Catherine Hayes
Although the literature regarding the etiology of oral clefts is extensive, unique causal factors remain unknown. The widely held belief is that oral clefts are of multifactorial etiology, with both genetic predisposition and environmental influence playing a role. In this chapter, the term environmental is used in its broadest sense to include characteristics of the individual, such as age, sex, and race; behaviors, such as smoking and alcohol use; as well as other exposures, such as occupational or environmental ones. While no strong risk factors for oral clefts have been identified, several potential environmental risk factors have been investigated, including smoking, alcohol, caffeine, benzodiazepines, and corticosteroids. In evaluating the role of specific risk factors or a category of risk factors in epidemiology, several criteria are examined for causality. These include consistency, strength of association, biologic plausibility, temporal sequence, and dose-response relationship. Often, there is a remarkable lack of consistency for several of the risk factors discussed in this chapter. Also, many of the measures of association are not strong. In addition, usually there is little information regarding the temporal sequence between exposure and outcome. A dose-response relationship can seldom be demonstrated. Therefore, making inferences about the causality of oral clefts is largely an inconclusive task. In this chapter, the current evidence regarding the role of environmental risk factors in the etiology of oral clefts is discussed.
Maternal Cigarette Smoking
Several studies examining the association between maternal cigarette smoking and oral clefts have found conflicting results. While some have concluded that smokers are more likely to have a child with an oral cleft (Ericson et al., 1979; Khoury et al., 1987, 1989; Shaw et al., 1996), others have not demonstrated a positive, statistically significant association (Saxen, 1975; Seidman et al., 1990; Kelsey et al., 1978; Evans, 1979; Werler et al., 1990; Lieff et al., 1999). A meta-analysis of these studies concluded that the overall odds ratio (OR) for cleft lip with or without cleft palate (CL/P) was 1.29 [95% confidence interval (CI) 1.18-1.42] and that for CP was 1.32 (95% CI 1.10-1.62), indicating a weak association between maternal smoking and oral clefts (Wyszynski et al., 1997).
In some of these studies, positive associations were found for some, but not all, oral cleft groups. This is not unexpected as it is often seen in studies of oral clefting; i.e., a risk factor may be associated with one type of cleft (e.g., CL/P) but not another [e.g., cleft palate (CP)]. This inconsistency is most likely related to the fact that oral clefts represent a heterogeneous group of anomalies. Therefore, specific environmental factors may not have the same effect on the various subgroups. Also, the sample sizes in the subgroups are typically unequal, resulting in a difference in the statistical power to detect an effect, if one exists.
A gene-environment interaction has been proposed in the occurrence of oral clefts in children with the rare allele C2 of the transforming growth factor a (TGF-a) gene born to women who smoked during pregnancy (Hwang et al., 1995; Shaw et al., 1996; Beaty et al., 1997; Romitti et al., 1999). This provided the first empirical evidence for the multifactorial etiology of oral clefts. More details on this may be found in Chapter 23.
In a study conducted in Sweden, infants with CL/P born between 1983 and 1992 (n = 1834) were identified from the Swedish Registry of Congenital Malformations and the Medical Birth Registry. Smoking was recorded as none, <10 cigarettes/day, or ≥10 cigarettes/day. Four groups of oral clefts were evaluated: cleft lip (CL), CL/P, CP, and the Robin sequence. A statistically significant increased risk was observed for isolated CP with an OR of 1.35 and a 95% CI of 1.12-1.63 (Kallen, 1997).
Khoury et al. (1987, 1989) consistently reported an increased risk for oral clefts among smokers in two reports using data from the Maryland Birth Defects Reporting and Information System. In both studies, cases were ascertained at birth and reported by hospitals on a special form that included demographic information, obstetric variables, and prenatal illnesses and exposures. Smoking history was obtained by an obstetric nurse directly from the mother by asking (1) “Did you smoke at any time during pregnancy?” and (2) “If yes, how many cigarettes a day (1-5, 6-10, 11-20, >20).” Statistically significant increased risks were obtained for both CL/P (OR = 1.55, 95% CI 1.10-2.18) and CP (OR = 1.96, 95% CI 1.10-3.50) in the latter study and for CL/P (OR = 3.33, 95% CI 1.3-8.4) in the former.
