Radiofrequency Catheter Modification of Sinus Pacemaker Function Guided by Intracardiac Echocardiography
Background The sinus P wave arises from a pacemaker complex distributed along the crista terminalis. We investigated the feasibility of modification of sinus pacemaker function using graded applications of radiofrequency energy along the crista terminalis in dogs to achieve sinus rate control.
Methods and Results Modification of sinus pacemaker function (30±5% reduction in intrinsic heart rate with retention of a normal P-wave axis) was performed in 11 dogs (group 1). Total sinus pacemaker ablation (>50% reduction in intrinsic heart rate with development of a low ectopic atrial or a junctional rhythm) was performed in 4 dogs (group 2). Intracardiac echocardiography was used to identify the crista terminalis as an anatomic marker of sinus node location. Sinus pacemaker modification caused a significant decrease in intrinsic heart rate (31% reduction, P<.001), heart rate responsiveness to isoproterenol (30% reduction, P<.0001), and average (20% reduction, P=.0002) and maximal (22% reduction, P=.0007) heart rates during 24-hour Holter monitoring. In 6 of the 11 animals, the targeted rate reduction of 30±5% was accurately achieved (mean, 31.6±4.3%; P<.001), and in the other 5, significant reduction of intrinsic heart rate was achieved but with greater variation (28.0±17.3%, P<.005). Corrected sinus node recovery time was not prolonged. After modification, earliest activation was mapped to the crista terminalis inferior to the lesion in all animals. In long-term follow-up (3.7±1.0 months), effects were maintained. After total sinus pacemaker ablation, junctional and low atrial escape pacemakers were unstable.
Conclusions This study demonstrates the feasibility of modification of sinus pacemaker function for sinus rate control using catheter-based radiofrequency ablation guided by intracardiac echocardiography. This can be done while pacemaker stability and attenuated responsiveness to autonomic influences are preserved. Intracardiac echocardiography accurately defined the crista terminalis and provided a reliable means to anatomically localize catheter position in relation to the sinus node.
The past decade has witnessed enormous advances in the use of radiofrequency catheter ablative procedures for cure or control of a variety of cardiac arrhythmias. However, no one has described a technique for reliable modification of the sinus rate. At present, this must be achieved with pharmacological therapy, which may frequently be limited by side effects, patient intolerance, or treatment failure related to an associated disease process.
The unique anatomic1 2 3 and functional4 5 characteristics of the sinus node have been well characterized. The sinus P wave arises from a pacemaker complex distributed over an area extending from the junction of the superior vena cava and right atrial appendage in the inferior direction along the sulcus terminalis almost to the inferior vena cava. A close correspondence between the change in heart rate and the change in sites of impulse origin within this complex in response to certain autonomic manipulations has been shown.4 5 This site-specific differential sensitivity to autonomic inputs allows the possibility of targeted modification of sinus pacemaker function. Indeed, successful modification of sinus pacemaker function by use of epicardial laser irradiation recently was described in dogs.6 However, this approach requires thoracotomy, epicardial mapping, and epicardial laser application and therefore, a clinical application has not been developed.
In the present study, we investigated the feasibility of modification of sinus pacemaker function using graded applications of radiofrequency energy along the crista terminalis in dogs. Chu et al7 recently demonstrated that intracardiac echocardiography can be used to guide anatomic placement of radiofrequency lesions. We hypothesized that intracardiac echocardiography could be used to accurately position the ablation electrode on the crista terminalis and supplemented this anatomic localization with endocardial activation mapping to define the sites of earliest activation during total autonomic blockade. We also performed total sinus pacemaker ablation in another group of animals to characterize any difference in lesion extent required for total ablation, to define the “margin for error” during an attempted modification procedure, and to compare the physiological characteristics of the escape pacemakers.
Fifteen adult mongrel dogs of either sex were included in the study. Group 1 consisted of 11 animals that underwent sinus pacemaker modification, and group 2 consisted of 4 animals that had total sinus pacemaker ablation. All animals underwent the same series of pharmacological and electrophysiological tests both before and after ablation, as described below. All studies conformed to the “Position of the American Heart Association on Research Animal Use” adopted November 11, 1984, and the protocol was approved by the University of California Committee on Animal Research.
Modification of sinus pacemaker function (group 1) was defined by a targeted reduction of intrinsic heart rate of 30±5% with retention of a normal P-wave axis in the frontal and horizontal planes (Fig 1⇓, top).
Total sinus pacemaker ablation (group 2) was defined by a reduction of intrinsic heart rate of >50% with a low atrial (inverted P waves in inferior leads II, III, and aVF) or a junctional (P waves merged with the QRS complex) escape rhythm (Fig 1⇑, bottom).
Surgical Technique and Recordings
Animals were premedicated with Innovar-vet (0.1 mL/kg SC) and anesthetized with sodium pentobarbital (7 to 15 mg/kg IV). After intubation, animals were ventilated with room air, and the level of anesthesia was continuously monitored. Body temperature was maintained at 37°C with a heating mattress.
Venous access was obtained by cutdown technique over both femoral veins and over the right external jugular vein. Each vein was cannulated with appropriately sized hemostatic sheaths. Throughout the study, three surface ECG leads (I, II, and III) and five bipolar electrograms recorded from the decapolar catheter were displayed simultaneously on a multichannel oscilloscope. The distal bipolar recording from the ablation catheter was also displayed during endocardial mapping. Femoral arterial blood pressure was continuously monitored throughout the procedure. Recording was performed with an Electronics for Medicine recorder (PPG Inc) at a paper speed of 100 mm/s during autonomic testing and at 150 mm/s during activation mapping. A 12-lead ECG was recorded before and after each pharmacological manipulation at a paper speed of 50 mm/s.
