Chronic Rapid Atrial Pacing
Structural, Functional, and Electrophysiological Characteristics of a New Model of Sustained Atrial Fibrillation
Background Despite the clinical importance of atrial fibrillation (AF), the development of chronic nonvalvular AF models has been difficult. Animal models of sustained AF have been developed primarily in the short-term setting. Recently, models of chronic ventricular myopathy and fibrillation have been developed after several weeks of continuous rapid ventricular pacing. We hypothesized that chronic rapid atrial pacing would lead to atrial myopathy, yielding a reproducible model of sustained AF.
Methods and Results Twenty-two halothane-anesthetized mongrel dogs underwent insertion of a transvenous lead at the right atrial appendage that was continuously paced at 400 beats per minute for 6 weeks. Two-dimensional echocardiography was performed in 11 dogs to assess the effects of rapid atrial pacing on atrial size. Atrial vulnerability was defined as the ability to induce sustained repetitive atrial responses during programmed electrical stimulation and was assessed by extrastimulus and burst-pacing techniques. Effective refractory period (ERP) was measured at two endocardial sites in the right atrium. Sustained AF was defined as AF ≥15 minutes. In animals with sustained AF, 10 quadripolar epicardial electrodes were surgically attached to the right and left atria. The local atrial fibrillatory cycle length (AFCL) was measured in a 20-second window, and the mean AFCL was measured at each site. Marked biatrial enlargement was documented; after 6 weeks of continuous rapid atrial pacing, the left atrium was 7.8±1 cm2 at baseline versus 11.3±1 cm2 after pacing, and the right atrium was 4.3±0.7 cm2 at baseline versus 7.2±1.3 cm2 after pacing. An increase in atrial area of at least 40% was necessary to induce sustained AF and was strongly correlated with the inducibility of AF (r=.87). Electron microscopy of atrial tissue demonstrated structural changes that were characterized by an increase in mitochondrial size and number and by disruption of the sarcoplasmic reticulum. After 6 weeks of continuous rapid atrial pacing, sustained AF was induced in 18 dogs (82%) and nonsustained AF was induced in 2 dogs (9%). AF occurred spontaneously in 4 dogs (18%). Right atrial ERP, measured at cycle lengths of 400 and 300 milliseconds at baseline, was significantly shortened after pacing, from 150±8 to 127±10 milliseconds and from 147±11 to 123±12 milliseconds, respectively (P<.001). This finding was highly predictive of inducibility of AF (90%). Increased atrial area (40%) and ERP shortening were highly predictive for the induction of sustained AF (88%). Local epicardial ERP correlated well with local AFCL (R2=.93). Mean AFCL was significantly shorter in the left atrium (81±8 milliseconds) compared with the right atrium 94±9 milliseconds (P<.05). An area in the posterior left atrium was consistently found to have a shorter AFCL (74±5 milliseconds). Cryoablation of this area was attempted in 11 dogs. In 9 dogs (82%; mean, 9.0±4.0; range, 5 to 14), AF was terminated and no longer induced after serial cryoablation.
Conclusions Sustained AF was readily inducible in most dogs (82%) after rapid atrial pacing. This model was consistently associated with biatrial myopathy and marked changes in atrial vulnerability. An area in the posterior left atrium was uniformly shown to have the shortest AFCL. The results of restoration of sinus rhythm and prevention of inducibility of AF after cryoablation of this area of the left atrium suggest that this area may be critical in the maintenance of AF in this model.
Atrial fibrillation (AF) remains the most frequently encountered arrhythmia in the clinical setting.1 2 3 Despite this, the management of AF remains less than satisfactory. Most AF animal models have been developed in the short-term setting, and AF has been maintained in these models by pharmacological or electrical stimulation of the vagus nerve.4 5 6 7 Cox et al8 recently developed an AF model in anesthetized dogs after surgically inducing mitral regurgitation. AF was induced by programmed electrical stimulation in 19 of 22 (75%) dogs. The major limitation of these models is failure to sustain AF over prolonged periods.