In another case-control study, mothers were interviewed within 6 months of delivery. This study included 400 cases with CL/P and 215 cases with CP compared to 2710 controls with other malformations (Werler et al., 1990). Smokers were categorized into the following groups: 1-14, 15-24, or ≥25 cigarettes/day. No statistically significant increased ORs were observed for any cleft category after adjustment for age, education, race, maternal alcohol use, vitamin A supplementation, and medical and reproductive histories. Adjusted ORs were 0.70 (95% CI 0.3-1.6) for CL/P and 0.80 (95% CI 0.3-2.2) for CP. In a later extension of this study (Lieff et al., 1999), no significant increased risk for oral clefts were seen among cases in the highest smoking level. Adjusted ORs were 0.91 (95% CI 0.63-1.32) for CL/P and 0.99 (95% CI 0.54-1.91) for CP.
A large prospective cohort study resulted in no statistically significant increase in risk for smoking in either subcategory of clefts (Shiono et al., 1986). In this study, information was obtained from the Kaiser-Permanente Birth Defects Study and the Collaborative Perinatal Project. Smoking status was ascertained by self-report when women were seen for routine prenatal care. In the Kaiser-Permanente Birth Defects study, women were categorized as to smoking status and information regarding congenital malformations was ascertained from medical charts, discharge diagnoses, neonatologists' notes, and autopsy reports. In the Collaborative Perinatal Project, pregnancy outcomes were ascertained by physician examination conducted at birth, 4 months, and 1 year, as well as from medical and autopsy reports. A total of 54 congenital anomalies were evaluated in the pregnancy outcomes of 86,946 live births. The OR for CL/P was 1.1 (95% CI 0.5-2.4) based on 22 cases and that for 0.7 CP was (95% CI 0.3-1.8) based on 19 cases. These results may be due to insufficient power in both study groups.
In a population-based case-control study from the Washington State Birth Registry, no statistically significant increased risk for oral clefts was observed (Van den Eeden et al., 1990). Smoking information was collected from birth certificate data and categorized as “yes/no,” which likely resulted in misclassification bias. This information was available for 96% of controls and 97% of cases. One hundred five isolated CL/P and 37 CP cases were included in the analysis. Adjusted ORs were 1.5 (95% CI 1.0-2.3) for CL/P and 1.2 0.6-2.5) for CP. In a multicenter case-control study (Lorente et al., 2000b), the ORs for first-trimester maternal smoking were 1.79 (95% CI 1.07-3.04) for CL/P and 0.86 (95% CI 0.40-1.87) for CP. Smoking was categorized as regular smokers, never smokers, and ex-smokers; however, modest information was provided on how smoking history was ascertained. In addition, only 63% of eligible cases were interviewed, thereby limiting the interpretation and generalizability of the results.
Data from the U.S. Natality database of 1996 were evaluated in a case-control study (Chung et al., 2000). Information was available for 3,891,494 births, including 2207 cases with oral clefts. Controls (n = 4414) were randomly selected from newborns without congenital anomalies. The database included maternal demographic information, maternal health risk factors, and infant health characteristics. Smoking history, as well as the number of cigarettes smoked per day during pregnancy, was recorded in the database. The smoking categories were none, 1-10, 11-20, and ≥21 cigarettes/day. The overall adjusted OR for CL/P was 1.34 (95% CI 1.16-1.54), with dose-response rates reported for the three smoking categories of 1.50 (95% CI 1.28-1.76), 1.55 (95% CI 1.23-1.95), and 1.78 (95% CI 1.22-2.59), respectively. Unfortunately, birth certificates from the U.S. Natality database do not discriminate between CL/P and CP. This represents the largest study to date examining the association between smoking and oral clefts.
