Researchers have found that smoking during pregnancy increases the risk of birth defects, such as club foot and missing limbs, reports The Guardian.
The news report is based on a systematic review which assessed previous research on smoking during pregnancy to determine the risks of birth defects. Smoking while pregnant is already known to be harmful to the baby, increasing the risk of miscarriage, small babies and premature birth. This study was the first to specifically look at the risk of birth defects. It found that the risk of various birth defects increased for mothers who smoked, with the odds rising from between 9% and 50% for different abnormalities. The annual incidence of these sorts of defect is around 3 to 5% of births in the UK.
Overall, this was a well-conducted study, and its findings are convincing evidence that smoking increases the risk of some birth defects. Smoking during pregnancy is already known to be harmful to the baby. Women stop smoking before they become pregnant, or as early as possible into the pregnancy. Read our Pregnancy care planner for the benefits of stopping smoking and advice on quitting.
Where did the story come from?
The study was carried out by researchers from University College London. No external funding sources were reported. The study was published in the (peer-reviewed) medical journal Human Reproduction Update . The story was covered well by BBC News and The Guardian .
What kind of research was this?
This was a systematic review of observational studies (cohort, case-control studies and surveys) that investigated whether maternal smoking is associated with birth defects. Smoking during pregnancy is already known to be a risk factor for miscarriage, low birthweight, premature births and small foetuses. The researchers say that many of the 7,000 chemicals found within cigarettes can cross through the placenta barrier and directly affect the baby. However, they say that despite 50 years of research on the effects of smoking on pregnancy, a review on birth defects has not been carried out.
A systematic review is the best way to address this type of question. A systematic search identifies all of the research that is relevant to a topic and, usually, filters them for quality.
What did the research involve?
Using keywords related to smoking or birth defects, the reviewers searched two medical databases for English articles published between 1959 and February 2010. They also checked the reference lists of two US Surgeon General’s reports to ensure that they had not omitted any relevant articles.
The reviewers went through 9,328 abstracts and obtained the full scientific paper if the abstract referred to maternal smoking or risk factors. In total, the reviewers assessed 768 full papers. To be included in the review, the papers had to be based on an observational study of women who smoked during pregnancy, in which the article reported the odds ratio (OR) or relative risk (RR) of there being a defect among pregnant smokers compared to non-smokers. This left 172 relevant articles covering 101 different studies that were included in the analysis.
Of these 101 studies, 16 were prospective cohort studies, three were case-control studies where smoking status was recorded in early pregnancy, 62 were “retrospective” case-control studies where smoking status was recorded after delivery, and 20 were surveys. In all of these studies, maternal smoking status and other characteristics were obtained by questionnaires or interviews during early pregnancy or shortly after birth using surveys, interviews or birth certificates.
The data from the studies were pooled and differences between the studies (heterogeneity) was assessed. A statistical technique called a random effects model was used to calculate the odds ratio of having a birth defect (i.e. the odds of a child born to a mother who smoked during pregnancy having a birth defect relative to the odds of a birth defect in a child born to a non-smoker).
The researchers also did further analyses in which they only used the studies that were performed prospectively. This was to avoid possible reporting bias that retrospective studies might be subject to, in which smokers who had an affected baby may have been more likely to classify themselves as non-smokers.
What were the basic results?
Of the 172 publications, a total of 173,687 babies had been born with birth defects, leaving 11,674,332 babies to be classed as unaffected controls.
