Scientists have come closer to solving the riddle of cot death with the discovery that more than half of all babies who die unexpectedly may have had problems in the womb.

Scientists have come closer to solving the riddle of cot death with the discovery that more than half of all babies who die unexpectedly may have had problems in the womb.

A study of birth records in Scotland shows that mothers who had a placenta defect were up to three times more likely to have a baby who later suffered cot death.

The finding is a key piece in the cot death puzzle, which has mystified researchers for 30 years. A total of 354 babies died of sudden infant death syndrome (Sids) in 2003 in Britain and it is still the most common cause of death in infants in the industrialised world.

Researchers at Cambridge University and the Department of Public Health at the Greater Glasgow Health Board say the babies may have been deprived of oxygen or nutrients before birth, which affected the development of the brain functions controlling the heart and lungs.

This could put them at greater risk from environmental stresses after birth, such as overheating or cigarette smoke, which are known to increase the likelihood of cot death.

Campaigns during the 1990s to get parents to put their babies to sleep on their backs and to ban smoking from the bedroom helped to cut cot deaths by 70 per cent. But progress in reducing the cot death rate further has since slowed.

The distress caused by the tragedies is exacerbated by the lack of any explanation. It can also arouse suspicions which, in some cases, have led to accusations of murder.

If confirmed, the finding, published in the New England Journal of Medicine, could lead to a test that would enable doctors to identify vulnerable babies at birth and help parents to monitor them more closely during their crucial first year. Joyce Epstein, director of the Foundation for the Study of Infant Deaths, which funded the research, said: "This is a very important piece of work. It is another piece of the puzzle. The Holy Grail of Sids research is to predict which babies are at risk and put in place measures to protect them."

Previous research has shown that babies who grow poorly in the womb or who have difficult births are at greater risk of cot death, but the nature of the link has remained obscure.

For the study, researchers examined the records of 214,000 women who gave birth in Scotland, of whom 114 suffered cot deaths. All the mothers underwent routine blood tests for Down's syndrome and other abnormalities during the second trimester of pregnancy.

The results showed that there were 7.5 cot deaths per 10,000 births among the mothers with the highest levels of alpha-feto protein in their blood, compared with 2.5 per 10,000 births among those with the lowest levels.

Gordon Smith, professor of obstetrics and gynaecology at Cambridge University, who led the study, said: "In at least half of cot deaths, there is a biological vulnerability that puts the babies at increased risk. This is probably an underestimate and we could say there is a biological determinant for the majority of Sids deaths."

Profesor Smith stressed that, even in the high-risk group, the chances of a cot death were less than one in 1,000, and the finding did not undermine earlier advice to put babies to sleep on their back or avoid exposing them to cigarette smoke. "We do not want to terrify women who have a high alpha-feto protein level, nor are we saying that the environment after birth is not important. But we are saying these babies may be more vulnerable."

The presence of alpha-feto protein, which comes from the foetus, in the mother's blood suggests that the placenta is "leaky" and indicates a defect in placental function. If the baby is not getting the oxygen or nutrients it needs that could affect the development of the control mechanisms in the brain of the heart and respiration. "A bad environment in the womb may affect the baby's response to a bad environment after birth," Professor Smith said.

The next stage of the research would be to develop a more accurate test to detect babies at risk of Sids. "We are working on a more predictive model which would include gestational age, sex and birth weight. One should not take an isolated measure as a predictor," Professor Smith said.

'We will probably never know why Adam died'

By Kate Ocock

Adam was eight weeks old when he died on New Year's Day 2002. We had celebrated New Year's Eve quietly at home and I had fallen asleep with Adam on the sofa after giving him a feed. When my partner John woke me at about 5am I noticed Adam wasn't breathing. We called an ambulance and they took us to hospital and spent 20 minutes trying to resuscitate him, but it was too late.

It all seemed unreal. One thing I remember is going home and sitting on the front doorstep in the freezing cold. It was as if I had to prove to myself that I was still alive and this was actually happening.

The police came to investigate but they were very sensitive and considerate. We had to wait for the post-mortem because of the holidays but it showed nothing at all.

The question of why it happened was all-consuming at the time. One of the hardest things was that everywhere I went there seemed to be babies and I found myself asking 'Why not that one?' It was awful. You have to accept that you don't know and probably never will know.

Oliver was born in July last year and we have had a system of alarms and checks to monitor him. He did have an apnoea attack [he stopped breathing] at just the same age as Adam - eight weeks - and spent 24 hours in hospital, but all they could say was they didn't know why he did it. I have been more worried with Oliver. When he was at home I would think 'What if it happened again?'

If the research published today can help explain what happens then that must be better.

Every risk factor that can be prevented means fewer people will have to go through what we went through.