The aim of this study was to check the validity of the biochemical screening of pregnancies with Down's syndrome during the second trimester of pregnancy, in order to reduce the incidence of invasive diagnostic procedures. We used the optimal balance between sensitivity and specificity to determine the "cut off" values to estimate the results of the biochemical screening. Between January 1995 and December 2000, 2000 pregnancies were checked by double (determining hCG and AFP serum levels) and triple test, (determining hCG, AFP and uE3 serum levels). Competitive radioimmunochemical procedures (2nd trimester Amerlax-M, Ortho Clinical Diagnostics, USA) were used. The risk of Down's syndrome was calculated by Prenata program (Ortho Clinical Diagnostics, USA). The "cut off" median MoM values in pregnancies with Down's syndrome were 0.73 (AFP); 2.02 (hCG) and 0.74 (nE3). The calculated risk was compared with possibility 1:300 to estimate the results of biochemical screening. Our results were checked in the cytogenetic laboratory where samples of amniotic fluid, that we also took, were sent. We observed lower AFP levels (0.96 +/- 0.09 MoM), uE3 levels (0.65 +/- 0.1 MoM) and higher levels of hCG (1.57 +/- 0.27 MoM) in pregnancies with Down's syndrome, in comparison with euploid pregnancies of the corresponding gestational age. With 1:200 risk, the sensitivity of triple test is 80%, with acceptable number of false-positive results. This cut-off value showed to be acceptable for separating positive from negative results. Invasive procedures should be performed in pregnancies with positive screening result, with the aim of getting the tissue sample of the fetus for further cytogenetic analysis.