Inadvertent Atrioventricular Block During Radiofrequency Catheter Ablation
Results of the Pediatric Radiofrequency Ablation Registry
Background Inadvertent atrioventricular block is a complication of radiofrequency ablation. The present study is an analysis of the incidence, significance, and factors associated with inadvertent atrioventricular block during radiofrequency catheter ablation in childhood and adolescence.
Methods and Results The records of the Pediatric Radiofrequency Ablation Registry were reviewed. Between January 1, 1991, and April 1, 1994, atrioventricular block occurred in 23 of 1964 radiofrequency ablations (1.2%): 14 as third-degree block (3 transient) and 9 as second-degree block (5 transient). Atrioventricular block occurred from 5 seconds to 2 months (mean, 4.1 days; median, 15 seconds) after the onset of the energy application. Eight transient cases lasted 1 hour to 1 month (mean, 9.4 days; median, 7 days). Inadvertent atrioventricular block was related to the ablation anatomic site: 3 of 111 (2.7%) anteroseptal, 11 of 106 (10.4%) midseptal, and 2 of 197 (1.0%) right posteroseptal sites (P=.0007 for anteroseptal, P=.0001 for midseptal, and P=.17 for right posteroseptal versus nonright septal sites). Five of 314 (1.6%) ablations for atrioventricular nodal reentrant tachycardia resulted in atrioventricular block (P=.004 versus nonright septal sites). Compared with a matched subgroup, radiofrequency ablation experience was the only significant risk factor (32.7 versus 106.6, P=.002) for the occurrence of atrioventricular block.
Conclusions Inadvertent atrioventricular block may occur during or late after radiofrequency catheter ablation. It is associated with ablations for (1) anterior and midseptal accessory pathways and atrioventricular nodal reentry and (2) relative institutional inexperience.
Radiofrequency catheter ablation has become an accepted and routine treatment option in the management of patients with cardiac arrhythmias. Many centers, including pediatric centers, now consider radiofrequency catheter ablation to be the first-line treatment option for certain arrhythmias.1 However, before the usefulness of ablation can be evaluated and before it can be compared with other antiarrhythmic treatment options, its risks and benefits need to be better defined.
Inadvertent atrioventricular block is a major potential complication of radiofrequency catheter ablation for treatment of supraventricular tachycardia.2 3 4 5 6 7 8 The conduction block may occur during or late9 after the radiofrequency energy application, and the block may be transient.10 In children, this complication may be especially important given the presumed need for permanent cardiac pacing for an expected lifetime of ≥70 years.
The Pediatric Radiofrequency Ablation Registry was established in November 1990 by the Pediatric Electrophysiology Society.11 It is a voluntary registry composed of pediatric patients who underwent radiofrequency catheter ablation at institutions of Pediatric Electrophysiology Society members. The Pediatric Electrophysiology Society member contributes to the Registry by sending the procedure data/results to the coordinating center (University of Nebraska [Omaha]). The acute data are correlated and reviewed with the participating members every 6 months and with the entire Pediatric Electrophysiology Society on a yearly basis. Follow-up data are collected yearly and reviewed with the same groups.
In an attempt to define the incidence and long-term significance of the problem of inadvertent atrioventricular block during radiofrequency catheter ablation in children and the risk factors associated with this complication, the results of the Pediatric Radiofrequency Ablation Registry were reviewed, and they are presented in this report.
Patients were included in this study if they were ≤21 years of age at the time of radiofrequency catheter ablation, according to the definition proposed by the American Academy of Pediatrics.12 From January 1, 1991, through April 1, 1994, 29 centers (see “Appendix”) contributed acute data on 1749 patients with a total of 1964 radiofrequency catheter ablation procedures. All members of the Pediatric Electrophysiology Society were eligible to participate and submit data on children who had undergone radiofrequency catheter ablation. The goal of the Registry was to study all such patients treated at each center; however, no center was excluded because of incomplete participation. Follow-up data were collected on each patient in an attempt to provide information about the status of the acute complications, late complications (ie, complications not detected at the completion of the procedure), recurrence of the initial clinical tachycardia, development of new arrhythmias, and the use of medications and their effectiveness.
