Effect of Atrial Fibrillation on Atrial Refractoriness in Humans
BackgroundThe acute effect of atrial fibrillation (AF) on the atrial effective refractory period (ERP) in humans is unknown.
Methods and Results In 20 patients without structural heart disease, the atrial ERP was measured before and after pacing-induced AF at drive cycle lengths of 350 and 500 ms. Immediately after spontaneous AF conversion, the post-AF ERP was measured. The pre-AF ERPs at 350 and 500 ms were 206±23 and 216±17 ms, respectively. The time to spontaneous conversion of AF was 7.3±1.9 minutes. The first post-AF ERPs at drive cycle lengths of 350 and 500 ms were 175±30 ms (P<.0001 versus pre-AF) and 191±30 ms (P<.0001 versus pre-AF), respectively. The post-AF ERP returned to the pre-AF ERP value after a mean of 8.4±0.3 minutes. In 15 patients, during the determination of the post-AF ERP, secondary episodes of AF lasting 1±1.5 minutes were reinduced 6±3 times per patient. There was a significant inverse logarithmic relationship between the time to reinduction of AF and the duration of secondary episodes of AF (P<.0001, r=.5).
Conclusions In humans, several minutes of induced AF is sufficient to shorten the ERP for up to ≈8 minutes. The temporal recovery of the ERP is reflected in progressively shorter episodes of reinduced AF. These data imply that AF transiently shortens the atrial wavelength and suggest a mechanism by which AF may perpetuate itself.
Experimental studies in goats have demonstrated that atrial fibrillation (AF) induces remodeling of the atria and that this effect is time dependent and may promote perpetuation of AF.1 In humans, clinical observations suggest that AF promotes changes in the atria. Chronic AF is often preceded by episodes of paroxysmal AF,2 3 and chemical or electrical cardioversion of AF is more often successful when the AF is of recent onset.4 5 6 7 However, no previous studies have assessed the acute electrophysiological effects of AF on the atria in humans. The hypothesis of this study was that AF induces changes in atrial refractoriness that may facilitate its perpetuation. Therefore, the purpose of this study was to measure the changes in the right atrial effective refractory period (ERP) after pacing-induced AF.
Characteristics of the Study Population
The subjects of this study were 20 patients referred to the University of Michigan Hospital for electrophysiological testing and/or radiofrequency catheter ablation. Exclusion criteria included a baseline rhythm of AF, atrial flutter, or atrial tachycardia; a contraindication to the use of propranolol or atropine; the presence of structural heart disease determined by a transthoracic echocardiogram; the inability to achieve a stable electrode catheter position in the right atrial appendage; the inability to induce AF by pacing; or the failure of pacing-induced AF to convert spontaneously to sinus rhythm. Among 28 patients who were screened, 20 (8 men and 12 women; mean age, 40±15 years [±SD]) satisfied all criteria. The mean left ventricular ejection fraction was 0.59±0.08. The electrophysiology procedure was performed to evaluate and treat paroxysmal supraventricular tachycardia in 19 patients and to evaluate syncope in 1.
At the time of the electrophysiology procedure, 2 patients were being treated with a β-adrenergic antagonist, 1 with a calcium channel antagonist, and 1 with digitalis. In the other 16 patients, all antiarrhythmic drug therapy was discontinued at least 5 half-lives before the procedure. After informed consent was obtained, three 7F sheaths (Daig Corp) were placed in a femoral vein, and three quadripolar electrode catheters, which had an interelectrode spacing of 2-5-2 mm (Mansfield EP), were positioned in the high right atrium, the His bundle position, and the right ventricular apex. Patients were sedated with intravenous midazolam and received 3000 U heparin IV. Leads V1, I, II, and III and the intracardiac electrograms were recorded (Mingograph 7; Siemens-Elema AB). Pacing was performed with a programmable stimulator (Bloom Associates, Ltd).
