obtained (d Å 181 0 82). The last slow pathway conducted beat had an HA of 30 ms (uS-S(postablation)). When the S-Spostablation curve was translocated rightward and downward by this d, the S-Scpostablation curve became superimposed with the F-F curve. Although uS-Sc(preablation) of 197 ms (Å50 / 147) was longer than uF-F (170 ms), uS-Sc(postablation) of 129 ms (Å30 / 99) was shorter than uF-F. In all control patients, the F-F curves were closely associated with S-F curves that were obtained by coupling extrastimuli to a preceding slowly conducted fast pathway beat (exemplified by case 10, Figure 4). The slow pathway curve time constant tS-S (100 { 55 ms) in the study group was neither significantly different from the fast pathway curve time constant tF-F (88 { 28 ms) in the study group, nor significantly different from the fast pathway curve time constant tF-F (108 { 37 ms) in the control group. The corrected effective refractory period of the slow pathway uS-Sc (187 { 66 ms) in the study group was not significantly different from the effective refractory period of the fast pathway uF-F (170 { 62 ms) in the control group. However, the effective refractory period of the fast pathway in the study group uF-F (232 { 49 ms) was significantly longer than the latter (p õ0.04).
ARRHYTHMIAS AND CONDUCTION DISTURBANCES/AV NODAL PATHWAYS
/ 2w19 0766 Mp
427
Thursday Jan 23 04:47 PM
EL–AJC (v. 79, no. 4 ’97)
0766
427
FIGURE 2. Atrioventricular nodal recovery curves, corrected S-Sc curves, and F-F curve 5% to 95% confidence interval (C.I.) for cases 1 to 8.
DISCUSSION Our results can be summarized as following: (1) The S-S curve (true slow pathway recovery curve) had a similar time constant as the F-F curve (true fast pathway recovery curve). (2) Whereas the S-S curve was markedly shifted upward and leftward from the F-F curve, the degree of each shift was close to the difference of the basic fast and slow pathway conduction time (a constant). When the S-S curve is translocated rightward and downward by this value, the resultant S-Sc curve became similar to the F-F curve. (3) Whereas the effective refractory period of the fast pathway in patients with dual pathways was longer than that of the control patients, the effective refractory period of the slow pathway (when corrected) was close to that of the fast pathway of control patients. We hypothesized that (1) the leftward shift of the S-S curve from the F-S curve is due to a different degree of underestimation of the true nodal recovery 428
THE AMERICAN JOURNAL OF CARDIOLOGYT
/ 2w19 0766 Mp
428
VOL. 79
Thursday Jan 23 04:47 PM
period between fast and slow pathways (AHmin(slow) and AHmin(fast)) with a difference of d, and (2) the upward shift of the S-S curve from the F-F curve is due to the same difference of d. With these premises we expected that when the true slow pathway (S-S) curve was first translocated rightward by the observed d (e.g., equalizing the degree of underestimation of the true recovery period of fast and slow pathways), and then translocated downward by the same d (e.g., correcting the intrinsic conduction difference between fast and slow pathways), the resultant S-Sc (HA / d, AH 0 d) curve could be similar to the F-F curve if they have similar time-dependent recovery property. This was exactly what we found. In patients with dual AV nodal pathways, AHmin(fast) can be obtained by an extrastimulation protocol with a long coupling HA. Because of competition from fast pathway conduction, the observed AHmin(slow) could be different from the true AHmin(slow), which may be obtained in patients with pure slow FEBRUARY 15, 1997
EL–AJC (v. 79, no. 4 ’97)
0766
FIGURE 3. Atrioventricular nodal recovery curves for case 9 in which a single radiofrequency ablation eliminated the fast pathway (see text for discussion). Abbreviation as in Figure 1.
pathway conducted beats. In 1 such patient (case 9), although the preablation S-Sc curve (using preablation observed d) was only approximated to the F-F curve, when the post-fast pathway ablation AHmin(slow) was used to obtain the true d, the S-Scpostablation curve became completely superimposed on the latter. The underestimation of the true nodal coupling by the HA method is due to ‘‘time elapsed in activation of NH cells before occurrence of H’’ and ‘‘time elapsed after the A for the activation of late AN cells in the considered beat.’’12 If fast and slow pathways both enter the same node, when the input route of the pathway is shifted from fast to slow, the conduction time difference ‘‘from A to AN cells’’ should be the main cause for the different degree of underestimation. This could explain why the F-F and S-F coupled recovery curves are similar, but the SS curve is shifted leftward by a d from the S-F or FF curve. Although it is not entirely clear, the reason for a lesser degree of leftward shift of the F-S curve from the F-F curve might be: While the shift from fast to slow pathway causes an underestimation of the true nodal coupling and results in a leftward shift of the F-S curve, a counterbalance produced by a concealed slow pathway penetration of the basic
drive impulse into the node causes an overestimation of the true nodal coupling and results in a rightward shift of the curve (the degree of overestimation and rightward shift would be dependent upon the degree of concealed penetration). The finding that the remarkable difference between the fast pathway (F-F) curve and the slow pathway (S-S) curve can be resolved by a single constant of the basic conduction time difference between the fast and slow pathway, supports the theory that both slow and fast pathway input routes to the node exhibit little time-dependent recovery property (thus, both are likely extra nodal structures), and the body of the node per se is responsible for the component of time-dependent recovery property. The finding that the slow pathway effective refractory period in patients with a dual pathway is equal to the fast pathway effective refractory period of the control patients implies that when the effective refractory period of the slow pathway is reached, the