The intracardiac imaging system has been described previously.7 The imaging catheter was introduced via a 10F sheath in the right external jugular vein and positioned at the junction of the superior vena cava and right atrial appendage to obtain optimal images of the crista terminalis. When necessary, the ultrasound catheter was advanced into the body of the right atrium to obtain images from more caudal aspects of the crista terminalis and recorded on super-VHS videotape for review.
Electrophysiological and Pharmacological Evaluation
In both group 1 and group 2 dogs, electrophysiological and pharmacological evaluations of the sinus node pacemakers were performed under general anesthesia before radiofrequency ablation and at a follow-up study 14 days later. In addition, 3 group 1 animals underwent repeat electrophysiological and pharmacological evaluation at ≈3-month follow-up to evaluate the long-term effects of sinus pacemaker modification.
A 6F quadripolar catheter (0.5-cm interelectrode spacing) was positioned in the right atrial appendage for pacing. A 7F decapolar catheter (Elecath, 2-mm bipole spacing, 10-mm interbipole spacing) was inserted via the femoral vein and positioned with the tip located in the superior vena cava. The second bipole (electrodes 3 and 4) was positioned at the superior vena caval–right atrial junction (Fig 2⇓). In this manner, the decapolar catheter lay approximately parallel to the crista terminalis. Positioning was confirmed with intracardiac echocardiography, which was used to accurately define the superior vena cava–right atrial junction and was reconfirmed throughout the experimental procedure by use of both echocardiographic and fluoroscopic landmarks. Both intracardiac echocardiography and fluoroscopy were used in positioning of the decapolar catheter at the follow-up study to duplicate its location at the initial study. Bipolar electrograms were recorded from each electrode pair and displayed throughout the study. Simultaneous atrial activation sequence mapping from the decapolar bipoles was performed relative to the surface P wave recorded in three ECG leads. In cases in which activation times appeared to be simultaneous, the initial atrial activation vector was analyzed, with the point of electrogram polarity reversal8 noted to assist with definition of the pacemaker location.
Autonomic function testing with evaluation of heart rate, blood pressure, and site of earliest atrial activation on the decapolar catheter was then performed, both before ablation and at the follow-up study, in the following sequence, with time allowed for return to baseline after each intervention: (1) An intravenous infusion of nitroprusside in three incremental doses (1, 2, and 4 μg · kg−1 · min−1) was administered to evaluate the response of sinus node pacemaker cells to reflex sympathetic activation. (2) An intravenous infusion of isoproterenol in six incremental doses ranging between 1 and 6 μg/min was administered to evaluate the response of sinus node pacemaker cells to direct sympathetic stimulation. (3) Two incremental bolus doses of phenylephrine (0.2 and 0.4 mg) were administered to evaluate the pacemaker response during vagotonia. (4) Sequential intravenous administration of atropine (0.5 mg/kg) followed by propranolol (1.0 mg/kg) was performed to evaluate the maximum heart rate after vagolysis and then the intrinsic heart rate during total autonomic blockade.
After completion of autonomic function testing (with total autonomic blockade established), the maximal sinus node recovery time and maximal corrected sinus node recovery time were evaluated by right atrial pacing for periods of 30 seconds at decreasing pacing cycle lengths (decrements of 20 ms) commencing 20% below resting.
Radiofrequency catheter ablation was performed during autonomic blockade to allow evaluation of the effects of each radiofrequency application on heart rate without the variations produced by autonomic innervation. Ablation was performed with a 7F catheter with an 8F, custom-designed, 10-mm distal ablation electrode with a thermistor tip (EP Technologies). On the basis of previous studies,6 9 we assumed that the area of ablation required to produce modification of sinus pacemaker function with changes in maximal and average heart rate of ≈30% would be between 2 and 3 cm2. This is a much greater area of tissue desiccation than that provided by conventional ablation catheters with a 4- or 5-mm tip. If the power output is increased to maintain sufficient current density at the distal electrode, then lesion size will increase in proportion to electrode surface area.10
Radiofrequency energy (500 kHz) was delivered between the distal electrode and a large surface area backplate from a generator providing continuous closed-loop temperature feedback control of energy output (EP Technologies). This was programmed to adjust power to achieve an electrode tip and tissue temperature of 70°C. Energy was limited to a maximum output of 100 W and was delivered for a period of 60 seconds. Electrode temperature, power, and impedance were monitored during each application. If a temperature of 50°C was not achieved within 15 to 20 seconds, endocardial contact was considered inadequate and energy application was discontinued.
Intracardiac echocardiography was used to accurately position the ablation electrode tip, marked by a characteristic fan-shaped artifact,7 on the crista terminalis at the lateral junction of the superior vena cava with the right atrial appendage. The catheter tip was then moved in the anterosuperior or inferior direction along this ridge to locate the site of earliest endocardial activation as assessed by recording from the distal ablation bipole relative to the surface P wave in three ECG surface leads. Activation mapping was also performed from other right atrial sites to confirm that the earliest activation times were located on the crista terminalis. During each energy application, catheter tip position was monitored continuously with intracardiac echocardiography. After each application, a shift in the pacemaker site was detected by a change in the endocardial activation sequence and/or from a change in the surface ECG P-wave morphology. Right atrial endocardial mapping was repeated after each application.
