Objective: To determine if unexplained low second-trimester maternal serum human chorionic gonadotropin (MShCG) is a useful predictor of complications of pregnancy. Study design: Between 2/1/90 and 1/3/91, 3,116 patients underwent prenatal screening using second-trimester maternal serum α-fetoprotein (MSAFP), MShCG and maternal serum unconjugated estriol (MSuE3). Among these, there were 160 patients with complete obstetrical history who had second-trimester MShCG < 0.4 multiples of the median (MoM). These were compared to 261 controls with complete obstetrical history and a normal second-trimester MSAFP, MSuE3 and MShCG. Results: No differences were found in gestational age at delivery, neonatal weight, premature rupture of membranes or pregnancy loss. The relative risk of pregnancy-induced hypertension in the study group was 0.29 (p < 0.01) and that of gestational diabetes was 0.3 (p < 0.05). Only when low MShCG was associated with a high or low MSAFP or low MSuE3 was there a significantly increased loss of pregnancy (relative risk 11.7; p < 0.0001). Conclusion: The data suggest that second-trimester MShCG < 0.4 MoM by itself has no influence on the outcome of pregnancy