Evaluation of syncope by upright tilt testing with isoproterenol. A nonspecific test

Abstract

Objective: To determine the proportion of patients with syncope of unknown etiology who have a positive response to upright tilt testing and to determine the specificity of this test as a marker for vasodepressor syncope.

Design: Comparison of upright tilt testing with isoproterenol in patients with syncope of unknown etiology and in controls.

Setting: Outpatient clinics of a tertiary care center.

Patients: A total of 20 patients with syncope of unknown etiology and 40 controls matched by age, sex, and lack of underlying cardiovascular or other diseases had upright tilt testing with isoproterenol infusion. Controls consisted of two groups (groups I and II) who had slightly different methods of tilt testing in conjunction with isoproterenol.

Interventions: Upright tilt testing at 80 degrees from horizontal was done for up to 15 minutes. If end points were not reached, infusion of isoproterenol was started at 1 microgram/min and increased with graded increments in infusion rates up to 5 micrograms/min.

Measurements: The development of syncope or presyncope in association with hypotension, bradycardia, or both.

Main results: Positive response to tilt testing in patients with syncope was 75% (95% CI, 55% to 95%); it was 65% in control group I (CK, 44% to 86%) and 45% in control group II (CI, 20% to 70%). Case patients and controls with positive tilt-test responses were similar with respect to time to completion of the test, mean total isoproterenol dose, mean isoproterenol dose level at completion, average mean blood pressure at completion, mean heart rate at completion, mean decline in systolic blood pressure, and decline in mean blood pressure or heart rate (P greater than 0.05 for all comparisons). Further, symptoms during tilt testing in case patients with positive responses were similar to those in controls with positive tilt-test results.

Conclusions: The rate of positive tilt testing in patients with syncope is equivalent to that in controls without a history of syncope. The low specificity of this test makes its use as a marker for vasodepressor syncope questionable.