- Email: [email protected]
Transient entrainment and interruption of atrioventricular node tachycardia
769
JACC Vol. lJ. No 4 April 19K7:76lJ- 7)
Transient Entrainment and Interruption of Atrioventricular Node Tachycardia PEDRO BRUGADA, MD, ALBERT L. WALDO, MD. FACC,* HEIN J.J . WELLENS, MD, FACC Maastricht . The Netherlands
The possibility of transiently entraining and interrupting the common type of atrioventricular (AV) node tachycardia (anterograde slow, retrograde fast AV node pathway) was studied using atrial and ventricular pacing in 18 patients with paroxysmal AV node tachycardia. Transient entrainment occurred in all patients. During atrial pacing, localized block in the AV node for one beat followed by anterograde conduction over the fast pathway was obser ved in three patients. During ventr icular pacing , localized block for one beat followed by retrograde conduction over the slow pathway was not observed in any patient. Neither atrial nor ventricular fusion beats were obser ved during entrainment.
These observations indicate in a way not previously shown that reentry involving two functionally dissociated pathw ays in the AV node is the underlying mechanism of paroxysmal AV node tachycardia. The inability to demonstrate atrial or ventricular fusion beats during entrainment suggests a true intranodal location of the reentrant circuit. Finally, the abilit y to transiently entr ain intranodal tach ycardia demon strates that this electrophysiologic phenomenon is not exclusively limited to macr oreentrant circuits. o Am Coli Cardiol /9 87;9:769-75)
Transient entrainment and interrupt ion of tachycardias has now been studied in the human heart during many spontaneous and induced arrhythmias. The observations made during transient entrainment of atrial nutter ( I), ventricular tachycardia (2.3) and circus movement tachycardia using an accessory pathway (4,5) have increased understanding of the phenomenon of transient entrainment and have led to the suggestion that tachycardias that can be transiently entrained are the result of reentry with an excitable gap. It has recently been suggested that the demonstration of transient entrainment of a tachycardia not only suggests reentry. but also strongly suggests macrorcentry as the underlying mechanism (6). In some patients with paroxysmal supraventricular tachycardia. the presence and use of an accessory pathway during the arrhythmia can be excluded by detailed studies during
programmed electrical stimulation of the heart (7) . As summarized recently (8). there is strong evidence that in some of these patients. neither the atria nor the ventricles participate in the mechanism of the tachycardia, suggesting a true intranodal location of the reentrant circuit. This reentrant circuit is probably one of the smallest that occurs naturally in the human heart. We studied the possibility of transiently entraining the tachycardia in patients with atrioventricular (AV) node tachycardia, a putatively reentrant rhythm (8) to 1) further investigate and characterize entrainment and interruption of reentrant rhythms; 2) examine and emphasize the suggested criteria for the demonstration of entrainment; 3) dissect and characterize the components of the reentrant circuit of AV node tachycardia; and 4) further examine and provide insights into dual AV node pathways.
From the Unive rsity Hospital. Depa rtment of Cardio logy . Maast ric ht. The Net herlands . Th is work was suppo rted in part by G rant IROI HL293XI from the Natio na l Heart . Lung . and Blood Institute . Natio nal Institutes of Health . Bethesda, Mary land . *Prese nt address: Division of Ca rdio logy . Case Wes tern Reserve Universi ty . Unive rsity Hosp itals of Cleve land . Cleveland. Ohio 44106. Manuscript receiv ed Jun e 2. 1986: revised man uscript rece ived Se ptember 23. 1986. acce pted October 10 . 1986 . Address for rep rint s: Ped ro Bru gada . MD. Cli nica l Elect rophys iology Laboratory. University Hospital. Depart men t of Ca rdiolog y . Maa strich t. Th e Neth erland s.
Methods
I!) 1987
by the American College of Cardiology
Study patients. Eighteen patients with paroxysmal supraventricular tachycardia were studied . There were 10 women and 8 men whose ages ranged from 25 to 64 years (mean 44). After informed consent was obtained. the patients underwent programmed electrical stimulation of the heart in the absence of any antiarrhythmic drugs . Programmed stimulation was performed using our previously 07.\)-1 O(!7 /87/$\ .50
770
BRUGADA ET AL. TRANSIENT ENTRAINMENT OF AV NODE TACHYCARDIA
described methodsof stimulation, recording and analysisof tracings (7). Electrode catheters were placed at the right atrium, right ventricularapex, His bundle area and coronary sinus. Once the modesof initiationand termination of tachycardia and the site of origin of the arrhythmia had been identified, the possibility of transiently entraining and terminating the tachycardia by rapid atrial and ventricular pacing was analyzed. Criteria for AV node tachycardia. The "common" (anterograde slow, retrograde fast) form of AV node tachycardia was diagnosed in all patients by using the following criteria (8). I. Initiation of supraventricular tachycardia by atrial premature beats or atrial pacing only after conduction over the AV node at a critical delay in AV node conduction time, and after a discontinuity in the anterograde AV node conduction curve. 2. The temporal occurrence of both the first atrial echo beat and subsequent atrial activation during the tachycardia after AV node and His bundle activation, but preceding or simultaneous with ventricular activation. 3. A similar atrial activation pattern during retrograde conduction over the fast AV node pathway during ventricular pacing and during tachycardia. 4. A His to atrium (HA) interval during tachycardia independent of the His to ventricle (HV) interval and bundle branch block pattern. 5. Inabilityto advanceatrial activationduring tachycardia by a ventricular premature beat introduced at the time of refractoriness of the bundle of His. 6. No evidence for anterograde or retrograde conduction over an accessory pathway. Criteria for transient entrainment. Previously reported criteria were used to establish the presence of transient entrainment (1-5,9). These criteria were, in summary: I) constant fusion beats during the period of entrainment except for the last captured beat, 2) progressive fusion at different entraining rates, and 3) localized block for one beat followed by antidromic activation. Concealed entrainment was defined as previously reported (5). To study transient entrainment, trains of 20 stimuli were introduced at selected rates from the right atrium (proximal to the area of slow conduction) and subsequently from the right ventricle (distal to the area of slow conduction). In three patients, stimuli were also introduced from the proximal coronary sinus. The firsttrain of stimuli waseither 10beats/min faster or had a 10 ms shorter cycle length than the tachycardia rate. The pacing rate was increased by 5 beat/min steps for the subsequent trains until the tachycardia was terminated.