Overall, the results of the studies described above are inconsistent (Table 13.1). These inconsistencies are likely due to different study designs, case definitions, exposure assessments, and timing of exposure. The evidence to date suggests that maternal cigarette smoking during the first trimester of gestation may be weakly associated with risk of oral clefting in the offspring. However, considering the number of women who smoke during pregnancy, even a weak association would have a significant public health impact.
|
TABLE 13.1. Adjusted Odds Ratios (95% CIs) for Highest Smoking Category and Isolated Oral Clefts |
||||||||||||||||||||||||||||||||||||||||||
|
||||||||||||||||||||||||||||||||||||||||||
Alcohol
Several studies have examined the association between maternal alcohol intake and oral clefts in the offspring. For example, Werler and colleagues (1990) explored the association between maternal alcohol use and defects in neural crest-derived structures/tissues of the ear, face, anterior neck, and heart. The number of drinking days per week, as well as the average number of drinks in a drinking day, were recorded separately for beer, wine, and liquor during the first 4 lunar months of pregnancy. Data were collected between 1983 and 1987. Cases were defined as infants with defects of neural crest-derived structures/tissues such as the face, ear, mouth, anterior neck, thymus, thyroid, parathyroid, branchial arch arteries, ventricular septum, and truncus. Cases associated with known Mendelian disorders were excluded. A total of 1464 cases were available for analysis. Controls were infants with malformations of non-neural crest-derived tissues. Three measures of alcohol exposure were used: (1) maximum number of drinks in any 1 day (maximum intensity), (2) number of drinking days per week (average frequency), and (3) number of drinks per drinking day (average intensity). Mothers who consumed at least one drink were compared to mothers who drank less than one drink. The reference category included mothers who reported no intake or a fraction of a drink. Logistic regression analysis was used to estimate relative risks, adjusting for mother's age, smoking history, race, religion, seizure history, diabetes, stillbirth, pregnancy history of rubella or measles, and family history of malformation in first-degree relatives. Consumption of alcohol at the highest level, defined as an average of 5 or more drinks per day on five or more drinking days, was more frequent among cases than controls. The only statistically significant increased risk was seen among CL/P cases in the highest intake category, five or more drinks per drinking day (OR = 3.0, 95% CI 1.1-8.5). Although no significant increase in risk for CP was demonstrated, there were almost twice as many CL/P cases (n = 333) as CP cases (n = 188). Statistical power might have been too modest in the latter group.
In a population-based case-control study conducted in Iowa between 1987 and 1991, cases were obtained from the Iowa Birth Defects registry. The registry includes data on birth defects diagnosed during the first year of life among infants born to Iowa residents (Munger et al., 1996). Ascertainment sources for cases included medical records of all Iowa hospitals, hospitals of neighboring states serving Iowa residents, and prenatal clinics. Diagnoses were confirmed by a medical record specialist, nurse geneticist, and two board-certified geneticists. Cases were selected from all live births, stillbirths, and aborted fetuses. Controls were randomly selected from all Iowa-resident live births listed with the Iowa Department of Public Health. Alcohol use was obtained via telephone interview and reported as number of drinks per month. Mothers were questioned as to alcohol use during the period of 3 months before pregnancy until the end of pregnancy.
Alcohol intake was categorized as 1-3, 4-10, or >10 drinks per month. Participation rates were 74% for cases and 54.6% for controls. More mothers with lower levels of education were non-participants, potentially introducing bias if they had higher levels of alcohol intake. The only statistically significant increased risk was observed in the highest alcohol consumption group (>10 drinks/month) for isolated CL/P (n = 118), with an OR of 4.0 (95% CI 1.1-15.1). In the CL/P group, no significant ORs were recorded for the lower consumption groups (4-10 drinks/month, OR = 3.5, 95% CI 0.8-15.4; 1-3 drinks/month, OR = 1.5, 95% CI 0.9-2.4). No significant increased risk for isolated CP (n = 56) was demonstrated for any of the alcohol intake levels. The models were adjusted for household income, maternal education, maternal smoking, maternal vitamin use, and gender of the child. However, for CL/P the χ2 test for trend was statistically significant, indicating increased risk with increasing alcohol exposure.
