Orofacial clefts are among the most common major birth defects in the United States, with approximately 1 in 870 live births affected by cleft lip with or without cleft palate (CLP), and 1 in 1500 births affected by cleft palate only (CPO).1 Previous studies have identified a number of factors associated with orofacial clefts, including family history,2 genetic factors,3,4 birth order,5 occupational exposures,6 anticonvulsants,7 multivitamin intake,8,9 alcohol,10 and smoking.11–16
The association between maternal smoking and orofacial clefts has been assessed in many studies, and a meta-analysis of these studies suggests a modest positive association for both CLP and CPO.17 However, previous studies have had a number of limitations, including inability to examine a dose–response effect of smoking, inadequate sample size to control for confounders, lack of careful case review and classification of cases, insufficient sample size or clinical detail to examine bilateral and unilateral CLP separately, and insufficient numbers to examine CPO with and without Pierre Robin sequence. The ability to separate the cases into more homogeneous subgroups is critical to understanding the possible mechanisms of a smoking effect. In addition, few previous studies have been able to assess the role of environmental tobacco smoke (ETS) exposure in the etiology of orofacial clefts. The objective of this analysis is to assess the association between maternal smoking, maternal exposure to ETS, and the occurrence of orofacial clefts in a large population-based case-control study of major birth defects in the United States.
The National Birth Defects Prevention Study is an ongoing, multistate, case–control study of environmental and genetic risk factors for major birth defects coordinated by the Centers for Disease Control and Prevention (CDC). Eight Centers for Birth Defects Research and Prevention contributed data for this analysis: Arkansas, California, Iowa, Massachusetts, New Jersey, New York, Texas, and CDC (Atlanta, GA). Methods have been described in detail elsewhere.18 In brief, case infants are ascertained from birth defects surveillance systems, with 7 of the 8 sites employing active case-finding methods. All sites use standard case definitions and a careful clinical review of each case record by a clinical geneticist.19 Because the goal of the study is to understand causes of birth defects related to environmental exposures and genetic polymorphisms, we excluded infants with a recognized or strongly suspected single-gene condition or chromosome abnormality. All sites ascertain case infants diagnosed within 12 months after delivery, and some sites ascertain case infants diagnosed beyond one year of age. Control infants are a random sample of live births (with no major birth defects) from the same population as the case infants. The source population is defined by maternal residence at delivery. Control infants are selected proportionate to the number of births occurring in a given month; some sites select control infants from birth certificate files and some from delivery logs of birth hospitals. Trained interviewers conduct telephone interviews with each mother to collect information on the following exposures during the 3 months prior to conception and during pregnancy: maternal illnesses and medication; pregnancy history; nutrition; caffeine, tobacco, alcohol, and illicit drugs; occupation and environment; and demographics. Seventy-six percent of eligible orofacial cleft case mothers and 69% of eligible control mothers participated in the interview.
In addition to the routine case review by clinical geneticists at each site, all infants with a diagnosis of an orofacial cleft were reviewed by one clinical geneticist (S.A.R.) to ensure that they met the inclusion criteria for this analysis. Infants with an orofacial cleft secondary to another defect (eg, holoprosencephaly or amniotic band sequence) were excluded. All infants were classified as either having an isolated cleft (no additional major defect) or multiple defects (major unrelated defects in at least 2 different organ systems), because these groups likely have different causes, and infants with multiple defects are more likely to have an unrecognized genetic condition than infants with isolated defects.19
Orofacial clefts were analyzed in 2 major categories (CLP and CPO) because of the presumed etiologic and pathogenetic distinctions between these malformations. Further subgroups for analysis included CPO, with and without the diagnostic code for Pierre Robin sequence in their abstracted medical record, isolated and multiple, and males and females. Subanalyses of CLP included bilateral and unilateral, isolated and multiple, males and females and isolated/bilateral and isolated/unilateral. These further subanalyses enhance our capacity to explore possible etiologic mechanisms. For example, CPO in infants with Pierre Robin sequence might be a result of a pathogenetic mechanism involving the mandible, in contrast to cases having primary clefts of the palate.