Results The spontaneous tachycardia cycle length at rest ranged from 240 to 470 ms (mean 345). The mean pacing cycle
lACC Vol. 'I. No.4 April 1987:76'1- 75
length that terminated the tachycardia from the right atrium was 298 ms (range 220 to 410). There was a mean zone of 48 ms (range 15 to 110) in which the tachycardia was not terminated by rapid pacing and in which transient entrainment could be studied. In the nine patients in whom rapid ventricular pacing was performed, the pacing cycle length that terminated the
tachycardia ranged from 210 to 410 ms (mean 300). There was a mean zone of 58 ms (range 10 to 95) in which the tachycardia was not terminated by rapid ventricular pacing, and in which transient entrainment could be studied. Effects of atrial pacing at cycle lengths not terminating AV node tachycardia. During atrial pacing at cycle lengths not resulting in termination of AV node tachycardia, the following observations were made (Fig. I). I) The atrial and ventricular rates increased to the pacing rate, associated with a marked increase in AH interval in
all instances. The changes in the AH interval at different cycle lengths not terminating the tachycardia are illustrated for eight patients in Figure 2, although data were similar for all 18 patients. 2) Ventricular activation after the last paced atrial beat occurred at the pacing cycle length , and was thereafter
immediately followed by resumption of the original tachycardia cycle length. Atrial activation after the last paced atrial beat always occurred after an interval that was longer than the pacing or tachycardia cycle length. 3) No atrial or ventricularfusion beats were seen. Therefore, although it seems certain that transient entrainment had occurred (concealed entrainment), it could not be established in the absence of fusion beats (3-5). Effects of atrial pacing at cycle lengths resulting in termination of AV node tachycardia. In all 18 patients, the pacingcycle length resulting in termination of the tachycardia was associated with a Wenckebach conduction pattern in the anterograde slow pathway. When conduction block occurred in this pathway, the tachycardia was terminated. In 15 patients, continuation of pacing at the same cycle length resulted in a repeated Wenckebach conduction pattern in the slow pathway, so that if cessation of pacing did not occur with a blocked beat, but rather occurred after a conducted beat, the tachycardia was reinitiated (4,5). In three patients. continuation of pacing after termination of tachycardia resulted in immediate I: I anterograde AV node conduction (Fig. 3). The latter represents conduction block to the His bundle for one beat. followed by activation of the His bundle from a different direction (by way of the fast pathway) and with a shorter conduction time (that is. a shorter AH interval). For the other 15 patients, this did not happen because, presumably, the anterograde effective refractory period of the fast pathway was longer than the pacing cycle length. Effects of ventricular pacing at cycle lengths not terminating AV node tachycardia. During ventricularpacing
Figure I. Five electrocardiographic leads are simultaneously shown with bipolar electrograms from the right atrium (RA).distal and proximal coronary sinus (CSd, CSp) and a His bundle recording during a representative example of transient entrainment of atrioventricular (AV) node tachycardia during atrial pacing. The spontaneous tachycardia cycle length was 355 ms. Pacing is performed from the right atrium at a cycle length of 345 ms. The atrial and ventricular cyclelengths shorten to the pacing cycle length. Ventricular activation after the last paced atrial complex occurs at the pacing cycle length (345 ms) and is followed by immediate resumption of the original tachycardia cycle length. The interval between the last paced atrial complex and the next atrial activation during tachycardia is 430 rns, longer than the pacing cycle length and the spontaneous tachycardia cycle length. The AH interval during entrainment prolongs to 335 ms, as compared with an AH interval of 300 ms during tachycardia. In this and all subsequent fi gures. all values are in milliseconds.
771
BRUGADA ET AL. TRANSIENT ENTRAINMENT OF AV NODE TAC HYCA RDIA
l ACC Vol. 9. No, 4 April 1987:769-75
II
A
A
A
345
430
A
355
A
355
A
RA_~~~V"""'---'V'---I/'
j
HI5-L.M.'
I
;
H '! I i
AH.I 335 CS~
j
H ': '
H :
300
I
I
1I ~~
Hi ',. ,~I.-.----....,J'lI"f __ H 'j
n
I
---Ilt------ '----I~'---.....Jf'-
~,....-~'\Yr-'- - 1 -.....JIMr-·
at cycle lengths not terminating AV node tachycard ia, the following observations were made (Fig . 4) . I ) The atrial and ventricular rates increased 10 the pacing rates with minimal changes in the ventricle to atrium (VA) interval when different pacing rates were compared. Changes in the VA interval at different pacing cycle lengths are illustrated for six patients in Figure 5. 2) Atrial activation after the last paced ventricular beat occurred at the pacing cycle length, and was thereafter immed iately followed by resumption of the original tachycardia cycle length. Ventricular activation after the last paced ventricular beat always occurred after an inter val that was longer than the pacing or tachycardia cycle length (Fig. 4) . 3) No atrial or ventricular f usion beats were seen. Once again, although it seems certain that transient entrainment had occurred (concealed entrainment), it could not be established in the absence of fusion beats (4,5) . Effects of ventricular pacing at cycle lengths terminating AV node tachycardia. During ventricular pacing. the tachycardia was termin ated in four patients by block in VA conduction (block of the antidromic wave front of the pacing impulse in the slow pathway and block of the orthodromic wave front of the pacing impulse in the fast pathw ay) and in the remain ing five patients by both retrograde and anterograde block in the slow pathway . Contin uation of ventricular pacing at the same cycle length once the tachycardia had been terminated was never associated with retrograde conduction over the slow pathway.
A >/'-------...tv-
\
-/r--~ ,(Iv-r-- ............. , Ir--- ~ ,"{"I""----.,..Hi........ ~,
Ii" ',
A
~
I
I
Figure 2. Changes in AH interval during entrainment of atrioventricular (AV) node tachycardia are shown plotted against the pacing cycle length (PCl) . Data from 8 of the 18 patients were selected for illustrative purposes. The AV node conduction times over the slow pathway increased markedly with minorchanges in pacing cycle length (filled circles), demonstrating very steep AV node conduction curves of the anterograde slow pathway. Open circles represent the AH interval during the spontaneous tachycardia. A-H (ms) 450
•
400 III
o
.: 350
300
~ :
•
... ..