In a population-based case-control study of California births from 1987 to 1989, cases were ascertained by review of hospital and genetic center records (Shaw and Lammer, 1999). Cases were diagnosed within 1 year after birth, and controls were randomly selected from all live-born infants from the same geographic area as the cases born during the same time period. Telephone interviews were completed with 734 case mothers (84.7%) and 732 controls (78.2%). Women were asked to report information regarding exposures during the period of 1 month before pregnancy until 3 months after conception. They were asked how often during this period they had any alcoholic beverage, including wine, beer, whiskey, or any other alcoholic drink. Responses were categorized as never, 1-3 during the 4-month critical window period, 1-3/month, 1-4/week, or every day. Those who reported any drink were further asked how often they had five or more drinks at one sitting. No increased risk for oral clefts was found among those categorized as having low alcohol consumption (less than weekly, weekly, daily). However, for those who reported consuming five or more drinks per drinking occasion, the ORs were significantly elevated for isolated CL/P (n = 348, OR = 3.4, 95% CI 1.1-9.7), for CL/P associated with other malformations (n = 99, OR = 4.6, 95% CI 1.2-18.8), and for oral clefts occurring with known syndromes (n = 69, OR = 6.9, 95% CI 1.9-28.6). There was no significant increased risk for isolated CP (OR = 1.0, 95% CI 0.23-8.5). There were no infants with CP associated with other malformations in the highest intake category.
Another population-based case-control study in Iowa (Romitti et al., 1999) showed a significant increase in risk of CL/P among women reporting ≥4 drinks/month during the periconceptional period. Alcohol exposure was ascertained for the period of 3 months before conception to 9 months following conception. Information regarding the number of days per month and the typical number of drinks consumed on those days was used to calculate the maximum number of drinks consumed in a 24 h period. Cases were ascertained through the Iowa Birth Defects registry from 1987 to 1994. Controls were selected from all eligible Iowa live births during the same time period. Information on alcohol intake was assessed for the 1-month period prior to pregnancy through the 9-month period following conception. Women were asked to detail the number of days per month that they consumed alcohol, the typical number of drinks per day, and the maximum number of drinks consumed in a 24 h interval. The average number of drinks per month was then determined for each subject. Response rates were 64% (n = 142) for cases and 60.6% (n = 230) for controls. For CL/P, the ORs for women who reported one to three drinks/month were 1.3 (95% CI 0.9-2.0) and for those who reported four or more drinks per month 2.8 (95% CI 1.2-6.6). For CP, the ORs were 1.1 (95% CI 0.6-1.9) and 1.7 (95% CI 0.5-6.4), respectively.
In a multicenter case-control study in four European countries (France, United Kingdom, Italy, and The Netherlands), cases were defined as any live-born or stillborn child or fetus with a major congenital malformation diagnosed prenatally or during the perinatal period (birth to 6 days) (Lorente et al., 2000b). Controls were born immediately after cases and identified in maternity wards or from birth records. Sixty-three percent of eligible case mothers were interviewed. Alcohol intake was assessed via personal interview in the hospital after birth. Alcohol consumption was dichotomized as <70 g or ≥70 per week. Ex-drinkers included those who reported drinking one drink per day up until a few weeks before pregnancy and who reported stopping at the beginning of pregnancy. The only statistically significant increased risk was for CP (OR = 2.28, 95% CI 1.02-5.09). This increase was significant for isolated CP as well as for CP associated with other anomalies. This finding was not consistent with the studies discussed above, which showed no increased risk for CP.