conduction block occurs at the same level in dual pathway patients and in normal patients (e.g., node proper). The finding that the fast pathway effective refractory period in patients with a dual pathway is longer than that of control patients further indicates that the fast pathway input tract in patients with dual pathways has a longer effective refractory period than that of its node proper. Thus, we speculate that when the effective refractory period of the fast input tract is shorter than that of the node proper, a single atrioventricular recovery curve (fast curve) is observed. Otherwise, the premature atrial impulse is blocked in the fast input tract and continues to enter the node via the slow input tract, resulting in dual pathway physiology of similar time-dependent recovery property in both pathways. Study limitations: In only 1 of our study patients, the true AHmin(slow) could be obtained. Because current ablation therapy for dual pathways was directed toward elimination of the slow pathway, in most cases the true AHmin(slow) was unknown. Use of the observed AHmin(slow) may cause an inaccurate estimation of d, with subsequent erroneous translocation of the S-S curve. However, we could not justify a prospective study with intentional fast pathway ablation because of the risk of heart block. Moreover, our current protocol precludes a retrospective study. Additionally, ablation procedures may not be so selective as to leave one of the limbs unaffected. However, the exact correction factor will change only the degree of correction of the curve position on the plot, but will not change the shape of the curve nor alter the concept. Besides recovery property, AV nodal conduction time is also intrinsically controlled by facilitation and fatigue properties.11,18 – 25 Fatigue property has a slow time course and thus should have a minimal effect on our study protocol. Facilitation property has a short time constant. In our protocol, the use of a short coupling interval to produce a slow conducted beat before the scanning beats, may cause nodal facilitation and result in a slight leftward shift
ARRHYTHMIAS AND CONDUCTION DISTURBANCES/AV NODAL PATHWAYS
/ 2w19 0766 Mp
429
Thursday Jan 23 04:47 PM
EL–AJC (v. 79, no. 4 ’97)
0766
429
3. Wu D, Hung JS, Kuo CT. Determinants of sustained slow pathway conduction and relation to reentrant tachycardia in patients with dual atrioventricular nodal transmission. Am Heart J 1981;101:521–528. 4. Denes P, Wu D, Dhingra RC, Chuquimia R, Rosen KM. Demonstration of dual A-V nodal pathways in patients with paroxysmal supraventricular tachycardia. Circulation 1973;48: 549–555. 5. Wu D, Denes P, Dhingra R, Khan A, Rosen KM. The effects of propranolol on induction of A-V nodal reentrant paroxysmal tachycardia. Circulation 1974;50:665–677. 6. Denes P, Wu D, Dhingra RC, Amat-y-Leon F, Wyndham C, Rosen KM. Dual atrioventricular nodal pathways: a common electrophysiologic response. Br Heart J 1975;37:1069– 1076. 7. Rosen KM, Denes P, Wu D, Dhingra RC. Electrophysiological diagnosis and manifestation of dual A-V nodal pathways. In: Wellens HJJ, Lie KI, Janse MJ, eds. The Conduction System of the Heart. Philadelphia: Lea & Febiger, 1976:453– 466. 8. Moe GK, Preston JB, Burlington H. Physiologic evidence for a dual A-V transmission system. Circ Res 1956;4:357–375. 9. Wathen M, Natale A, Wolfe K, Lee R, Newman D, Klein G. An anatomically guide approach to atrioventricular node slow pathway ablation. Am J Cardiol 1992;70:886–889. FIGURE 4. Atrioventricular nodal recovery curves for case 10 in whom no slow ¨ ber die unvollsta¨ndige sto¨rung der erregungs10. Mobitz W. U pathway was demonstrated. Atrial extrastimulation protocol was performed u¨berleitung zwischen vorhof und kammer des menschlichen with the scanning beats coupled to a preceding fast pathway conducted beat herzens. Zeit gesamte exp Med 1924;41:180–237. with short (F-F curve) and long (S-F curve) A1H1 intervals (See text for discus11. Lewis T, Master AM. Observations upon conduction in sion). BCL Å basic cycle length. the mammalian heart: A-V conduction. Heart 1925;12: 209 – 269. 12. Billette J. Preceding His-atrial interval as a determinant of atrioventricular of the S-F curve from the F-F curve at short coupling nodal conduction time in the human and rabbit heart. Am J Cardiol intervals. This may also partially explain why in 1976;38:889–896. most cases the ‘‘corrected’’ slow pathway curve is 13. Young M. A reappraisal of atrioventricular nodal excitability during functional 2:1 block: what should be the gauge? PACE 1990;13:1666–1673. closely approximated but not ‘‘exactly like’’ the fast 14. Young M, Wolff GS, Castellanos A, Gelband H. Application of the Rosenblueth hypothesis to assess atrioventricular nodal behavior. Am J Cardiol pathway curve. 1986;57:131–134. 15. Young ML, Wolff GW, Castellanos A, Gelband H. Application of the RoAcknowledgment: We wish to thank Tsu-Juey Wu, senblueth hypothesis to assess cycle length effects on the refractoriness of the node. Am J Cardiol 1986;57:142–145. MD, and Shih-Ann Chen, MD, of Veterans General atrioventricular 16. Lehmann MH, Steinman RT, Meissner MD, Schuger CD. Quantitating AV Hospital, National Yang-Ming University, Taipei, nodal function: has A1A2 outlived its usefulness? PACE 1990;13:1674–1677. Taiwan in providing case 8 for this research, Jerome 17. Teague S, Collins S, Wu D, Denes P, Rosen K, Arzbaecher R. A quantitative of normal AV nodal conduction curve in man. J Appl Physiol Lightbourne, MD, of the Department of Pediatrics, description 1976;40:74–78. and Robert Chen, PhD, of the Department of Math- 18. Billette J. Short time constant for rate-dependent changes of atrioventricular conduction in dogs. Am J Physiol 1981;241:H26–H33. ematics and Computer Science, University of Miami, 19. Nayebpour M, Talajic M, Villemaire C, Nattel S. Vagal modulation of the for their collection of data and comments. rate-dependent properties of the atrioventricular node. Circ Res 1990;67:1152– 1166. 20. Nayebpour M, Talajic M, Nattel S. Quantitation of dynamic AV nodal ADDENDUM properties and application to predict rate-dependent AV conduction. Am J PhySince the submission of this manuscript, Amellal siol 1991;261:H292–H300. 