All mothers were asked whether they had ever smoked cigarettes and, if so, whether they had smoked cigarettes any time from 3 months before pregnancy until their infant's birth. Mothers who reported any smoking in this time period were asked specifically whether they smoked during each of the following 8 time periods: the third, second, or first months before pregnancy; the first, second, and third months of pregnancy; and the second and third trimesters of pregnancy. Mothers were also asked to report the number of cigarettes they smoked per day during each of these time periods.
Mothers reported their exposure to environmental tobacco smoke at home or work during the same time periods. Mothers were asked whether anyone in their household smoked cigarettes in their home during their pregnancy or the 3 preceding months and about each specific time period. Next, mothers were asked whether anyone smoked cigarettes near them at a workplace or school during the same time periods.
Infants were classified as exposed to periconceptional maternal smoking if their mother reported smoking at any time in the month before or the first 3 months of pregnancy. The referent or unexposed comparison group for our assessment of maternal smoking was infants whose mother reported no smoking and no exposure to ETS at home or work during the month before pregnancy or the first 3 months of pregnancy. Among women reporting periconceptional smoking, the level of smoking was classified as heavy (25+ cigarettes per day), medium (15–24 cigarettes per day), or light (1–14 cigarettes per day).
Our assessment of ETS was limited to nonsmoking mothers, defined as mothers who reported no active smoking during the month before or the first 3 months of pregnancy. Infants were classified as exposed to ETS if their mother reported any exposure to household or workplace/school tobacco smoke in the month before pregnancy or the first 3 months of pregnancy. Infants whose mothers reported no ETS exposure during the same period were the referent.
We included infants born on or after 1 October 1997 up through infants with an estimated date of delivery on or before 31 December 2001. Analyses were limited to mothers with completed interviews; 13 case mothers and 35 control mothers with partial interviews were excluded. All main effects were assessed by crude and stratified analyses using SAS version 9.0 (SAS Institute Inc., Cary, NC). To control for potential confounding, an unconditional logistic regression model was fit to the data. We excluded infants having a first-degree relative with an orofacial cleft (14 controls, 52 CLP, 28 CPO), according to maternal report. All effect estimates were adjusted for the following variables, selected a priori based on biologic plausibility and prior studies,14,15 self-reported maternal race/ethnicity (non-Hispanic white, non-Hispanic black, Hispanic, other), study center (8 sites), periconceptional folic acid use (any use in the month before pregnancy or the first month of pregnancy versus no use), maternal age (up to 30 years, 31 years or older), infant sex, prepregnancy obesity (body mass index ≥30 kg/m2, <30 kg/m2), first-trimester alcohol use (any use in the month before or the first 3 months of pregnancy versus no use), maternal education (0–12 years, >12 years), and gravidity (primigravid, multigravid).
Mothers of 933 infants with nonsyndromic CLP, 528 infants with nonsyndromic CPO, and 3390 infants with no major birth defects (controls) completed the maternal telephone interview. The majority of case infants had no other defects: 88% of infants with CLP and 80% of infants with CPO. Among infants with CLP, 67% had unilateral CLP, 22% had bilateral CLP, 1% had central CLP, and 10% had CLP with no information on laterality. One quarter of infants with CPO had a diagnosis of Pierre Robin sequence (Table 1).
The majority of infants with CLP were male (66%), and just over half of infants with CPO were female (53%). Hispanic ethnicity was associated with CLP and, consequently, the 2 study sites with the highest representation of Hispanic mothers (California and Texas) also had a higher prevalence of CLP. Both CPO and CLP were associated with preterm birth, with the strongest associations noted for very preterm births (<32 weeks’ gestation). Young maternal age was modestly associated with CLP, whereas advanced maternal age was associated with CPO. There was some suggestion of associations of low maternal education with both CLP and CPO. Being underweight before pregnancy was associated with CLP, but maternal body mass index was not associated with CPO. The strongest risk factor was having a first-degree family history of orofacial clefts: the odds of having a parent or sibling with an orofacial cleft were 13 to 14 times higher for infants with CLP or CPO than for control infants (Table 2).
