Author + information
- Received November 14, 2001
- Revision received March 26, 2002
- Accepted May 20, 2002
- Published online August 21, 2002.
- Frank Bode, MD*,†,
- Pamela Karasik, MD, FACC*,
- Hugo A Katus, MD, FESC† and
- Michael R Franz, MD, PhD, FACC*,* ()
- ↵*Reprint requests and correspondence:
Dr. Michael R. Franz, Cardiology Division, VAMC, 50 Irving Street, NW, Washington, DC 20422, USA.
Objective The purpose of this study was to test the hypothesis that a ventricular tachycardia (VT) induction site has a shorter action potential duration (APD) and effective refractory period (ERP) than a noninducing site, resulting in collision against longer ERP (“upstream”) as opposed to shorter ERP (“downstream,” no collision).
Background Induction of sustained VT is often feasible at one stimulation site while application of an identical pacing protocol to another site fails to provoke VT.
Methods Sixty-nine patients undergoing programmed stimulation for VT inducibility had monophasic action potential recording/pacing catheters placed in the right ventricular outflow tract (RVOT) and right ventricular apex (RVA) simultaneously. Up to three extra-stimuli were introduced in 5 to 10 ms decrements until ERP was reached. Upon completion of a drive cycle at one stimulation site, it was repeated at the other.
Results Thirty-eight patients had inducible VT, nine exclusively by RVA pacing and nine exclusively by RVOT pacing. Action potential duration and ERP at the induction site were significantly shorter (12 ± 15 ms, p <0.05 and 22 ± 14 ms, p < 0.01, respectively, at 600 ms basic cycle length) than at the noninduction site. Dispersion of repolarization between corresponding APD at the two sites was 58 ± 41 ms during baseline stimulation (S1) at the inducing site but only 37 ± 23 ms at the noninducing site (p < 0.05). Dispersion increased during extra-stimulus application (p < 0.05), reaching a maximum of 75 ± 45 ms during VT induction, but only 53 ± 33 ms during extra-stimulation at the noninduction site.
Conclusions Site specificity of VT induction underscores the role of dispersion of repolarization and refractoriness in facilitating re-entry arrhythmias. Upstream stimulation at a site with short repolarization produces larger dispersion and facilitates VT induction.
Initiation of ventricular tachycardia (VT) in a clinical setting is usually accomplished by right ventricular pacing, introducing up to three premature extra-stimuli (1–5). The introduction of premature impulses onto the preceding repolarization phase has been shown to shorten action potential duration (APD) and to result in local activation from less repolarized membrane potentials, leading to impaired impulse propagation (6). Multisite monophasic action potential (MAP) recordings in experimental protocols have also emphasized the importance of dispersion of ventricular repolarization for the onset of ventricular tachyarrhythmia (7–9). Simultaneous MAP recordings from the human ventricle showed an increased dispersion of repolarization during extra-stimulus application (10). It is still unclear why programmed electrical stimulation at one right ventricular site might fail to induce VT, whereas application of the same stimulation protocol to another right ventricular site might induce the arrhythmia. The present study sought to determine by MAP recordings which factors contribute to the site-specificity during programmed stimulation.