21. Rosenblueth A. Functional refractory period of cardiac tissues. Am J Physiol F. and Billette J. (Circulation 1996;94:824–832) 1958;194:171–183. published a report on studies in the rabbit heart. This 22. Merideth J, Mendez C, Mueller WJ, Moe GK. Electrical excitability of atrioventricular nodal cells. Circ Res 1968;23:69–85. report supports our findings. 23. Billette J, Me´tayer R, St-Vincent M. Selective functional characteristics of rat-induced fatigue in rabbit atrioventricular node. Circ Res 1988;62:790–799. 24. Billette J, Me´tayer R. Origin, domain, and dynamics of rate-induced variations of functional refractory period in rabbit atrioventricular node. Circ Res 1. Rosen KM, Mehta A, Miller RA. Demonstration of dual atrioventricular 1989;65:164–175. nodal pathways in man. Am J Cardiol 1974;33:291–294. 25. Talajic M, Papadatos D, Villemaire C, Glass L, Nattel S. A unified model 2. Wu D, Denes P, Dhingra R, Pietras RJ, Rosen KM. New manifestations of of atrioventricular nodal conduction predicts dynamic changes in Wenckebach dual A-V nodal pathways. Eur J Cardiol 1975;2:459–466. periodicity. Circ Res 1991;68:1280–1293.
430
THE AMERICAN JOURNAL OF CARDIOLOGYT
/ 2w19 0766 Mp
430
VOL. 79
Thursday Jan 23 04:47 PM
FEBRUARY 15, 1997
EL–AJC (v. 79, no. 4 ’97)
0766
MD, MPH,
Chi-Tai Kuo,
MD,
Vikas Kohli,
MD,
and Grace S. Wolff,
MD
The purpose of this study was to determine whether fast and slow atrioventricular (AV) nodal pathways have the same recovery property. AV nodal recovery property is studied by delivering atrial extrastimuli coupled to atrial beats and plotting nodal coupling intervals against nodal conduction time. In patients with dual pathways the resultant curves will include a fast to fast (F-F) and a fast to slow (F-S) pathway coupled curves. Although fast pathway recovery property can be represented by the former, slow pathway recovery property requires further assessment by studying slow to slow (S-S) pathways coupled curve. In 9 patients with dual pathways F-F, FS, S-F, and S-S curves were obtained by pacing protocols. In 8 patients (control) without dual pathways, F-F curve and atrial extrastimuli coupled to a preceding slowly conducted fast pathway beat (also designated as
S-F curve) were obtained. (1) The S-S curve had a similar time constant as the F-F curve. (2) Although the S-S curve was markedly shifted upward and leftward from the FF curve, the degree of leftward and upward shifts of the S-S curve from the F-F curve were both close to the difference of the basic fast and slow pathway conduction time (a constant). (3) Although the effective refractory period of the fast pathway in dual pathway patients was longer than that of the control patients, the slow pathway effective refractory period when corrected was close to that of fast pathway in control patients. These results suggest that the fast and slow AV nodal pathways have a similar time-dependent recovery property. Q 1997 by Excerpta Medica, Inc. (Am J Cardiol 1997;79:424–430)
trioventricular (AV) nodal recovery property A can be studied by delivering atrial extrastimuli coupled to basic atrial beats and plotting nodal rest-
node, with the lower portion of the node serving as a final common pathway and implying the existence of recovery properties in both fast and slow pathways,8 our results suggest that fast and slow pathway input routes have different conduction time and refractory periods, and converge on the body of the AV node which, in turn, is responsible for the component of time-dependent recovery property.
ing intervals against their resultant nodal conduction time. If the plot shows 2 nonoverlapping ranges of conducting times at the same resting intervals1 – 3 or discontinuous curves1,4 – 6 dual nodal pathways can be recognized. Because basic atrial beats usually conduct through the fast pathway (F), the resultant recovery curves will include an F-F coupling curve and a fast to slow pathway (S) coupling (F-S) curve. Whereas the former represents the true fast pathway recovery curve, the latter does not represent the true slow pathway recovery curve. This was pointed out by Rosen KM et al7 in 1976: ‘‘A truer assessment of slow pathway properties would be provided if the preceding conduction is also via the slow pathway.’’ Therefore, in addition to F-F and F-S curves, we obtained S-F and S-S curves by coupling extrastimuli to a preceding slow pathway conducted beat in patients with dual nodal pathways. We found that although the S-S curve was leftward and upward shifted from the F-F curve, their time-dependent recovery properties were similar. Whereas dual nodal pathways were conventionally interpreted as longitudinal dissociation within the upper portion of the
METHODS Nine patients with demonstrated dual AV nodal pathways in whom dual pathway conduction was not atrial limited formed the study group. Eight patients without dual pathways formed the control group. Informed consent for the electrophysiologic study was obtained from patients or their parents. Studies were performed under either general anesthesia or deep sedation (midazolam and fentanyl) in the postabsorptive state. An electrode catheter was positioned at the His bundle region to record the His bundle electrogram that had recognizable atrial (A), His bundle (H), and ventricular potentials. Study group: EXTRASTIMULI WITH A PRECEDING FAST PATHWAY COUPLING: A progressively shorter
From the Department of Pediatrics, University of Miami, Miami, Florida, and First Division of Cardiology, Department of Medicine, Chang Gung Memorial Hospital, Taipei, Taiwan. This work was supported in part by the Children’s Cardiac Research Foundation of Florida and Grant NSC314-B-182A-016 from the National Science Council of Academia Sinica in Taiwan. Manuscript received July 5, 1996; revised manuscript received and accepted August 29, 1996. Address for reprints: Ming-Lon Young, MD, Pediatric Cardiology, Department of Pediatrics, (R-76), University of Miami, P.O. Box 016960, Miami, Florida 33101.
