To investigate whether a single serum beta-hCG in pregnancies achieved by assisted reproductive technologies (ART) can accurately predict pregnancy viability and, in viable pregnancies, multiple gestation. Four hundred sixty-one consecutive successful ART pregnancies were studied retrospectively. Seventy-one of the 461 patients were excluded because their beta-hCG was either drawn on the incorrect day or outside our facility. Three hundred ninety subjects had a serum beta-hCG drawn 14 days after ET or 16 days after gamete transfer. The beta-hCG samples were analyzed by immunoradiometric assay based on the Third International Reference Standard (IRS) (First International Reference Preparation (IRP)). Pregnancy status was followed, at minimum, through the first trimester. One hundred fifty (38%) of the 390 were found to be nonviable, resulting in spontaneous abortion (n = 38, 10%), ectopic pregnancy (n = 27, 6%), or biochemical pregnancies (n = 85, 22%). A statistically significant difference by the Scheffe F-test was found between the mean beta-hCG value of the nonviable (115 mIU/mL) (conversion factor to SI unit, 1.00) and viable (428 mIU/mL) pregnancies. The positive predictive value of a single beta-hCG > 100 mIU/mL in distinguishing viable from nonviable pregnancies was 0.83 (sensitivity 91%, specificity 71%). Of the 240 viable pregnancies, 74 (32%) were multiple gestations (57 twins, 14 triplets, and 3 quadruplets). The mean beta-hCG of the singleton pregnancies (266 mIU/mL) was significantly different from that of the multiple gestations (792 mIU/mL). The positive predictive value of a single serum beta-hCG < or = 400 mIU/mL in distinguishing singleton from multiple gestations was 0.92 (sensitivity 86%, specificity 82%). A single early serum beta-hCG may be used in ART pregnancies to predict which pregnancies will continue beyond the first trimester and to identify multiple gestations. Early reassuring tests may reduce anxiety.