In both group 1 and group 2 dogs, 24-hour Holter monitoring was performed for 2 consecutive days before the ablation procedure and for 2 consecutive days after the procedure in the second postoperative week. In three group 1 animals, Holter monitoring was also performed for 2 consecutive days before the long-term follow-up study. The average and maximum heart rates were determined for each 1-hour period and for the full 24 hours by a Zymed Holter analyzer. Assessment was also made of the duration of time for which the heart rate was >110 beats per minute (prospectively defined as tachycardia) during the 24-hour period. The analysis of minimum heart rate and development of pauses is complicated by the fact that normal dogs may demonstrate pauses of >2 seconds at times of high vagal tone. Therefore, we evaluated the maximal pause duration and the number of pauses of >2 seconds in a 24-hour period.
All hearts were fixed in 10% buffered formalin. Before dissection, lesions created by the ablation were identified on the epicardial aspect and measured. The chambers were opened in standard fashion along the lines of blood flow. The thickness of the damage within the atrial wall, as well as the endocardial extent, was also measured. Representative full-thickness pieces of the heart across the junction between the superior vena cava and right atrium laterally (both at the lesion and more caudally where the myocardium appeared normal) were taken and submitted for histology. For localization of sympathetic nerves, a block of fresh tissue from the same area was fixed in 4% buffered paraformaldehyde for 4 hours and then kept in 30% sucrose buffer until the immunohistochemical procedure, performed on floating thick sections according to previously described methods.11
All results are given as mean±SD in the text and as mean±SEM in the figures. Statistical analyses were made using a paired t test for two-group comparisons and with ANOVA for comparisons involving three or more groups. Between-group analysis was by Scheffé’s F test. Statistical significance was accepted at P<.05. In all figures, the preablation data from animals in groups 1 and 2 are pooled for the sake of visual clarity. There were no significant differences between the two groups in any of the preablation parameters.
Preablation Study: Pharmacological Evaluation (Groups 1 and 2)
There was no statistically significant difference in heart rate, blood pressure, or decapolar activation sequence between group 1 and group 2 dogs during any of the preablation autonomic interventions, and these parameters are presented as pooled data in this section.
Baseline heart rate (71.1±19.1 beats per minute) showed considerable beat-to-beat variation in cycle length, consistent with vagal predominance. At this heart rate, the site of earliest activation in the decapolar catheter was usually located between the second and third bipoles, and spontaneous shifts frequently occurred. At this time, P-wave morphology in the 12-lead ECG typically demonstrated low amplitude in the inferior leads.
With incremental infusion of nitroprusside, there was a shift in site of earliest activation on the decapolar catheter to a more superior location (between the first and second bipoles; Fig 3⇓) and a peaking of P waves in inferior ECG leads. Blood pressure decreased significantly (20.2±8.0%, P<.0001) and was accompanied by a progressive increase in heart rate (maximum 93±43% increase, P<.0001; Fig 3⇓). The effects of isoproterenol infusion on heart rate, activation sequence, and P-wave morphology were similar to those observed with nitroprusside (215±79% increase in maximal rate, P<.0001; Fig 3⇓). In contrast, phenylephrine produced an increase in blood pressure (19.1±7.5% increase, P<.0001) accompanied by a significant decrease in heart rate (34±11% reduction, P<.0001, Fig 3⇓). Earliest activation shifted in the inferior direction toward the region of the third bipole (Fig 3⇓), and P-wave morphology was of low amplitude in the inferior ECG leads. After autonomic blockade, the earliest activation site was again shifted in the superior direction (Fig 3⇓), with a significant increase in heart rate compared with baseline (154.5±36.8 versus 71.1±19.1 beats per minute, P<.0001), confirming the marked vagal predominance in the anesthetized animals. An example of the effects of autonomic interventions on the decapolar activation sequence and P-wave morphology is shown in Fig 4⇓.
Ablation Procedure (Groups 1 and 2)
In all dogs, the initial targeted site during ablation was on the crista terminalis at the lateral junction of the superior vena cava with the right atrial appendage. This was accurately localized with intracardiac echocardiography (Fig 5⇓), and earliest activation in this region after complete autonomic blockade was confirmed with right atrial endocardial mapping in all animals. A general correspondence between decreasing heart rate and progressively inferior pacemaker activity was noted. However, this was not a simple linear relation. To achieve the first decrease in heart rate, it was necessary to apply a mean of 3.8±1.8 energy applications in the region of the lateral junction of the superior vena cava with the right atrial appendage in group 1 animals and 4.2±2.2 energy applications in group 2 animals (P=NS). In two group 1 animals, it was also necessary to extend the lesion medially along the crista terminalis, anterior to the superior vena cava. Successful ablation of cranial pacemaking groups was marked by a variable initial reduction in heart rate associated with a shift of the pacemaker in the inferior direction along the crista terminalis. To achieve the target heart rate reduction for animals in each of the two groups, further energy applications needed to be delivered for a variable distance in the inferior direction along the crista terminalis. This distance was ≈2 to 3 cm for sinus pacemaker modification and 3 to 4 cm (virtually extending to the junction of the right atrium with the inferior vena cava) for total sinus pacemaker ablation.
