AbstractObjectiveThe objective of this study was to examine in theory the clinical utility of a prenatal algorithm that uses rapid aneuploidy detection in all cases and G‐banded analysis for selected cases (RAD/G algorithm).MethodsOver a 4‐year period, amniotic fluid samples were prospectively assigned into RAD (limited analysis) or RAD/G (intensive analysis) categories based upon the likelihood of the fetus having a chromosome anomaly. The samples were cultured and analyzed by standard cytogenetic methods. The rates of clinically significant chromosomal anomalies potentially undetectable by the RAD/G algorithm were calculated.ResultsThe karyotype was normal in 3861/4054 (95.24%) cases and abnormal in 193 (4.76%). From these data, the detection rate of the RAD/G algorithm was 87.6% if all abnormalities detected by G‐banding were taken into consideration and 97.6% if abnormalities having reduced predictive value were excluded (balanced rearrangements and most mosaic cases).ConclusionsCompared to G‐banding alone, the RAD/G algorithm has a reduction in sensitivity due to undetectable abnormalities and mosaicism in the RAD group. However, it provides a rapid and inexpensive alternative to traditional G‐banded analysis, and might be more appropriate for patients with uncomplicated, low risk pregnancies. Copyright © 2008 John Wiley & Sons, Ltd.