Conversely, the study of ventricular arrhythmias, particularly ventricular fibrillation, has been strengthened by the development of reproducible animal models. Tachycardia-induced cardiomyopathy achieved by chronic rapid ventricular pacing has been reported recently.9 10 11 In this model, marked structural, functional, and electrophysiological changes were documented and are associated with ventricular fibrillation.10 We therefore hypothesized that chronic rapid atrial pacing would lead to atrial myopathy, thus yielding a reproducible model of sustained AF.
Twenty-two adult mongrel dogs weighing 16 to 30 kg each were used. Baseline 12-lead ECG, white blood cell count, and hemoglobin, serum creatinine, and electrolyte levels were measured to exclude unhealthy animals. The protocol was approved by the University Council on Animal Care of the University of Western Ontario and was completed in accordance with the Guidelines to the Care and Use of Experimental Animals of the Canadian Council on Animal Care.
The dogs were tranquilized with an injection of acepromazine (0.3 to 0.4 mg/kg body wt IM) for transportation to the laboratory and then anesthetized with a mixture of 4% halothane, 2 L/min nitrous oxide, 1 L/min oxygen, and 2 L/min medical air administered by mask. The dogs were then intubated and ventilated at a rate of 10 to 12 breaths per minute and a tidal volume of 12 to 15 mL/kg with a Hospital 300 Anesthesia Ventilator (Hospital Medical Corporation). The halothane level was subsequently reduced to 0.5% to 1%. A circulating water blanket and controller were used to maintain body temperature at 37±1°C. Lactated Ringer’s solution was infused throughout the procedure.
Two-dimensional echocardiography was performed in 11 dogs. Atrial size was assessed by calculating the atrial area by planimetry in the apical four-chamber view (Hewlett Packard 77025A U/S system, 2.5-mHz transducer). At least four repeated measurements were performed for both atria, and an average was calculated. Doppler ultrasound flow studies of the mitral and tricuspid valves were also performed.
Arterial pressure was continuously monitored by an indwelling catheter placed in the right femoral artery through a cut-down. Two 6F quadripolar catheters were inserted through direct cut-down of the left internal jugular vein and were fluoroscopically advanced to the right atrial (RA) appendage and lower RA, respectively. All data were continuously displayed on an Electronics for Medicine VR-16 switched-beam oscilloscope and were recorded periodically on photographic paper at 50 to 100 mm/s. The data were also recorded simultaneously on FM tape with a 16-channel Ampex PR 2230 recorder. The RA was stimulated at a pulse duration of 2 ms and twice diastolic threshold with a Grass S88 stimulator (Grass Medical Instruments). A custom-designed timer was used to trigger the stimulator.
Baroreceptor reflex sensitivity was assessed in all dogs by the method of Smyth et al.12 The dogs were given 10 to 40 μg/kg phenylephrine HCl (Neo-Synephrine, Winthrop Laboratories) until an increase of 30 to 40 mm Hg was achieved.13 The dose that achieved the targeted increase in blood pressure was repeated three times. Each RR interval was plotted as a function of the preceding systolic blood pressure, and beat-by-beat analysis was performed when the RR interval changed. A least-squares-fit linear regression was performed, and reflex control of the heart rate was expressed as the slope of the linear regression line. The three slopes obtained were averaged, and a mean baroreceptor slope was calculated for each dog.
P-wave duration was measured from the surface ECG at baseline and after pacing. The PA interval was measured from the onset of the P wave on the surface ECG to the onset of the atrial electrogram recorded at the lower RA.
The atrial effective refractory period (ERP) was measured at two endocardial sites, the RA appendage and the lower RA, by delivering a train of eight atrial paced beats S1 at two cycle lengths (400 and 300 ms), followed by an extrastimulus (S2) introduced at coupling intervals decremented by 10 ms to scan the entire atrial diastolic interval. Responses were monitored on a Tektronix 5513 storage oscilloscope. Atrial ERP was defined as the longest S1-S2 interval that failed to result in atrial depolarization.