The researchers found significant positive associations between maternal smoking and several birth defects for the offspring when compared to pregnant non-smokers:
- Cardiovascular/heart defects: pregnant smokers had a 9% increased odds (Odds Ratio [OR] 1.09, 95% confidence interval (CI) 1.02 to 1.17)
- Muscle/skeletal defects: pregnant smokers had a 16% increased odds (OR 1.16, 95% CI 1.05 to 1.27)
- Limb reduction defects: pregnant smokers had a 26% increased odds (OR 1.26, 95% CI 1.15 to 1.29)
- Missing/extra digits: pregnant smokers had an 18% increased odds (OR 1.18, 95% CI 0.99 to 1.41)
- Clubfoot: pregnant smokers had a 28% increased odds (OR 1.28, 95% confidence interval 1.10 to 1.47)
- Craniosynostosis (a condition resulting in an abnormal head shape): pregnant smokers had a 33% increased odds (OR 1.33, 95% CI 1.03 to 1.73)
- Facial defects: pregnant smokers had a 19% increased odds (OR 1.19, 95% CI 1.06 to 1.35)
- Eye defects: pregnant smokers had a 25% increased odds (OR 1.25, 95% CI 1.11 to 1.40)
- Cleft palate: pregnant smokers had a 28% increased odds (OR 1.28, 95% CI 1.20 to 1.36)
- Gastrointestinal defects: pregnant smokers had a 27% increased odds (OR 1.27, 95% CI 1.18 to 1.36)
- Gastroschis (protrusion of the intestines near umbilical cord): pregnant smokers had a 50% increased risk (OR 1.50, 95% CI 1.28 to 1.76)
- Rectum abnormalities: pregnant smokers had a 20% increased risk (OR 1.20, 95% CI 1.06 to 1.36)
- Hernia: pregnant smokers had a 40% increased risk (OR 1.40, 95% CI 1.23 to 1.59)
- Undescended testes: pregnant smokers had a 13% increased risk (OR 1.13, 95% CI 1.02 to 1.25)
There was a reduced odds of pregnant smokers’ babies having hypospadias, a condition where the urethra in the penis is in the wrong position (OR 0.90, 95% CI 0.85 to 0.95) or skin defects (OR 0.82, 95% CI 0.75 to 0.89).
This increased risk was only present for individual defects, and not for all the defects combined. When the reviewers combined all of the defects together (including defects where no difference had been observed between the smokers or non-smokers) there was no overall difference in the odds of a non-smoker and a smoker having a child with a birth defect (OR 1.01, 95% CI 0.96 to 1.07).
The researchers also found that when they pooled data from prospective studies only, similar odds ratios were found.
How did the researchers interpret the results?
The researchers said that maternal smoking is an important risk factor for several major birth defects. They say that public health information should make women aware of these risks and encourage more women to quit smoking before or early on in pregnancy.
This large systematic review has found that women who smoke during pregnancy are more likely to have babies with certain birth defects than women who don’t smoke. The researchers list the comparative difference in risk between smokers and non-smokers for each of these abnormalities.
It is not possible to use these findings to work out the absolute incidence of these birth defects, i.e. how many women actually have babies with these birth defects. This is mainly due to the design of the studies that the review looked at, and the rarity of some of these defects. However, the researchers do say that 3,759 babies were born with these sorts of congenital anomalies in England and Wales in 2008, a year when there were about 708,000 births. This would make the annual incidence of all these defects 5%.
Although a systematic review is the best way to try to answer a question such as this, the individual studies that it reviewed were observational and many were retrospective (looking back at whether babies with birth defects had mothers who smoked, rather than watching smokers to see whether they had babies with birth defects [a prospective study]). This could have led to a potential limitation of recall bias in which women may have inaccurately reported their smoking status dependent on whether their child had a birth defect. The researchers did a sub-analysis including only prospective studies, which showed similar results. They say these analyses suggested that recall bias had not affected their results to a great extent. However, as there may also be a social stigma of smoking while pregnant even in prospective studies it is not possible to know whether reporting bias may have occurred (i.e. some smokers may have reported that they were non-smokers).
The researchers discussed some other potential limitations of their review that they tried to address, including:
- “Publication bias” could potentially have occured if studies that found little or no association between maternal smoking and birth defects were less likely to be published. However, the researchers performed some statistical tests to assess whether this had happened and found that there was none.
- An inherent problem with conducting a systematic review is that data from a wide variety of studies is pooled, which may have differences in their study designs. There are also many other factors that may affect the risk of birth defects, such as maternal age and alcohol use. The researchers pooled risk calculations from the included studies that had taken into account such potential confounding factors. However, the factors that were taken into account may have varied between the included studies.
However, overall, this was a well-conducted study, and its findings are convincing evidence that smoking does increase the risk of some birth defects.
Smoking during pregnancy is already known to be harmful to the baby. Women who smoke who want to become pregnant should consult their GP, midwife or NHS smoking cessation services. Women should stop smoking before they become pregnant, or as early as possible into the pregnancy.
Analysis by Bazian
Edited by NHS Website
Links to the headlines
BBC News, 16 July 2011
The Guardian, 16 July 2011
Links to the science
Human Reproduction Update 2011, First published online: July 11