••
;" -~bo
o
:
:~--b6 250
'0 0
300
350
pel (ms)
400
450
BRUGADA ET AL TRANSIENT ENTRAINMENT OF AV NODE TACHYCARDIA
772
lACC Vol. 9, No.4 April 19H7:769-7:i
II~
335
A
A
A
A
A
A
A
A
A
A
A
A
A
HRA -~ ~ ..,r------v---v--~..;______ J'--v-J'--v-J---.r-J'--v- J---.r- ..,.r----v-------,~'---v--
in activation of the His bundle from a different direction (previously, it was activated by way of the slow AV node pathway: now it is activated by way of the fast AV node pathway: this change is represented by a change in the direction of the arrow in VI) and with a shorter conduction time (that is. a shorter AH interval). When atrial pacing is discontinued, sinus rhythm resumes (last beat). These observations are further evidence for the presence of two distinct pathways for anterograde AV node conduction, S stimulus artifact: HRA = high right atrium.
Figure 3. Same organization of electrocardiographic and electrographic recordings in the same patient as in Figure I during and immediately after cessation of atrial pacing at a cycle length of 335 ms. The corresponding ventricular cycle lengths (in ms) arc indicated above the surface electrocardiographic lead V/>. During an AV node Wenckebach sequence, localized conduction block occurs in the AV node for one beat (see arrows in lead VI and His recordings) thereby terminating the underlying AV node tachycardia, Continued atrial pacing at a cycle length of 335 ms results
AIII
HRA____..t'
cs---··v 1
.'. I'
.....i~
i
~'
~1
·v
·v
1sd
'"'I"
360
~ i
~
340
t
~
}~
~ Figure 4. Three panels illustrating transient entrainment of AV node tachycardia (A and B) and its termination by ventricular pacing (C). Lead III is shown with three bipolar electrograrns. The original tachycardia cycle length was 340 ms. Pacing from the right ventricular apex with a shorter cycle length results in shortening in the atrial and ventricular cycle lengths to the pacing cycle length. There is a minimal increase in the VA interval from 150 (A) to 160 ms (B), and then to 180 ms (C) when the pacing cycle length is shortened from 335 to 295 to 270 ms, respectively. In A and B, the tachycardia resumes at its original cJ cle length after cessation of pacing. In C, block of conduction in the anterograde slow pathway results in termination of the tachycardia.
BRUGADA ET AL. TRANSIENT ENTRAINMENT OF AV NODE TACHYCA RDIA
lACC Vol. 9. No.4 April 1987:769- 75
there was a trend toward narrower entrainment zones for tachycardias with shorter cycle lengths, the data points were quite scattered, particularly for ventricular pacing. During atrial and ventricular pacing, fusion beats at the atrial or ventricular level were never observed.
V-A (rna)
250
200
.. .... ......
150
250
300
350
Discussion 400
450
pel (rna)
Figure 5. IIlustration of the change in the VA interval during transient entrainment of AV node tachycardia during ventricular pacing in relation to the ventricular pacing cycle length (peL). Six patients were selected for illustrative purposes. Only minimal changes inthe VA interval (maximum 30ms)occurred onchanging the pacing cycle length , indicating a rather flat retrograde conduction curve for the retrograde fast AV node pathway. Data were simi lar for all patients studied with ventricular pacing.
Relation between tach ycardia cycle length and entrainment zone. The entrainment zone (difference between the spontaneous tachycardia cycle length and the pacing cycle iength terminating the tachycardia) was compared with the spontaneous tachycardia cycle length. Results are illustrated for atrial and ventricular pacing in Figure 6. Although
Figure 6. Relation between the spontaneous tachycardia cycle length (Cl.) and the entrainment zone (see text for definition). Closed circles, Entrainment zones during atrial pacing. Open circles, Entrainment zones during ventricularpacing. Double open circles, Patients whose tachycardia was terminated by VA block in the retrograde fast pathway during ventricular pacing. There was a trend for narrower entrainment zones when thespontaneous tachycardia cycle length was shorter, particularly during atrial pacing . Results were, however. scattered. ENTRAINMENT ·zone" (rna )
• ATRIUM o VENTRICLE
130
•
110
70
•
•
o ©
90
0
. 0 0
•
50
• • • ••e • • • • © 0
30 10 0
773
i
..
200
250
300
350
0
• ©•
• 400
TACHYCARDIA Cl (rna)
450
The data in this study I) demonstrate conclusively for the first time that transient entrainment of AV node tachycardia occurs, 2) confirm in a manner not previously shown that reentry involving two functionally dissociated pathways in the AV node is the most likely mechanism of AV node tachycardia, 3) extend our observations of the phenomenon of concealed transient entrainment, and 4) support the notion that there is a true intranodal location of the reentrant circuit of AV node tachycardia. Demonstration of transient entrainment of AV node tachycardia. Transient entrainment of a tachycardia has been defined as an increase in the rate of the tachycardia to a faster pacing rate. with resumption of the intrinsic rate of the tachycardia, on either abrupt cessation of pacing or slowing of the pacing rate below the intrinsic rate of the tachycardia (1-5) . At pacing cycle lengths not resulting in termination of AV node tachycardia, this was observed in all 18 patients during atrial pacing, confirming the observations of Portillo et al. (10), and in all nine patients who were studied with ventricular pacing. However. the conditions, extensively discussed elsewhere (1-5,9), that have to be fulfilled to permit the demonstration of transient entrainment were not shown in the study of Portillo et al. (10), and in our study were shown only in the three patients who demonstrated localized conduction block for one beat followed by activation of the blocked site from a different direction and with a shorter conduction time (the third criterion for the demonstration of transient entrainment). This demonstration of transient entrainment of AV node tachycardia illustrates that the phenomenon of transient entrainment is not limited to a macroreentrant circuit and that. under appropriate conditions, it can be demonstrated for any reentrant circuit. whatever its size. Dual AV node pathways. As summarized recently (8), the presence of two functionally dissociated pathways in the AV node was first suggested by studies in the canine heart ( I 1- 13) and subsequently by studies in humans (14- 17). Conclusions from these studies were based primarily on the response of AV node conduction to spontaneous or induced premature beats, both atrial and ventricular. The present study demonstrating localized conduction block to the His bundle for one beat during atrial pacing followed by activation of the His bundle by subsequent paced beats with a shorter conduction time provides strong new evidence of the presence of two functionally dissociated pathways, one " fast" and one "slow ," in the AV node. This confirms in
774
BRUGADA ET AL. TRANSIENT ENTRAINMENT OF AV NODE TACHYCARDIA
