Tranquillisers and tocolytics can depress the appetite, creating even more problems. Anti hypertensive drugs can help maintain a mother's blood pressure but cannot reduce it, nor can they prevent protein appearing in her urine, growth retardation, pre term birth or caesarean section.

But the most controversial treatment of all concerns low dose aspirin. Aspirin acts as an antiplatelet and anti coagulant. For the last 10 years, results of trials with aspirin have been mixed. At least one meta analysis concluded that low dose aspirin reduced the risk of pregnancy induced hypertension and severe low birth weight (JAMA, 1991; 266: 260-4). Another suggested that, while it may reduce the risk of hypertension , aspirin also increases the risk of abruptio placentae (the placenta peeling off the uterine wall) and does not reduce the risk of illness in the newborn (N Eng J Med, 1993; 329: 1213-8).

Still later trials have shown that it does not benefit either mother or baby. One, involving 1,066 women taking 50 mg aspirin daily showed no difference in outcome between those women treated with aspirin and those without (Lancet, 1993; 341: 396-400). Another revealed that women taking aspirin needed more blood transfusions after delivery (BMJ, 1994; 308: 1250-1). The most recent trial, however, has finally put the nail in aspirin's coffin. The study, involving 2,539 "high risk" women, showed that aspirin therapy just doesn't work and should now be abandoned, since it brings with it damaging side effects, such as stomach irritation, spontaneous bleeding and premature placental separation (New Eng J Med, 1998; 338: 701-5).

We need to be aware of bias and habit, even among those who espouse conventional treatment. For instance, in one randomised, controlled trial, women being cared for by obstetricians were 8.8 times more likely to be admitted to hospital for treatment and 11.4 times more likely to be diagnosed as having protein in their urine than those in a hospital day care unit (Lancet, 1992; 339: 224-7). There were no differences in anti hypertensive drug use between the groups. The obstetric group were also 4.9 times more likely to have labour induced. Although obstetricians have long justified their high intervention rates by claiming they look after a higher risk population, in this study there were no differences in Apgar scores, birth weight and rates of admission to neonatal intensive care units between the babies of the two groups, suggesting that aggressive treatment makes little difference.

Many doctors believe that nutrition is the stuff of home economics classes, not science. As a result, feeding pregnant women is a low medical priority long after prodding and scanning. There is also extreme social resistance, among men and women, to the idea. After all, women are supposed to feed their families, not be told what to eat.

Medicine can detect pre-eclampsia with varying degrees of success, but cannot cure it. Prevention is the only cure and nutrition appears to be the only prevention.

It behoves us to remember that eating for two refers not to quantity, but quality. Because the mother and baby are one large system, it is vital that women eat to their appetite and that the food they eat be of the best possible quality.

Pre-eclampsia is only the tip of the iceberg. Many adult illnesses can be traced back to the environment in the womb. Maternal nutrition, when viewed from this perspective, is neither fad nor fancy, but a profound responsibility.