Sixty-nine consecutive male patients underwent electrophysiologic study for evaluation of documented or clinically suspected ventricular tachyarrhythmia. After approval by the Institutional Review Board, written informed consent to the protocol was obtained from all patients. Antiarrhythmic drugs had been withdrawn for at least five half-lives. Two MAP recording/pacing catheters (EP Technologies Inc., San Jose, California) were placed in the right ventricular apex (RVA) and right ventricular outflow tract (RVOT) through a femoral vein. These catheters allowed us to pace the heart and to simultaneously record MAPs from the stimulation site, as described previously (11). Monophasic action potential and a 12-lead surface electrocardiogram were stored on a Bard electrophysiology recording system (Lowell, Massachusetts). Programmed electrical stimulation was performed with 2 ms pulse duration at twice diastolic threshold strength by a Bloom Electrophysiology (Fischer Imaging, Denver, Colorado) stimulator. Hearts were paced at a basic cycle length (BCL) of 600 ms (S1-S1), starting randomly at one of the sites. Following eight S1-beats, a premature extra-stimulus (S2) was introduced after completion of the last S1-repolarization, and a pause of 2 s duration was allowed before the next drive cycle was started. The S1-S2 coupling interval was shortened in 10 ms decrements until it began to encroach onto the S1-repolarization phase, followed by 5 ms decrements until the effective refractory period (ERP) was reached. ERP was defined as the largest coupling interval that failed to induce a propagated ventricular response. The S2 was then held at a coupling interval of 30 ms above the ERP, and further extra-stimuli were added. A second extra-stimulus (S3) was introduced with an initial coupling interval 50 ms longer than the preceding coupling interval, followed by 10 ms and 5 ms decrements, respectively. Both the S1-S2 and the S2-S3 coupling intervals were decreased until the shortest intervals were reached that still resulted in capture. Introduction of a third extra-stimulus (S4) was performed in the same fashion. The stimulation protocol was repeated on the other right ventricular site, before each additional extra-stimulus was introduced. When stimulation at 600 ms BCL failed to induce a sustained ventricular arrhythmia (duration ≥15s), an identical protocol was applied at 400 ms BCL. Once VT was induced, the study was ended. The end point of the study was reached when a stimulation sequence had been completed at the initial pacing site and no sustained arrhythmia occurred, while at the other pacing site a corresponding stimulation sequence induced sustained VT. According to standard clinical practice, reinduction from the other site was not attempted in patients with tachycardia already induced at the initial pacing site. This also ruled out an influence of intermittent high rates and cardioversion on electrophysiologic properties thereafter. Catheters were required to remain in the same positions throughout the study as verified by stable, high fidelity MAP recordings.
Simultaneously recorded MAPs from both ventricular sites were evaluated at BCL (S1) and upon introduction of one to three extra-stimuli (S2-S4). The MAP durations (APDs) were measured from the upstroke in phase 0 to the 90% repolarization level (Fig. 1A). When a premature action potential fell within the repolarization phase of the preceding action potential, the slope of phase 3 repolarization was extended to the 90% repolarization level to determine APD. During extra-stimulus pacing, cumulative APD of a drive train was determined from the upstroke of the final S1 MAP to the repolarization of the last extra-stimulus MAP. The site-specific response to stimulation was evaluated at two particular pacing sequences. The VT-inducing train was compared to an activation sequence with the same BCL and same number of extra-stimuli applied to the remote, noninducing site. Here, the sequence with the closest coupling intervals still resulting in capture was analyzed (Fig. 1B). The repolarization level at which each extra-stimulus was imposed onto the preceding action potential was determined. Conduction time between stimulation site and the remote recording site was defined as the delay between the upstrokes of both simultaneously recorded MAPs. Intersite dispersion of repolarization was determined as the time difference between corresponding action potentials of the two distant sites at 90% repolarization level (Fig. 1). Dispersion of repolarization resulted from APD differences plus the activation delay to the remote site, that is conduction time. Ventricular tachycardia was required to last for at least 15 s to be sustained.
Data are presented as mean value ± SD. Intersite differences for APD, ERP, conduction time, repolarization dispersion and repolarization level were statistically evaluated by two-way repeated-measures analysis of variance with interactions, followed by post hoc analysis via Bonferroni adjustment. The unit of analysis was the patient. Values of p < 0.05 were defined as statistically significant.