a unique way that reentry involvingat least two functionally dissociated pathways in the AV node is the most likely mechanism of this supraventricular tachycardia. Thus, the data from this study support and expand the generally accepted concept (8) that in patientswith AV node tachycardia of the common variety, a slow AV node pathway is used anterogradely and a fast AV node pathway retrogradely during the arrhythmia. The behavior of tachycardia intervals during atrial and ventricular pacing is explained by the functional properties of these two pathways. The slow AV node pathway has, in the anterograde direction, a relatively long conduction time and a short refractory period. The retrograde properties of the slow pathway are very difficult to study in these patients because, retrogradely, the fast pathway has a short conduction time and a short refractory period. The properties intrinsic to the slow pathway can be thought of as a long conduction time and either a short or a long refractory period, but it may be that the slow pathway does not conduct at all retrogradely in the majority of these patients. Accepting these properties, one can easily understand the different observations made during atrial and ventricular pacing that occurred during transient entrainment of AV node tachycardia. During atrial pacing, the anterograde conduction times over the slow pathway increased markedly and, in all patients, tachycardiawas interrupted by a Wenckebach type of conduction block in the slow pathway. When this block occurred, anterograde conduction over the fast pathway then took place in three patients. This phenomenon has been previously described as "localizedconduction block for one beat followed by I: I conduction to the blocked site from a different direction and with a shorter conduction time" (I ,5,9). The reason for not observing this phenomenon at the time of termination of AV node tachycardia by ventricular pacing was the presence of unidirectional block in the slow pathway, probably caused by a very long retrograde refractory period of the slow pathway. In addition, on the basis of previous work (5) showing that pacing from a site orthodromically distal to the area of slow conduction of a reentrant loop will not allow the demonstration of any of the criteria of transient entrainment, it is not surprising that, in this study, ventricular pacing (that is, pacing orthodromically distal to the area of slow conduction of the AV node tachycardia) failed to show localized conduction block for one beat. Concealed entrainment. Because of the impossibility of recording AV node fusion beats in the electrocardiogram (during transient entrainment, the antidromic wave front of each pacing impulse presumably entered into the AV node fast pathway where it collided with the orthodromic wave front of the previous beat), it was impossible to fulfill either the first criterion for the demonstration of transient entrainment (constant fusion beats during the period of entrainment except for the last captured beat, which is entrained but not
lAce Vol. 9. No.4 April 1987:769-75
fused) or the second criterion (progressive fusion) . Nevertheless, in the absence of the ability to show any criteria for the demonstration of transient entrainment, these observations are consistent with concealed transient entrainment (5) and are well explained by the dual AV node pathway concept. Intranodal location of the reentrant circuit AV node tachycardia. As discussed, fusion beats were not observed in these patients during atrial pacing and were not expected to be presentduring ventricular pacing at the time that transient entrainment was occurring. Although it probably would have been difficult to see atrial fusion beats in the electrocardiogram had they been present, this observation suggests both a true intranodal location of the reentrant circuit and the lack of participation of atrial or ventricular structures in the mechanism of the AV node tachycardia. Thus, these observations completeand are in agreementwith those made by Portillo et at. (10) during their studies of transient entrainment and interruption of AV node tachycardia. Such a reentrant circuit limited to the AV node is probably one of the smallest reentrant circuits naturallyoccurring in the human heart.
References I.
Waldo Al., Maclean WAH, Karp RB, Kouchoukos NT. lames TN. Entrainment and interruption of atrial flutterwith atrial pacing. Studies in man following open heart surgery. Circulation 1977;56:737--45.
2. Maclean WAH, Plump Vl, Waldo AL. Transient entrainment and interruption of ventricular tachycardia. PACE 1981 ;4:358-66 . 3. Waldo Al., Henthorn RW, Plump Vl , MacLean WAH. Demonstration of the mechanism of transient entrainment and interruption of ventricular tachycardia with rapid atrial pacing. 1 Am Coli Cardiol 1984 ;3:422-30.
4. Waldo AL, Plump Vl. Arciniegas lG , et al. Transient entrainment and interruption of the atrioventricular bypass pathway type of paroxysmal atrial tachycardia. A model for understanding and identifying reentrant arrhythmias . Circulation 1983;67:73-83. 5. Okumura K. Henthorn RW, Epstein AE. Plumb VJ, Waldo AL. Further observationson transient entrainment: importanceof pacing site and properties of the components of the reentry circuit. Circu. lation 1985;72:1293-307. 6.
Anderson KP, SwerdlowCD, Mason lW . Entrainment of ventricular tachycardia . Am 1 Cardiol 1984;53:335-40 .
7. Wellens HJJ. Value and limitations of programmed electrical stimulation of the heart in the study and treatment of tachycardia. Circulation 1978;57:845-53. 8.
9.
Brugada P. Wellens HJJ. Electrophysiology. mechanism. diagnosis and treatmentof paroxysmal recurrentatrioventricularnodalreentrant tachycardia. In: Surawicz B, Reddy CP, Prystowsky EN, eds. The Tachycardias. Boston: Martinus Nijhoff, 1984:131-58.
Brugada P. Wellens HJJ. Entrainment as an electrophysiologic phenomenon. 1 Am Coli Cardiol 1984;3:451-4 . 10. Portillo B, Mejias J. Leon-Portillo N. Zaman L, Myerburg RJ, Castellanos A. Entrainment of atrioventricular nodal reentrant tachycardias during overdrive pacing from high right atrium and coronary sinus. With special reference to atrioventricular dissociation and 2: I retrograde block during tachycardias. Am1 Cardiol 1984;53: 1570-6. II . Moe GK. Preston JB. Burlington H. Physiologic evidence for dual AV transmission system. Circ Res 1956;4:357-75.
lACC Vol. 9, No.4 April 1987:769-75
12. Rosenblueth A. Ventricular 'echoes.' AmJ Physiol 1958; 195:53-60. 13. MendezC, Han1, Garciade Jalon PO, MoeGK. Somecharacteristics of ventricular echoes. Circ Res 1965; 16:562-81. 14. Schuilenburg RM, Durrer D. Ventricular echo beats in the human heart elicited by induced ventricular premature beats. Circulation 1969;40:337-47. 15. Goldreyer BN, Damato AN. The essential role of atrioventricular
BRUGADA ET AL. TRANSIENT ENTRAINMENT OF AV NODE TACHYCARDIA
775
conduction delay in the initiation of paroxysmal supraventricular tachycardia. Circulation 1971 ;43:679-87. 16. Denes P, Wu 0, Dhingra RC, Chuquimia R, Rosen KM. Demonstration of dual AV nodal pathways in patients with paroxysmal supraventricular tachycardia. Circulation 1973;48:549-55. 17. Rosen KM, Mehta A, Miller RA. Demonstration of dual atrioventricular nodal pathways in man. Am J Cardiol 1974;33:291-4.
