Pregnancy: Brief Report
Orofacial clefts, particularly cleft lip with or without cleft palate (CL/P) are among the most common congenital malformations worldwide.1 Many studies have suggested a positive association between maternal smoking and the risk of CL/P.2 Few studies, however, have investigated the effect of passive smoking on risk of CL/P. In China, about two-thirds of Chinese adult men are regular smokers, whereas few Chinese women smoke; therefore women's passive smoking is widespread.3 We examined the association between maternal passive smoking and the risk of CL/P in a Chinese population with high prevalence of orofacial clefts (about 3/1000).4
Data came from a population-based case-control study of external malformations in 4 counties (Pingding, Xiyang, Taigu, Zezhou) of Shanxi Province in China, which has been described elsewhere.5 Briefly, cases were live born or stillborn infants with major external birth defects identified by an active surveillance system. Controls were infants without identified malformations, matched to cases by county, sex, maternal ethnic group, and approximate date of conception. Exposure information was collected by trained health workers through face-to-face interviews within 1 week after delivery. The study was approved by the Institutional Review Board of Peking University Health Science Center.
For this analysis, we used data collected between January 2003 and December 2006. The cases were liveborn or stillborn infants assigned in the surveillance system an ICD-9 diagnosis of 749.1 (cleft lip) or 749.2 (cleft lip and palate). All cases were reviewed by 3 pediatricians at Peking University Health Science Center. Passive smoking was defined as exposure to smoking on average at least once per week and at least 1 cigarette each time, from other people at home or in public places, from 1 month before to 2 months after pregnancy. For women who reported passive smoking, we also investigated the frequency of the passive smoking. To increase the power of this study, all controls, including controls for other external malformations as well as controls for CL/P, were compared with CL/P cases. We excluded infants of women who reported active smoking during the periconceptional period. Risks were estimated by the odds ratio (OR) and its 95% confidence interval (95% CI). Multivariable logistic regression was used to estimate the adjusted OR. Potential confounders adjusted for in the analysis included the matching variables (infant sex, season of conception, resident county), plus maternal age, education, occupation, gravidity, parity, history of birth defect in a previous pregnancy, flu or fever in early pregnancy, and periconceptional folic acid use. Maternal occupation, periconceptional flu or fever and infant sex were included in the final model. Data analysis was carried out using SPSS 11.5 (SPSS, Chicago, IL).
We collected questionnaires for 89 CL/P cases and 680 controls. Participation was about 70% for cases and 90% for controls. We excluded 1 case and 23 controls whose mother reported active smoking during the periconceptional period or did not report smoking status. We further excluded 2 infants because of missing data on maternal passive smoking, and 4 infants whose mother was not Han ethnicity, the largest ethnic group in China. This left 88 cases and 651 controls in the final analysis. Nine cases (10%) had other major external malformations.
Case mothers were less likely to be farmers and more likely to have had a periconceptional flu or fever than control mothers. Infants with CL/P were more often boys compared with control infants. The distribution of other characteristics was similar for the 2 groups (Table 1). The OR of CL/P with maternal passive smoking was 1.8 (95% CI = 1.1–2.8). After adjustment for maternal occupation, periconceptional flu or fever and infant sex, the OR was 2.0 (1.2–3.4). There was a positive dose-response relationship between exposure frequency and the risk of CL/P: adjusted ORs were 1.6 and 2.8 for exposure frequencies of 1–6 times per week and more than 6 times per week, respectively (P for trend = 0.0012, Table 2). The adjusted ORs were slightly higher among boys (2.3 [1.2–4.5]) than among girls (1.7 [0.8–3.9]). Neither the exclusion of the 9 CL/P cases with other malformations or exclusion of subjects with missing covariates in multivariable analyses changed the results substantially (not shown).
