The prenatal miracle of ultrasound has taken another knock in one of the largest studies to be conducted on its efficacy.

To aid this six year study in Oxford, a register was set up of congenital malformations suspected prenatally and all congenital malformations, including chromosomal anomalies confirmed at birth. Out of 33,376 babies, 725 (2 per cent) were judged abnormal at delivery but prenatal ultrasound only picked up 55 per cent of these.

Even more striking was the fact that with 174 pregnancies where an anomaly was identified prenatally, the mother went on to deliver a perfectly healthy child. In the context of this study, roughly one in three prenatal diagnoses of abnormality was wrong, or "false positive". False positive results are on the rise, and the authors attribute this phenomenon to so called "soft markers" variations on normal foetal physiology, which the sonographer looks for and which may or may not indicate a serious malformation.

By including soft markers in the examination, the detection rate of malformations increased by 4 per cent, but the number of false positives increased 12 fold. This study was carried out at a centre of excellence, where professional standards were high.

As the accompanying commentary points out, in hospitals where ultrasound is performed by individuals who are not so experienced and well trained, the false positive rate could be even higher (Lancet, 1998; 352: 1568-8, 1577-81).