JACC Vol. lJ. No 4 April 19K7:76lJ- 7)
Transient Entrainment and Interruption of Atrioventricular Node Tachycardia PEDRO BRUGADA, MD, ALBERT L. WALDO, MD. FACC,* HEIN J.J . WELLENS, MD, FACC Maastricht . The Netherlands
The possibility of transiently entraining and interrupting the common type of atrioventricular (AV) node tachycardia (anterograde slow, retrograde fast AV node pathway) was studied using atrial and ventricular pacing in 18 patients with paroxysmal AV node tachycardia. Transient entrainment occurred in all patients. During atrial pacing, localized block in the AV node for one beat followed by anterograde conduction over the fast pathway was obser ved in three patients. During ventr icular pacing , localized block for one beat followed by retrograde conduction over the slow pathway was not observed in any patient. Neither atrial nor ventricular fusion beats were obser ved during entrainment.
These observations indicate in a way not previously shown that reentry involving two functionally dissociated pathw ays in the AV node is the underlying mechanism of paroxysmal AV node tachycardia. The inability to demonstrate atrial or ventricular fusion beats during entrainment suggests a true intranodal location of the reentrant circuit. Finally, the abilit y to transiently entr ain intranodal tach ycardia demon strates that this electrophysiologic phenomenon is not exclusively limited to macr oreentrant circuits. o Am Coli Cardiol /9 87;9:769-75)
Transient entrainment and interrupt ion of tachycardias has now been studied in the human heart during many spontaneous and induced arrhythmias. The observations made during transient entrainment of atrial nutter ( I), ventricular tachycardia (2.3) and circus movement tachycardia using an accessory pathway (4,5) have increased understanding of the phenomenon of transient entrainment and have led to the suggestion that tachycardias that can be transiently entrained are the result of reentry with an excitable gap. It has recently been suggested that the demonstration of transient entrainment of a tachycardia not only suggests reentry. but also strongly suggests macrorcentry as the underlying mechanism (6). In some patients with paroxysmal supraventricular tachycardia. the presence and use of an accessory pathway during the arrhythmia can be excluded by detailed studies during
programmed electrical stimulation of the heart (7) . As summarized recently (8). there is strong evidence that in some of these patients. neither the atria nor the ventricles participate in the mechanism of the tachycardia, suggesting a true intranodal location of the reentrant circuit. This reentrant circuit is probably one of the smallest that occurs naturally in the human heart. We studied the possibility of transiently entraining the tachycardia in patients with atrioventricular (AV) node tachycardia, a putatively reentrant rhythm (8) to 1) further investigate and characterize entrainment and interruption of reentrant rhythms; 2) examine and emphasize the suggested criteria for the demonstration of entrainment; 3) dissect and characterize the components of the reentrant circuit of AV node tachycardia; and 4) further examine and provide insights into dual AV node pathways.
From the Unive rsity Hospital. Depa rtment of Cardio logy . Maast ric ht. The Net herlands . Th is work was suppo rted in part by G rant IROI HL293XI from the Natio na l Heart . Lung . and Blood Institute . Natio nal Institutes of Health . Bethesda, Mary land . *Prese nt address: Division of Ca rdio logy . Case Wes tern Reserve Universi ty . Unive rsity Hosp itals of Cleve land . Cleveland. Ohio 44106. Manuscript receiv ed Jun e 2. 1986: revised man uscript rece ived Se ptember 23. 1986. acce pted October 10 . 1986 . Address for rep rint s: Ped ro Bru gada . MD. Cli nica l Elect rophys iology Laboratory. University Hospital. Depart men t of Ca rdiolog y . Maa strich t. Th e Neth erland s.
Methods
I!) 1987
by the American College of Cardiology
Study patients. Eighteen patients with paroxysmal supraventricular tachycardia were studied . There were 10 women and 8 men whose ages ranged from 25 to 64 years (mean 44). After informed consent was obtained. the patients underwent programmed electrical stimulation of the heart in the absence of any antiarrhythmic drugs . Programmed stimulation was performed using our previously 07.\)-1 O(!7 /87/$\ .50
770
BRUGADA ET AL. TRANSIENT ENTRAINMENT OF AV NODE TACHYCARDIA
described methodsof stimulation, recording and analysisof tracings (7). Electrode catheters were placed at the right atrium, right ventricularapex, His bundle area and coronary sinus. Once the modesof initiationand termination of tachycardia and the site of origin of the arrhythmia had been identified, the possibility of transiently entraining and terminating the tachycardia by rapid atrial and ventricular pacing was analyzed. Criteria for AV node tachycardia. The "common" (anterograde slow, retrograde fast) form of AV node tachycardia was diagnosed in all patients by using the following criteria (8). I. Initiation of supraventricular tachycardia by atrial premature beats or atrial pacing only after conduction over the AV node at a critical delay in AV node conduction time, and after a discontinuity in the anterograde AV node conduction curve. 2. The temporal occurrence of both the first atrial echo beat and subsequent atrial activation during the tachycardia after AV node and His bundle activation, but preceding or simultaneous with ventricular activation. 3. A similar atrial activation pattern during retrograde conduction over the fast AV node pathway during ventricular pacing and during tachycardia. 4. A His to atrium (HA) interval during tachycardia independent of the His to ventricle (HV) interval and bundle branch block pattern. 5. Inabilityto advanceatrial activationduring tachycardia by a ventricular premature beat introduced at the time of refractoriness of the bundle of His. 6. No evidence for anterograde or retrograde conduction over an accessory pathway. Criteria for transient entrainment. Previously reported criteria were used to establish the presence of transient entrainment (1-5,9). These criteria were, in summary: I) constant fusion beats during the period of entrainment except for the last captured beat, 2) progressive fusion at different entraining rates, and 3) localized block for one beat followed by antidromic activation. Concealed entrainment was defined as previously reported (5). To study transient entrainment, trains of 20 stimuli were introduced at selected rates from the right atrium (proximal to the area of slow conduction) and subsequently from the right ventricle (distal to the area of slow conduction). In three patients, stimuli were also introduced from the proximal coronary sinus. The firsttrain of stimuli waseither 10beats/min faster or had a 10 ms shorter cycle length than the tachycardia rate. The pacing rate was increased by 5 beat/min steps for the subsequent trains until the tachycardia was terminated.