Each of the above studies used different definitions for highest alcohol intake, ranging from ≥4 drinks/month to ≥5 drinks/drinking occasion (Table 13.2). Despite these differences, these studies consistently demonstrated, with one exception (Lorente et al., 2000b), an increased risk for CL/P with high alcohol intake, with significant point estimates ranging from 2.8 to 4.0. The fact that there was no increased risk demonstrated for CP in all but one study may be explained by inadequate statistical power to detect an increased risk.
|
TABLE 13.2. Adjusted Odds Ratios (95% CIs) for Highest Alcohol Intake Category and Isolated Oral Clefts* |
||||||||||||||||||||||||
|
||||||||||||||||||||||||
Caffeine
Only one case-control study has examined the association between caffeine and oral clefts (Rosenberg et al., 1982). In this study, information on beverage consumption was obtained by asking mothers about their daily consumption of decaffeinated coffee, coffee containing caffeine, tea, cola, and other soft drinks during pregnancy. Those who drank between one and four cups during their entire pregnancy were classified as “occasional drinkers,” and those who drank between five and six cups per week were considered “daily drinkers.” A total of 2030 mothers of malformed infants were interviewed between March 1976 and December 1980. Caffeine consumption was estimated by summing the intake from tea (35 mg), caffeinated coffee (100 mg), and cola (45 mg). There were 299 cases of CL/P and 120 of CP, which were compared to 712 controls with other malformations. Caffeine consumption, examined separately for cola, tea, and coffee, was categorized as none, 1-2, or ≥3 beverages/day or occasional use. Overall daily caffeine consumption was categorized as none, 1-199 mg, 200-399 mg, or ≥400 mg. None of the ORs were significantly greater than 1.0, indicating that caffeine use is not associated with oral clefts. However, no definitive conclusion should be made based on a single study.
Epilepsy
Epileptic women are at increased risk of having a child with an oral cleft (Niswander and Wertelecki, 1973; Dronamraju et al., 1970; Speidel and Meadow, 1972; Monson et al., 1973; Friis, 1979; Hecht et al., 1989; Abrishamchian et al., 1994). It is unclear if it is the epilepsy per se, the drug therapy used to treat the epilepsy, or an underlying genetic link between epilepsy and clefting that accounts for this increased prevalence of clefting disorders among children of epileptic women. Use of anticonvulsants, several of which are known folic acid antagonists, may affect the developing fetus, resulting in congenital malformations, most notably oral clefts (Meadow et al., 1970; Millar and Nevin, 1973; Monson et al., 1973; Kelly et al., 1984; Dansky et al., 1992). Although some authors have pointed out that it may be the mother's epilepsy and not the use of anticonvulsants that leads to the formation of oral clefts, this has not been adequately demonstrated (Dronamraju, 1970; Monson et al., 1973; Friis, 1979; Friis et al., 1981; Kelly et al., 1984). Further, there may be a genetic relationship between epilepsy and oral clefts (Durner et al., 1992).
A review of studies investigating clefts among families with a history of epilepsy found that there was no familial occurrence of clefts in individuals with epilepsy (Hecht and Annegers, 1990). A case-control study by Abrishamchian et al. (1994) demonstrated an increased risk for CL/P among women with epilepsy (OR = 3.78, 95% CI 1.65-7.88). A statistically significant increased risk for CP, however, was not demonstrated.
An increased risk associated with drug therapy for epilepsy during pregnancy was demonstrated, especially for polytherapy, with a 10-fold increased risk for CL/P (OR = 10.5, 95% CI 1.52-59.9). These results come from a population-based study and might represent the most convincing evidence to date of an association between epilepsy, its treatment, and oral clefts. The authors pointed out, however, that the attributable risks for clefts associated with epilepsy and its treatment are very low (3.3% for CL/P and 0.9% for CP). The association between antiepileptic drugs and oral clefts may be related to the fact that many anti-epileptic drugs reduce plasma folate levels (Schwaninger et al., 1999).
