Value of Tilt Table Testing in Syncope

Abstract & Commentary

Synopsis: Intermittent bradycardia accounts for many but not all cases of syncope in patients with no or minimal structural heart disease.

Source: Moya A, et al. Circulation. 2001;104: 1261-1267.

Moya and colleagues report the results of the multicenter study evaluating the use of implantable loop recorders (ILRs) and tilt-table testing in patients with syncope of uncertain origin. This report focuses on 2 groups of patients. The isolated syncope group included 82 patients with no or minimal structural heart disease who also had no baseline intraventricular conduction defects and normal tilt-table studies. The tilt- positive group included 29 patients with syncope or near syncope during tilt-table study. All patients had experienced at least 3 episodes of syncope during the previous 2 years and had an interval between the first and last episode of syncope that was longer than 6 months. Prior to enrollment, all patients had undergone a history, physical examination, baseline electrocardiogram, echocardiogram, and 24 hour ambulatory monitoring that did not provide a diagnosis for the patient’s syncope. Tilt-table testing was done with either intravenous isoproterenol or sublingual nitroglycerin if the baseline tilt-table study was negative. At the conclusion of these tests, an ILR (Reveal, Medtronic) was implanted subcutaneously. Recordings obtained during each episode of syncope or presyncope during follow-up were retrieved, printed, and analyzed by Moya et al and the diagnosis was confirmed by an events committee. The primary end point of the study was the electrocardiographic findings at the first episode of recurrent syncope. Electrocardiographic findings associated with episodes of presyncope were a secondary finding.

The isolated syncope group and the tilt-positive syncope group had relatively similar clinical characteristics. The mean (± standard deviation) age in the isolated syncope group was 63 ± 17 years vs. 64 ± 15 years in the tilt-positive group. In the tilt-positive group of 29, 21% had syncope during the passive phase, and 79% had syncope during the drug phase. Six patients had an asystolic response, 14 patients had mixed bradycardia and vasodepression, and 9 patients had a pure vasodepressor response during the study.

In the isolated syncope group, an ILR-documented syncopal event occurred in 24 patients after a median follow-up of 105 days. Four additional patients who had syncope were unable to activate the ILR. The actuarial estimates for recurrent syncope were 15%, 34%, and 41% at 3, 9, and 15 months, respectively. The most frequent documented finding at the time of syncope, observed in 11 of 24 patients, was 1 or more prolonged asystolic pauses with a median duration of 31 seconds. Two other patients had severe sinus bradycardia. One patient had syncope related to an ectopic atrial tachycardia. Normal sinus rhythm (9 patients) or sinus tachycardia (1 patient) were seen in the remaining patients.

Presyncope occurred in 19 patients in the isolated syncope group and was documented by the ILR during 20 episodes. Four of the episodes were associated with relative bradycardia, 8 with normal sinus rhythm, 4 with supraventricular tachycardia, and 4 with sinus tachycardia. Three of the patients with presyncope also went on to have syncope.

The results of ILR monitoring in the tilt-positive group were similar to those in the isolated syncope group. A documented syncopal event occurred in 8 of 39 tilt-positive patients after a median of 59 days. Asystolic pauses were associated with syncope in 5 of the 8 patients, 1 patient had sinus bradycardia and 1 had sinus rhythm at the time of the episode. Two additional patients in the tilt-positive group had syncope but were unable to activate the ILR. In the tilt-positive group, the actuarial estimates for recurrent syncope were 25%, 30% and 34% at 3, 9, and 15 months, respectively. The estimates of recurrence for the tilt-positive and the isolated syncope groups were not different. Thirteen episodes of presyncope were documented with ILR tracings in 7 patients in this group. Presyncope was associated with either relative bradycardia (2 episodes), normal sinus rhythm (5 episodes), or a supraventricular arrhythmia (6 episodes).

Moya et al suggest that intermittent bradycardia accounts for many, but not all, cases of syncope in patients with no or minimal structural heart disease. Supraventricular arrhythmias are uncommon and ventricular arrhythmias were not observed. They also conclude that presyncope cannot be used as a surrogate finding in place of true syncopal episodes. Because of these observations, they feel that interventions, including pacemaker therapy, should be postponed until a definite diagnosis can be made. Tilt-table study evaluation was not helpful in these patients.

Comment by John P. DiMarco, MD, PhD

Recurrent unexplained syncope continues to be one of the more frustrating problems faced by cardiologists. Fifteen years ago tilt-table studies were introduced for the evaluation of patients with unexplained syncope but their value remains controversial. Data presented in this paper suggest that tilt-table studies reveal results that are neither sensitive nor specific in patients with recurrent unexplained syncope. The recurrence rate was the same in both the tilt-negative and tilt-positive groups. The spectrum of cardiac rhythms documented at the time of syncope were also similar in both groups.

It remains puzzling why tilt-table studies are of such limited value in patients with recurrent unexplained syncope. Clearly, observations made during tilt-table studies have been helpful for understanding of the mechanisms and pathophysiology of neurocardiac syncope in general. Although a tilt-table study can be used to trigger an episode in susceptible individuals, it does not appear to be useful for making a diagnosis or guiding therapy in individual patients.

The patients in this study were relatively old. It is possible that tilt-table studies might be much more helpful in young individuals in whom neurocardiac therapy is more common. However, the data presented here suggest that in older patients, an ILR will provide more useful diagnostic information. It should also be noted that electrophysiologic studies had been performed in 73 of the 111 patients and this may explain the low incidence of tachyarrhythmias noted.