peer reviewed ; BACKGROUND: The objective was to evaluate the diagnostic value of RHD fetal genotyping from the plasma of D- mothers as soon as 10 weeks' gestation in a routine clinical practice in Belgium. STUDY DESIGN AND METHODS: A prospective study was conducted between November 2002 and December 2006. DNA extraction was performed in an automated closed tube system. Fetal RHD/SRY genotypes were detected in the plasma of 563 pregnant mothers by real-time polymerase chain reaction (PCR) targeting multiple exons 4, 5, and 10 of the RHD gene and targeting an SRY gene sequence. These were compared to the D phenotypes determined in the 581 babies they delivered. RESULTS: By combining amplification of three exons, the concordance rate of fetal RHD genotypes in maternal plasma and newborn D phenotypes at delivery was 100 percent (99.8% including one unusual false-positive). The presence of nonfunctional RHD genes and the absence of a universal fetal marker, irrespective of fetal sex, did not influence the accuracy of fetal RhD status prediction. The RHD genotyping from 18 twin pregnancies was also assessed. Five weak D women were excluded from the RHD fetal genotyping prediction. Three discrepant results (0.5%) between predicted fetal genotype and cord blood phenotype were not confirmed by the baby phenotypes from venipuncture blood. CONCLUSION: Prenatal prediction of fetal RHD by targeting multiple exons from the maternal plasma with real-time PCR is highly sensitive and accurate. Over 4 years, this experience has highly modified our management of D- pregnant women.