Acute results were defined as the status of the patient and the arrhythmia at the termination of the procedure. Second-degree atrioventricular block was defined as Mobitz type I or II second-degree atrioventricular nodal block or higher-grade block with intermittent atrioventricular nodal conduction. Third-degree or complete atrioventricular block was defined as the absence of anterograde atrioventricular nodal conduction. Transient atrioventricular block was defined as atrioventricular block that was present at the end of the procedure that eventually returned to normal. Episodes of transient atrioventricular block that occurred during the procedure but had resolved by the end of the case were not reported in a uniform fashion and could not be used for analysis. Indications for the procedures have been previously defined.11
Analysis of Data
Twenty-three patients were identified from the Registry records as having second- or third-degree atrioventricular block. A detailed questionnaire was then completed by the individual centers regarding the identified cases. Cases were then compared with an arrhythmia mechanism and anatomic radiofrequency energy application site-matched control group (n=25).
Group mean±SD and median values were calculated. Comparisons were made by χ2 analysis with continuity correction according to Yates13 and by unpaired Student's t tests. Differences with a value of P<.05 were considered statistically significant.
Patient characteristics and details of the radiofrequency ablation procedures are presented in Table 1⇓⇓. Inadvertent atrioventricular block occurred in 23 procedures (1.2%): 14 as third-degree block (3 transient) and 9 as second-degree block (5 transient). The difference in the proportion of cases that were transient between third-degree atrioventricular block (3 of 14; 21.4%) and second-degree atrioventricular block (5 of 9; 55.6%) was not significant (P=.22).
Among patients, the mean age was 10.9±5.1 years (range, 9 months to 20.1 years), and the mean weight was 41.3±21.5 kg (range, 8.2 to 87.5 kg). Three patients had heart disease: one congenital ventricular septal defect, one idiopathic dilated cardiomyopathy, and one tachycardia-induced cardiomyopathy. Seventeen patients had orthodromic reciprocating tachycardia, including 3 with persistent junctional reciprocating tachycardia and 2 with multiple accessory pathways. Five patients had atrioventricular nodal reentrant tachycardia, and 1 patient had both orthodromic reciprocating tachycardia and atrioventricular nodal reentrant tachycardia.
Sixteen centers reported 1 to 3 patients with inadvertent atrioventricular block. Four laboratories were dedicated to electrophysiology procedures only, whereas 12 laboratories performed both electrophysiology and cardiac catheterization with angiography. In 6 laboratories, the support staff was dedicated to electrophysiology only, whereas the staff had combined responsibilities in the other 10 laboratories. The centers had previously performed a mean of 32.7±29.2 pediatric ablation procedures (range, 2 to 33 procedures) before the index case resulting in atrioventricular block.
One to three physicians (mean, 2.25±0.72 physicians) participated in each procedure. The primary physicians (those who decided to initiate the application) were board certified for all procedures, pediatric trained for 20 procedures, and internal medicine trained for 3 procedures. The primary physicians had been the primary physician in 60.8±46.8 (range, 12 to 150) previous procedures and assisted in 46.5±40.0 (range, 1 to 100) additional radiofrequency ablations. The assistant (second physician) was board certified for 9 procedures and not board certified for 9 procedures. The second physician was pediatric trained for 10 procedures and internal medicine trained for 8 procedures.
The procedures were performed for life-threatening arrhythmias in 4 patients, for refractory medical management in 7, for an adverse drug reaction in 1, and for patient choice in 11. Fourteen patients received general endotracheal anesthesia, and 9 received intravenous sedation.
The ablation catheters were steerable in all patients. The electrode tips of the catheters were 6F in 6 patients, 7F in 16, and 8F in 1. They were domed in 19 and grooved in 4. No long sheaths were used.
The mean number of radiofrequency energy applications preceding atrioventricular block during the procedure was 10±9.4 (range, 1 to 30). The energy applications that produced or were the last before the development of atrioventricular block lasted 35.3±21.1 seconds (range, 10 to 85 seconds) using 52.3±6.0 V (range, 40 to 60 V) and 30.9±9.1 W (range, 16 to 46 W).