Atrial vulnerability was defined as the ability to induce sustained atrial repetitive responses during programmed electrical stimulation. Inducibility of AF was determined by programmed electrical stimulation at basic cycle lengths of 400 and 300 ms with up to three extrastimuli (S4) delivered. The second extrastimulus (S3) was introduced with the S1-S2 interval fixed at 30 ms longer than the atrial ERP. The S2-S3 interval was set at 80% of the basic cycle length, introducing the extrastimulus with 10-ms scanning decrements. If double extrastimuli failed to induce arrhythmia, a third extrastimulus (S4) was introduced by use of the same protocol. If the latter protocol failed to induce AF, burst pacing at a cycle length of 100 ms for 20 to 30 seconds was tried. Nonsustained AF was defined as repetitive atrial responses lasting <5 minutes and spontaneously terminating. Sustained AF was defined as irregular repetitive atrial responses lasting >15 minutes.
After the programmed stimulation protocol was completed, all catheters were withdrawn and the right femoral artery was ligated. A unipolar screw-in Medtronic J pacing lead was inserted through the left jugular vein incision, and the tip of the lead was fluoroscopically placed and fixed in the RA appendage. The proximal end of the pacing lead was connected to a custom-modified Medtronic programmable pulse generator (5941,8322,8329) and subsequently implanted in a subcutaneous pocket in the neck. The pacemaker was set at 70 beats per minute (bpm) in the demand mode. After all the incisions were closed, the dogs were allowed to recover from anesthesia and were returned to the animal quarters.
Twenty-four hours were allowed for lead stabilization, and then the pacemaker was programmed to 400 bpm (150 ms) and maintained at this rate for 6 weeks. This invariably resulted in a mean ventricular rate of approximately 130 bpm. The surface ECG was checked at 2-week intervals to ensure constant pacing.
Six weeks after the initial study, the pacemaker was programmed to the lowest rate attainable and the dogs were studied by use of the same procedure described for the baseline study. Two-dimensional echocardiography was assessed in the 11 dogs with baseline echo. P-wave duration and PA interval were measured before electrophysiological study. Baroreceptor reflex sensitivity was evaluated in all dogs by use of the protocol previously described. Atrial ERP and AF vulnerability were determined by insertion of a catheter through the right jugular vein. Once sustained AF was induced, the chest was opened by means of a median sternotomy and the heart was exposed and suspended in a pericardial sling. Sinus rhythm was restored by electric cardioversion delivered with epicardial paddles when necessary. Ten quadripolar plaque electrodes (interelectrode distance, 10 mm) were surgically attached to the RA and left atrial (LA) walls (Fig 1⇓). Bipolar epicardial electrograms were continuously recorded on magnetic tape and stored on diskette for off-line signal processing by a computerized mapping system (CMS 1000, Biomedical Instrumentation). Local epicardial AF cycle length (AFCL) was calculated by measuring the interval between the steepest negative deflection of each activation point in a 20-second window and averaged to obtain the mean AFCL. In 8 dogs, local epicardial ERP was measured at each site and correlated with the local AFCL. In 11 dogs, cryoablation at −30° to −45°C with a 10-mm probe (Frigitronics model CCS 100) for 120 seconds was serially applied to the area with shortest AFCL.
After the restudy was completed, the heart was excised and examined for size and gross abnormalities. Both atria were removed and sent to the pathology department for light and electron microscopic studies.
Values are expressed as mean±SD. Paired Student’s t test was used to compare mean values between baseline and restudy. Linear regression analysis was used to evaluate the relation between epicardial ERP and local mean AFCL. Baroreceptor reflex sensitivity was assessed by a least-squares-fit linear regression, and reflex control of heart rate was expressed as the slope of the linear regression line. Only slopes with a correlation coefficient of ≥.80 were accepted for analysis. The slopes were averaged, and a mean baroreceptor slope was calculated for each dog by one-way ANOVA. The null hypothesis was rejected at the P=.05 level.
After 6 weeks of continuous rapid atrial pacing, biatrial enlargement was documented in the 11 dogs assessed by echocardiography (Fig 2⇓). The areas of both atria were markedly increased: the LA increased from 7.8±1 to 11.3±1 cm2 (45%) and the RA from 4.3±0.7 to 7.2±1.3 cm2 (67%, P<.001). Mild tricuspid regurgitation was documented after pacing in 3 of 11 dogs (25%). No evidence of significant mitral or tricuspid valve dysfunction was observed at either baseline or restudy.