424
pacing impulse (with corresponding A2 and H2) coupled to 8 atrial paced beats (with stable A1H1 intervals) was delivered to the right atrium (until atrial tissue became refractory). Both fast and slow pathway recovery curves with a preceding fast pathway coupling (F-F and F-S curves) were obtained. EXTRASTIMULI WITH A PRECEDING SLOW PATHWAY COUPLING: The protocol was similar except that the
basic beats were conducted to the ventricle via the slow nodal pathway. If the immediate preceding conducted beat to the extrastimuli was conducted
Q1997 by Excerpta Medica, Inc. All rights reserved.
/ 2w19 0766 Mp
424
Thursday Jan 23 04:47 PM
0002-9149/97/$17.00 PII S0002-9149(96)00780-1
EL–AJC (v. 79, no. 4 ’97)
0766
through the fast pathway (this occurred in most cases), an extra atrial stimulus was added with a short coupling interval to force conduction to the ventricle via the slow pathway without causing reentrant beat(s). Thus, both fast and slow pathway recovery curves with a preceding slow pathway coupling (SF and S-S curves) were obtained. Evidences of dual nodal pathways were provided by: (1) 2 distinctively different ranges of A2H2 at equivalent H1A21 – 3; (2) the pacing-induced AV nodal echoes or tachycardia showed anterograde slow-retrograde fast-type nodal reentry4,5; or, (3) a sudden prolongation of A2H2 of ¢50 ms in a 10-ms decrement of H1A2.1,4,6,7 In patients with AV nodal reentrant tachycardia, direct proof of a slow pathway was provided by a later successful slow pathway ablation performed by an anatomically guided approach.9 Control group: In control patients, similar protocols were applied except that an extra beat of a slowly conducted fast pathway impulse was introduced after the last beat of basic train with a short coupling interval that was close to the fast pathway effective refractory period. Thus, fast pathway curves with a preceding fast pathway (F-F curve) or a preceding slowly conducted fast pathway (also designated as S-F curve) were obtained. Construction of nodal recovery curves (HA method):
With the following reasoning, we chose HA as the index of nodal resting period to construct nodal recovery: Mobitz in 192410 and Lewis and Master11 first noticed that PR intervals depend on the recovery time measured from the preceding RP intervals. Because P (or, equivalently A) is the earliest input to the node while R (or H) is the latest output from the AV node, PP intervals do not take into account the conduction time change of the preceding beat, Mobitz proposed that an intermediate point between P and R should be used as standard for measurement of the recovery period. Whereas AA would overestimate and HA would underestimate the true nodal recovery period, the degree of overestimation in the former could be varied depending on the degree of prolongation of the preceding AH interval (especially when dealing with the long AH of a slow pathway conducted beat). As stated by Billete in 1976,12 the underestimation of the nodal resting period by HA method is due to: ‘‘The HA interval preceding a given atrial beat includes neither the time elapsed in activation of NH cells before the occurrence of the H complex in the preceding beat, nor the time elapsed after the A complex for the activation of late AN cells of the node in the considered beat. These 2 values should be added, respectively, at the left and right ends of the HA interval in order to obtain the exact value of the phase of excitability at which the impulse entered the delay-providing part of the node.’’ Because the sum of ‘‘the time elapsed after A for the activation of AN cells’’ and ‘‘the time elapsed in activation of NH cells before the occurrence of H’’ is equal to basic AH after an infinite long recovery time,13 – 16 use of the HA method to study nodal recovery property could underestimate
the recovery period by a constant of the basic AH. Thus, for the reason of simplicity we chose HA as the index of nodal resting period. After plotting A2H 2 against H1A2, every (H1A2,A2H2) point was classified into 4 groups: FF, F-S, S-F, and S-S couplings. In each group the individual recovery curve was fitted to the equation12,17: AH Å AHmin / exp (a 0 HA/t), for HA ¢ u, where AH (a simplified notation for A2H2) is the AH at any given HA (a simplified notation for H1A2). AHmin is the AH after an infinite recovery time (or the shortest observed A2H2). The shortest AH obtainable at the longest cycle length was taken as the fast pathway AHmin (AHmin (fast)). The shortest recognizable slow pathway conducted AH was taken as the slow pathway AHmin (AHmin(slow)). a is a positive constant, t is the recovery time constant, and u is the minimum H1A2 at which a conducted A2H2 is registered (e.g., nodal effective refractory period). The curve-fitting process was performed using a commercially available software (SigmaPlott, Jandel Scientific, San Rafael, California). As stated above, using the HA method to study the nodal recovery property could underestimate the recovery period of the fast pathway by a constant of AHmin(fast) and the slow pathway by a constant of AHmin(slow). Therefore, the difference of underestimation of true nodal coupling for recovery curves between fast and slow pathways could be equal to a d, where d Å AHmin(slow) 0 AHmin(fast). To compare fast and slow pathway recovery properties, the S-S curve was translocated by this d rightward (HA/d,AH) and rightward and downward (HA/d, AH-d) to form corrected S-S curves (S-Sc curves)(this S-Sc curve will have the same time constant as the original S-S curve). In using HA as a gauge to compare the effective refractory period of fast and slow pathways, the different degree of underestimation of true nodal coupling between fast and slow pathway should also be considered. Thus, the same observed d was used to correct the slow pathway effective refractory period (uS-Sc Å uS-S / d) to be compared with that of the fast pathway (uF-F). Five to 95% confidence intervals of the F-F curve were calculated by a commercially available software (SigmaStatt, Jandel Scientific). Mean { SD were obtained and paired Student’s t tests were applied when appropriate. A p value õ0.05 was considered significant.