The study protocol was approved by the Human Research Committee and was performed after completion of the clinically indicated portion of the electrophysiology procedure. Quadripolar electrode catheters were positioned in the right atrial appendage and against the right atrial free wall. In the atrial appendage, electrodes 1 and 2 were used for pacing and electrodes 3 and 4 were used to record the local right atrial electrogram. The other atrial catheter was used to record the local atrial electrogram with electrodes 1 and 2 and to confirm atrial capture during programmed atrial stimulation. Autonomic blockade was achieved by infusion of atropine (0.04 mg/kg) and propranolol (0.2 mg/kg) over a period of ≈5 minutes.8 The mean patient weight was 77±21 kg, and the mean atropine and propranolol doses were 2.9±0.7 and 14.8±3.8 mg, respectively. There were no significant differences between the pre-AF and the post-AF sinus cycle length (647±104 versus 650±107 ms, P=.15) or atrial-His interval (91.5±28 versus 89.3±30 ms, P=.45).
The mean atrial capture threshold was 0.7±0.2 mA. Pacing was performed at three times threshold. The atrial ERP was measured by an incremental technique in 5-ms steps at basic drive cycle lengths of 350 and 500 ms for eight beats with a 1-second pause between pacing trains. The ERP was defined as the longest S1-S2 coupling interval that failed to result in atrial capture. The pre-AF atrial ERP was measured three times and averaged.
AF was induced by bursts of atrial pacing at cycle lengths of 160 to 190 ms. The duration of pacing required to induce sustained AF was quantified. After at least 5 minutes of AF, the AF was allowed to spontaneously convert to sinus rhythm. Immediately upon conversion to sinus rhythm and until the atrial ERP returned to within 5 ms of the baseline atrial ERP, the post-AF atrial ERP was measured at alternating drive cycle lengths of 350 and 500 ms. To assess the temporal changes in the ERPs, the time from conversion of AF to determination of each post-AF atrial ERP was measured to the nearest second. Whenever secondary episodes of AF were unintentionally induced during measurement of the post-AF ERP, the time at which the AF was induced was noted and the duration of the episode was measured to the nearest second. In one patient, a secondary episode of AF was persistent and did not revert to sinus rhythm until electrical cardioversion. In this patient, only the data collected before the onset of the secondary episode of AF were used for analysis.
To control for the possible effects of repeated refractory period determinations on the atrial ERP, the atrial ERP was measured repeatedly after autonomic blockade in a separate group of 16 subjects. The atrial ERP was measured in sequential fashion at alternating basic drive cycle lengths of 350 and 500 ms, as in the study population, for a total of up to 20 determinations.
The control population (5 men and 11 women; mean age, 48±19 years) included patients referred to the University of Michigan Hospital for electrophysiological testing and/or radiofrequency catheter ablation. Exclusion criteria included a baseline rhythm of AF, atrial flutter, or atrial tachycardia; a contraindication to the use of propranolol or atropine; the presence of structural heart disease determined by a transthoracic echocardiogram; or the inability to achieve a stable electrode catheter position in the right atrial appendage. The mean left ventricular ejection fraction was 0.64±0.05. The study protocol was performed after radiofrequency ablation of an accessory pathway or AV nodal reentrant tachycardia.
Continuous variables are expressed as mean±SD. Continuous variables were compared by a paired t test, and categorical variables were compared by χ2 analysis. Serial measurements of the post-AF atrial ERP were analyzed by ANOVA with repeated measures. Linear interpolation of the plotted serial measurement data was used to generate data for analysis of temporal changes of the atrial ERP in both the control and study patients.9 10 Nonlinear regression analysis was used to correlate the duration of secondary episodes of AF to the logarithm of the time to induction of secondary AF. A value of P<.05 was considered significant.
Changes in the Atrial ERP
Before induction of AF, the right atrial ERP at a drive cycle length of 350 ms was 206±23 ms and at a drive cycle length of 500 ms was 216±17 ms. The duration of induced AF, including the time required for atrial pacing, was 7.3±1.9 minutes. The time from termination of rapid atrial pacing to measurement of the first post-AF atrial ERP was 6.6±2.1 minutes. The post-AF atrial ERP was measured 165 times (8±3 times per patient) at a drive cycle length of 350 ms and 154 times (7±3 times per patient) at a drive cycle length of 500 ms.
