To the Editor: Murray et al recently suggested that after an initial screening urinalysis, routine urinalysis could be eliminated from antenatal care without adverse outcomes for women.1 Their conclusions are based on a prospective observational study of 1000 women, 26 of whom developed pre-eclampsia, with 6/24 (25%) developing new proteinuria before the onset of hypertension. We have concerns about the authors' claims, which, we believe, are not justified by their findings.

Pre-eclampsia remains a major cause of maternal and perinatal mortality.2 A study of 26 cases of pre-eclampsia is clearly underpowered to evaluate important morbidity and mortality outcomes. Therefore, one must place great emphasis on potentially undetected cases of pre-eclampsia, such as those that repeatedly appear as examples of substandard care in the Report on confidential inquiries into maternal deaths in the United Kingdom.3 That the initial screening urinalysis detected only 2/26 (8%) women who subsequently developed pre-eclampsia is no surprise. In contrast, routine urinalysis at antenatal visits detected 25% of cases of pre-eclampsia before the onset of hypertension. In the UK, this would amount to 4500 women per annum. The detection of proteinuria provokes further antenatal visits earlier than would normally be planned, facilitating early detection and management of pre-eclampsia. Intervals of 2–4 weeks between visits are usual in the early part of the third trimester when the risks to mother and fetus of severe early onset pre-eclampsia are greatest.4 By eliminating routine urinalysis from antenatal care, a quarter of affected women are potentially exposed to the risk of ongoing undetected pre-eclampsia for up to four weeks.

Murray et al suggest that most women would be detected by risk assessment at the initial visit. While there are known risk factors for pre-eclampsia, the largest group who develop pre-eclampsia are primigravid women with no previous history of note. In addition, there are no reliable studies to justify this claim of Murray et al. Recent randomised controlled trials have suggested that the frequency of antenatal visits could be reduced without demonstrating harmful effects.5 None of these trials, however, have the power to demonstrate safety. Even the recent large World Health Organization trial involving 24 526 women6 would have required 490 000 women for 80% power, and 649 910 women for 90% power, to exclude the 40% increase in maternal mortality actually observed with reduced antenatal care. Changes in well-tried methods of antenatal care need to be assessed with more stringency before it is concluded that they are safe.