Pre-pregnancy handling of antineoplastic drugs and risk of miscarriage in female nurses.

OBJECTIVE To examine the association between handling of antineoplastic drugs (AD), use of exposure controls, and risk of miscarriage. METHODS Women in the Nurses' Health Study 3 self-reported AD administration and use of engineering controls (EC) and personal protective equipment (PPE) at baseline. Nurses who reported pregnancies after baseline were included in this analysis. We estimated the Hazard Ratio (HR) of miscarriage in relation to baseline AD handling using multivariable Cox proportional regression modified for discrete time data. RESULTS 2,440 nurses reported 3,327 pregnancies within a median of 3 years after baseline (range:1-8 years), of which 550 (17%) ended as miscarriages. Mean (standard deviation) age at baseline was 29.7 years (4.3). At baseline, 12% of the nurses self-reported currently handling AD and 28% previously handling. Compared to nurses who never handled AD, nurses who handled AD at baseline had a HR of miscarriage of 1.26 (95% CI: 0.97, 1.64) after adjusting for age, body mass index, and smoking. This association was stronger for losses after 12 weeks gestation (HR=2.39 [95% CI: 1.13, 5.07]), and among nurses who did not always use EC and PPE. Nurses who did not always use gloves had a HR of 1.51 (95% CI:0.91, 2.51) compared to 1.19 (95% CI:0.89, 1.60) for those always using gloves; nurses who did not always use gowns had a HR of 1.32 (95% CI:0.95, 1.83) compared to 1.19 (95% CI:0.81, 1.75) for nurses always using gowns. AD handling prior to baseline was unrelated to risk of miscarriage. CONCLUSION(S) We observed a suggestive positive association between AD handling and miscarriage, particularly among nurses who did not consistently use PPE and EC. These associations appeared to be more evident among second trimester losses.