Baroreceptor Reflex Sensitivity
After 6 weeks of continuous rapid atrial pacing, no changes in the baroreceptor reflex sensitivity slopes were observed (26.39±14.87 versus 28.08±12.38 ms/mm Hg at baseline, P=NS).
Focal and early hypertrophy were observed in both atria. Also, in most cases, large mural thrombi were also found in either the LA or RA. No evidence of increased connective tissue content was documented.
Severe changes in the architecture of both LA and RA were documented by electron microscopy after 6 weeks of rapid atrial pacing. These changes were characterized by a marked increase in the number and size of the mitochondria and by disruption of the sarcoplasmic reticulum (Fig 3⇓). Enlarged nuclei and dilatation of the rough endoplasmic reticulum were also observed. These changes are consistent with hypertrophy and perturbation of the cellular metabolic activity. No changes were observed in ventricular samples assessed randomly in eight dogs.
P-wave duration and PA interval were assessed at baseline and after 6 weeks of chronic rapid atrial pacing. Significant increases in P-wave duration and PA interval were observed after pacing (55.5±7.5 versus 88.8±8.5 ms and 36.8±10.6 versus 55.5±9.8 ms, respectively, P<.05). A significant reduction in endocardial RA ERP was documented after chronic rapid atrial pacing at cycle length drives of 400 and 300 ms (150±8 to 127±10 ms and 147±11 to 123±12 ms, respectively, P<.001). Sustained AF was not induced by programmed electrical stimulation at baseline study in any dog. In contrast, sustained AF was readily inducible by programmed electrical stimulation in 11 of 22 dogs (Fig 4⇓). Of 11 total dogs, AF was induced by a single extrastimulus in 6 dogs, by two extrastimuli in 3, and by three extrastimuli in 2. Burst pacing was necessary to induce AF in another 7 dogs. AF was induced at least twice with the same extrastimulation protocol in 15 of 18 (83%) dogs. Nonsustained atrial flutter was induced in 2 of the remaining 4 dogs. The mean ventricular response during sustained AF was 595±85 ms. Nonsustained AF was induced in 2 dogs (9%). After programming the pacemaker to the lowest heart-rate setting, AF occurred spontaneously in 4 dogs (18%). These episodes of spontaneous AF were not initiated or preceded by premature atrial contractions. Overall, sustained AF was induced at restudy in 18 of 22 (82%) dogs at a mean duration of 45±7 minutes.
An increase in atrial area of at least 40% was strongly correlated with the inducibility of sustained AF (r=.87). A 15% reduction in ERP was associated with increased atrial vulnerability. When these variables were combined, the positive predictive value for the induction of sustained AF was 88%.
Mean AFCL was significantly shorter in the LA (81±8 ms) compared with the RA (94±9 ms, P<.05). Mean AFCL was correlated with the local epicardial refractory period in 8 dogs, and a strong correlation (R2=.93) between these two measurements was achieved (Fig 5⇓). Epicardial AFCL showed a significant variance (SD2) between sites (LA, 36.5 ms2; RA, 30.5 ms2), suggesting increased dispersion in refractoriness. Detailed analysis of the mean AFCL showed that an area localized to the posterior LA (74±6 ms) had a consistently faster AFCL than the rest of the epicardial sites (Fig 6⇓). This area was always localized to the inferoposterior LA adjacent to the left inferior pulmonary vein. Cryoablation of this area was attempted in 11 dogs (Fig 7⇓), with a mean of 9 applications.5 6 7 8 9 10 11 12 13 14 The area ablated was between ≈5 and 8 cm2. Before AF was terminated, AFCL was significantly prolonged in both atria (LA, 81±8 to 132±5 ms; RA, 94±9 to 136±4 ms, P<.0001). AF was terminated in 9 of 11 (82%) dogs (Fig 8⇓). Sustained AF was no longer inducible by either programmed electrical stimulation or burst pacing; instead, nonsustained atrial flutter was induced in 7 of 9 (77%) dogs.