The atrioventricular block occurred from 5 seconds to 2 months (mean, 4.1 days; median, 15 seconds) after the onset of the radiofrequency energy application. The 8 transient cases lasted 1 hour to 1 month (mean, 9.4 days; median, 7 days).
The anatomic site of the radiofrequency energy application was related to the occurrence of atrioventricular block (Fig 1⇓). Inadvertent atrioventricular block was induced in 2 of 1236 (0.2%) nonright septal radiofrequency catheter ablations in the entire registry (one right posterior [second-degree, transient] and one posteroseptal from within the coronary sinus [third-degree, permanent]). The other 21 cases all occurred during applications along the right septal aspect of the tricuspid annulus for accessory atrioventricular connections and/or atrioventricular nodal reentrant tachycardia.
Sixteen procedures to interrupt accessory connections resulted in second- or third-degree atrioventricular block. The inadvertent atrioventricular block occurred in 3 of 111 (2.7%) anteroseptal, 11 of 106 (10.4%) midseptal, and 2 of 197 (1.0%) posteroseptal sites (P=.0007 for anteroseptal, P=.0001 for midseptal, and P=.17 for right posteroseptal versus nonright septal sites).
In the entire registry, 5 of 314 procedures (1.6%) for atrioventricular nodal reentrant tachycardia resulted in atrioventricular block (P=.004 versus nonright septal pathways): 1 (a midseptal application) of 17 patients (5.9%) for fast pathway ablations, and 4 (two midseptal, two anteroseptal applications) of 297 patients (1.3%) for slow pathway ablations (P=.6, fast versus slow approach) (Fig 2⇓).
The efficacy of the procedures (success rate versus the complication rate) is presented in Table 2⇓.
Seventeen patients have been followed for a mean of 9.7±9.9 months (range, 1 to 30 months). Their management and outcomes are shown in Fig 3⇓. Three of the 6 patients who were lost to follow-up received permanent pacemakers after the initial ablation procedure. One patient with dilated cardiomyopathy died 2 days after the ablation. He had been critically ill before the ablation, and it was elected to withhold further therapy.
The comparison of the 23 patients with an arrhythmia-matched population (n=25) is shown in Table 3⇓. Fewer previous procedures had been performed at the institutions when block occurred than had been performed in the control group (32.7 versus 106.6, P=.002). During the follow-up period of the control subjects (1 to 36 months; mean, 12.2 months), 5 of the 25 control subjects (20%) developed a recurrence of tachycardia. All of the recurrences occurred within 1 month of the ablation procedure.
Inadvertent or unintentionally induced atrioventricular block is a major potential complication of radiofrequency catheter ablation. In the present study, the results of the Pediatric Radiofrequency Catheter Ablation Registry were reviewed to help define the incidence of this problem and assess the risk factors associated with inadvertent atrioventricular block. We also sought to define the significance of transient atrioventricular block and develop strategies to prevent unintentional block.
The anatomic site of the application of the radiofrequency energy was clearly related to unintentional atrioventricular block. Only 2 of 1236 (0.2%) of nonright septal applications developed block, whereas 21 of 728 (2.9%) right septal applications produced block. The risk of inadvertent atrioventricular block can be further stratified according to the position within the septal region: 2.7% anteroseptal, 10.4% midseptal, and 1.0% posteroseptal. An additional 1.6% of procedures for modification of atrioventricular nodal reentrant tachycardia resulted in block.