Thirty-eight patients had inducible VT. In 20 patients VT was induced from the first attempted stimulation site and the study was ended. Eighteen patients had VT induced from the second stimulation site after a corresponding stimulation sequence had already been applied unsuccessfully to the first attempted site, thereby fulfilling our inclusion criteria. Underlying heart disease was coronary artery disease (n = 14) and idiopathic dilated cardiomyopathy (n = 4). Ejection fraction was 39 ± 10%. Monomorphic VT was elicited in 13 subjects from a stimulation site in the RVA (n = 6) or RVOT (n = 7). VT cycle length averaged 301 ± 82 ms. In five patients, polymorphic VT was induced by RVA (n = 3) or RVOT pacing (n = 2) and VT cycle length was 204 ± 15 ms. Successful pacing sequences included one (n = 3), two (n = 6) and three extra-stimuli (n = 9) at drive train cycle length of 600 ms (n = 10) and 400 ms (n = 8). Arrhythmias were terminated by overdrive pacing (n = 10) or direct current shock (n = 8).
Action potential duration at the inducing site was significantly shorter than APD at the noninducing site. Figure 2 illustrates the APD difference between the two sites during stimulation at BCLs of 600 ms and 400 ms. Premature extra-stimuli (S2-S4) could be introduced at progressively shorter coupling intervals, resulting in progressively shorter APD. A significant difference in APD between inducing and noninducing site persisted with each extra-stimulus introduced. Durations of corresponding premature action potentials were consistently shorter at the inducing site (Fig. 3). The difference was reflected by the cumulative APD of the drive trains, measured from the upstroke of the last S1 response to the repolarization of the last extra-stimulus. Upon VT induction, cumulative APD at the inducing site was 31 ± 47 ms (p < 0.05) less than the cumulative APD at the noninducing site during closest coupled extra-stimulation.
The ERP was closely correlated to APD (r = 0.83; p < 0.01). Accordingly, ERP at the inducing stimulation site was shorter than the ERP at the noninducing site (Fig. 2). The difference in refractoriness influenced the prematurity at which extra-stimuli could be introduced. During VT induction, S2 was introduced at a S1-S2 coupling interval 22 ± 16 ms beyond the local ERP. This was explained by the fact that 12 patients had VT induced even prior to maximal shortening of the S1-S2 interval. At the noninducing site a corresponding S1-S2 coupling interval approached the local ERP by 1 ± 14 ms.
Conduction times between the two ventricular sites were independent of the site of stimulation. During baseline (S1) stimulation, impulses required 55 ± 24 ms to propagate from the inducing site to the remote site and 57 ± 23 ms in the opposite direction (p = NS). Premature extra-stimuli resulted in an increased conduction time between the two sites (Fig. 4). The last extra-stimulus introduced during VT induction required an additional 15 ± 19 ms (22 ± 24%; p < 0.01) to propagate from the stimulation site to the remote recording site. A corresponding extra-stimulus introduced to the noninducing site resulted in delayed conduction to the inducing site by 11 ± 4 ms (20 ± 25%; p < 0.01) that was not significantly different from that in the opposite direction. Likewise, the conduction delay produced by an equivalent number of additional extra-stimuli did not differ between the two propagation pathways (Fig. 4).
Dispersion of repolarization
Dispersion of repolarization—defined as the APD difference between corresponding action potentials at the two ventricular sites plus conduction time—was 58 ± 41 ms during baseline (S1) stimulation at the inducing site but only 37 ± 23 ms during pacing at the noninducing site (p < 0.05; Fig.5). Extra-stimulus application during VT induction increased the dispersion of repolarization by 17 ± 24 ms at the inducing site (p < 0.01). At the noninducing site, dispersion also increased during extra-stimulation (16 ± 21 ms; p < 0.01). Yet, maximum dispersion of 53 ± 33 ms at the noninducing site only approached the amount of intersite dispersion that was present at the inducing site even during baseline (S1) stimulation. At the inducing site, extra-stimulation produced a markedly larger dispersion (75 ± 45 ms during VT induction; p < 0.05).