Results The spontaneous tachycardia cycle length at rest ranged from 240 to 470 ms (mean 345). The mean pacing cycle
lACC Vol. 'I. No.4 April 1987:76'1- 75
length that terminated the tachycardia from the right atrium was 298 ms (range 220 to 410). There was a mean zone of 48 ms (range 15 to 110) in which the tachycardia was not terminated by rapid pacing and in which transient entrainment could be studied. In the nine patients in whom rapid ventricular pacing was performed, the pacing cycle length that terminated the
tachycardia ranged from 210 to 410 ms (mean 300). There was a mean zone of 58 ms (range 10 to 95) in which the tachycardia was not terminated by rapid ventricular pacing, and in which transient entrainment could be studied. Effects of atrial pacing at cycle lengths not terminating AV node tachycardia. During atrial pacing at cycle lengths not resulting in termination of AV node tachycardia, the following observations were made (Fig. I). I) The atrial and ventricular rates increased to the pacing rate, associated with a marked increase in AH interval in
all instances. The changes in the AH interval at different cycle lengths not terminating the tachycardia are illustrated for eight patients in Figure 2, although data were similar for all 18 patients. 2) Ventricular activation after the last paced atrial beat occurred at the pacing cycle length , and was thereafter
immediately followed by resumption of the original tachycardia cycle length. Atrial activation after the last paced atrial beat always occurred after an interval that was longer than the pacing or tachycardia cycle length. 3) No atrial or ventricularfusion beats were seen. Therefore, although it seems certain that transient entrainment had occurred (concealed entrainment), it could not be established in the absence of fusion beats (3-5). Effects of atrial pacing at cycle lengths resulting in termination of AV node tachycardia. In all 18 patients, the pacingcycle length resulting in termination of the tachycardia was associated with a Wenckebach conduction pattern in the anterograde slow pathway. When conduction block occurred in this pathway, the tachycardia was terminated. In 15 patients, continuation of pacing at the same cycle length resulted in a repeated Wenckebach conduction pattern in the slow pathway, so that if cessation of pacing did not occur with a blocked beat, but rather occurred after a conducted beat, the tachycardia was reinitiated (4,5). In three patients. continuation of pacing after termination of tachycardia resulted in immediate I: I anterograde AV node conduction (Fig. 3). The latter represents conduction block to the His bundle for one beat. followed by activation of the His bundle from a different direction (by way of the fast pathway) and with a shorter conduction time (that is. a shorter AH interval). For the other 15 patients, this did not happen because, presumably, the anterograde effective refractory period of the fast pathway was longer than the pacing cycle length. Effects of ventricular pacing at cycle lengths not terminating AV node tachycardia. During ventricularpacing
Figure I. Five electrocardiographic leads are simultaneously shown with bipolar electrograms from the right atrium (RA).distal and proximal coronary sinus (CSd, CSp) and a His bundle recording during a representative example of transient entrainment of atrioventricular (AV) node tachycardia during atrial pacing. The spontaneous tachycardia cycle length was 355 ms. Pacing is performed from the right atrium at a cycle length of 345 ms. The atrial and ventricular cyclelengths shorten to the pacing cycle length. Ventricular activation after the last paced atrial complex occurs at the pacing cycle length (345 ms) and is followed by immediate resumption of the original tachycardia cycle length. The interval between the last paced atrial complex and the next atrial activation during tachycardia is 430 rns, longer than the pacing cycle length and the spontaneous tachycardia cycle length. The AH interval during entrainment prolongs to 335 ms, as compared with an AH interval of 300 ms during tachycardia. In this and all subsequent fi gures. all values are in milliseconds.
771
BRUGADA ET AL. TRANSIENT ENTRAINMENT OF AV NODE TAC HYCA RDIA
l ACC Vol. 9. No, 4 April 1987:769-75
II
A
A
A
345
430
A
355
A
355
A
RA_~~~V"""'---'V'---I/'
j
HI5-L.M.'
I
;
H '! I i
AH.I 335 CS~
j
H ': '
H :
300
I
I
1I ~~
Hi ',. ,~I.-.----....,J'lI"f __ H 'j
n
I
---Ilt------ '----I~'---.....Jf'-
~,....-~'\Yr-'- - 1 -.....JIMr-·
at cycle lengths not terminating AV node tachycard ia, the following observations were made (Fig . 4) . I ) The atrial and ventricular rates increased 10 the pacing rates with minimal changes in the ventricle to atrium (VA) interval when different pacing rates were compared. Changes in the VA interval at different pacing cycle lengths are illustrated for six patients in Figure 5. 2) Atrial activation after the last paced ventricular beat occurred at the pacing cycle length, and was thereafter immed iately followed by resumption of the original tachycardia cycle length. Ventricular activation after the last paced ventricular beat always occurred after an inter val that was longer than the pacing or tachycardia cycle length (Fig. 4) . 3) No atrial or ventricular f usion beats were seen. Once again, although it seems certain that transient entrainment had occurred (concealed entrainment), it could not be established in the absence of fusion beats (4,5) . Effects of ventricular pacing at cycle lengths terminating AV node tachycardia. During ventricular pacing. the tachycardia was termin ated in four patients by block in VA conduction (block of the antidromic wave front of the pacing impulse in the slow pathway and block of the orthodromic wave front of the pacing impulse in the fast pathw ay) and in the remain ing five patients by both retrograde and anterograde block in the slow pathway . Contin uation of ventricular pacing at the same cycle length once the tachycardia had been terminated was never associated with retrograde conduction over the slow pathway.
A >/'-------...tv-
\
-/r--~ ,(Iv-r-- ............. , Ir--- ~ ,"{"I""----.,..Hi........ ~,
Ii" ',
A
~
I
I
Figure 2. Changes in AH interval during entrainment of atrioventricular (AV) node tachycardia are shown plotted against the pacing cycle length (PCl) . Data from 8 of the 18 patients were selected for illustrative purposes. The AV node conduction times over the slow pathway increased markedly with minorchanges in pacing cycle length (filled circles), demonstrating very steep AV node conduction curves of the anterograde slow pathway. Open circles represent the AH interval during the spontaneous tachycardia. A-H (ms) 450
•
400 III
o
.: 350
300
~ :
•
... ..