The first atrial ERP measured immediately after conversion of AF at a drive cycle length of 350 ms was 175±30 ms (P<.0001 versus pre-AF ERP). A significant reduction in the atrial ERP persisted until the fifth measurement of the post-AF atrial ERP at a drive cycle length of 350 ms (Table 1⇓). For the atrial ERP measured at a drive cycle length of 500 ms, the first atrial ERP after conversion of AF was 191±30 ms (P<.0001 versus pre-AF ERP), and this significant reduction in the atrial ERP persisted until the fifth measurement of the post-AF atrial ERP (Table 1⇓).
Temporal Recovery of the Atrial ERP
The temporal recovery of the atrial ERP at drive cycle lengths of 350 and 500 ms in the study population is summarized in Fig 1⇓. A significant reduction in the post-AF atrial ERP compared with the pre-AF atrial ERP persisted for 5.5 minutes at a drive cycle length of 350 ms and for 8.0 minutes at a drive cycle length of 500 ms. There was no significant difference in the temporal recovery of the post-AF atrial ERP between the basic drive cycle lengths of 350 and 500 ms.
The percent change of the first post-AF atrial ERP compared with the pre-AF atrial ERP did not correlate with the duration of the pacing-induced AF either at a drive cycle length of 350 ms (P=.9, r=.02) or at a drive cycle length of 500 ms (P=.9, r=.05). In 7 patients, the duration of pacing-induced AF was <6 minutes (mean, 5.6±0.3 minutes), and in 13 patients, the duration of pacing-induced AF was >6 minutes (mean, 8.3±1.6 minutes, P<.001). There was no difference in the percent change of the first post-AF atrial ERP compared with the pre-AF atrial ERP between patients with >6 or <6 minutes of AF with a drive cycle length of either 350 ms (16±10% versus 12±7%, P=.4) or 500 ms (13±9% versus 9±6%, P=.4).
Influence of Atrial Pacing on the Atrial ERP
The total duration of rapid atrial pacing required to induce sustained AF was 47±39 seconds (range, 11 to 154 seconds). Nine patients required <30 seconds of pacing (mean, 16±4 seconds) to induce AF, and the remaining 11 patients required >30 seconds of pacing (mean, 73±36 seconds, P=.001). Comparing patients requiring >30 and <30 seconds of pacing, there was no difference in the percent change of the first post-AF atrial ERP compared with the pre-AF atrial ERP at a drive cycle length of 350 ms (15±6% versus 12±10%, P=.5) or 500 ms (12±9% versus 10±8%, P=.5).
Induction of Secondary Episodes of AF
During the measurement of the post-AF atrial ERP in the study population, 84 secondary episodes of AF were unintentionally induced in 15 patients (6±3 episodes per patient). Forty-one episodes of secondary AF were induced at a drive cycle length of 350 ms, and 43 episodes were induced at a drive cycle length of 500 ms. Among 40 initial measurements of the post-AF atrial ERP at both drive cycle lengths of 350 and 500 ms, 22 secondary episodes of AF (56%) were induced at a mean interval of 32±18 seconds after spontaneous conversion of AF. Fifteen (38%), 11 (29%), 10 (26%), 7 (21%), 9 (32%), 5 (18%), 3 (11%), 2 (11%), 0, and 0 secondary episodes of AF were induced at 144±52, 223±91, 292±117, 378±145, 404±149, 535±192, 525±181, 557±213, 618±203, and 634±224 seconds after spontaneous conversion of AF, respectively (P<.0001, r=.94, Fig 2⇓). Secondary episodes of AF lasted 1±1.5 minutes, with a range of 2 seconds to 14 minutes.
Secondary episodes of AF in the study population were induced at a mean A1-A2 coupling interval of 176±31 ms (P<.005 versus pre-AF atrial ERP) at a drive cycle length of 350 ms. At a drive cycle length of 500 ms, the mean A1-A2 coupling interval that induced secondary episodes of AF was 193±29 ms (P<.005 versus pre-AF atrial ERP).