Most mothers (97%) who smoked in the month before pregnancy continued during at least the first part of pregnancy. Maternal smoking in the periconceptional period was associated with CLP (adjusted odds ratio [aOR] = 1.3; 95% confidence interval [CI] = 1.0–1.6). Among subtypes of CLP, the strongest associations were for bilateral CLP (1.7; 1.2–2.6) and isolated CLP (1.4; 1.1–1.7). Periconceptional smoking was only weakly associated with CPO (1.2; 0.9–1.5). The results were similar for infants with CPO with and without Pierre Robin sequence. The effects of maternal smoking were similar for male and female infants (Table 3).
Effect estimates were greater for infants exposed to the heaviest levels of maternal smoking (Table 4). Mothers who smoked heavily in the periconceptional period were about twice as likely to have an infant with any orofacial cleft (1.8; 95% CI 1.1–2.9) than were women who did not smoke during this period. The association with heavy smoking was strongest for infants with bilateral CLP (4.2; 95% CI 1.7–10) and CLP with multiple defects (3.1;1.0–10). CPO with Pierre Robin sequence was associated with heavy maternal smoking (2.5; 95% CI 0.9–7.0), whereas a much weaker association was observed among infants with CPO without Pierre Robin sequence (1.4; 0.6–3.2). Among mothers with no folic acid use periconceptionally, the association with smoking was stronger (1.4; 1.1–1.8) than among mothers who took folic acid (1.2; 0.9–1.5).
Among nonsmoking mothers, exposure to any ETS at home or work in the periconceptional period was not associated overall with either CLP or CPO (Table 5). There was a modest association with ETS among infants with CPO with multiple defects (1.7; 1.0–3.0), and a weak association among female infants with CPO (1.3; 0.9–2.0). No dose information was available to quantify the level of ETS exposure.
In this study, periconceptional maternal smoking was associated with orofacial clefts and in particular CLP. The effect was strongest for bilateral CLP and isolated CLP. Among infants of the heaviest smoking mothers, bilateral CLP was 4 times more likely and CPO with Pierre Robin sequence was over twice as likely as among nonsmoking mothers. However, only 2.2% of case-mothers and 1.3% of control-mothers reported this heaviest level of smoke exposure. Overall, the heaviest smokers were about twice as likely to have an infant with an orofacial cleft. There was no overall impact of ETS exposure on the occurrence of orofacial clefts. However, we were not able to assess ETS exposure by dose of the exposure, and any true effect might have been masked by mixing light and heavy ETS exposures. Our findings are consistent with a recent meta-analysis of smoking and orofacial clefts that included 24 studies published between 1974 and 2001.17 For CLP and CPO, the meta-analysis yielded effect estimates essentially the same as our study. Assuming the relation between maternal smoking and orofacial clefts is causal, 4% of all orofacial clefts and 12% of bilateral CLP can be attributed to periconceptional maternal smoking.
Heavy maternal smoking was associated with CPO in the presence of the Pierre Robin sequence but not with other cases of CPO. Palatal clefts arise from failure of the palatal shelves to fuse, which might be due to several mechanisms, for example, a failure of elevation of the palatal shelves or micrognathia and resultant posterior placement of the tongue impairing closure of the shelves (Pierre Robin sequence). Evaluation of the effect of smoking in the presence and absence of Pierre Robin may help to identify the mechanism responsible for the association. Heavy maternal smoking was also associated with CLP (OR 1.8) and most strongly associated with bilateral CLP (OR 4.2). Bilateral clefting is a more severe form, and a 2-fold higher sibling recurrence risk has been reported among infants with bilateral compared with unilateral cleft lip.20 In addition, some evidence suggests that genetic risk factors for clefts may differ by laterality.21 Therefore, bilateral clefts might be more likely among persons with a certain gene variant that interacts with tobacco exposure. If smoking (and particularly higher doses of smoking) is more strongly associated with bilateral clefts, this might direct research efforts to identify gene variants related to tobacco detoxification among bilateral cases compared with unilateral cases or controls.