The present study describes the characteristics of a new canine model of sustained AF. This model may be the experimental counterpart of clinically observed cardiomyopathy induced by tachycardia.14 15 16 Our model is highly reproducible, and sustained AF is readily inducible and can be maintained for prolonged periods of time.
The marked changes in atrial architecture that were documented by electron microscopy, such as fiber disarray and early hypertrophy, may contribute to the maintenance of AF in this model. Disorganization of atrial fiber orientation may slow the rate of conduction, facilitating reentry.17 Similarly, alterations observed at the cellular level, such as the increase in mitochondrial size and disruption of the sarcoplasmic reticulum, may create ionic alterations that increase atrial vulnerability, triggering AF.
A marked increase in atrial size was documented by two-dimensional echocardiography. A 40% increase in atrial area was strongly correlated with the inducibility of sustained AF. This finding may contribute to the maintenance of AF in this model. This finding provides further support for the critical atrial mass hypothesis.18
A high incidence of mural thrombi in both atria was observed by light microscopy. However, two-dimensional echocardiography did not show any evidence of mural thrombi, and no clinical findings suggesting systemic embolization were documented. This model may further expand our understanding of the development of systemic emboli in the presence of chronic nonvalvular AF.
Baroreceptor Reflex Sensitivity
Depressed baroreceptor reflex sensitivity has been reported in the paced model of induced heart failure.19 Interestingly, no changes in baroreceptor function were noted after 6 weeks of continuous rapid atrial pacing. It is possible that the development of baroreceptor alterations in the paced induced heart failure model may be mediated by the presence of heart failure and not by the tachycardia per se. Finally, increased atrial vulnerability in our AF model cannot be attributed to alterations in autonomic balance, as evidenced by a preserved baroreceptor reflex sensitivity after pacing.
Prolonged P-wave duration and PA interval were noted after pacing. A significant shortening of the RA ERP was also documented at restudy. A minimal reduction of 15% in ERP associated with an increase of at least 40% in atrial area was highly predictive of sustained AF. Similarly, local epicardial ERP measured at five sites in each atrium showed a marked dispersion of refractoriness. Local epicardial ERP was strongly correlated with local AFCL, a measure previously used in experimental20 21 22 and human studies23 as an index of local refractoriness. We found a significantly shorter AFCL in the LA compared with the RA. Further analysis of the mean AFCL demonstrated that an area localized to the posterior LA was consistently faster than the rest of the epicardial sites analyzed (Fig 6⇑). The significant difference between the AFCL of the LA and RA may be explained by a shorter refractory period in the LA, modulated in part by differences in autonomic innervation.24 However, the exact mechanism leading to this difference remains unclear. The use of the AFCL as an index of refractoriness is based on the concept that, during fibrillation, a wandering wavelet will reexcite the cells as soon as they recover their excitability.20 21 22 23 Cryoablation of this area significantly prolonged AFCL in both atria and successfully restored sinus rhythm in most dogs (82%), rendering sustained AF noninducible. The ablated area may have been large enough to prevent reentry of multiple wavelets. These findings suggest that the ability to maintain AF in this model may be related to an area localized in the posterior LA that can sustain rapid atrial rates. It is possible to speculate that the increased susceptibility to AF in this model was triggered by alterations in the wavelength mediated by a shortened refractoriness and conduction depression. Although we did not determine the wavelength in this set of experiments, we did note a dispersion of refractoriness of 30.5 ms2 in the RA and 36.5 ms2 in the LA and increased right intra-atrial conduction delay, suggested by the increase in PA interval. The role of triggered activity as a potential mechanism in the generation of AF in this model cannot be entirely ruled out. Alterations in the microarchitecture and anisotropic properties may cause inhomogeneous and discontinuous propagation of the impulse.25 26 27 Depression of conduction in our model may be expected and could be related to the marked ultrastructural changes documented by electron microscopy. Altogether, these findings may be responsible for the increased atrial vulnerability noted in this model.