The atrial ERP measured immediately after spontaneous conversion of secondary episodes of AF in the study population at a drive cycle length of 350 ms was not significantly different from the A1-A2 that induced secondary episodes of AF (190±30 versus 193±29 ms, P=.3). At a drive cycle length of 500 ms, there was no significant difference between the atrial ERP determined immediately after conversion of secondary episodes of AF and the A1-A2 that induced secondary episodes of AF (193±29 versus 187±28 ms, P=.6).
The elapsed time after spontaneous conversion of the pacing-induced AF to the induction of secondary episodes of AF was correlated with the duration of the secondary episodes of AF. There was a significant inverse logarithmic relationship between the time to induction of secondary AF and the duration of secondary AF (P<.0001, r=.5, Fig 3⇓).
Effect of Secondary Episodes of AF on Temporal Recovery of the ERP
During the measurement of the post-AF atrial ERP in the study population, secondary episodes of AF were unintentionally induced in 15 patients, and no episodes were induced in 5 patients. There were no significant differences in the atrial ERP after spontaneous conversion of AF between patients with and without secondary episodes of AF at a drive cycle length of 350 ms (Fig 4⇓) or 500 ms (Fig 5⇓).
To assess the effect of long versus short episodes of secondary AF on the temporal recovery of the atrial ERP, patients with secondary episodes of AF were stratified according to the duration of secondary episodes of AF. Eight patients had 18 episodes of secondary AF lasting >90 seconds (mean, 114±25 seconds), and 7 had 66 episodes of secondary AF lasting <90 seconds (mean, 31±17 seconds; P<.0001). There was no difference in temporal recovery of the post-AF atrial ERP at a drive cycle length of 350 or 500 ms between patients with secondary episodes of AF lasting <90 or >90 seconds (Table 2⇓).
Repeated Measurement of the Atrial ERP
In the control patients, the atrial ERP was measured 17±4 times per patient, over a mean of 9.2±1.9 minutes, at alternating drive cycle lengths of 350 and 500 ms. The first atrial ERP at a drive cycle length of 350 ms measured after autonomic blockade was 200±11 ms. For the atrial ERP measured at a drive cycle length of 500 ms, the first atrial ERP was 203±18 ms (P=.9 versus baseline ERP). The temporal changes in the atrial ERP at drive cycle lengths of 350 and 500 ms are summarized in Fig 6⇓. There is no significant difference between the atrial ERP measured at baseline and the atrial ERPs determined during repeated measurement.
The main findings of this study are that pacing-induced AF significantly shortens the right atrial ERP, that this shortening occurs after only a few minutes of AF, and that temporal recovery of the AF-induced shortening of the atrial ERP occurs over ≈5 to 8 minutes. In addition, upon termination of a several-minute episode of AF, there is an increased propensity for the induction of secondary episodes of AF, which diminishes progressively as the elapsed time after the conversion of the primary episode of AF increases. Furthermore, the duration of secondary episodes of AF becomes progressively shorter as the elapsed time after conversion of the primary episode of AF increases. The temporal relationship between changes in the atrial ERP, the propensity for reinduction of AF, and the duration of episodes of reinduced AF after conversion of a primary episode of AF demonstrates that AF, even only several minutes in duration, induces transient changes in atrial physiology that promote its perpetuation.
High-density mapping studies of AF in humans suggest that a determinant of AF is the presence of a critical number of wandering reentrant atrial wavelets.11 12 To sustain these multiple-reentry circuits, there must be local conduction block and the wavelength of the depolarizing wave front must be sufficient to allow recovery of the myocardium proximal to the conduction block. Wavelength has been defined as the distance traveled by the depolarizing wave front during the refractory period (wavelength=conduction velocity×refractory period).13 In AF, if the atrial wavelength is long, reentry may not be maintained and AF may self-terminate. However, if the atrial wavelength is relatively short because of either depressed or anisotropic conduction, a short refractory period, or both, then a greater number of wave fronts can circulate through the atria and AF may be sustained.14 15 16 As demonstrated in this study, a brief episode of AF shortens the atrial refractory period, thereby shortening the wavelength. A reduction in wavelength would be expected to increase the propensity for AF, as was observed in this study.