The anatomy of the atrial septal region and its structures is complex.14 Previous reports of both radiofrequency2 3 7 8 9 10 15 16 17 and surgical18 19 20 ablation have demonstrated the variable and frequently relatively high incidence of both atrioventricular block and/or failure to cure arrhythmias in this region of the heart. Kuck et al16 reported a 0% (0 of 6) incidence of atrioventricular block in patients with midseptal accessory pathways, and Schlu¨ter and Kuck17 reported an incidence of 0% (0 of 12) in patients with anteroseptal accessory pathways. However, Yeh et al7 reported a 36% (5 of 14) incidence of transient or permanent atrioventricular block in patients with intermediate septal accessory pathways. Greene et al8 found inadvertent atrioventricular block occurred in 25% (1 of 4) of anteroseptal pathway, 50% (1 of 2) of midseptal pathway, and 3.2% (4 of 124) of atrioventricular nodal reentry ablations. Goyal et al3 reported an 8% (7 of 85) and Fenelon et al10 reported a 6.1% (12 of 194) incidence of inadvertent atrioventricular block in patients with typical atrioventricular nodal reentrant tachycardia. In a multicenter North American Society of Pacing and Electrophysiology survey, unintentional atrioventricular block was induced in 0.1% (6 of 5423) of procedures for atrioventricular nodal reentrant tachycardia.21
The risk of atrioventricular block and the efficacy of the procedure may be related in part to the aggressiveness of the ablationist. The overall success rate of 86% (Table 2⇑) suggests that in general, an aggressive approach was used despite the young age and small size of many of the patients. The efficacy can also be stratified according to the mechanism and location of the arrhythmia. The relatively high success and low complication rates of “nonseptal” procedures make them attractive treatment modalities. However, with the higher rates of inadvertent atrioventricular block in the anterior and mid septum, one might question the appropriateness of radiofrequency ablation, at least as a first-line option. Obviously, the seriousness of the arrhythmia and the likelihood of its spontaneous resolution, the efficacy of the procedure, the special situation of the patient (ie, airplane pilots, underwater divers, and so on) must be considered when choosing a treatment plan.
It is not surprising that atrioventricular block occurs during ablation procedures in the anterior and mid septum. Direct extension of the area of coagulation necrosis22 into the bundle of His or the compact atrioventricular node would be expected to interrupt atrioventricular conduction. The occurrence of atrioventricular block during ablation procedures in the posterior septum and outside the right septal region is more speculative. Whether it is secondary to catheter movement during energy application, cardiac motion, vascular injury,23 or an anomalously placed atrioventricular conduction axis has not been defined. It is interesting that unintentional complete atrioventricular block may occur in ≤50% of radiofrequency catheter ablations for modulation of atrioventricular conduction in patients with supraventricular arrhythmias.24 Possibly, an anomalously placed atrioventricular conduction axis may predispose to the development of supraventricular arrhythmias and then result in atrioventricular block when a posterior approach is used.
Of note is the return to sinus rhythm in 3 of 14 patients with acute third-degree and in 5 of 9 patients with second-degree atrioventricular block. Only 2 patients had return of atrioventricular nodal conduction within the first 24 hours after the application of the radiofrequency energy, resulting in atrioventricular block, with late return of the conduction occurring as late as 1 month after the acute block. Conversely, there were no reported (recognized) cases of late-onset third-degree atrioventricular block after the catheter ablation procedure. This is in contrast to the report by Fenelon et al,10 in which transient complete atrioventricular block was associated with the late occurrence of atrioventricular block in adults. Similarly, two of the cases with second-degree atrioventricular block occurred 24 hours and 2 months after the procedure and thus emphasize the need for late follow-up in these patients.
The role of permanent pacing in these patients has not been completely defined. Clearly, patients with persistent symptomatic bradycardia require permanent pacemakers. However, the need for permanent pacemakers in asymptomatic patients, those with transient atrioventricular block, and those with well-tolerated persistent block is controversial. In 6 patients (2 with third-degree and 4 with second-degree block), the rhythm returned to sinus and remained so at late follow-up (Figs 1 and 3⇑⇑). Five patients with persistent atrioventricular block (3 third-degree and 2 second-degree block) remained asymptomatic without permanent pacing. The 1 patient who developed late-onset second-degree block at 2 months was also asymptomatic. These data suggest that in the asymptomatic patient, cautious observation is a valid treatment option. It is important to recognize that our patient series differs from previous reports of late symptomatic atrioventricular block in adults.8 9 10 Series with more patients stratified according to age; site and degree of block; and response to exercise, atropine, isoproterenol, and programmed stimulation are needed to clarify the appropriate management of patients with asymptomatic atrioventricular block after radiofrequency ablation.
