Mechanism of Syncope in Patients With Bundle Branch Block

Abstract & Commentary

Synopsis: Intermittent AV block may occur in patients with pre-existing bundle branch block even if they have a negative electrophysiologic study.

Source: Brignole M, et al. Circulation. 2001;104: 2045-2050.

Brignole and colleagues, on behalf of the International Study on Syncope of Uncertain Etiology (ISSUE) Investigators, describe the use of implantable loop recorders (ILRs) in patients with recurrent syncope, bundle branch block, and a negative conventional evaluation that included electrophysiologic (EP) studies. ISSUE is a multicenter, international, prospective study to determine the optimal methods for evaluating patients with syncope of uncertain origin. The data concern patients with a baseline bundle branch block in whom syncope was still suspected to be due to bradycardia despite a negative EP study. Patients could be entered into the trial if they had recurrent syncope. They underwent a careful history, physical examination, baseline ECG, carotid sinus massage, echocardiogram, 24-hour ambulatory ECG monitoring, and a complete EP study including atrial and ventricular stimulation. The criteria for a positive EP study were as follows: sinus bradycardia and an abnormal sinus node recovery time; a baseline HV interval 70 msec or greater; high degree His-Purkinje block after intravenous ajmaline, induction of sustained monomorphic ventricular tachycardia, or a supraventricular arrhythmia with hypotension. Patients also underwent a tilt-table study but because of the presence of bundle branch block, the tilt-table study results were not regarded as being adequate for diagnosis.

Patients who had completed these evaluations with no etiology for their syncope determined received an ILR (Reveal, Medtronic). They were then followed, and the results of monitoring are reported in this paper.

Fifty-two patients are included in the study. Their age was 71 ± 8 years and 33/52 were male. The mean QRS duration was 134 ± 18 msec. Twenty-eight of 52 had some form of associated heart disease. However, only 5 of 52 had an ejection fraction less than 40%. At baseline EP study, the mean HV interval was 55 ± 9 msec and the maximum HV interval after ajmaline infusion was 79 ± 19 msec. Tilt-table studies had been positive for hypotension in 7 of 51 but none had asystole or severe bradycardia.

An ILR-documented syncopal event occurred in 19 patients after a median of 48 days. One or more prolonged asystolic pauses attributed to atrioventricular (AV) block or sinus arrest were the most frequent findings observed in 17 patients. In 2 patients, only sinus rhythm or sinus tachycardia was documented at the time of syncope. In addition to these 19 patients, another 9 patients had clinical events during the study. Three patients developed persistent third-degree AV block without syncope; 3 patients had syncope but did not activate the ILR; 2 patients had AV block associated with presyncope and 1 patient died suddenly during a medical procedure after he developed atrial fibrillation that progressed to bradycardia.

AV block was the most frequent finding at the time of recurrent syncope. The actuarial estimates of AV block occurrence were 24%, 34% and 34% at 3, 9, and 15 months’ follow-up, respectively. Tilt-table testing results were not predictive of bradyarrhythmias. Three of the 15 patients with AV block and 1 of the 4 patients who had sinus arrest associated with syncope had manifest hypotension without bradycardia during their tilt-table study. No other clinical finding or test results predicted the occurrence of AV block during follow-up. Brignole et al conclude that intermittent AV block may occur in patients with preexisting bundle branch block even if they have a negative EP study. An ILR-based strategy seems a safe and appropriate approach for documentation of the need for pacing in these patients.

Comment by John P. DiMarco, MD, PhD

Recurrent unexplained syncope is often a frustrating condition to evaluate, and this becomes particularly true when it occurs in patients with baseline conduction defects. It was recognized many years ago that pacemaker therapy often failed to prevent recurrent syncope in such patients, and guidelines were developed to prevent the overuse of pacemakers. These guidelines were based on longitudinal studies of patients with bundle branch blocks but without syncope or other symptoms. These studies showed that the rate of progression to clinical AV block was slow in such patients. In addition, it was recognized that patients with bundle branch block were also likely to have ventricular arrhythmias or even supraventricular arrhythmias as the cause of syncope. Unfortunately, the data presented in this paper by Brignole et al illustrate how little we know about the development of AV block in patients with bundle branch block.

A few comments should be made about Brignole et al’s methodology. In this paper, they considered patients to have positive EP studies if they manifest inducible ventricular or supraventricular tachyarrhythmias or an HV interval of greater than 70 msec. An HV interval longer than 70 msec was considered an indication for pacing in these patients with recurrent syncope. In the new ACC-AHA guidelines for pacemaker implantation,1 however, only the finding of an HV interval greater than 100 msec was an indication for pacing. This is an extreme value that is rarely seen in patients unless they have had documented AV block. Several years ago, Englund and associates also reported that programmed ventricular stimulation in patients with bifascicular block could also be only a nonspecific finding.2 Therefore, significant questions still remain about the value of EP study in these patients.

The current paper is part of a large international study on syncope. When the final study results are reported, it would be interesting to see how many patients fell into in which of Brignole et al’s 4 classifications. Data from this study should clarify the optimal approaches for diagnosis in patients with different clinical characteristics and presentations.


1. Gregoratos G, et al. Circulation. 1998;97:1325-1335.

2. Englund A, et al. J Am Coll Cardiol. 1995;26: 1508-1515.