The effect estimates for intermediate smoking exposure across the various subtypes of clefts did not follow a trend consistent with a linear dose-response (Table 4). This might indicate a higher degree of exposure misclassification from the lowest to the middle exposure categories, due to underreporting of any smoking as well as of the amount of smoking. Alternatively, this could suggest a threshold effect, with maternal smoking having an impact only at higher levels of exposure.
Maternal smoking also might impact birth outcomes differentially for male and female infants. Previous studies have suggested that male infants exposed to maternal smoking are more likely to exhibit fetal distress22; deficits in birth weight and length23–25; and birth defects,26 including CLP and CPO.27 In the current study, the effects of smoking on CLP were similar for males and females. For CPO, for the risk among females was slightly higher than among males. Thus, our data do not support an excess risk among male infants.
Approximately 4000 compounds have been identified in tobacco smoke, including aromatic amines, which can damage DNA and proteins.28,29 A recent study found an association of polymorphic variants of NAT1 (a gene that codes for N-acetyl transferase 1, a key enzyme associated with aromatic amine biotransformation in the first trimester) and the risk of orofacial clefts associated with maternal smoking.30 Maternal exposure to cadmium, another component of cigarette smoke, has been associated with cleft palate in animal models.31,32 Another possible mechanism is that the risk could be mediated through folate levels, since some studies have shown that folic acid-containing vitamins decrease the risk for clefts,8,9 and one study has shown an interaction between smoking and multivitamin use, with the effect for smoking most evident among mothers who did not take multivitamins.33 Both active smokers and nonsmokers with high ETS exposure have lower serum folate and red blood cell folate levels than nonsmokers.34,35 However, in our data, periconceptional folic acid intake did not markedly change the association between smoking and clefts. Another possible mechanism is related to maternal hypoxia, which has been shown to increase the risk for cleft lip in certain strains of mice.36–38 Vasoconstriction of fetal and maternal blood vessels, caused by nicotine and hypoxia,39 could also result in an increased risk for orofacial clefts, mediated by a decreased supply of nutrients to embryonic tissues.40
In this study, ETS exposure was associated with CPO in the presence of multiple defects, with no overall impact of ETS on the occurrence of orofacial clefts. Both associations might simply reflect chance findings. One limitation was our inability to quantify ETS exposure. Studies using cotinine levels to measure dose of ETS exposure have documented adverse reproductive outcomes at the highest levels of ETS exposure.41
The validity of the information reported by mothers for both active smoking and ETS exposure is of concern, both because of the time interval between the exposure (early pregnancy) and the exposure assessment (6 weeks to 24 months after the estimated date of delivery), and because of the social undesirability of smoking during pregnancy. Despite this limitation, studies that have compared repeated measures of self-reported smoking during pregnancy have demonstrated specificity of reported nonsmoking at 94% or greater.42,43 Comparing reported maternal smoking during pregnancy with serum or urine cotinine levels has demonstrated a high validity for maternal interview smoking reports, with interview reports of smoking confirmed for 85% and nonsmoking confirmed for 95% of pregnant women who enrolled in a study in the United States in 1992.44 However, some recent studies have suggested that accuracy of reporting is significantly lower among populations of lower socioeconomic status.45,46
Recall bias might have played a role if case-mothers recalled and reported their exposures more completely than control-mothers, and residual confounding could remain from unmeasured confounders. Although the participation rate of the study is relatively high for this type of study (76% for case-mothers and 69% for control-mothers), nonparticipants differed from participants, limiting the generalizability of our findings. Finally, some diagnostic information, such as Pierre Robin sequence, may be inconsistently documented in medical records.
This analysis provides confirmatory evidence of the adverse effects of maternal smoking on infants, which adds to the previously documented effects of increasing infant mortality, stillbirth, and preterm delivery.47 Although the effect estimates are weak for low levels of maternal smoking, future gene-environment studies will likely identify subgroups of the population at highest risk from smoking during pregnancy. The costs of maternal smoking are substantial and must be measured both in terms of the direct health care costs and by the resulting morbidity and mortality.48 These results support the importance of smoking prevention and cessation programs among all women of childbearing potential. Additional research is needed to determine how best to communicate to reproductive-aged women the risk of orofacial clefts associated with smoking. However, this effort might actually serve to strengthen existing efforts to reduce smoking during pregnancy because women could be more receptive to messages about the risk of orofacial clefts. The consistency of findings for orofacial clefts and smoking suggest an opportunity for prevention of these serious defects.
