Expectant mothers with a Rhesus (RhD) negative blood group who carry an RhD positive fetus can develop antibodies to the RhD antigen when fetal red blood cells escape into the mother’s circulation. This can happen at any time but is most common during the last three months of pregnancy and during childbirth, and rarely does it cause illness in the first child. However, if the mother has a second RhD positive conception, her antibody levels increase greatly and can cross the placenta to the growing baby and destroy red cells, causing anaemia, jaundice and sometimes death.

In 2005 about 65,000 RhD positive infants were born to RhD negative mothers in the UK, about 10% of all births. Since the 1960s all RhD negative expectant mothers who developed antibodies have been offered injections of anti-RhD immunoglobulin in subsequent pregnancies to ‘mop up’ fetal cells and prevent her forming more antibodies. This programme together with improved neonatal care has reduced Rhesus related death rate by 90%. In 2002 the National Institute for Health and Clinical Excellence (NICE) in the UK recommended that all RhD negative mothers should be offered treatment, even in their first pregnancy. Since in mainly white populations only about 38% of the fetuses are RhD negative a method for predicting the fetal RhD status was needed.

Since 1995, the International Blood Group Reference Laboratory of the English National Blood Service in Bristol has analysed the DNA of fetal cells obtained from babies’ blood taken by needle biopsy from sensitized RhD negative mothers. In 1997, Lo and colleagues from Oxford showed that fetal DNA could be detected in the mothers’ blood, and this has now almost replaced the need for needle biopsy. This analytical method is however both labour intensive and expensive and therefore unsuitable for the mass screening of all RhD negative women.

In a BMJ paper published on 12 April 2008, the Bristol Blood Group Laboratory reported their experience with a high-throughput automated system that extracts DNA and dispenses it for analysis by quantitative polymerase chain reaction. Each step takes about three hours. They analysed samples from RhD negative pregnant women attending clinics for routine testing at about 28 weeks of pregnancy. The results predicted by DNA analysis were compared with the results in cord blood in 1869 deliveries. The correct result was predicted in almost 96% of pregnancies, with inconclusive or unobtainable results in 3.4% and only 0.16% false negatives. (The three false negative samples were analysed more than 14 days after collection because of transport problems and in routine practice would have been discarded.) The paper concludes that if this test had been applied as a guide to treatment, only 2% of women would have received injections unnecessarily compared with 38% without the test.

The Bristol workers are now exploring the use of samples obtained earlier in pregnancy than 28 weeks. In the editorial that accompanied the BMJ paper, Dr Sailesh Kumar commented ‘Automated techniques should make mass testing easier and cost effective, and they should minimise the risks of receiving an unnecessary blood product. If these techniques are shown to be as reliable earlier in pregnancy the arguments for universal testing will be compelling’.