In group 1 dogs, after sinus pacemaker modification, earliest endocardial activation was at the crista terminalis inferior to the radiofrequency lesion in the vicinity of the third and fourth decapolar bipoles (Figs 3⇑ and 4D⇑). With this shift, there was also a flattening of P waves in the inferior ECG leads. In no group 1 animals did an atrioventricular junctional rhythm emerge. The intrinsic heart rate decreased from an initial 154.5±36.8 to 107.0±26.2 beats per minute (30.5±11.6% reduction, P<.001) immediately after the ablative procedure and to an intrinsic heart rate of 94.7±23.2 beats per minute (38.7±12.0% reduction, P<.001) at the 2-week follow-up study (Fig 6⇓). The difference between the intrinsic heart rate immediately after the procedure with that at the follow-up study was not statistically significant (P=NS, Scheffé’s F test). The narrow prospective definition of heart rate reduction (30±5%) was achieved in 6 of the 11 animals undergoing sinus pacemaker modification (31.6±4.3% reduction immediately and 38.6±10.1% reduction at the 2-week follow-up study; by ANOVA, P<.001 compared with baseline). In the other 5 animals, there was also a highly significant reduction in intrinsic heart rate, although results were more variable (28.0±17.3% reduction immediately and 37.6±15.1% reduction at the 2-week follow-up study; by ANOVA, P<.005 compared with baseline). These more variable results occurred in the initial animals during development of the technique. The prospective heart rate definition was achieved in 5 of the last 6 consecutive animals in which sinus pacemaker modification was attempted.
In group 2 dogs, after total ablation, the rhythm was either junctional (n=2) or arose from the vicinity of the low right atrium (n=2) and was associated with deeply inverted P waves in the inferior ECG leads. The prospective definition of total ablation was met in all four animals. The intrinsic heart rate decreased from 161.0±6.3 beats per minute initially to 68.0±11.5 beats per minute immediately after the procedure (57.8±7.8% reduction) and to 56.0±20.5 beats per minute (65.3±12.1% reduction) at the 2-week follow-up study (P<.001, ANOVA; Fig 6⇑). As in group 1, the difference between intrinsic heart rate immediately after the procedure and that at the 2-week follow-up study was not statistically significant (P>.05, Scheffé’s F test).
In no instance in group 1 or 2 animals was it necessary to apply radiofrequency energy to sites not located on the crista terminalis to achieve the required heart rate reduction.
The mean number of radiofrequency applications required was 6.2±3.2 in group 1 (range, 2 to 14) and 9.8±3.0 in group 2 (range, 7 to 14, P=.01). In none of the group 2 animals was a total ablation achieved without the criteria for modification having first been observed on a preceding application. In these group 2 animals, the mean additional number of radiofrequency applications required to achieve a total ablation after a sinus pacemaker modification had been observed was 2.8±0.5 applications (range, 2 to 4). An initial increase in sinus rate occurred during 81% of radiofrequency energy applications. This was followed by a decrease in heart rate during energy delivery in 23% of applications. The mean activation time for radiofrequency applications that caused heart rate reduction was −21±9 ms before onset of the surface P wave.
Intracardiac Echocardiographic Observations
Intracardiac echocardiography clearly demonstrated the prominent ridge of the crista terminalis at the junction of the superior vena cava and right atrial appendage in all animals (Fig 5⇑). This ridge became less prominent as it extended in the inferior direction along the posterolateral right atrium but could still be clearly defined as the junction of the posterior smooth-walled atrium with the anterior trabeculated atrium. Direct imaging of the electrode catheter tip–tissue contact was successful in 93% of energy applications. Interference of the imaging catheter with the ablation catheter necessitating removal of the imaging catheter did not occur in any animals.
The site of lesion formation could be detected after most radiofrequency applications by the appearance of local irregular swelling that produced ultrasound shadowing artifact (Fig 5⇑). Subsequent pathological examination demonstrated superficial adherent thrombus at the endocardial lesion site that was presumed to be partly responsible for the swelling and ultrasound artifact observed (Fig 7⇓).
Group 1. In group 1 dogs, after sinus pacemaker modification, the heart rate remained responsive to both reflex-mediated and direct sympathetic activation but with the peak attainable heart rate significantly reduced. Nitroprusside infusion caused a decrease in blood pressure similar to that at the initial study but with a reflex-mediated increase in heart rate at both 2 and 4 μg · kg−1 · min−1, which was significantly less than before ablation (24±21% reduction, P=.04, and 27±18% reduction, P=.006, respectively; Fig 3⇑). Similarly, although heart rate remained responsive to isoproterenol infusion, the maximal attainable rate was significantly less than that observed before sinus pacemaker modification (30±9% reduction, P<.0001, Fig 3⇑).
Right atrial activation mapping in the baseline state demonstrated that earliest endocardial activation occurred on the crista terminalis in the region of the third and fourth bipolar pairs of the decapolar catheter in all group 1 animals (Fig 3⇑). With sympathetic activation (either direct or reflex mediated), in contrast to before ablation, there was no cranial shift along the crista terminalis in the site of earliest activation.
Phenylephrine caused an increase in systolic blood pressure associated with a decrease in heart rate that was not significantly different from the premodification response (Fig 3⇑). In no instance did sinus pauses or sinus arrest occur during administration of phenylephrine in group 1 animals. The site of earliest endocardial activation on the crista terminalis was similar to the preablation site after administration of phenylephrine (Fig 3⇑).
During complete autonomic blockade, intrinsic heart rate (94.7±23.2 beats per minute) was significantly greater than the baseline rate (62.8±18.1 beats per minute, P=.01) at the commencement of the follow-up study. However, the magnitude of the increase (50±16%) was significantly attenuated compared with the preablation increase (145±26%, P<.01).