Another principal finding of this study was the learning curve for the avoidance of atrioventricular block during radiofrequency catheter ablation. There was a significant difference in the number of previous procedures performed for the ablations resulting in atrioventricular block versus the control matched group. In a prior study of the Pediatric Radiofrequency Catheter Ablation Registry,25 Danford et al demonstrated a learning curve for both the success of the procedure and the occurrence of complications.
What did we learn? When compared with an arrhythmia-matched control population (Table 3⇑), we could demonstrate no differences for the number or training of physicians, catheter type or size, use of long sheaths, anesthesia, rhythm during radiofrequency energy application, or patient size. It is important to note, however, that in our study, the power to demonstrate differences is low. Thermistor-guided radiofrequency ablation has been reported to be beneficial in pediatric patients.26 During the period of our study (1991 through 1994), tip temperature–guided ablations had not yet gained widespread exposure, and we could not show a difference in the rate of induced atrioventricular block (0 of 22 patients versus 2 of 21 control subjects). Also, in a comparison of the control group (the nonblocked septal cases) with right and left free wall accessory pathway ablations, there were no differences demonstrated for the power and duration of energy applied or for the use of “insurance burns.” However, during follow-up in the control group, 20% of patients developed recurrences of the initial arrhythmia. This is a greater recurrence rate than has been reported for older patients with accessory connections (6.4% to 12%).27 28 29 It is not clear whether the greater recurrence rate in this pediatric series is related to a “learned” less aggressive approach in the septal region or to inherent differences in septal pathways in children. It is also possible that Registry participation is more conducive to diligent recurrence reporting.
In 2 patients, unsuspected ablation catheter movement during the application resulted in permanent atrioventricular block. In both patients, the catheters dislodged from the posteroseptal region to a position adjacent to the His bundle catheter. We have learned that it is essential to visualize under fluoroscopy the ablation catheter position during energy applications when the catheter is positioned in such a way that it could dislodge and move to the septal region.
In conclusion, catheter application of radiofrequency energy in children and adolescents in the right septal region for accessory pathways was associated with a 2.9% risk of transient or permanent atrioventricular block and a 1.6% risk during ablations for atrioventricular nodal reentrant tachycardia. Relative inexperience was the principal variable associated with inadvertent atrioventricular block. Return of sinus rhythm occurred in 7 of 23 patients with atrioventricular block. Conversely, 2 patients had onset of second-degree atrioventricular block ≥24 hours after the procedure. These relative risks need to be considered in selection of management strategies for young patients with septal accessory pathways or atrioventricular nodal reentrant tachycardia. Also, the long-term prognosis for inadvertent atrioventricular block is variable, with the potential return of normal atrioventricular conduction in up to 33% of patients.
Appendix: Participants in Pediatric Radiofrequency Ablation Registry
Children's Hospital, Boston (Mass): J. P. Saul, E. Walsh; South Carolina Children's Heart Center (Charleston): C. Case, P. C. Gillette; Oregon Health Science University (Portland): M. Silka; University of California, San Francisco: G. Van Hare; Texas Children's Hospital (Houston): R. Friedman; Michigan University Medical Center (Ann Arbor): M. Dick; Duke University Medical Center (Durham, NC): R. Kanter; Atlanta Children's Heart Center (Ga): R. Campbell, J. E. Hulse; Cleveland Clinic Foundation (Ohio): R. Sterba; St Louis Children's Hospital (Mo): B. Bromberg; Children's Memorial Hospital, Chicago (Ill): D. W. Benson, B. Deal; Toronto Hospital for Sick Children (Ontario, Canada): R. Hamilton, R. Gow; Vanderbilt University (Nashville, Tenn): F. Fish; University of Nebraska Medical Center (Omaha): J. Kugler; Denver Children's Hospital (Colo): M. Schaffer; Miami University School of Medicine (Fla): G. Wolff, M. L. Young; Columbia University Babies Hospital (New York, NY): A. J. Hordof; Children's Hospital of the King's Daughter (Norfolk, Va): B. Ross; Children's Hospital of San Diego (Calif): J. Perry; Minneapolis Children's Hospital (Minn): D. Burton; Children's Hospital of Michigan: P. Karpawich; Arkansas Children's Hospital (Little Rock): C. Erickson; Rainbow Babies and Children's Hospital (Cleveland, Ohio): G. Van Hare; University of Florida, College of Medicine (Gainesville): M. Epstein; Georgetown University Hospital (Washington, DC): K. Kuehl, S. Beder; Long Island Jewish Medical Center (Queens, NY): C. Kurer; Iowa University (Iowa City): D. Atkins; Geisinger Medical Center (Danville, Pa): M. Cohen; Loma Linda University (Calif): J. McCormack.