RESULTS Table I lists the demographic features and parameter data of nodal recovery curves in our study and in control patients. Of the 6 patients with nodal reentrant tachycardia, the slow pathway was successfully ablated in 5 and the fast pathway in 1. In Figure 1 (case 1, upper panel), the shortest conducted beat of the fast pathway was 85 ms (AHmin(fast)). When HA was shortened from 200 ms (u of the fast pathway with a F-F coupling, uF-F) to 190 ms, there was a sudden jump in AH. The last
ARRHYTHMIAS AND CONDUCTION DISTURBANCES/AV NODAL PATHWAYS
/ 2w19 0766 Mp
425
Thursday Jan 23 04:47 PM
EL–AJC (v. 79, no. 4 ’97)
0766
425
426
THE AMERICAN JOURNAL OF CARDIOLOGYT
/ 2w19 0766 Mp
426
VOL. 79
Thursday Jan 23 04:47 PM
FEBRUARY 15, 1997
EL–AJC (v. 79, no. 4 ’97)
0766
9M 12F 15F 10F 9M 5F 14M 14F
JET WPW WPW WPW WPW WPW SD NCS
Slow area 4 attempts Slow area 2 attempts Slow area 2 attempts Slow area 1 attempt Not applicable Not applicable
Ablation
460 700 680 600 620 470 920 650 AP AP AP AP AP
ablation ablation ablation ablation ablation
u
a
t
F-S Curve
232
170
210 280
200 330 240 190 240 230
8.64
8.05
6.26 6.70 7.46 7.60 6.52 10.40
40
58
123 126 56 55 81 29
u
a
6.75
5.60
7.07 6.38
7.17 7.18 4.78 7.64 8.14 6.77
6.66
Mean 81 {SD {13
108
81 95 121 48 118 92 169 140 170
130 150 130 190 170 110 170 310
{1.04 {37 {62
6.60 6.50 5.70 8.76 6.26 5.54 6.38 7.55
70 80 100 70 80 70 100 75
108
122
114 118
103 103 205 69 63 77
137
5.90
5.58
5.90 4.63
4.87 5.46 4.80 4.77 6.39 10.67
a
100
108
64 176
115 161 83 139 33 20
t
151
130 180 100 120 170 20 120 370
187
197
310 110
170 190 140 105 220 245
uc
{75 {66
20
50
80 0100
30 20 040 90 120
070
u
S-S Curve
{40 {1.89 {55
211
130
180 230
220 270 210 200 250 210
u
{0.85 {43 {101
5.98
5.48 6.14 5.05 6.54 5.56 5.30 6.07 7.69
143 103 186 80 178 86 180 138
{31 {1.03 {42
160
135
180
100 165 180 160 200 160
t
S-F Curve
Control Group (patients without dual AV nodal pathways)
71
88
76
64 123
95 44 113 94 119 63
{36
7.63
6.95
8.63 6.81
7.37 11.75 6.42 6.89 5.96 7.85
7.71
249
t
F-F Curve
Study Group (patients with dual AV nodal pathways)
a
{51 {1.73 {28 {49 {1.35
Mean 82 {SD {22
300 290
325 250 180 230 260 180
229
70 80
85 90 60 85 130 55
Slow
AHmin Fast
82
578 Fast area 1 attempt
730 AP ablation 950 Slow area 1 attempt
880 960 660 650 800 580
SCL
Atrioventricular Nodal Recovery Curve Before Ablation
181
AHmin Slow
85 80 60 70
AHmin Fast
AP Å accessory pathway; AVNRT Å atrioventricular nodal reentrant tachycardia; JET Å junctional ectopic tachycardia; NCS Å neurocardiac syncope; SCL Å sinus cycle length (ms); SD Å aborted sudden death; WPW Å Wolff-Parkinson-White syndrome; uc Å corrected u value.