Group 2. All group 2 dogs showed a return of some periods of sinus rhythm with a normal P-wave axis at the follow-up study. The pacemaker site was usually located on the crista terminalis in the vicinity of the third or fourth decapolar bipole. However, this rhythm was inherently unstable, with pacemaker shift, most frequently to a low atrial or a junctional site, occurring either spontaneously or in response to autonomic interventions (Fig 8⇓). Although heart rate response to either nitroprusside or isoproterenol was not significantly different from that in group 1 animals (Fig 3⇑), this was due to responsiveness of a junctional or low atrial focus leading to sympathetically mediated junctional or atrial tachycardia (Fig 8⇓). In two group 2 animals, phenylephrine produced prolonged pauses during the follow-up study.
At the follow-up study, intrinsic heart rate (56.0±10.3 beats per minute) did not increase significantly from the baseline heart rate (60.3±12.7 beats per minute). The intrinsic heart rate at follow-up study was significantly slower in group 2 than in group 1 animals (P<.05).
Sinus Node Recovery Time
Group 1. In group 1, maximum sinus node recovery time at the follow-up study (939.0±280.2 ms) was significantly longer than the maximum preablation recovery time (579.5±123.6 ms, P=.001). However, when expressed as a percentage of the resting cycle length, there was no significant difference (135.7% and 134.7%, respectively, P=NS).
Group 2. The maximum preablation sinus node recovery time in group 2 animals (541.7±86.9 ms) was not significantly different from that in group 1 (579.5±123.6 ms, P=NS). However, after total ablation, prolonged postpacing pauses (range, 5.5 to 8.4 seconds; mean, 6.8±0.7 seconds) occurred in all group 2 animals.
Holter Monitoring (Groups 1 and 2)
Holter monitoring data from dogs in both groups are presented in Table 1⇓. In both groups, there were highly significant reductions in maximum and average heart rates and also in time spent in tachycardia. A comparison of Holter parameters after total ablation in group 2 animals with parameters after sinus pacemaker modification in group 1 animals demonstrated a significantly shorter time spent in tachycardia (P=.003) and a nonsignificant trend toward a lower maximum heart rate (P=.07) in the group 2 animals. There was no difference in average heart rate. In group 1 animals, there was no significant increase in the maximal pause duration after sinus pacemaker modification (3.1±1.2 seconds premodification versus 3.8±1.0 seconds postmodification; P=NS), in the number of pauses of >2 seconds in duration during the 24-hour period of monitoring (7.2±5.6 versus 10.1±10.1, respectively, P=NS), or in the minimum heart rate (32.7±19.3 versus 28.7±13.4 beats per minute, respectively, P=NS). Holter recordings from group 2 animals were characterized by prolonged episodes of junctional rhythm and frequent changes in the P-wave morphology. In addition, group 2 animals demonstrated a significant increase in both the maximal pause duration (2.7±0.6 seconds preablation versus 5.0±0.5 seconds postablation; P<.01) and in the number of pauses of >2 seconds in the 24-hour period of monitoring (4.8±2.1 versus 16.5±7.8, P<.05). None of the animals in either group died suddenly during the follow-up period.
In the three group 1 animals followed for a mean period of 3.7±1.0 months, there were no significant differences at the late study compared with the 2-week follow-up study for intrinsic heart rate, maximal heart rate on isoproterenol, maximal heart rate or average heart rate on 24-hour Holter monitor, or time spent in tachycardia on 24-hour Holter monitor. Intrinsic heart rate was decreased by 39% at 2 weeks and by 35% at the long-term follow-up study (P=NS). Maximal heart rate on isoproterenol was decreased by 31% at 2 weeks and by 36% at the long-term follow-up study (P=NS). Maximal heart rate on Holter monitor was decreased by 27% at 2 weeks and by 25% at the long-term follow-up study (P=NS). Average heart rate on Holter monitor was decreased by 17% at 2 weeks and 19% at the long-term follow-up study (P=NS). Time in tachycardia on 24-hour Holter monitor was decreased by 80% at 2 weeks and 88% at the long-term follow-up study (P=NS).
All three animals demonstrated a stable sinus pacemaker P-wave morphology, with earliest atrial activation mapped to the mid–crista terminalis (earliest activation at the third [n=1] or fourth [n=2] bipole of the decapolar catheter at the long-term follow-up study), and overdrive pacing did not cause abnormal sinus node pauses. There was no cranial shift of the activation focus with sympathetic activation.
In all dogs from groups 1 and 2, pathological examination demonstrated that lesions were accurately located at the lateral right atrial–superior vena caval junction and extended a variable distance along the crista terminalis (Fig 9⇓). In the two animals that required radiofrequency applications to be extended medially along the crista terminalis, lesions in this location were demonstrated at pathological examination.
Table 2⇓ demonstrates a comparison of lesion size in group 1 versus group 2 animals. The mean endocardial lesion length in group 2 animals was significantly greater than in group 1 animals, but there was no significant difference in maximal lesion width.
In addition, pathological examination demonstrated the accuracy of lesion localization (Fig 9⇑). The maximal distance from the edge of the crista terminalis that lesions were found was 0.8±0.2 cm.
Histological examination confirmed that radiofrequency lesions were transmural. No residual sinus node pacemaker cells were seen in any sections from the right atrial–superior vena caval junction (Fig 7⇑). In addition, none of the sections taken through the crista terminalis at sites inferior to the lesion demonstrated typical sinus node pacemaker cells. Neural staining did not demonstrate any sympathetic nerve terminals in the lesion area, although these were present in normal appearance from sections taken immediately inferior to the lesion (Fig 7⇑). Histological evaluation in the 3 animals undergoing long-term follow-up also demonstrated that lesions were transmural with fibrosis and accurately targeted to the crista terminalis. In 2 of these animals, there was some cartilaginous metaplasia.
