The aim of this work was to assess the hemodynamic responses to the tilt test of 51 patients with syncope (n = 31) or presyncope (n = 20) of unknown etiology. A protocol with an inclination of 70 degrees (20 min) with or without isoproterenol (mean dose of 3.6 +/- 0.3 ug/min), was used. Forty five percent of patients had a positive test, 18 with isoproterenol at 70 degrees (group 1A) and 5 without isoproterenol (group 1B); 28 patients had a negative test (group 2). These groups did not differ in age or sex distribution. Basal heart rate was 76.2 +/- 2x'. At the end of the test it was 73.4 +/- 5.7 in group 1A, 78.0 +/- 7.5 in group 1B and 120.4 +/- 3.3 in group 2 (p < 0.01). Systolic blood pressure decreased to 78.1 +/- 4.8 mmHg in group 1A, to 76.2 +/- 9.9 in group 1B and did not decrease in group B (130.0 +/- 5.7 mmHg, p < 0.01). The required dose of isoproterenol was higher in group 1A than in group 2 (4.4 +/- 0.3 vs 3.1 +/- ug/min, p < 0.01). It is concluded that the tilt test reproduced symptoms and accompanying hemodynamic mechanisms in a high proportion of patients with syncope of unknown etiology. This test should be incorporated in the diagnostic workup of these patients.