10 11 12 13 14 15 16 17
AVNRT
9 65F
AVNRT AVNRT AVNRT AVNRT JET NCS
WPW AVNRT
14F 17M 13F 31F 10M 15M
Diagnosis
7 14M 8 55F
1 2 3 4 5 6
Age (yr) No. & Sex
TABLE I Demographic Data and Atrioventricular Nodal Recovery Curve Parameters
a 5.73
74
t
u
30
u
110 220 230 130
S-S Curve
102 110 128 100
t
F-F Curve
6.65 6.55 5.83 6.22
a
After Ablation
FIGURE 1. Atrioventricular nodal recovery curves for case 1. Upper panel, individual (HA,AH) points of different coupling types and F-F, F-S, S-F, S-S, and postablation (p.a.) F-F recovery curves. Lower panel, corrected S-S (S-Sc) curve and F-F curve 5% to 95% confidence interval (C.I.) together with the above curves (postablation F-F curve is not shown). BCL Å basic driven cycle lengths (see text for discussion).
conducted beat of the slow pathway had an HA of 100 ms (uF-S). When the preceding conducted beat was via the slow pathway and the HA was shortened from 220 ms (uS-F) to 210 ms, there was a sudden AH jump from the fast pathway conducted beat to the slow pathway conducted beat of 325 ms (AHmin(slow)). The last slow pathway conducted beat had an HA of 070 ms (uS-S). Although F-F and SF curves were found to coincide (Figure 1, lower panel), the S-S curve was shifted leftward from the F-S curve. When the S-S curve was rightward translocated by the d (240 Å AHmin(slow) 325 0 AHmin(fast) 85), the S-Sc curve became rightward to the F-S curve. When the S-S curve was rightward and downward translocated by the d (e.g., [H1A2/240,A2H20240]), the S-Sc curve became superimposed on the F-F curve and was within its 95% confidence intervals. The uS-Sc (170 ms Å uS-S of 070 ms / d value of 240 ms) was shorter than the uF-F (200 ms).
Figure 2 shows nodal recovery curves of case 1 to 8 (in case 7, only a few beats of slow pathway conducted beats of F-S coupling were induced; thus, the F-S curve was not provided). In each case, although the F-F and S-F curves were closely associated, the S-S curve was shifted leftward from the F-S curve. When the S-S curve was translocated rightward by their individual d, the S-Sc curve became either rightward (cases 1 and 6) or close to (cases 2 to 5 and 8) the F-S curve. When the S-S curve was translocated rightward and downward by their individual d, the SSc curve became close to or superimposed to the FF curves (except in case 3 in which the S-Sc curve was shifted leftward of the F-F curve and was closely approximated to the S-F curve) and were within their 5% to 95% confidence intervals. In case 9 (Figure 3, upper panel) a single ablation attempt to eliminate the slow pathway resulted in fast pathway ablation. The postablation slow pathway curve (S-Spostablation curve) was obtained by atrial extrastimulation coupled to 2 basic driven cycle lengths (all A1H1 were conducted via the slow pathway). Before the fast pathway ablation, the minimum slow pathway conducted AH (AHmin(slow)preablation of 229 ms) was obscured by the fast pathway uF-F or uS-F. The last slow pathway conducted beat had an HA of 50 ms (uS-S(preablation)). When the S-S curve was translocated rightward by the d (147 Å 229 0 AHmin(fast) 82), the S-Sc curve became rightward to the F-S curve (Figure 3, lower panel). When the S-S curve was translocated rightward and downward by the d, the S-Sc curve became closely associated with the F-F curve within its 5% to 95% confidence interval. When the AHmin(slow) that was observed after fast pathway ablation (AHmin(slow)postablation of 181 ms) was taken as AHmin(slow), a d value of 99 ms was obtained (d Å 181 0 82). The last slow pathway conducted beat had an HA of 30 ms (uS-S(postablation)). When the S-Spostablation curve was translocated rightward and downward by this d, the S-Scpostablation curve became superimposed with the F-F curve. Although uS-Sc(preablation) of 197 ms (Å50 / 147) was longer than uF-F (170 ms), uS-Sc(postablation) of 129 ms (Å30 / 99) was shorter than uF-F. In all control patients, the F-F curves were closely associated with S-F curves that were obtained by coupling extrastimuli to a preceding slowly conducted fast pathway beat (exemplified by case 10, Figure 4). The slow pathway curve time constant tS-S (100 { 55 ms) in the study group was neither significantly different from the fast pathway curve time constant tF-F (88 { 28 ms) in the study group, nor significantly different from the fast pathway curve time constant tF-F (108 { 37 ms) in the control group. The corrected effective refractory period of the slow pathway uS-Sc (187 { 66 ms) in the study group was not significantly different from the effective refractory period of the fast pathway uF-F (170 { 62 ms) in the control group. However, the effective refractory period of the fast pathway in the study group uF-F (232 { 49 ms) was significantly longer than the latter (p õ0.04).
ARRHYTHMIAS AND CONDUCTION DISTURBANCES/AV NODAL PATHWAYS
/ 2w19 0766 Mp
427
Thursday Jan 23 04:47 PM
EL–AJC (v. 79, no. 4 ’97)
0766
427
FIGURE 2. Atrioventricular nodal recovery curves, corrected S-Sc curves, and F-F curve 5% to 95% confidence interval (C.I.) for cases 1 to 8.