Secondary Episodes of AF
The frequent induction of secondary episodes of AF after a primary episode of pacing-induced AF supports the hypothesis that AF induces changes in atrial physiology that promote perpetuation of AF. AF-induced shortening of the atrial ERP is one possible mechanism for the induction of secondary episodes of AF; however, other changes in atrial physiology, such as an increase in the conduction velocity or dispersion of atrial refractoriness, may also facilitate the induction of secondary episodes of AF.
The atrial ERPs that induced secondary episodes of AF did not differ significantly from the ERPs measured immediately after spontaneous termination of secondary episodes of AF. This absence of a significant shortening in ERP may be explained by the difference in duration of secondary episodes of AF compared with pacing-induced AF. Pacing-induced AF converted after ≈7 minutes, compared with a mean of ≈1 minute for secondary episodes of AF. This suggests that the phenomenon of AF-induced shortening of the atrial ERP may require several minutes of AF.
Temporal Recovery of Atrial Properties
In each patient in this study, pacing-induced AF spontaneously converted to sinus rhythm, suggesting that the wavelengths of the reentrant circuits sustaining AF increased until the required number of circuits could not be maintained and AF stopped. Yet despite conversion to sinus rhythm, the electrophysiological effects of AF on atrial tissue persisted. The temporal recovery of AF-induced changes in atrial electrophysiological properties is demonstrated by a progressive increase in the atrial ERP, a decrease in vulnerability to the reinduction of AF, and a decrease in the duration of episodes of reinduced AF. These data imply that pacing-induced AF transiently alters atrial properties and that this effect persists after spontaneous conversion to sinus rhythm and diminishes with time.
Mechanisms of AF-Induced Shortening of the Atrial ERP
A likely mechanism for the AF-induced shortening of the atrial ERP is a physiological, rate-dependent shortening of the action potential duration. Rate-dependent shortening of the refractory period is well documented in ventricular myocardium.17 18 19 20 Activation and deactivation of the gated outward potassium current during the action potential plateau phase can result in beat-to-beat potassium ion accumulation and shortening of the action potential duration and the ERP.21 22 23 24 25
A second possible mechanism for shortening of the atrial ERP after AF is “cardiac memory.” Cardiac memory refers to T-wave changes induced by rapid ventricular pacing that persist after resumption of normal AV conduction. It is possible that similar repolarization changes may occur in the atria after spontaneous conversion of AF. Experimental studies assessing cardiac memory after 20 minutes of high-rate ventricular pacing have demonstrated that 4-aminopyridine, a drug that blocks the transient outward potassium current, abolishes changes in T-wave morphology, but lidocaine does not.26 27 This finding implies that one mechanism for cardiac memory is changes in potassium ion current. A similar alteration in potassium current may occur as a result of AF. Change in the potassium current may explain post-AF alteration of atrial repolarization as well as shortening of the atrial ERP after only a few minutes of AF.
A third possible mechanism for persistent changes in atrial properties after an episode of AF is anisotropic capacitive coupling.28 29 The multiple wavelets that sustain AF propagate simultaneously parallel and perpendicular to the atrial fiber axis, and this type of tissue activation may result in changes in cellular coupling and in action potential duration. After resumption of isotropic conduction during sinus rhythm, these coupling changes may persist. The inverse logarithmic relationship between the time to induction of secondary episodes of AF and the duration of the secondary AF is consistent with the expected pattern of temporal decay in capacitive coupling (Fig 3⇑).
Another possible mechanism for a shortening of the atrial ERP after AF is a change in autonomic tone, since either vagal or adrenergic activation can shorten atrial refractoriness.30 31 32 33 In this study, propranolol and atropine were administered to minimize the possibility that the measured changes in atrial refractoriness were caused by changes in autonomic tone. A constant degree of autonomic tone was confirmed by the absence of any significant differences in the sinus cycle length or atrial-His interval before and after completion of the study protocol. It should be noted that spontaneous episodes of AF are likely to be associated with sympathetic activation, which would further shorten atrial refractoriness. Therefore, the results of this study may underestimate the degree of AF-induced shortening of the atrial ERP that may occur after clinical episodes of AF.