With progressive shortening of the premature stimulus coupling intervals, stimuli began to encroach onto the repolarization phase of the preceding action potential. Maximal encroachment occurred when the coupling interval was closest to the ERP. The site-specific response (VT induction or not) was not related to the degree of encroachment. For instance, in patients that required three extra-stimuli for VT induction (n = 9), the S2, S3 and S4 stimuli during induction were applied at repolarization levels of 87 ± 11%, 84 ±1 5% and 86 ± 16%, respectively. At the noninducing site, with closest coupled stimulation, extra-stimuli encroached to 81 ± 11%, 76 ± 11% and 72 ± 18%, respectively (p = NS compared to inducing site).
The present study shows several major findings. In patients with VT inducible at only one of two ventricular sites, the successful site showed a shorter APD and a shorter ERP. Stimulation at a site with shorter APD produced larger baseline dispersion of repolarization than stimulation at a site with longer APD. The introduction of premature extra-stimuli prolonged intraventricular conduction time between the two recording sites and increased the dispersion of ventricular repolarization. Prior to VT induction at the site with short APD, repolarization dispersion was found to reach a maximal value. Closest coupled extra-stimulation at the site with longer repolarization only produced an amount of dispersion that was comparable to that at the site with shorter repolarization during baseline (S1) stimulation. This has implications for the inducibility of VT based on repolarization dispersion and for the practical approach of VT induction during electrophysiologic testing.
Relation between repolarization and refractoriness
In normal myocardium, local repolarization determines local refractoriness. Although both parameters may vary considerably throughout the ventricle (12,13), a close correlation between APD and ERP has been demonstrated in human myocardium at a given site (11,14). A single S2 extra-stimulus at twice diastolic threshold was found to capture local myocardium when the previous action potential had repolarized to 75% to 85% (6,11,14–16). This is in accordance with the present study where excitability recurred at comparable levels of repolarization. The degree of extra-stimulus encroachment onto the preceding action potential showed no intersite difference. Therefore, successful VT induction could not be explained by stimulation from less repolarized membrane potentials.
Effect of premature extra-stimuli
During programmed electrical stimulation, one to three premature extra-stimuli are commonly applied to a right ventricular endocardial site. Introduction of each additional extra-stimulus as well as decreasing the coupling interval of a premature stimulus toward the ERP of the preceding beat is considered to increase the aggressiveness of the protocol and to facilitate VT induction. This seems due to two important effects. First, shortening the coupling interval decreases the APD and refractoriness of a premature response. Additional extra-stimuli can be introduced with even shorter coupling intervals than previous ones, resulting in progressive APD shortening. Second, intraventricular conduction time increases when extra-stimuli are closely coupled to previous activation. The more a premature extra-stimulus encroaches onto the repolarization phase of the preceding action potential (6), the more sodium channels have not yet recovered from inactivation (17,18), which might explain impaired pulse propagation. A decrease in APD and in conduction velocity both contribute to a shortening of the wavelength of excitation. A shorter wavelength increases the spatial inhomogeneity of activation and facilitates the development of arrhythmias based on re-entrant circuits.
Possible mechanism of site specificity of VT induction
To initiate re-entry, a premature impulse must encounter a zone of unidirectional block, be conducted around that zone through alternate pathways, activate the zone distal to the block with delay to retrogradely invade the zone of block and re-excite the tissue proximal to it. Stimulation at a site with earlier repolarization increases the likelihood of the propagating wave front to collide with sites not yet repolarized and to result in functional block. Depolarizing waves originating from sites with longer APD are less likely to encounter refractoriness and to create functional block. An impulse from a site with earlier repolarization, by way of its prematurity, may even propagate into the left ventricle first and thus be the first (earliest) to initiate an anatomical re-entry circuit which is predominantly located there. While the inducibility of VT by our programmed stimulation protocols support a re-entrant arrhythmia mechanism, other mechanisms like triggered activity or abnormal automaticity cannot be ruled out. However, they should not be affected by repolarization dispersion between stimulation sites.
