Authors
Korrick SA, Chen C, Damokosh AI, Ni J, Liu X , Cho S, Altshul L, Rayan L, Xu X.

Title:
Association of DDT with spontaneous abortion: a case-control study

Source:
Annals of Epidemiology 11(7): 491-496, 2001.

Summary:
Approximately 15% of clinically recognized pregnancies and an estimated 50% of all pregnancies result in spontaneous abortion (SAB). Advanced maternal age and smoking are the only well established risk factors for SAB. Hypotheses of an association between maternal infection, alcohol, caffeine, parity, and occupational exposures have been tested, but consistent findings have not been demonstrated. Animal models suggest that increased fetal resorption is associated with exposure to dichlorodiphenyltrichloroethane (DDT). This chemical, which has now been banned in most countries, was used as an agricultural and residential insecticide. DDT is highly persistent and continues to be used in some developing areas of the world to control malaria. This pilot case control study set out to investigate the potential association between exposure to DDT and SAB.

Cases and controls were derived from a pre-existing prospective cohort study of women working in the textile industry in China. Recruitment for this parent study began in 1996 and consisted of women who were married, between the ages of 22 and 34 years, nulliparous, without history of SAB or infertility, and planning to have a child. These women were followed forward in time and details of their pregnancies and pregnancy outcomes recorded. Fifteen eligible cases and 15 controls were derived from this population and agreed to participate in the study. A case was defined as a woman whose first pregnancy in the study period resulted in a clinically recognized SAB, and a control was defined as a woman whose first study pregnancy resulted in a live birth. Exposure was assessed in in 1998 by taking a blood sample and testing the serum for isomers of dichlorodiphenlydichloroethylene (DDE) (the metabolite of DDT).

Cases had significantly higher serum levels of DDE than did the controls. Each ng/g serum increase of p,p'-DDE and o,p'-DDE in were associated with a 1.13 (95% confidence interval (CI) 1.02-1.26) and 1.6 (95% CI 1.1-2.3) increase in the odds of SAB respectively, after adjusting for age and body mass index.

In this study, exposure assessment was based on a biological measurement; a strength over other studies that used indirect methods of exposure assessment such as survey data which may be subject to recall bias, or estimation based on occupation. However, it should be considered that the serum samples were collected two years after the initiation of the study, and thus after the index pregnancy and subsequent breastfeeding experience for many subjects. This may have introduced a differential bias in exposure measurements between cases and controls as lactation is an important excretory pathway for DDT, and cases were more likely to breastfeed than controls. The study investigators did attempt to compensate for this discrepancy by estimating the serum DDE levels before the index pregnancy. However, this incorporated several assumptions and may not have been sufficient to compensate for the fact that the controls were more likely to have breastfed than the cases.

Small numbers were also a limitation of this study; a sample of only 15 cases and 15 controls was analyzed. This was in part due to the fact that this was a pilot study. In addition, concentrations of DDE were not expressed on a per lipid weight basis. This makes serum levels difficult to compare with population based studies such as the US National Health and Nutrition Examination Survey, and precludes generalizing the results to other populations.

In summary, this study suggests that SAB may be associated with maternal serum DDE levels. However, larger studies with prospective exposure assessment are required to confirm this finding.