Possible Effects of Pacing
The rapid atrial pacing that was used to induce AF may in itself have resulted in changes in atrial refractoriness, either locally or globally. However, because there were usually several minutes between the termination of pacing and the onset of the ERP determinations, it is unlikely that rapid pacing was responsible for the changes in atrial refractoriness observed after conversion of AF to sinus rhythm.
Because atrial ERPs were measured repeatedly after termination of AF, it is conceivable that the repeated determinations of refractoriness by themselves affected the atrial ERP. However, the absence of a change in atrial ERP over time in the control group rules out this possibility.
In a goat model of AF, Wijffels et al1 demonstrated a shortening of the atrial ERP in response to chronic AF. In this study, AF was repetitively induced by bursts of atrial pacing until the duration of AF increased from a few seconds to being persistent. After chronic AF converted to sinus rhythm, the ERP and the intra-atrial conduction velocity were measured. The atrial ERP shortened significantly after episodes of AF, without a significant change in atrial conduction velocity. Therefore, in this goat model, changes in atrial wavelength were attributable primarily to a decrease in refractoriness. In contrast to the study by Wijffels et al, the AF-induced changes in atrial refractoriness in the present study occurred after only a few minutes of AF.
A preliminary study in humans evaluated changes in atrial electrophysiological properties after pacing.34 This study compared changes in the atrial ERP after rapid pacing for 3 minutes during sinus rhythm in patients with paroxysmal AF and in patients with AV conduction disturbances but without atrial arrhythmias. There was a significant shortening of the atrial ERP after rapid pacing in both control and paroxysmal AF patients. Although this study did not control for changes in autonomic tone or describe the temporal recovery of the ERP, it does provide evidence that one mechanism for shortening of the atrial ERP during AF is a rate-dependent shortening of the action potential duration.
A limitation of this study is that the findings may be specific to pacing-induced AF in subjects with structurally normal atria and may not apply to spontaneous episodes of AF, episodes of AF occurring in patients with paroxysmal or chronic AF, or episodes of AF occurring in patients with structurally abnormal atria. A second limitation is that only the right atrial ERP was measured, and therefore the responses of other areas of the atrium to pacing-induced AF are unknown. Finally, conditioning pacing trains, which have been demonstrated to be useful in improving the reproducibility of ventricular refractory period determinations,35 were not used in this study. The use of conditioning pacing trains would have precluded the frequent measurements of refractoriness needed to detect temporal changes.
In conclusion, brief episodes of AF may be sufficient to shorten atrial refractoriness for several minutes. The physiological significance of this shortening of atrial refractoriness may be reflected in a heightened propensity for the reinduction of AF after conversion to sinus rhythm. As the atrial ERP gradually lengthens to a baseline level, the propensity to reinduce AF diminishes, as does the duration of the reinduced AF episodes, possibly because of progressive lengthening of the atrial wavelength. These findings suggest a mechanism by which AF may promote its own perpetuation.
- Received January 30, 1996.
- Revision received March 25, 1996.
- Accepted April 9, 1996.
- Copyright © 1996 by American Heart Association
Wijffels MCEF, Kirchhof CJHJ, Dorland R, Allessie MA. Atrial fibrillation begets atrial fibrillation: a study in awake chronically instrumented goats. Circulation. 1995;92:1954-1968.
Godtfredsen J. Atrial fibrillation: course and prognosis: a follow-up study of 1212 cases. In: Kulbertus ZHE, Olsson SB, Schlepper M, eds. Atrial Fibrillation. Molndal, Sweden: AB Hassell; 1982:134-145.
Crijns HJGM, Van Wijk LM, Van Gilst WH, Kingma JH, Van Gelder IC, Lie KI. Acute conversion of atrial fibrillation to sinus rhythm: clinical efficacy of flecainide acetate: comparison of two regimens. Eur Heart J. 1988;9:634-638.
