Tuesday October 2 2007
A healthy diet in pregnancy is best
An abnormal cholesterol level during pregnancy may increase the risk of premature birth, reported the Daily Mirror. It was found that “low levels of cholesterol were just as bad as high ones when it comes to the chances of causing an early birth”, the newspaper added.
BBC News explained that women were at a 5% risk of having a premature birth if they had normal cholesterol levels, but that this increased to over 12% in women with the highest cholesterol levels, and to over 21% in Caucasian women with the lowest cholesterol levels.
These reports are based on a study that investigated women’s cholesterol levels and their pregnancy outcome. These findings are preliminary and cannot be applied to all pregnant women.
Where did the story come from?
Drs Robin Edison, Maximilian Muenke, and colleagues from the National Institutes of Health (NIH) and other medical research centres in the US, carried out this research. The study was funded by the Division of Intramural Research, National Human Genome Research Institute, NIH, Department of Health and Human Services and was published in the peer-reviewed medical journal Pediatrics.
What kind of scientific study was this?
This was a retrospective cohort study designed to examine whether low cholesterol during pregnancy is linked to premature birth or other poor birth outcomes.
The researchers identified women for this study from 9,938 women who attended a routine second trimester screening at clinics in western South Carolina between 1996 and 2001. At the time, medical data were collected and blood samples given.
From this large group of women, the researchers selected those women who were aged 21-34, non-smokers, not substance abusers, did not have diabetes, who were carrying only one baby, had not had a previous abnormal pregnancy, and who had a live birth at one of the two hospitals near to the research centre.
The blood samples were then tested for their cholesterol levels, and researchers selected 118 women from those who met the criteria above, who had blood cholesterol in the lowest 10% of measured levels and 940 women who had normal to high levels of cholesterol.
Researchers looked at these women’s medical charts to find out whether the growth of the baby in the womb was normal, the baby had abnormalities, and the baby was born prematurely (before 37 weeks). They then compared these outcomes for women with the very low cholesterol levels with those for the women with normal to high cholesterol levels. They adjusted these analyses for factors that might affect pregnancy outcome, such as maternal age, ethnic group, baby’s sex and weight, and evidence of abnormally slow growth in the baby.
What were the results of the study?
Researchers found that women with low cholesterol levels were at an increased risk of premature birth compared with women who had normal (mid-range) cholesterol levels. When these analyses were carried out separately for white and black women, this increase in risk was only seen in white women, whose odds of having a premature birth were about 5 to 6 times higher if they had low cholesterol. White women with high cholesterol were also at increased risk of a premature birth. Low cholesterol was not associated with an increased risk of serious abnormalities in the baby.
What interpretations did the researchers draw from these results?
They concluded that having a blood cholesterol level in the lowest 10% of measurements increased the risk of premature birth in white women who did not have other high risk factors for premature birth.
What does the NHS Knowledge Service make of this study?
This study was reasonably well conducted, but does have some limitations, some of which are acknowledged by its authors.
- This study had very strict inclusion criteria, as the researchers were trying to restrict their analyses to women who did not have any obvious risk factors leading to a premature birth or other pregnancy complications. This means that the results observed in this highly selected group of women may not be representative of what would happen in the population as a whole.
- Among the eligible women, researchers only selected about half to a third for analysis. It is not clear why they did not analyse all eligible women, or how they selected the women that they did. Unless these women were selected completely at random, they may not be representative of all eligible women.
- Because the data were collected retrospectively, there is the possibility that it will not be as accurate as if it were collected contemporaneously. For example, the blood samples had been stored in a freezer and blood cholesterol measurements may not be as accurate in stored samples as in fresh ones. In addition, medical chart data may be incomplete or recorded inaccurately.
- Cholesterol levels were only measured on one blood sample for each woman. It would have been more accurate to take multiple readings over time, one sample may not be truly representative of the mother’s cholesterol levels.
- It is not possible to say from this study by itself whether the association between low maternal cholesterol and premature birth is causal. Other factors associated with low maternal cholesterol may be responsible, for example, having a generally poor diet, which is lacking in minerals and vitamins. The authors did take steps to try to account for factors that might play a role.
The authors of the study themselves report that their study results “require validation” by other studies, but that it is an “important preliminary finding”. Regardless of whether these findings are replicated in further studies, common sense and good evidence suggest that a healthy balanced diet is important for all, particularly pregnant women.
Sir Muir Gray adds...
When researchers say that something increases risk, they should really say that the there is a statistical association between the two things.
Just as there is a statistical association between the sun and pubs – there are more pubs open at sunset then at sunrise – but this does not mean that the sun setting increases ‘the risk’ of pubs opening.
This research does not change the advice to a woman who is pregnant.