DISCUSSION Our results can be summarized as following: (1) The S-S curve (true slow pathway recovery curve) had a similar time constant as the F-F curve (true fast pathway recovery curve). (2) Whereas the S-S curve was markedly shifted upward and leftward from the F-F curve, the degree of each shift was close to the difference of the basic fast and slow pathway conduction time (a constant). When the S-S curve is translocated rightward and downward by this value, the resultant S-Sc curve became similar to the F-F curve. (3) Whereas the effective refractory period of the fast pathway in patients with dual pathways was longer than that of the control patients, the effective refractory period of the slow pathway (when corrected) was close to that of the fast pathway of control patients. We hypothesized that (1) the leftward shift of the S-S curve from the F-S curve is due to a different degree of underestimation of the true nodal recovery 428
THE AMERICAN JOURNAL OF CARDIOLOGYT
/ 2w19 0766 Mp
428
VOL. 79
Thursday Jan 23 04:47 PM
period between fast and slow pathways (AHmin(slow) and AHmin(fast)) with a difference of d, and (2) the upward shift of the S-S curve from the F-F curve is due to the same difference of d. With these premises we expected that when the true slow pathway (S-S) curve was first translocated rightward by the observed d (e.g., equalizing the degree of underestimation of the true recovery period of fast and slow pathways), and then translocated downward by the same d (e.g., correcting the intrinsic conduction difference between fast and slow pathways), the resultant S-Sc (HA / d, AH 0 d) curve could be similar to the F-F curve if they have similar time-dependent recovery property. This was exactly what we found. In patients with dual AV nodal pathways, AHmin(fast) can be obtained by an extrastimulation protocol with a long coupling HA. Because of competition from fast pathway conduction, the observed AHmin(slow) could be different from the true AHmin(slow), which may be obtained in patients with pure slow FEBRUARY 15, 1997
EL–AJC (v. 79, no. 4 ’97)
0766
FIGURE 3. Atrioventricular nodal recovery curves for case 9 in which a single radiofrequency ablation eliminated the fast pathway (see text for discussion). Abbreviation as in Figure 1.
pathway conducted beats. In 1 such patient (case 9), although the preablation S-Sc curve (using preablation observed d) was only approximated to the F-F curve, when the post-fast pathway ablation AHmin(slow) was used to obtain the true d, the S-Scpostablation curve became completely superimposed on the latter. The underestimation of the true nodal coupling by the HA method is due to ‘‘time elapsed in activation of NH cells before occurrence of H’’ and ‘‘time elapsed after the A for the activation of late AN cells in the considered beat.’’12 If fast and slow pathways both enter the same node, when the input route of the pathway is shifted from fast to slow, the conduction time difference ‘‘from A to AN cells’’ should be the main cause for the different degree of underestimation. This could explain why the F-F and S-F coupled recovery curves are similar, but the SS curve is shifted leftward by a d from the S-F or FF curve. Although it is not entirely clear, the reason for a lesser degree of leftward shift of the F-S curve from the F-F curve might be: While the shift from fast to slow pathway causes an underestimation of the true nodal coupling and results in a leftward shift of the F-S curve, a counterbalance produced by a concealed slow pathway penetration of the basic
drive impulse into the node causes an overestimation of the true nodal coupling and results in a rightward shift of the curve (the degree of overestimation and rightward shift would be dependent upon the degree of concealed penetration). The finding that the remarkable difference between the fast pathway (F-F) curve and the slow pathway (S-S) curve can be resolved by a single constant of the basic conduction time difference between the fast and slow pathway, supports the theory that both slow and fast pathway input routes to the node exhibit little time-dependent recovery property (thus, both are likely extra nodal structures), and the body of the node per se is responsible for the component of time-dependent recovery property. The finding that the slow pathway effective refractory period in patients with a dual pathway is equal to the fast pathway effective refractory period of the control patients implies that when the effective refractory period of the slow pathway is reached, the conduction block occurs at the same level in dual pathway patients and in normal patients (e.g., node proper). The finding that the fast pathway effective refractory period in patients with a dual pathway is longer than that of control patients further indicates that the fast pathway input tract in patients with dual pathways has a longer effective refractory period than that of its node proper. Thus, we speculate that when the effective refractory period of the fast input tract is shorter than that of the node proper, a single atrioventricular recovery curve (fast curve) is observed. Otherwise, the premature atrial impulse is blocked in the fast input tract and continues to enter the node via the slow input tract, resulting in dual pathway physiology of similar time-dependent recovery property in both pathways. Study limitations: In only 1 of our study patients, the true AHmin(slow) could be obtained. Because current ablation therapy for dual pathways was directed toward elimination of the slow pathway, in most cases the true AHmin(slow) was unknown. Use of the observed AHmin(slow) may cause an inaccurate estimation of d, with subsequent erroneous translocation of the S-S curve. However, we could not justify a prospective study with intentional fast pathway ablation because of the risk of heart block. Moreover, our current protocol precludes a retrospective study. Additionally, ablation procedures may not be so selective as to leave one of the limbs unaffected. However, the exact correction factor will change only the degree of correction of the curve position on the plot, but will not change the shape of the curve nor alter the concept. Besides recovery property, AV nodal conduction time is also intrinsically controlled by facilitation and fatigue properties.11,18 – 25 Fatigue property has a slow time course and thus should have a minimal effect on our study protocol. Facilitation property has a short time constant. In our protocol, the use of a short coupling interval to produce a slow conducted beat before the scanning beats, may cause nodal facilitation and result in a slight leftward shift
