Combining inhibin A with existing second-trimester markers in maternal serum screening for Down's syndrome.

To assess the value of inhibin A as an additional second-trimester maternal serum marker of Down's syndrome we studied 56 affected and 280 unaffected pregnancies matched for gestational age. The median level in the cases was 1.62 multiples of the gestation-specific median (MOM) in the controls, with 95 per cent confidence limits of 1.34-1.96. The distribution of inhibin levels in affected and unaffected pregnancies was approximately log Gaussian, with means about 1 standard deviation apart. This degree of separation was similar to that for human chorionic gonadotropin (hCG), free beta-hCG, and unconjugated oestriol (uE3), but about double that of alpha-fetoprotein (AFP) measured in the same samples. Inhibin was largely uncorrelated with AFP and uE3, whereas the log correlation coefficient with hCG was 0.29 (P = 0.19) for Down's syndrome and 0.41 (P < 0.0001) for unaffected pregnancies; with free beta-hCG, it was 0.18 (P = 0.38) and 0.38 (P < 0.0001), respectively. On the basis of these results and other published studies, we estimate that measuring inhibin A in addition to AFP and hCG or free beta-hCG (with or without uE3) will increase the detection rate for a fixed 5 per cent false-positive rate by about 7 per cent.