The authors thank the scientists and staff of all participating sites of the National Birth Defects Prevention Study, in particular Cynthia A. Moore for her assistance with this manuscript.
1. Shaw GM, Carmichael SL, Yang W, et al. Congenital malformations in births with orofacial clefts among 3. 6 million California births, 1983–1997. Am J Med Genet
2. Bixler D, Fogh-Andersen P, Conneally PM. Incidence of cleft lip and palate in the offspring of cleft parents. Clin Genet
3. Murray JC. Gene/environment causes of cleft lip and/or palate. Clin Genet
4. Spritz RA. The genetics and epigenetics of orofacial clefts. Curr Opin Pediatr
5. Vieira AR, Orioli IM. Birth order and oral clefts: a meta analysis. Teratology
6. Shaw GM, Nelson V, Iovannisci D, et al. Maternal occupational chemical exposures and biotransformation genotypes as risk factors for selected congenital anomalies. Am J Epidemiol
7. Alsdorf R, Wyszynski DF. Teratogenicity of sodium valproate. Expert Opin Drug Saf
8. Itikala PR, Watkins ML, Mulinare J, et al. Maternal multivitamin use and orofacial clefts in offspring. Teratology
9. Shaw GM, Lammer EJ, Wasserman CR, et al. Risks of orofacial clefts in children born to women using multivitamins containing folic acid periconceptionally. Lancet
10. Shaw GM, Lammer EJ. Maternal periconceptional alcohol consumption and risk for orofacial clefts. J Pediatr
11. Khoury MJ, Gomez-Frias M, Mulinare J. Does maternal cigarette smoking during pregnancy cause cleft lip and palate in offspring? Am J Dis Child
12. Khoury MJ, Weinstein A, Panny S, et al. Maternal cigarette smoking and oral clefts: a population-based study. Am J Public Health
13. Shaw GM, Wasserman CR, Lammer EJ, et al. Orofacial clefts, parental cigarette smoking, and transforming growth factor-alpha gene variants. Am J Hum Genet
14. Lorente C, Cordier S, Goujard J, et al. Tobacco and alcohol use during pregnancy and risk of oral clefts. Am J Public Health
15. Chung KC, Kowalski CP, Kim HM, et al. Maternal cigarette smoking during pregnancy and the risk of having a child with cleft lip/palate. Plast Reconstr Surg
16. Little J, Cardy A, Arslan MT, et al. Smoking and orofacial clefts: a United Kingdom-based case-control study. Cleft Palate Craniofac J
17. Little J, Cardy A, Munger RG. Tobacco smoking and oral clefts: a meta analysis. Bull World Health Organ
18. Yoon PW, Rasmussen SA, Lynberg MC, et al. The National Birth Defects Prevention Study. Public Health Rep
. 2001;116(1 Suppl):32–40.