••
;" -~bo
o
:
:~--b6 250
'0 0
300
350
pel (ms)
400
450
BRUGADA ET AL TRANSIENT ENTRAINMENT OF AV NODE TACHYCARDIA
772
lACC Vol. 9, No.4 April 19H7:769-7:i
II~
335
A
A
A
A
A
A
A
A
A
A
A
A
A
HRA -~ ~ ..,r------v---v--~..;______ J'--v-J'--v-J---.r-J'--v- J---.r- ..,.r----v-------,~'---v--
in activation of the His bundle from a different direction (previously, it was activated by way of the slow AV node pathway: now it is activated by way of the fast AV node pathway: this change is represented by a change in the direction of the arrow in VI) and with a shorter conduction time (that is. a shorter AH interval). When atrial pacing is discontinued, sinus rhythm resumes (last beat). These observations are further evidence for the presence of two distinct pathways for anterograde AV node conduction, S stimulus artifact: HRA = high right atrium.
Figure 3. Same organization of electrocardiographic and electrographic recordings in the same patient as in Figure I during and immediately after cessation of atrial pacing at a cycle length of 335 ms. The corresponding ventricular cycle lengths (in ms) arc indicated above the surface electrocardiographic lead V/>. During an AV node Wenckebach sequence, localized conduction block occurs in the AV node for one beat (see arrows in lead VI and His recordings) thereby terminating the underlying AV node tachycardia, Continued atrial pacing at a cycle length of 335 ms results
AIII
HRA____..t'
cs---··v 1
.'. I'
.....i~
i
~'
~1
·v
·v
1sd
'"'I"
360
~ i
~
340
t
~
}~
~ Figure 4. Three panels illustrating transient entrainment of AV node tachycardia (A and B) and its termination by ventricular pacing (C). Lead III is shown with three bipolar electrograrns. The original tachycardia cycle length was 340 ms. Pacing from the right ventricular apex with a shorter cycle length results in shortening in the atrial and ventricular cycle lengths to the pacing cycle length. There is a minimal increase in the VA interval from 150 (A) to 160 ms (B), and then to 180 ms (C) when the pacing cycle length is shortened from 335 to 295 to 270 ms, respectively. In A and B, the tachycardia resumes at its original cJ cle length after cessation of pacing. In C, block of conduction in the anterograde slow pathway results in termination of the tachycardia.
BRUGADA ET AL. TRANSIENT ENTRAINMENT OF AV NODE TACHYCA RDIA
lACC Vol. 9. No.4 April 1987:769- 75
there was a trend toward narrower entrainment zones for tachycardias with shorter cycle lengths, the data points were quite scattered, particularly for ventricular pacing. During atrial and ventricular pacing, fusion beats at the atrial or ventricular level were never observed.
V-A (rna)
250
200
.. .... ......
150
250
300
350
Discussion 400
450
pel (rna)
Figure 5. IIlustration of the change in the VA interval during transient entrainment of AV node tachycardia during ventricular pacing in relation to the ventricular pacing cycle length (peL). Six patients were selected for illustrative purposes. Only minimal changes inthe VA interval (maximum 30ms)occurred onchanging the pacing cycle length , indicating a rather flat retrograde conduction curve for the retrograde fast AV node pathway. Data were simi lar for all patients studied with ventricular pacing.
Relation between tach ycardia cycle length and entrainment zone. The entrainment zone (difference between the spontaneous tachycardia cycle length and the pacing cycle iength terminating the tachycardia) was compared with the spontaneous tachycardia cycle length. Results are illustrated for atrial and ventricular pacing in Figure 6. Although
Figure 6. Relation between the spontaneous tachycardia cycle length (Cl.) and the entrainment zone (see text for definition). Closed circles, Entrainment zones during atrial pacing. Open circles, Entrainment zones during ventricularpacing. Double open circles, Patients whose tachycardia was terminated by VA block in the retrograde fast pathway during ventricular pacing. There was a trend for narrower entrainment zones when thespontaneous tachycardia cycle length was shorter, particularly during atrial pacing . Results were, however. scattered. ENTRAINMENT ·zone" (rna )
• ATRIUM o VENTRICLE
130
•
110
70
•
•
o ©
90
0
. 0 0
•
50
• • • ••e • • • • © 0
30 10 0
773
i
..
200
250
300
350
0
• ©•
• 400
TACHYCARDIA Cl (rna)
450
The data in this study I) demonstrate conclusively for the first time that transient entrainment of AV node tachycardia occurs, 2) confirm in a manner not previously shown that reentry involving two functionally dissociated pathways in the AV node is the most likely mechanism of AV node tachycardia, 3) extend our observations of the phenomenon of concealed transient entrainment, and 4) support the notion that there is a true intranodal location of the reentrant circuit of AV node tachycardia. Demonstration of transient entrainment of AV node tachycardia. Transient entrainment of a tachycardia has been defined as an increase in the rate of the tachycardia to a faster pacing rate. with resumption of the intrinsic rate of the tachycardia, on either abrupt cessation of pacing or slowing of the pacing rate below the intrinsic rate of the tachycardia (1-5) . At pacing cycle lengths not resulting in termination of AV node tachycardia, this was observed in all 18 patients during atrial pacing, confirming the observations of Portillo et al. (10), and in all nine patients who were studied with ventricular pacing. However. the conditions, extensively discussed elsewhere (1-5,9), that have to be fulfilled to permit the demonstration of transient entrainment were not shown in the study of Portillo et al. (10), and in our study were shown only in the three patients who demonstrated localized conduction block for one beat followed by activation of the blocked site from a different direction and with a shorter conduction time (the third criterion for the demonstration of transient entrainment). This demonstration of transient entrainment of AV node tachycardia illustrates that the phenomenon of transient entrainment is not limited to a macroreentrant circuit and that. under appropriate conditions, it can be demonstrated for any reentrant circuit. whatever its size. Dual AV node pathways. As summarized recently (8), the presence of two functionally dissociated pathways in the AV node was first suggested by studies in the canine heart ( I 1- 13) and subsequently by studies in humans (14- 17). Conclusions from these studies were based primarily on the response of AV node conduction to spontaneous or induced premature beats, both atrial and ventricular. The present study demonstrating localized conduction block to the His bundle for one beat during atrial pacing followed by activation of the His bundle by subsequent paced beats with a shorter conduction time provides strong new evidence of the presence of two functionally dissociated pathways, one " fast" and one "slow ," in the AV node. This confirms in
774
BRUGADA ET AL. TRANSIENT ENTRAINMENT OF AV NODE TACHYCARDIA
