OBJECTIVE
To investigate the association between maternal cigarette smoking and serum levels of alpha-fetoprotein (AFP), unconjugated estriol (uE3), and hCG and to determine whether it is appropriate to take smoking status into account when screening for fetal Down syndrome.
METHODS
Smoking information was obtained from 23,668 pregnant women at the time of routine screening for fetal Down syndrome. Serum levels of AFP, uE3, and hCG were stratified by smoking status and compared. Individual risks for fetal Down syndrome were analyzed both before and after adjusting the hCG levels for the effect of smoking. The prevalence of Down syndrome in the study population was calculated for both smokers and non-smokers.
RESULTS
The average AFP, uE3, and hCG levels in women who smoked cigarettes were 3% greater (95% confidence interval [CI] 2-4%), 3% less (95% CI 2-4%), and 23% less (95% CI 22-24%), respectively, compared with levels in non-smokers. At comparable maternal ages, women who smoked cigarettes were identified as being at high risk for fetal Down syndrome 40% less often than non-smokers (4.0 versus 7.1% at a risk level of 1:250). Adjusting hCG levels for smoking status reduced the overall false-positive rate from 6.4 to 6.1% (using a term risk cutoff level of 1:250). The predicted effect on detection was small (a 0.7% increase). The prevalence of Down syndrome was lower among smokers, but the difference was not statistically significant (odds ratio 0.51, 95% CI 0.15-1.5).
CONCLUSIONS
Adjusting the serum markers used for Down syndrome screening for the effect of maternal smoking has a small effect on overall screening performance. Given the uncertainty over whether there is a lower birth prevalence among women who smoke cigarettes, such adjustment is not currently justified.