Our results demonstrate the feasibility of modification of sinus pacemaker function for sinus rate control using catheter-based radiofrequency ablation guided by intracardiac echocardiography. This can be done while pacemaker stability and responsiveness to autonomic influences are preserved. The effects of sinus pacemaker modification were maintained during long-term follow-up of 3 months.
Anatomic and Physiological Considerations
Using endocardial mapping, we were able to demonstrate that under autonomic influences, the origin of the dominant sinus pacemaker group may be shifted over a distance of up to 3 cm along the region of the sulcus terminalis. Sympathetic activation caused a shift to a cranial pacemaker location, usually at the lateral junction of the superior vena cava with the right atrial appendage but also extending medially along the crista terminalis in two cases. Reflex vagal stimulation caused a shift of the pacemaker site to a more caudal location on the crista terminalis. The demonstration of shifts in origin of the sinus node impulse along the sulcus terminalis under autonomic influences was first observed by Lewis et al12 and by Meek and Eyster13 early in the century and has since been reported by other investigators using either mapping or microelectrode techniques.14 15 16 17 Most recently, Boineau et al5 18 used a computerized epicardial mapping system to show that over a physiological range of spontaneous heart rates, the dominant pacemaker may occur over a wide distribution, as far cranially as the right atrial–superior vena caval junction or as far caudally as the right atrial–inferior vena caval junction. These sites of origin were centered about the long axis of the sulcus terminalis and, as in our study, produced a P-wave axis on the surface ECG within the normal sinus spectrum. In addition, in response to autonomic manipulations, a close correspondence existed between the heart rate and the site of impulse origin within the sulcus terminalis, consistent with a graduated site-specific differential sensitivity to autonomic inputs.4 5 19 Vagal stimulation produced an increase in cycle length associated with change in location of the dominant pacemaker to a lower atrial site. Isoproterenol infusion resulted in dominance of a more cranial and anterior pacemaker with associated decrease in cycle length. The existence of a widely distributed pacemaker complex initially shown in dogs has more recently also been demonstrated in humans.20
This widely distributed physiological pacemaker complex contrasts with the more localized and constant anatomic location of the histologically defined sinus node.3 21 The human sinus node lies immediately beneath the epicardium within the sulcus terminalis of the right atrium at the junction of the anterior trabeculated appendage with the posterior smooth-walled venous component.3 21 The endocardial aspect of the sulcus terminalis is marked by the crista terminalis. In most cases, the node lies lateral to the crest of the atrial appendage, and specialized cells extend in the inferior direction in the sulcus terminalis for ≈10 mm. In up to 10% of cases, the node extends across the crest of the appendage anterior to the superior vena cava.3 In most respects, the anatomy of the sinus node of the dog is very similar to that of the sinus node of humans.1 2 In our study and those of Boineau et al,5 19 the spatial distribution of the physiologically defined pacemaker complex exceeded the dimensions of the histologically defined sinus node by a factor of 3 to 4.
Sinus Pacemaker Modification
After selective ablation of the cranial pacemakers, the sinus node pacemaker was located in a more caudal position along the crista terminalis. With adrenergic stimulation, the heart rate was significantly less responsive than before ablation and was no longer accompanied by cranial movement of pacemaker origin. The pacemaker cells at this site were stable, and the relation between sinus node recovery time and prepacing cycle length was unchanged from the preablation study. Changes in heart rate parameters during autonomic manipulation were also paralleled by similar reductions in ambulatory heart rate indexes. The feasibility of modification of sinus pacemaker function was first demonstrated in dogs by surgical excision.9 22 23 24 Randall et al9 23 found that surgical modification of sinus pacemaker function required excision of a block of tissue ≈2.0 cm long by 1.5 cm wide, dimensions similar to those required in our study. The escape rhythm was localized to the sulcus terminalis at the caudal end of the excised area. This rhythm was considered to arise from subsidiary atrial pacemakers that were shown to be stable and responsive to autonomic maneuvers but with a significant reduction in the maximal attainable heart rate. The description of these pacemakers as “subsidiary atrial” is consistent with histological studies showing that the specialized sinus node pacemaker tissue is relatively localized.3 21 In our study, histological sections taken through the crista terminalis immediately inferior to the ablative lesion at the presumed site of postablation pacemaking failed to demonstrate the characteristic histological appearance of the sinus node in any animal. However, this does not preclude cells in this region from having pacemaker function.
Recently, the feasibility of modification of sinus pacemaker function with epicardial laser irradiation in open-chest dogs has been demonstrated.6 In this study, epicardial laser photocoagulation was applied to the earliest site of activation (defined by epicardial mapping) during isoproterenol infusion and repeated to remapped activation points until a 30±5% decrease in heart rate occurred. After ablation, the rhythm was shown to arise from a site within the sulcus terminalis immediately caudal to the ablative lesion. Pacemaker cells in this region produced a stable rhythm responsive to autonomic inputs but with a significant reduction in maximal, average, and intrinsic heart rates of a magnitude similar to those we observed. In contrast to Randall et al,9 the authors of Reference 6 characterized the pacemaker cells in this region of the sulcus terminalis as part of the sinus node pacemaker complex. This characterization is well supported by the physiological studies of Boineau et al4 5 and by our observations of changes in sinus node pacemaker origin in the present study.
In all 11 animals in group 1, we were able to achieve sinus pacemaker modification without inadvertently causing a total sinus node ablation. In 6 animals, the narrow prospective intrinsic heart rate definition was achieved. In 5 others, there was more variability, although in these 5, intrinsic, maximal, and average heart rates were all significantly decreased without a total sinus node ablation being produced. This variability reflected the learning curve during development of the technique. The effects of sinus pacemaker modification were maintained during long-term follow-up of up to 3 months.
