Ninety two patients with unexplained syncope and 52 normal controls without history of syncope were divided into three groups respectively. Baseline tilt table test (BTTT), multistage isoproterenol tilt table test (MITTT), and single stage isoproterenol tilt table test (SITTT) were performed in all of them. Logistics regression was used with the result of TTT as independent variable; dependent variables included sex, age, basal heart rate, basal blood pressure, the index of syncope (i) and the difference between basal heart rate and the rate before tilting (h12). The results were as follows: (1) According to the regression for the total population tested, the main factors influencing the result of TTT were i and h12 (P = 0.012 and P = 0.001). Similar result was found in the patient group (P = 0.052 and P = 0.032). (2) In the control group, sex and h12 were the predictors of the result (P = 0.090 and P = 0.016). It is concluded that: (1) The result of TTT is mainly affected by the frequency and degree of syncope, as well as by the use of isoproterenol; (2) Isoproterenol increases the sensitivity of TTT, but decreases the specificity; (3) Sex is probably an important factor affecting the result of TTT.