a unique way that reentry involvingat least two functionally dissociated pathways in the AV node is the most likely mechanism of this supraventricular tachycardia. Thus, the data from this study support and expand the generally accepted concept (8) that in patientswith AV node tachycardia of the common variety, a slow AV node pathway is used anterogradely and a fast AV node pathway retrogradely during the arrhythmia. The behavior of tachycardia intervals during atrial and ventricular pacing is explained by the functional properties of these two pathways. The slow AV node pathway has, in the anterograde direction, a relatively long conduction time and a short refractory period. The retrograde properties of the slow pathway are very difficult to study in these patients because, retrogradely, the fast pathway has a short conduction time and a short refractory period. The properties intrinsic to the slow pathway can be thought of as a long conduction time and either a short or a long refractory period, but it may be that the slow pathway does not conduct at all retrogradely in the majority of these patients. Accepting these properties, one can easily understand the different observations made during atrial and ventricular pacing that occurred during transient entrainment of AV node tachycardia. During atrial pacing, the anterograde conduction times over the slow pathway increased markedly and, in all patients, tachycardiawas interrupted by a Wenckebach type of conduction block in the slow pathway. When this block occurred, anterograde conduction over the fast pathway then took place in three patients. This phenomenon has been previously described as "localizedconduction block for one beat followed by I: I conduction to the blocked site from a different direction and with a shorter conduction time" (I ,5,9). The reason for not observing this phenomenon at the time of termination of AV node tachycardia by ventricular pacing was the presence of unidirectional block in the slow pathway, probably caused by a very long retrograde refractory period of the slow pathway. In addition, on the basis of previous work (5) showing that pacing from a site orthodromically distal to the area of slow conduction of a reentrant loop will not allow the demonstration of any of the criteria of transient entrainment, it is not surprising that, in this study, ventricular pacing (that is, pacing orthodromically distal to the area of slow conduction of the AV node tachycardia) failed to show localized conduction block for one beat. Concealed entrainment. Because of the impossibility of recording AV node fusion beats in the electrocardiogram (during transient entrainment, the antidromic wave front of each pacing impulse presumably entered into the AV node fast pathway where it collided with the orthodromic wave front of the previous beat), it was impossible to fulfill either the first criterion for the demonstration of transient entrainment (constant fusion beats during the period of entrainment except for the last captured beat, which is entrained but not
lAce Vol. 9. No.4 April 1987:769-75
fused) or the second criterion (progressive fusion) . Nevertheless, in the absence of the ability to show any criteria for the demonstration of transient entrainment, these observations are consistent with concealed transient entrainment (5) and are well explained by the dual AV node pathway concept. Intranodal location of the reentrant circuit AV node tachycardia. As discussed, fusion beats were not observed in these patients during atrial pacing and were not expected to be presentduring ventricular pacing at the time that transient entrainment was occurring. Although it probably would have been difficult to see atrial fusion beats in the electrocardiogram had they been present, this observation suggests both a true intranodal location of the reentrant circuit and the lack of participation of atrial or ventricular structures in the mechanism of the AV node tachycardia. Thus, these observations completeand are in agreementwith those made by Portillo et at. (10) during their studies of transient entrainment and interruption of AV node tachycardia. Such a reentrant circuit limited to the AV node is probably one of the smallest reentrant circuits naturallyoccurring in the human heart.
References I.
Waldo Al., Maclean WAH, Karp RB, Kouchoukos NT. lames TN. Entrainment and interruption of atrial flutterwith atrial pacing. Studies in man following open heart surgery. Circulation 1977;56:737--45.
2. Maclean WAH, Plump Vl, Waldo AL. Transient entrainment and interruption of ventricular tachycardia. PACE 1981 ;4:358-66 . 3. Waldo Al., Henthorn RW, Plump Vl , MacLean WAH. Demonstration of the mechanism of transient entrainment and interruption of ventricular tachycardia with rapid atrial pacing. 1 Am Coli Cardiol 1984 ;3:422-30.
4. Waldo AL, Plump Vl. Arciniegas lG , et al. Transient entrainment and interruption of the atrioventricular bypass pathway type of paroxysmal atrial tachycardia. A model for understanding and identifying reentrant arrhythmias . Circulation 1983;67:73-83. 5. Okumura K. Henthorn RW, Epstein AE. Plumb VJ, Waldo AL. Further observationson transient entrainment: importanceof pacing site and properties of the components of the reentry circuit. Circu. lation 1985;72:1293-307. 6.
Anderson KP, SwerdlowCD, Mason lW . Entrainment of ventricular tachycardia . Am 1 Cardiol 1984;53:335-40 .
7. Wellens HJJ. Value and limitations of programmed electrical stimulation of the heart in the study and treatment of tachycardia. Circulation 1978;57:845-53. 8.
9.
Brugada P. Wellens HJJ. Electrophysiology. mechanism. diagnosis and treatmentof paroxysmal recurrentatrioventricularnodalreentrant tachycardia. In: Surawicz B, Reddy CP, Prystowsky EN, eds. The Tachycardias. Boston: Martinus Nijhoff, 1984:131-58.
Brugada P. Wellens HJJ. Entrainment as an electrophysiologic phenomenon. 1 Am Coli Cardiol 1984;3:451-4 . 10. Portillo B, Mejias J. Leon-Portillo N. Zaman L, Myerburg RJ, Castellanos A. Entrainment of atrioventricular nodal reentrant tachycardias during overdrive pacing from high right atrium and coronary sinus. With special reference to atrioventricular dissociation and 2: I retrograde block during tachycardias. Am1 Cardiol 1984;53: 1570-6. II . Moe GK. Preston JB. Burlington H. Physiologic evidence for dual AV transmission system. Circ Res 1956;4:357-75.
lACC Vol. 9, No.4 April 1987:769-75
12. Rosenblueth A. Ventricular 'echoes.' AmJ Physiol 1958; 195:53-60. 13. MendezC, Han1, Garciade Jalon PO, MoeGK. Somecharacteristics of ventricular echoes. Circ Res 1965; 16:562-81. 14. Schuilenburg RM, Durrer D. Ventricular echo beats in the human heart elicited by induced ventricular premature beats. Circulation 1969;40:337-47. 15. Goldreyer BN, Damato AN. The essential role of atrioventricular
BRUGADA ET AL. TRANSIENT ENTRAINMENT OF AV NODE TACHYCARDIA
775
conduction delay in the initiation of paroxysmal supraventricular tachycardia. Circulation 1971 ;43:679-87. 16. Denes P, Wu 0, Dhingra RC, Chuquimia R, Rosen KM. Demonstration of dual AV nodal pathways in patients with paroxysmal supraventricular tachycardia. Circulation 1973;48:549-55. 17. Rosen KM, Mehta A, Miller RA. Demonstration of dual atrioventricular nodal pathways in man. Am J Cardiol 1974;33:291-4.