Total Ablation of the Sinus Pacemaker
In this study, total ablation of sinus pacemaker function required creation of an extensive lesion extending along the length of the crista terminalis from the right atrial junction with the superior vena cava caudally to its junction with the inferior vena cava. Both the number of radiofrequency energy applications and the pathological lesion size were significantly greater than those required for sinus pacemaker modification. The escape pacemaker was initially located in the junctional or low right atrial region, and atrial overdrive pacing produced prolonged pauses. At the follow-up study 2 weeks later, the rhythm had returned to the caudal aspect of the crista terminalis near the inferior vena caval junction, but it was unstable, and pacemaker shift occurred spontaneously and in response to autonomic manipulations. Euler et al25 demonstrated that complete removal of the sinus node and crista terminalis in dogs required excision of a block of tissue ≈4.0 cm long by 1.5 cm wide, dimensions similar to those required for total ablation in this study. The characteristics of the escape pacemaker were also similar.
Modification Versus Total Ablation: Characteristics of the Escape Pacemaker
Despite the difference in lesion extent between animals in the two groups, there was no significant difference in heart rate response to sympathetic stimulation or in maximum or average heart rate on ambulatory monitoring. Although the sinus pacemaker was unable to respond to adrenergic stimulation in group 2 animals, junctional and low atrial pacemakers remained responsive. The implication is that the degree of heart rate control achievable with sinus pacemaker ablation may be limited by the sympathetic responsiveness of junctional (or other atrial) foci and is likely to be on the order of 20% to 30%, a rate reduction similar to that achieved with β-adrenergic blockade. Therefore, substantial additional rate control may not be achieved by total ablation rather than a more limited modification. Differentiation between these two outcomes is of obvious importance, in view of the noted difference in stability of the escape rhythm.
Ablation of Pacemaker Tissue or of Sympathetic Inputs?
The sinus node is richly innervated with both sympathetic and parasympathetic nerve fibers. Recent evidence has suggested that cardiac autonomic neural inputs may be modified by radiofrequency lesions, with consequent effects on sinus rate.26 Thus, it is feasible that changes in sinus node function as a result of application of radiofrequency energy may result from destruction of pacemaker tissue, of sympathetic inputs, or of both. Evidence from our study suggests that both effects are likely to have occurred. First, modification in sympathetic responsiveness was demonstrated in response to both direct and reflex-mediated sympathetic activation. Second, lesions produced by radiofrequency application via a 10-mm electrode tip involved the full thickness of the atrial wall, and no residual pacemaker tissue was demonstrated in any animal. Finally, histological examination using special staining techniques to highlight sympathetic nerve terminals failed to demonstrate any sympathetic innervation in the lesion area.
Use of Intracardiac Echocardiography
In a previous study of total sinus node ablation using radiofrequency energy, anatomic localization was performed with fluoroscopic guidance alone.27 Successful ablation was achieved in 8 of 10 animals and required a mean of >16 applications. Pathological evaluation demonstrated that radiofrequency energy was not well targeted, with some noncontiguous lesions being located in the right atrial appendage, the superior vena cava, and the right atrial free wall. We previously demonstrated that intracardiac echocardiography can be used to accurately localize the ablation electrode tip within the right atrium and can guide radiofrequency lesion placement to within a mean of 1.9 mm of the designated site.7 The anatomic location of the sinus pacemaker is clearly defined by the crista terminalis, and in all cases, pathological examination confirmed that intracardiac echocardiography provided accurate anatomic localization. In addition, the area of ablation at pathological examination in both group 1 and group 2 animals was very similar to that previously demonstrated in surgical studies.9 25 This suggests that unnecessary radiofrequency lesion application was limited. In no case was it necessary to apply radiofrequency energy away from this anatomically defined ridge to achieve the result of modification of sinus pacemaker function.
It is also worth noting that inadvertent conversion of a successful sinus pacemaker modification to a total sinus pacemaker ablation with an unstable escape rhythm (necessitating permanent pacemaker implantation) may require only several poorly targeted lesions. Fluoroscopy does not allow accurate visualization of the crista terminalis.
Since our primary aim was to demonstrate the feasibility of sinus pacemaker modification by use of radiofrequency energy, we did not compare intracardiac imaging with standard fluoroscopy for anatomic localization. Therefore, the notion that intracardiac imaging provides better anatomic localization of the sinus pacemaker than fluoroscopy, thereby allowing a reduction in procedure time, fluoroscopy time, and number of energy applications, must still be considered speculative.
Extensive lesion creation in the atrium could potentially increase the risk of development of atrial arrhythmias in the long term. Indeed, this may potentially occur with any ablative procedure in the atrium. However, if the lesion is extended to the junction of the superior vena cava and if care is taken, by use of intracardiac echocardiography, to make a continuous lesion, the risk of reentry occurring around the lesion is in theory reduced.
We used the decapolar catheter to evaluate changes in pacemaker location and supplemented this with endocardial activation mapping using the ablation catheter. A more accurate evaluation of the earliest activation site might have been facilitated by use of a multipolar basket catheter.28 In addition, we did not perform mapping of the left atrium.
