Objective: Second trimester total hCG and free βhCG levels in maternal serum samples of 33 pregnancies affected by fetal trisomy 21 and of 188 matched controls were compared in a retrospective study. To find out differences of discriminating efficacy by using one of these markers a multivariate discriminant analysis was performed. Method: Statistical evaluation was performed for hCG/free βhCG frequency distributions. Discriminant analysis was carried out using the status ‘affected’ or ‘unaffected’ as the group variable and the serum markers unconjugated estriol (uE3), alpha-fetoprotein (AFP), and alternatively, hCG or free βhCG, as discriminant variables. Results: The median of free βhCG MoM values in affected pregnancies was slightly higher (1.90 MoM) than the median of total hCG MoM values (1.72 MoM) but a lower standard deviation was stated for the logarithmic hCG MoM values (SD = 0.49) compared with free βhCG MoM values (SD = 0.51). A two-tailed Student’s t test revealed no significant differences of hCG and free βhCG MoM values in both the affected and unaffected pregnancies. By inclusion of free βhCG the discriminant analysis classified 26 out of 33 affected cases correctly and 45 out of 188 unaffected cases incorrectly. For the inclusion of hCG these ratios were 25/33 and 41/188, respectively. Taking in account the individual maternal age risks at a defined false-positive rate of 5% including free βhCG yielded a higher detection rate than including hCG. However, using 1:380 (age-related at-term risk of a 35-year-old woman) as a cut-off risk including hCG yielded a higher detection rate than including free βhCG. Conclusion: For the observed cases none of the markers, hCG or free βhCG, was superior in Down syndrome screening.