Previous AF models have been developed in dogs with structurally normal hearts. In these models, AF was generally sustained by electric or pharmacological stimulation of the parasympathetic nervous system.4 5 6 7 Allessie et al28 recently reported the induction of sustained AF by periodic burst pacing of the LA in five chronically instrumented dogs. These dogs were restudied 3 days after implanting a series of epicardial electrodes. Continuous atrial pacing was not performed in this model, and no evidence of structural changes in the atria was reported. These authors reported a significantly shorter AFCL in the LA, comparable to that reported in the present study. The same group recently developed a model of chronic AF by repeatedly inducing AF in chronically instrumented goats.29 In this model, AF was maintained by an external fibrillation pacemaker. Interestingly, the development of chronic AF was associated with a significant shortening of the atrial ERP and shortening of the AFCL. These findings are similar to those in the present study and provide further support for our model. Similarly, a model of acute AF achieved by rapid atrial pacing recently has been used to evaluate different atrial defibrillation waveforms.30 31 Cox et al8 recently developed an AF model in anesthetized dogs subjected to surgically inducing mitral regurgitation. AF was induced by programmed electrical stimulation in 19 of 22 (75%) dogs 3 months after the surgical procedure. We are aware of only one other “chronic” AF model, which used continuous bipolar 60-Hz AC atrial stimulation for the development of AF.32 In this model, in 4 of 6 dogs, AF persisted after the stimulation protocol was terminated. LA enlargement documented by two-dimensional echocardiography and comparable to that observed in our study was also reported. Although this model was reported in abstract form several years ago, no detailed data or a follow-up on this model is presently available. The model of AF described herein is unique, in that AF was associated with marked biatrial myopathy and increased atrial vulnerability in dogs without valvular disease.
Insight into the mechanism leading to sustained AF in this model is limited by the lack of extensive atrial mapping and interelectrode conduction-time data. Nonetheless, we describe a reproducible new model of sustained AF and the functional, structural, and electrophysiological characteristics of this model.
The present study demonstrated the feasibility of developing a reproducible model of sustained AF without valvular disease. AF without overt underlying disease has been increasingly recognized in humans, and the possibility of expanding our understanding of the mechanisms leading to AF in this setting may be enhanced by the present model. Also, new therapeutic modalities may be facilitated by the use of the present novel model.
This work was supported by the Heart and Stroke Foundation of Canada. Dr C.A. Morillo is the recipient of a research fellowship award from the Heart and Stroke Foundation of Canada, Ottawa. Dr G.J. Klein is the recipient of a distinguished professorship of the Heart and Stroke Foundation of Ontario, Toronto, Canada. The authors thank G.K. Wood and R. Kaleda for excellent technical assistance.
- Received June 1, 1994.
- Revision received August 24, 1994.
- Accepted September 5, 1994.
- Copyright © 1995 by American Heart Association
Kannel WB, Wolf PA. Epidemiology of atrial fibrillation. In: Falk RH, Podrid PJ, eds. Atrial Fibrillation: Mechanisms and Management. New York, NY: Raven Press; 1992:81-92.
Kannel WB, Abbott RD, Savage DD, McNamara PM. Epidemiologic features of atrial fibrillation. N Engl J Med. 1982;306: 1018-1022.
Onundarson PT, Thorgeirsson G, Jonmundsson E, Sigfusson N, Hardarson T. Chronic atrial fibrillation: epidemiologic features and 14 year follow-up: a case control study. Eur Heart J. 1987;8:521-527.
Hoff HE, Geddes LA. Cholinergic factor in atrial fibrillation. J Appl Physiol. 1955;8:177-192.
Scher D. Studies on auricular tachycardia caused by aconitine administration. Proc Soc Exp Biol Med N Y. 1947;64:223-239.
Moe GK, Abildskov JA. Atrial fibrillation as a self-sustaining arrhythmia independent of focal discharge. Am Heart J. 1959;58: 59-70.