Our study suggests that maternal passive smoking during periconceptional period may increase risk of CL/P in offspring. The risk appeared to rise with more frequent exposure. Few previous studies have explored the effect of passive smoking on CL/P, although the link between active maternal smoking and CL/P has been well established.2 Shaw et al6 found nonsmoking women who reported exposure to smokers in their home at a close distance (within 6 feet) were at increased risk for isolated CL/P (adjusted OR = 2.0 [95% CI = 1.2–3.4]). Little et al7 suggested a weak effect of passive smoking in nonsmokers. The adjusted OR for the highest tertile of hours of passive smoking, compared with no exposure, was 1.5 (95% CI = 0.7–3.2) for CL/P. Lie et al8 observed a 1.6-fold increased risk of isolated CL/P associated with passive smoking (adjusted OR = 1.6 [95% CI = 1.0–2.5]). A recent nested case-control study9 observed an increase in risk of CL/P (OR = 3.3 [95% CI = 0.9–12]) for women with levels of cotinine that indicate exposure to passive smoking. However, based on a larger case-control study, Honein et al10 did not find an association between exposure to any passive smoking at home or work and the risk of CL/P (OR = 1.2 [95% CI = 0.7–1.8]) among nonsmoking mothers. The discrepancy of these results may result from different definitions of passive smoking, the different ethnic/racial groups, or other factors.
We found the risk of CL/P associated with maternal passive smoking tended to be higher for boys than for girls. This is similar to Romitti et al's report,11 which suggested a greater effect of active smoking in boys. However, Little et al7 found that the CL/P risk with smoking was greater in girls, and Honein et al10 suggested the effects of maternal smoking on CL/P were similar for boys and girls.
Several limitations should be considered in interpreting our results. Although the study was population-based, differential participation between cases and controls (70% vs. 90%) may have created selection bias. Recall bias is the main concern in a case-control study. In our study, women were interviewed within the first week after delivery, and local women would have little knowledge about a possible association between passive smoking and CL/P. These factors should have reduced the problem of recall bias. The prevalence of passive smoking (54%) among control mothers is similar to that reported in the Chinese National Study.7 For the associated cases, other malformations were limited to only external defects, and therefore we could not identify syndromes or sequences.
Our study has several strengths. The study was conducted in a high clefts prevalence area, which increased the likelihood of identifying important risk factors. The high prevalence of men's smoking and low prevalence of women's smoking in the study population, combined with the restriction to nonsmoking mothers, allowed us to examine the effect of maternal passive smoking unconfounded by maternal active smoking.
Women's passive smoking is an important public health problem in China owing to the high proportion of male smokers and poor awareness of smoking damage. Our results, together with evidence of the association of CL/P with maternal active smoking, suggest that passive smoking could be an important contribution to CL/P risk in China.
1. Mossey PA, Little J. Epidemiology of oral clefts: an international perspective. In: Wyszynski DF, ed. Cleft Lip and Palate: From Origin to Treatment
. New York: Oxford University Press; 2002:127–158.
2. Little J, Cardy A, Munger RG. Tobacco smoking and oral clefts: a meta analysis. Bull World Health Organ
3. Yang G, Fan L, Tan J, et al. Smoking in China: findings of the 1996 National Prevalence Survey. JAMA
4. Li Z, Ren A, Liu J, et al. High prevalence of orofacial clefts in Shanxi Province in northern China, 2003–2004. Am J Med Genet Part A
5. Li Z, Ren A, Zhang L, Guo Z, Li Z. A population-based case-control study of risk factors for neural tube defects in four high-prevalence areas of Shanxi province, China. Paediatr Perinat Epidemiol
6. Shaw GM, Wasserman CR, Lammer EJ, et al. Orofacial clefts, parental cigarette smoking, and transforming growth factor-alpha gene variants. Am J Hum Genet
7. Little J, Cardy A, Arslan MT, Gilmour M, Mossey PA. Smoking and orofacial clefts: a United Kingdom-based case-control study. Cleft Palate Craniofac J
8. Lie RT, Wilcox AJ, Taylor J, et al. Maternal smoking and oral clefts: the role of detoxification pathway genes. Epidemiology
9. Shaw GM, Carmichael SL, Vollset SE, et al. Mid-pregnancy cotinine and risks of orofacial clefts and neural tube defects. J Pediatr
10. Honein MA, Rasmussen SA, Reefhuis J, et al. Maternal smoking and environmental tobacco smoke exposure and the risk of orofacial clefts. Epidemiology
© 2010 Lippincott Williams & Wilkins, Inc.
11. Romitti PA, Lidral AC, Munger RG, Daack-Hirsch S, Burns TL, Murray JC. Candidate genes for nonsyndromic cleft lip and palate and maternal cigarette smoking and alcohol consumption: evaluation of genotype-environment interactions from a population-based case-control study of orofacial clefts. Teratology