Quantitative fluorescent polymerase chain reaction (QF-PCR) has recently entered the field of prenatal diagnosis to overcome the need to culture fetal cells, hence to allow rapid diagnosis of some selected chromosomal anomalies. We reviewed the studies on the accuracy of QF-PCR in detecting chromosomal anomalies at prenatal diagnosis. Overall, 22 504 samples have been analysed. The detection rate of aneuploidies of the selected chromosomes (13, 18 and 21, and X and Y) was 98.6% (95% confidence interval 97.8-99.3). QF-PCR might play a major role and be considered a valid alternative to the full karyotype. Being less expensive, and almost entirely automated, more women could undergo invasive prenatal diagnosis without significant increase in health expenditure. By using QF-PCR as a stand-alone test, the chances of non diagnosing the commonest, and the only chromosome anomalies which do increase in frequency with maternal age, are approximately one in 150 abnormal karyotypes, or one in 10-30 000 samples, based on the age distribution. These error rates might be deemed acceptable, although most structural chromosomal anomalies will be missed. At present, women are rarely informed about the full spectrum of the conditions which might be diagnosed via amniocentesis or chorionic villous sampling. Some of these anomalies might be acceptable, in view of their limited or uncertain clinical relevance, and decision analysis might, in the majority of cases, confine the full karyotype to selected women who have specific indications.