Potential Clinical Implications
If the results of our animal study can be duplicated in patients, the potential clinical implications of modification of sinus pacemaker function with radiofrequency energy might include (1) control of heart rate in patients with the syndrome of inappropriate sinus tachycardia29 and (2) nonpharmacological heart rate control in patients with intractable angina intolerant to medications. If average heart rate is reduced and rate response to exertion is blunted, myocardial oxygen requirement would be expected to decrease. This may also be applicable in selected patients with heart failure in whom excessive sinus tachycardia appears to be detrimental to hemodynamics and β-adrenergic blockade is not tolerated.30
This study demonstrates the feasibility of modification of sinus pacemaker function for sinus rate control using catheter-based radiofrequency ablation guided by intracardiac echocardiography. This can be done while pacemaker stability and responsiveness to autonomic influences are preserved. If these results can be duplicated in patients, this may represent a useful nonpharmacological method of achieving sinus rate control in a spectrum of conditions in which the sinus rate is inappropriately rapid.
Intracardiac echocardiography accurately defined the crista terminalis and provided a reliable means to anatomically localize catheter position in relation to the sinus node.
Dr Kalman was funded as the Ralph Reader Overseas Research Fellow of the National Heart Foundation of Australia and was the recipient of a Telectronics traveling grant of the Royal Australasian College of Physicians. Dr Fisher was funded by the US Navy. We gratefully acknowledge Margret Mayes for excellent technical assistance and Zaida Rizzo for expert help with Holter analysis.
- Received June 7, 1994.
- Revision received June 5, 1995.
- Accepted June 23, 1995.
- Copyright © 1995 by American Heart Association
James TN, Sherf L, Fine G, Morales AR. Comparative ultrastructure of the sinus node in man and dog. Circulation. 1966;34:139-163.
Anderson KR, Ho SY, Anderson RH. Location and vascular supply of sinus node in human heart. Br Heart J. 1979;41:28-32.
Boineau JP, Schuessler RB, Cain ME, Corr PB, Cox JL. Activation mapping during normal atrial rhythms and atrial flutter. In: Zipes DP, Jalife J, eds. Cardiac Electrophysiology: From Cell to Bedside. Philadelphia, Pa: WB Saunders Co; 1990:537-547.
Boineau JP, Schuessler RB, Hackel DB, Miller CB, Brockus CW, Wylds AC. Widespread distribution and rate differentiation of the atrial pacemaker complex. Am J Physiol. 1980;239:H406-H415.
Littmann L, Svenson RH, Gallagher JJ, Bharati S, Lev M, Linder KD, Tatsis GP, Nichelson C. Modification of sinus node function by epicardial laser irradiation in dogs. Circulation. 1990;81:350-359.
Chu E, Fitzpatrick AP, Chin MC, Sudhir K, Yock PG, Lesh MD. Radiofrequency catheter ablation guided by intracardiac echocardiography. Circulation. 1994;89:1301-1305.
Randall WC, Rinkema LE, Jones SB, Moran JF, Brynjolfsson G. Functional characterization of atrial pacemaker activity. Am J Physiol. 1982;242:H98-H106.
Langberg JJ, Gallagher M, Strickberger SA, Amirana O. Temperature-guided radiofrequency catheter ablation with very large distal electrodes. Circulation. 1993;88:245-249.
Lewis T, Oppenheimer BS, Oppenheimer A. The site of origin of the mammalian heart beat: the pacemaker in the dog. Heart. 1911;2:147-169.
Meek WJ, Eyster JAE. Experiments on the origin and propagation of the impulse in the heart, IV: the effect of vagal stimulation and of cooling on the location of the pacemaker within the sino-auricular node. Am J Physiol. 1914;34:368-383.
Geesbreght JM, Randall WC. Area localization of shifting cardiac pacemakers during sympathetic stimulation. Am J Physiol. 1971;220:1522-1527.
Goldberg JM. Intra SA nodal pacemaker shifts induced by autonomic stimulation in the dog. Am J Physiol. 1975;229:1116-1123.
Boineau JP, Schuessler RB, Mooney CR, Wylds AC, Miller CB, Hudson RD, Borremans JM, Brockus CW. Multicentric origin of the atrial depolarization wave: the pacemaker complex: relation to dynamics of atrial conduction, P-wave changes and heart rate control. Circulation. 1978;58:1036-1048.
Boineau JP, Schuessler RB, Roeske WR, Autry LJ, Miller CB, Wylds AC. Quantitative relation between sites of atrial impulse origin and cycle length. Am J Physiol. 1983;245:H781-H789.
Boineau JP, Canavan TE, Schuessler RB, Cain ME, Corr PB, Cox JL. Demonstration of a widely distributed atrial pacemaker complex in the human heart. Circulation. 1988;77:1221-1237.
Randall WC, Talano J, Kaye MP, Euler D, Jones S, Brynjolfsson G. Cardiac pacemakers in absence of the SA node: responses to exercise and autonomic blockade. Am J Physiol. 1978;234:H465-H470.
Euler DE, Jones SB, Gunnar WP, Loeb JM, Murdock DK, Randall WC. Cardiac arrhythmias in the conscious dog after excision of the sinoatrial node and crista terminalis. Circulation. 1979;59:468-475.
Kocovic DZ, Harada T, Shea JB, Soroff D, Friedman PL. Alterations of heart rate and of heart rate variability after radiofrequency catheter ablation of supraventricular tachycardia: delineation of parasympathetic pathways in the human heart. Circulation. 1993;88:1671-1681.
Bauernfeind RA, Amat-y-Leon F, Dhingra RC, Kehoe R, Wyndham C, Rosen KM. Chronic nonparoxysmal sinus tachycardia in otherwise healthy persons. Ann Intern Med. 1979;91:702-710.