19. Rasmussen SA, Olney RS, Holmes LB, et al. Guidelines for case classification for the National Birth Defects Prevention Study. Birth Defects Res A Clin Mol Teratol
20. Mitchell LE, Risch NE. Correlates of genetic risk for non-syndromic cleft lip with or without cleft palate. Clin Genet
21. Farina A, Wyszynski DF, Pezzetti F, et al. Classification of oral clefts by affection site and laterality: a genotype-phenotype correlation study. Orthod Craniofac Res
22. Umapathysivam K, Haskard KA, Meffin E, et al. Correlation of umbilical cord whole blood viscosity and the presence of fetal distress at birth: a mathematical modelling study. Aust NZ J Obstet Gynaecol
23. Bremme K, Lagerström M, Andersson O, et al. Influences of maternal smoking and fetal sex on maternal serum oestriol, prolactin, hCG, and hPl levels. Arch Gynecol Obstet
24. Abell TD, Baker LC, Ramsey CNJr. The effects of maternal smoking on infant birth weight. Fam Med
25. Spinillo A, Capuzzo E, Nicola SE, et al. Factors potentiating the smoking-related risk of fetal growth retardation. Br J Obstet Gynaecol
26. Alderman BW, Takahashi ER, LeMier MK. Risk indicators for talipes equinovarus in Washington State, 1987–1989. Epidemiology
27. Romitti PA, Lidral AC, Murray JC, et al. 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
28. Nelson E. The miseries of passive smoking. Hum Exp Toxicol
29. Hein DW. Molecular genetics and function of NAT1 and NAT2: role in aromatic amine metabolism and carcinogenesis. Mutat Res
30. Lammer EJ, Shaw GM, Iovannisci DM, et al. Maternal smoking and the risk of orofacial clefts: susceptibility with NAT1 and NAT2 polymorphisms. Epidemiology
31. Ferm VH. Developmental malformation induced by cadmium. A study of timed injections during embryogenesis. Biol Neonate
32. Chernoff N. Teratogenic effects of cadmium in rats. Teratology
33. Shaw GM, Nelson V, Carmichael SL, et al. Maternal periconceptional vitamins: interactions with selected factors and congenital anomalies? Epidemiology
34. Mannino DM, Mulinare J, Ford ES, et al. Tobacco smoke exposure and decreased serum and red blood cell folate levels: data from the Third National Health and Nutrition Examination Survey. Nicotine Tob Res
35. McDonald SD, Perkins SL, Jodouin CA, et al. Folate levels in pregnant women who smoke: an important gene/environment interaction. Am J Obstet Gynecol
36. Millicovsky G, Johnston MC. Hyperoxia and hypoxia in pregnancy: simple experimental manipulation alters the incidence of cleft lip and palate in CL/Fr mice. Proc Natl Acad Sci USA
37. Bronsky PT, Johnston MC, Sulik KK. Morphogenesis of hypoxia-induced cleft lip in CL/Fr mice. J Craniofac Genet Dev Biol
38. Bailey LJ, Johnston MC, Billet J. Effects of carbon monoxide and hypoxia on cleft lip in A/J mice. Cleft Palate Craniofac J
39. Ross EM, Butler NR, Goldstein H. Smoking hazards to the fetus [letter]. BMJ
40. van Rooij IALM, Wegerif MJM, Roelofs HMJ, et al. Smoking, genetic polymorphisms in biotransformation enzymes, and nonsyndromic oral clefting: a gene-environment interaction. Epidemiology
41. Kharrazi M, DeLorenze GN, Kaufman FL, et al. Environmental tobacco smoke and pregnancy outcome. Epidemiology
42. Christensen AE, Tobiassen M, Jensen TK, et al. Repeated validation of parental self-reported smoking during pregnancy and infancy: a prospective cohort study of infants at high risk for allergy development. Paediatr Perinat Epidemiol
43. Tomeo CA, Rich-Edwards JW, Michels KB, et al. Reproducibility and validity of maternal recall of pregnancy-related events. Epidemiology
44. Klebanoff MA, Levine RJ, Morris CD, et al. Accuracy of self-reported cigarette smoking among pregnant women in the 1990s. Paediatr Perinat Epidemiol
45. Webb DA, Boyd NR, Messina D, et al. The discrepancy between self-reported smoking status and urine cotinine levels among women enrolled in prenatal care at four publicly funded clinical sites. J Public Health Manag Pract
46. Britton GR, Brinthaupt J, Stehle JM, et al. Comparison of self-reported smoking and urinary cotinine levels in a rural pregnancy population. J Obstet Gynecol Neonatal Nurs
47. U.S. Department of Health and Human Services, Public Health Service. Women and Smoking: a Report of the Surgeon General
. Washington (DC): Office of the Surgeon General; 2001.
48. Centers for Disease Control and Prevention (CDC). State estimates of neonatal health-care costs associated with maternal smoking—United States, 1996. MMWR Morb Mortal Wkly Rep