Effect of smoking status on inhibin‐A in second‐trimester prenatal screening for Down syndrome

Abstract

Inhibin-A is a useful marker in antenatal screening for Down syndrome and forms part of the quadruple test (alphafetoprotein, unconjugated oestriol, human chorionic gonadotrophin and inhibin-A) and the integrated test (nuchal translucency and pregnancy associated plasma protein-A in the first trimester of pregnancy and the quadruple test in the second trimester).Rudnicka et al. 1 reported that the level of inhibin-A among women who smoke at the time of their screening test is about 60% higher than that seen in women who do not smoke. They recommended that a simple adjustment could be used to ensure that the screening performance of the quadruple test was similar among smokers and non-smokers. Cowans and Spencer 2 reported that the level of inhibin-A in the first trimester of pregnancy was only around 10% higher among women who smoked compared with women who did not, which suggests that the difference increases with gestation and a single adjustment in the second trimester maybe too simplistic. Further, it has recently (personal communication) been suggested that the increase among women who smoke was not as high as 60%.To investigate this, we used data from 216 568 women screened as part of the Wolfson Institute screening programme between April 1999 and March 2013 with data on smoking status (the data in Rudnicka et al. are a subset of this). Figure 1 shows the ratio of smoker to non-smoker median inhibin-A MoM according to the gestational age (completed weeks) at the time of the screening test with 95% confidence intervals and the number of women who smoke at each gestational week of pregnancy. Overall, the inhibin-A MoM was 51% higher among smokers, and the effect of smoking increased with gestational age (1.23 at 14 weeks to 1.75 at 22 weeks). However, when we evaluated other second-trimester markers used in prenatal screening for Down syndrome, we found no significant gestational effect. Therefore, the reasons that inhibin-A demonstrate a gestational effect and other markers of placental origin, such as hCG, do not is unknown.Adjusting for smoking status has little overall effect on screening performance, 1 but if an adjustment were applied, it would be more accurate to apply a gestationspecific adjustment, as this would maintain screening performance of the quadruple test across the entire gestational age range for smokers as similar to that expected among non-smokers. The proportion of women who attend for their quadruple test at 14 completed weeks rather than later has increased. In 2006, 4.5% attended at 14 completed weeks compared with 12.7% in 2012, which may explain the decrease in the overall ratio of inhibin-A in smokers to non-smokers. Figure 1 Effect of smoking on inhibin-A according to gestational age Prenatal Diagnosis 2014, 34, 406-407