We wish to thank Kris Houston, RN, MSN, and Gary Felix, MS, for their invaluable help with data management and editorial assistance and Julie A. Marshall, PhD, for help with the statistical analysis.
Reprint requests to Michael S. Schaffer, MD, 1056 E 19th Ave, B 100, Denver, CO 80218. E-mail email@example.com.
*See “Appendix” for list of participants.
- Received March 13, 1996.
- Revision received July 30, 1996.
- Accepted July 30, 1996.
- Copyright © 1996 by American Heart Association
Hebe J, Schlu¨ter, Kuck KH. Catheter ablation in children with supraventricular tachycardia mediated by accessory pathways: use of radiofrequency current as a first line of therapy. Cardiol Young. 1994;4:28-36.
Engelstein ED, Kunze LP, Duckeck W, Kuck KH. Radiofrequency current modulation of atrioventricular nodal conduction: implications of unintentional third-degree conduction block. J Am Coll Cardiol. 1990;15:134A. Abstract.
Goyal R, Langberg JJ, Kim YN, Calkins H, Sousa J, El-Atassi R, Morady F. Complete atrioventricular (AV) block complicating radiofrequency catheter modification of AV nodal reentrant tachycardia (AVNRT). PACE. 1991;14(Suppl II):658. Abstract.
Jackman WM, Wang X, Friday KJ, Roman CA, Moulton KP, Beckman KJ, McClelland JH, Twidale N, Hazlitt HA, Prior MI, Margolis PD, Calame JD, Overholt E, Lazarra R. Catheter ablation of accessory atrioventricular pathways (Wolff-Parkinson-White syndrome) by radiofrequency current. N Engl J Med. 1991;324:1605-1611.
Langberg J, Leon A, Borganelli M, Kalbfleisch S, El-Atassi R, Calkins H, Morady F. A randomized prospective comparison of anterior and posterior approaches to radiofrequency catheter ablation of atrioventricular nodal reentry tachycardia. Circulation. 1993;87:1551-1556.
American Academy of Pediatrics Council on Child and Adolescent Health. Age limits of pediatrics. Pediatrics. 1988;81:736.
Baily NTJ. Statistical Methods in Biology. London, UK: English Universities Press; 1974:58-61.
Kuck KH, Schlu¨ter M, Gursoy S. Preservation of atrioventricular nodal conduction during radiofrequency current catheter ablation of midseptal accessory pathways. Circulation. 1992;86:1743-1752.
Gallagher JJ, Selle JG, Sealy WC, Fedor JM, Svenson RH, Simmern SH. Intermediate septal accessory pathways (IS-AP): a subset of preexcitation at risk for complete heart block/failure during WPW surgery. Circulation. 1986;74(suppl II):II-387. Abstract.
Nath S, Whayne JG, Kaul S, Goodman NC, Jayaweera AR, Haines DE. Effects of radiofrequency catheter ablation on regional myocardial blood flow: possible mechanism for late electrophysiological outcome. Circulation. 1994;89:2667-2672.
Engelstein ED, Kunze KP, Duckeck W, Kuck HK. Radiofrequency current modulation of atrioventricular nodal conduction: implications of unintentional third-degree conduction block. J Am Coll Cardiol. 1990;15:134A. Abstract.
Fenrich AL Jr, Marlow D, Cecchin F, Johnsrude C, Friedman RA. Thermistor guided RF ablation (TRFA): high overall success with low recurrence. Pediatr Cardiol. 1995;16:252. Abstract.