Cox JL, Canavan TE, Schuessler RB, Cain ME, Lindsay BD, Stone C, Smith PK, Corr PB, Boineau JP. The surgical treatment of atrial fibrillation, II: intraoperative electrophysiologic mapping and description of the electrophysiologic basis of atrial fibrillation. J Thorac Cardiovasc Surg. 1991;101:406-426.
Armstrong PW, Stopps TP, Ford SE, DeBold AJ. Rapid ventricular pacing in the dog: pathophysiologic studies of heart failure. Circulation. 1986;74:1075-1084.
Wilson JR, Douglas P, Hickey WF, Lanoce V, Ferraro N, Muhammad A, Reichek N. Experimental congestive heart failure produced by rapid ventricular pacing in the dog: cardiac effects. Circulation. 1987;75:857-867.
Smyth HS, Sleight P, Pickering GW. Reflex regulation of arterial pressure during sleep in man: a quantitative method of assessing baroreceptor sensitivity. Circ Res. 1969;24:109-121.
Billman GE, Schwartz PJ, Stone LH. Baroreceptor reflex control of heart rate: a predictor of sudden cardiac death. Circulation. 1982;66:874-880.
Weber KT, Pick R, Silver MA, Moe GW, Janicki JS, Zucker IH, Armstrong PW. Fibrillar collagen and remodeling of dilated canine left ventricle. Circulation. 1990;82:1387-1401.
Garrey WE. The nature of fibrillary contraction of the heart: its relation to tissue mass and form. Am J Physiol. 1914;33:397-414.
Chen JS, Wang W, Bartholet T, Zucker IH. Analysis of baroreflex control of heart rate in conscious dogs with pacing-induced heart failure. Circulation. 1991;83:260-267.
Lammers WJEP, Allessie MA, Rensma PL, Schalij MJ. The use of fibrillation cycle length to determine spatial refractoriness in electrophysiological properties and to characterize the underlying mechanism of fibrillation. New Trends Arrhythmias. 1986;2:109-112.
Opthof T, Ramdat Misier AR, Coronel R, Vermeulen JT, Verberne HJ, Frank RGJ, Moulijn AC, van Capelle FJL, Janse MJ. Dispersion of refractoriness in canine ventricular myocardium: effects of sympathetic stimulation. Circ Res. 1991;68:1204-1215.
Kim KB, Rodefeld MD, Schuessler RB, Boineau JP, Cox JL. Minimum atrial fibrillation interval estimates refractory period in the isolated canine bi-atrial preparation. PACE Pacing Clin Electrophysiol. 1994;17:839. Abstract.
Rensma PL, Allessie MA, Lammers WJEP, Bonke FIM, Schalij MJ. Length of excitation wave and susceptibility to reentrant atrial arrhythmias in normal conscious dogs. Circ Res. 1988;62:395-410.
Spach MS, Kootsey JM, Sloan JD. Active modulation of electrical coupling between cardiac cells of the dog: a mechanism for transient and steady state variations in conduction velocity. Circ Res. 1982;51:347-362.
Spach MS, Dolber PC. Relating extracellular potentials and their derivative to anisotropic propagation at a microscopic level in human cardiac muscle. Circ Res. 1986;58:356-371.
Spach MS, Miller WT, Geselowitz DB, Barr RC, Kootsey JM, Johnson EA. The discontinuous nature of propagation in normal canine cardiac muscle. Circ Res. 1981;48:39-54.
Allessie M, Kirchhof C, Scheffer GJ, Chorro F, Brugada J. Regional control of atrial fibrillation by rapid pacing in conscious dogs. Circulation. 1991;84:1689-1697.
Wijffels M, Kirchhof C, Fredericks J, Boersma L, Allessie M. Atrial fibrillation begets atrial fibrillation. Circulation. 1993;88(suppl I):I-18. Abstract.
Cooper RAS, Alferness CA, Smith WM, Ideker RE. Internal cardioversion of atrial fibrillation in sheep. Circulation. 1993;87: 1673-1683.
Salmon DR, McPherson DD, Augustine DE, Holida MD, White CW. A canine model of chronic atrial fibrillation: echocardiographic and electrocardiographic validation. Circulation. 1985;72(suppl III):III-250. Abstract.