The site specificity of VT induction, while commonly experienced in the clinical setting, has never been explained. Our data provide first evidence that a stimulus site with shorter APD or ERP is more likely to induce VT than a site with intrinsically longer APD or ERP. There simply is a greater probability of impulses to encounter functional block when they originate from a shorter ERP site to encounter refractoriness at other sites (“upstream” collision) as compared to impulses originating from sites with longer APD or ERP (“downstream” propagation).
As is common during clinical electrophysiologic studies, our data were derived from only two simultaneously stimulated and recorded sites of the right ventricle. Our study compared a VT initiating sequence of extra-stimuli with an equivalent sequence of extra-stimuli that had been applied unsuccessfully to another ventricular site immediately before. Only 18 of 38 patients with inducible VT fulfilled this criterion. We intentionally followed standard clinical protocols that take as an end point VT induction at whichever site it occurs first. Also, electrical properties of the myocardium may change after a VT episode due to adaptation processes to high rates and especially after electrical cardioversion. No attempts were made to reinitiate VT from either site applying additional extra-stimuli. The effect of a pacing protocol on left ventricular myocardium, where an arrhythmia substrate is most likely located, also was not evaluated due to adherence to standard protocol.
For the above reasons, the prevalence of VT induction of one site over the other was limited in our study group. Despite these limitations, immediate comparison of site-specific electrical properties by the chosen protocol detected a significantly shorter APD and ERP at VT inducing sites and proved that even small disparities in APD or ERP between two RV sites can make a difference as to whether VT is induced or not.
Our data suggest that during programmed electrical stimulation, VT is inducible more readily at a site with relatively short repolarization and refractoriness. Because APD is closely related to ERP (11,14), recording of MAP signals might be valuable to determine such site even without applying premature stimuli. Using standard catheters, locating a site with a short ERP by application of single extra-stimuli should be helpful to increase the likelihood of VT induction.
Dispersion of ventricular repolarization has long been implicated as a mechanism of VT occurrence (7–10). It may be of interest to further explore these site-specific properties under the influence of antiarrhythmic drugs, or in atrial myocardium.
The authors thank John C. Pezzullo, PhD, Biostatistics Core Director, General Clinical Research Center at Georgetown University Medical Center, for statistical advice.
- action potential duration
- basic cycle length
- effective refractory period
- monophasic action potential
- right ventricular apex
- right ventricular outflow tract
- ventricular tachycardia
- Received November 14, 2001.
- Revision received March 26, 2002.
- Accepted May 20, 2002.
- American College of Cardiology Foundation
- Doherty J.U.,
- Kienzle M.G.,
- Waxman H.L.,
- Buxton A.E.,
- Marchlinski F.E.,
- Josephon M.E.
- Doherty J.U.,
- Kienzle M.G.,
- Buxton A.E.,
- Marchlinski F.E.,
- Waxman H.L.,
- Josephon M.E.
- Koller B.S.,
- Karasik P.E.,
- Solomon A.J.,
- Franz M.R.
- Han J.,
- Moe G.K.
- Kuo C.S.,
- Munakata K.,
- Reddy C.P.,
- Surawicz B.
- Yuan S.,
- Wohlfart B.,
- Olsson S.B.,
- Blomstrom-Lundqvist C.
- Franz M.R.,
- Chin M.C.,
- Sharkey H.R.,
- Griffin J.C.,
- Scheinman M.M.
- Franz M.R.,
- Bargheer K.,
- Rafflenbeul W.,
- Haverich A.,
- Lichtlen P.R.
- Lee R.J.,
- Liem L.B.,
- Cohen T.J.,
- Franz M.R.
- Sager P.T.,
- Nademanee K.,
- Antimisiaris M.,
- et al.
- Bargheer K.,
- Bode F.,
- Klein H.U.,
- Trappe H.J.,
- Franz M.R.,
- Lichtlen P.R.