ARRHYTHMIAS AND CONDUCTION DISTURBANCES/AV NODAL PATHWAYS
/ 2w19 0766 Mp
429
Thursday Jan 23 04:47 PM
EL–AJC (v. 79, no. 4 ’97)
0766
429
3. Wu D, Hung JS, Kuo CT. Determinants of sustained slow pathway conduction and relation to reentrant tachycardia in patients with dual atrioventricular nodal transmission. Am Heart J 1981;101:521–528. 4. Denes P, Wu D, Dhingra RC, Chuquimia R, Rosen KM. Demonstration of dual A-V nodal pathways in patients with paroxysmal supraventricular tachycardia. Circulation 1973;48: 549–555. 5. Wu D, Denes P, Dhingra R, Khan A, Rosen KM. The effects of propranolol on induction of A-V nodal reentrant paroxysmal tachycardia. Circulation 1974;50:665–677. 6. Denes P, Wu D, Dhingra RC, Amat-y-Leon F, Wyndham C, Rosen KM. Dual atrioventricular nodal pathways: a common electrophysiologic response. Br Heart J 1975;37:1069– 1076. 7. Rosen KM, Denes P, Wu D, Dhingra RC. Electrophysiological diagnosis and manifestation of dual A-V nodal pathways. In: Wellens HJJ, Lie KI, Janse MJ, eds. The Conduction System of the Heart. Philadelphia: Lea & Febiger, 1976:453– 466. 8. Moe GK, Preston JB, Burlington H. Physiologic evidence for a dual A-V transmission system. Circ Res 1956;4:357–375. 9. Wathen M, Natale A, Wolfe K, Lee R, Newman D, Klein G. An anatomically guide approach to atrioventricular node slow pathway ablation. Am J Cardiol 1992;70:886–889. FIGURE 4. Atrioventricular nodal recovery curves for case 10 in whom no slow ¨ ber die unvollsta¨ndige sto¨rung der erregungs10. Mobitz W. U pathway was demonstrated. Atrial extrastimulation protocol was performed u¨berleitung zwischen vorhof und kammer des menschlichen with the scanning beats coupled to a preceding fast pathway conducted beat herzens. Zeit gesamte exp Med 1924;41:180–237. with short (F-F curve) and long (S-F curve) A1H1 intervals (See text for discus11. Lewis T, Master AM. Observations upon conduction in sion). BCL Å basic cycle length. the mammalian heart: A-V conduction. Heart 1925;12: 209 – 269. 12. Billette J. Preceding His-atrial interval as a determinant of atrioventricular of the S-F curve from the F-F curve at short coupling nodal conduction time in the human and rabbit heart. Am J Cardiol intervals. This may also partially explain why in 1976;38:889–896. most cases the ‘‘corrected’’ slow pathway curve is 13. Young M. A reappraisal of atrioventricular nodal excitability during functional 2:1 block: what should be the gauge? PACE 1990;13:1666–1673. closely approximated but not ‘‘exactly like’’ the fast 14. Young M, Wolff GS, Castellanos A, Gelband H. Application of the Rosenblueth hypothesis to assess atrioventricular nodal behavior. Am J Cardiol pathway curve. 1986;57:131–134. 15. Young ML, Wolff GW, Castellanos A, Gelband H. Application of the RoAcknowledgment: We wish to thank Tsu-Juey Wu, senblueth hypothesis to assess cycle length effects on the refractoriness of the node. Am J Cardiol 1986;57:142–145. MD, and Shih-Ann Chen, MD, of Veterans General atrioventricular 16. Lehmann MH, Steinman RT, Meissner MD, Schuger CD. Quantitating AV Hospital, National Yang-Ming University, Taipei, nodal function: has A1A2 outlived its usefulness? PACE 1990;13:1674–1677. Taiwan in providing case 8 for this research, Jerome 17. Teague S, Collins S, Wu D, Denes P, Rosen K, Arzbaecher R. A quantitative of normal AV nodal conduction curve in man. J Appl Physiol Lightbourne, MD, of the Department of Pediatrics, description 1976;40:74–78. and Robert Chen, PhD, of the Department of Math- 18. Billette J. Short time constant for rate-dependent changes of atrioventricular conduction in dogs. Am J Physiol 1981;241:H26–H33. ematics and Computer Science, University of Miami, 19. Nayebpour M, Talajic M, Villemaire C, Nattel S. Vagal modulation of the for their collection of data and comments. rate-dependent properties of the atrioventricular node. Circ Res 1990;67:1152– 1166. 20. Nayebpour M, Talajic M, Nattel S. Quantitation of dynamic AV nodal ADDENDUM properties and application to predict rate-dependent AV conduction. Am J PhySince the submission of this manuscript, Amellal siol 1991;261:H292–H300. 21. Rosenblueth A. Functional refractory period of cardiac tissues. Am J Physiol F. and Billette J. (Circulation 1996;94:824–832) 1958;194:171–183. published a report on studies in the rabbit heart. This 22. Merideth J, Mendez C, Mueller WJ, Moe GK. Electrical excitability of atrioventricular nodal cells. Circ Res 1968;23:69–85. report supports our findings. 23. Billette J, Me´tayer R, St-Vincent M. Selective functional characteristics of rat-induced fatigue in rabbit atrioventricular node. Circ Res 1988;62:790–799. 24. Billette J, Me´tayer R. Origin, domain, and dynamics of rate-induced variations of functional refractory period in rabbit atrioventricular node. Circ Res 1. Rosen KM, Mehta A, Miller RA. Demonstration of dual atrioventricular 1989;65:164–175. nodal pathways in man. Am J Cardiol 1974;33:291–294. 25. Talajic M, Papadatos D, Villemaire C, Glass L, Nattel S. A unified model 2. Wu D, Denes P, Dhingra R, Pietras RJ, Rosen KM. New manifestations of of atrioventricular nodal conduction predicts dynamic changes in Wenckebach dual A-V nodal pathways. Eur J Cardiol 1975;2:459–466. periodicity. Circ Res 1991;68:1280–1293.
430
THE AMERICAN JOURNAL OF CARDIOLOGYT
/ 2w19 0766 Mp
430
VOL. 79
Thursday Jan 23 04:47 PM
FEBRUARY 15, 1997
EL–AJC (v. 79, no. 4 ’97)
0766