Jose AD, Taylor RR. Autonomic blockade by propranolol and atropine to study intrinsic myocardial function in man. J Clin Invest. 1969;48:2019-2031.
Konings KTS, Kirchhof CJHJ, Smeets JRLM, Wellen HJJ, Penn OC, Allessie MA. High-density mapping of electrically induced atrial fibrillation in humans. Circulation. 1994;89:1665-1680.
Cox JL, Canavan TE, Schuessler RB, Cain ME, Lindsay BD, Stone C, Smith PK, Corr PB, Boineau JP. The surgical treatment of atrial fibrillation: intraoperative electrophysiologic mapping and description of the electrophysiologic basis of atrial flutter and atrial fibrillation. J Thorac Cardiovasc Surg. 1991;101:406-426.
Rensma PL, Allessie MA, Lammers WJEP, Bonke FIM, Schalij MJ. The length of the excitation wave as an index for the susceptibility to reentrant atrial arrhythmias. Circ Res. 1988;62:395-410.
Michelucci A, Padeletti L, Fradella GA. Atrial refractoriness and spontaneous or induced atrial fibrillation. Acta Cardiol. 1982;5:333-344.
Smeets JLRM, Allessie MA, Lammers WJEP, Bonke FIM, Hollen J. The wavelength of the cardiac impulse and reentrant arrhythmias in isolated rabbit atrium. Circ Res. 1986;58:96-108.
Gibbs CL, Johnson EA. Effect of changes in frequency of stimulation upon rabbit ventricular action potential. Circ Res. 1981;9:165-170.
Denes P, Wu D, Dhingra R. The effects of cycle length on cardiac refractory periods in man. Circulation. 1974;49:32-41.
Guss S, Kastor J, Josephson M. Human ventricular refractoriness. Circulation. 1976;53:450-456.
Hauswirth O, Noble D, Tsien RW. The dependence of plateau currents in cardiac Purkinje fibers on the interval between action potentials. J Physiol. 1972;222:22-49.
Colatsky TJ, Hogan PM. Effects of external calcium, calcium channel-blocking agents, and stimulation frequency on cycle length-dependent changes in canine cardiac action potential duration. Circ Res. 1980;46:543-552.
Boyett MR, Jewell BR. Analysis of the effects of rate and rhythm upon electrical activity in the heart. Prog Biophys Mol Biol. 1980;361:1-52.
Geller JC, Rosen MR. Persistent T-wave changes after alteration of the ventricular activation sequence: new insights into cellular mechanisms of ‘cardiac memory.’ Circulation. 1993;88:1811-1819.
del Balzo U, Rosen MR. T-wave changes persisting after ventricular pacing in canine heart are altered by 4-aminopyridine but not by lidocaine: implications with respect to phenomenon of cardiac ‘memory.’ Circulation. 1992;85:1464-1472.
Goyal R, Vikstrom B, Hero A, Morady F. Simulation of cardiac memory in a computer model utilizing capacitive coupling. J Electrocardiol. In press.
Spach MS, Dolber PC, Heidlage JF, Kootse JM, Johnson EA. Propagating depolarization in anisotropic human and canine cardiac muscle: apparent directional differences in membrane capacitance. Circ Res. 1987;60:206-219.
Coumel P. Neural aspects of paroxysmal atrial fibrillation. In: Falk RH, Podrid PJ, eds. Atrial Fibrillation: Mechanisms and Management. New York, NY: Raven Press Ltd; 1992:109-125.
Lewis T, Drury AN, Bulger HA. Observations upon flutter and fibrillation, VII: the effect of vagal stimulation. Heart. 1921;8:141-169.
Nahum LH, Hoff HE. Production of auricular fibrillation by application of acetyl-beta-methyl-choline chloride to localized region of the auricular surface. Am J Physiol. 1940;129:428-436.
Attuel P, Leclercq JF, Coumel P. Atrial electrophysiological substrate remodeling after tachycardia in patients with and without atrial fibrillation. Pacing Clin Electrophysiol. 1995;18(pt 2):804. Abstract.