Maternal smoking: age distribution, levels of alpha-fetoprotein and human chorionic gonadotrophin, and effect on detection of Down syndrome pregnancies in second-trimester screening

Objectives To study the levels of maternal serum alpha‐fetoprotein (AFP) and human chorionic gonadotrophin (hCG) in the second trimester in smokers and non‐smokers with unaffected and Down syndrome pregnancies; to examine the rate of smoking in different maternal age groups in a population having routine prenatal screening; and to assess the effect of smoking on the detection rates for Down syndrome and corresponding false‐positive rates, both overall and in different maternal age groups. Methods Information on maternal smoking status, maternal age and serum marker levels was collected from case note searches and the screening programme database on 2272 unaffected singleton pregnancies, 36 unaffected twin pregnancies and 103 singleton Down syndrome pregnancies. Results In unaffected pregnancies the smokers had a median age 3.3 years less than the non‐smokers, while in the Down syndrome cases the corresponding age difference was 2.0 years. Median analyte levels in multiples of the median (MoM) in the unaffected singleton pregnancies were, for non‐smokers: AFP=0.97, hCG=1.04; and for smokers, AFP=1.04, hCG=0.80. In the Down syndrome pregnancies the medians were, for non‐smokers: AFP=0.69, hCG=2.49; and for smokers, AFP=0.70, hCG=1.53. Correction for smoking status gave median MoMs of 1.0 for both AFP and hCG in the unaffected pregnancies in both smokers and non‐smokers. In the Down syndrome cases the corrected medians were, for non‐smokers: AFP=0.67, hCG=2.29; and for smokers, AFP=0.73, hCG=1.99. Before correction for maternal smoking the overall detection rate for Down syndrome was 66.7% with a false‐positive rate of 6.2%. After correction the detection rate was 67.7% with a false‐positive rate of 4.9%. Between the smoking and non‐smoking groups there was a significant difference in the detection rate (37.5% versus 76.0%) and the false‐positive rate (1.8% versus 8.1%), which disappeared after correction for smoking status (detection rate 62.5% versus 69.3%, false‐positive rate 3.9% versus 5.4%). No evidence of a lower incidence of Down syndrome in smokers was found. Conclusions While correcting AFP and hCG results for maternal smoking status will have little impact on the overall detection rate for Down syndrome, it may reduce the false‐positive rate and will improve the accuracy of the risks given to individual women. Copyright © 2002 John Wiley & Sons, Ltd.