OBJECTIVE To evaluate the discriminatory ability of maternal serum creatine kinase (SCK) as a test for ectopic pregnancy (EP). METHODS Serum creatine kinase concentrations were obtained prospectively from symptomatic patients being evaluated for early abnormal pregnancy. Serum creatine kinase concentrations from all patients and from a subset of these patients with maternal serum beta-hCG concentrations < 6,500 mIU/mL (conversion factor to SI unit, 1.00) were analyzed with descriptive statistics, receiver operator characteristic (ROC) curve analysis, and calculations of predictive values. METHODS A university hospital emergency room. METHODS Fifty-six patients with intrauterine gestations (25 with beta-hCG concentrations < 6,500 mIU/mL) and 23 patients with EP (20 with beta-hCG concentrations < 6,500 mIU/mL) were studied. RESULTS For all patients and the subgroup with beta-hCG concentrations < 6,500 mIU/mL, mean SCK levels were not significantly different between ectopic and intrauterine gestations. For all patients and the subgroup with beta-hCG concentrations < 6,500 mIU/mL, the areas under the ROC curves did not demonstrate discriminatory ability of the SCK test. The highest positive predictive value of an elevated SCK for EP was 52% using the SCK concentration of 70 U/L, and this was seen in the subgroup of patients with beta-hCG values < 6,500 mIU/mL. CONCLUSIONS Maternal SCK concentrations do not reliably predict EP.