The rate of positive head-up tilt (HUT), specificity, and same-day reproducibility of an HUT at 60 degrees combined with a low-dose isoproterenol infusion was assessed in the following patients: 120 consecutive patients with recurrent unexplained syncope, 30 healthy patients in a control group, and 30 patients with documented syncope not related to a vasodepressor reaction. HUT was positive in 61% (73 of 120) of patients with unexplained syncope. The false-positive rate in both the control and documented syncope groups was 6.6%. The mean isoproterenol dose infused was 1.4 +/- 0.5 microgram/min, 1.3 +/- 0.4 micrograms/min, 1.3 +/- 0.5 microgram/min, respectively (p = NS). HUT was positive during the drug free stage in 30 (25%) of 120 patients, and isoproterenol infusion was necessary in the remaining 43 (36%) patients. Immediate reproducibility was assessed in 75 patients, and HUT response was reproduced in 37 (82%) of 45 patients with a baseline positive HUT and in 28 (93%) of 30 patients with a baseline negative response. Overall, "sensitivity," specificity, and reproducibility were 61%, 93%, and 86%, respectively. Clinical variables that increased the probability of a positive outcome were age < or = 50 years and two or more syncopal episodes in the preceding 6 months in the absence of structural heart disease. These data support the use of an HUT protocol with low-dose isoproterenol infusion for the assessment of patients with recurrent syncope.