OBJECTIVE This study aims to assess the suitability of non-invasive prenatal RHD genotyping in non-immunized midtrimester pregnant women from a mixed ethnic population, to prevent unnecessary anti-D immunoglobulin prophylaxis and to identify RHD variants METHODS Rhesus D-negative pregnant women were offered fetal RHD genotyping at 24 gestational weeks. A total of 284 samples were tested for RHD status using multiplex rt-PCR amplification of exons 5 and 7 of the RHD gene and exons 6 and 10 in selected cases. Women carrying RHD-negative fetuses were counseled about their option to avoid routine antenatal anti-D immunoglobulin administration. Diagnostic accuracy of RHD genotyping was compared with postnatal Rhesus D serotyping. RESULTS A total of 184 positives (65%), 91 negatives (32%) and 7 cases (2.5%) compatibles with RHD variants were detected by RHD genotyping. No false negative results were found, and a single false positive was observed in a twin pregnancy. Genotyping was accepted when offered by 94% of women (284/302), and anti-D immunoglobulin was avoided in 95% (90/95) of RHD-negative fetuses. CONCLUSIONS Non-invasive routine antenatal RHD genotyping at 24 weeks of pregnancy is a highly accurate method, resulting in the avoidance of 95% of unnecessary administrations of anti-D immunoglobulin, with no false negative results.