Although the underlying mechanism for the development of oral clefts in the offspring of women with seizure disorders is not well understood, the risk of having a child with an oral cleft among women with seizures has been reported to be four to 11 times that of women with no seizure disorders (Dansky and Finnell, 1991). The conflicting reports, however, make it difficult to draw any strong conclusions as to whether it is the use of anticonvulsants by epileptic women during early pregnancy that increases the risk of having a child with an oral cleft. The lack of evidence for a strong familial occurrence may lead to the conclusion that an environmental influence plays a significant role in the association between epilepsy and oral clefts. In summary, the data suggest a strong association between seizure disorders in the mother, and perhaps the use of medications to treat them, and the occurrence of oral clefts in the offspring.
Benzodiazepines
Use of benzodiazepines during pregnancy and oral clefts in the offspring has been the subject of several reports. Benzodiazepines are prescribed to decrease anxiety, induce sedation, and treat eclampsia and pre-eclampsia (Bergman et al., 1992). In one of the first analytical epidemiologic studies, information was obtained from the Metropolitan Atlanta Birth Defects Program (Safra and Oakley, 1975). Drug use information was collected through open-ended questionnaires with specific questions regarding the use of “tranquilizers.” Mothers were also shown a card with samples of several tranquilizers, e.g., chlordiazepoxide (Librium), diazepam (Valium), prochlorperazine (Compazine), amitriptyline (Elavil), and haloperidol (Haldol). All interviews were completed within 4 months after delivery. Cases were mothers of infants with CL/P, and controls were mothers of infants with other congenital anomalies. Increased risks for first-trimester use of diazepam were demonstrated for all CL/P cases (OR = 4.1, 95% CI 1.5-11.5) as well as isolated CL/P (OR = 4.9, 95% CI 1.6-14.7).
In another case-control study (Rosenberg et al., 1983), no association was demonstrated between use of diazepam during pregnancy and oral clefts. Due to the timing of oral cleft development, the authors did not include first exposure to diazepam after the fourth lunar month in their exposed category. In this study, there were 445 CL/P cases, 166 CP cases, and 2498 controls. The control group consisted of mothers of infants with other malformations. Reported exposure to diazepam during the first 4 lunar months of pregnancy was compared between case subgroups and controls. No statistically significant increased risk was observed for any oral cleft subgroup. All relative risks approximated 1.0.
A study conducted in Hungary from 1980 to 1984 did not demonstrate an increased risk for oral clefts among women exposed to benzodiazepines during pregnancy (Czeizel, 1987). In this study, 630 cases with CL/P and 179 cases with CP were compared to matched controls with no significant increase in risk for any oral cleft demonstrated regardless of timing of diazepam use during pregnancy.
A prospective study reported no increase in risk of oral clefts for first-trimester diazepam use for any oral cleft (OR = 1.22, 95% CI 0.17-8.95) (Shiono and Mills, 1984). In a case-control study (Laegreid et al.,1990), benzodiazepine use was assessed by serum analysis from blood samples taken in early pregnancy (prior to 12 weeks). Eighteen cases of congenital malformation were compared to 60 controls. Of these, six cases were classified as CLP. The OR for exposure to benzodiazepines during early pregnancy was 14.5 (p = 0.04). A subgroup analysis was not presented, nor was it clearly stated whether these were isolated clefts or clefts associated with other anomalies.
A meta-analysis of case-control and cohort studies concluded that cohort studies did not demonstrate an association between benzodiazepines and oral clefts, with a summary OR of 1.19 (95% CI 0.34-4.15), while case-control studies did demonstrate a slightly increased risk for oral clefts, with a summary OR of 1.79 (95% CI 1.13-2.82) (Dolovich et al., 1998). These results were not broken down for cleft subgroups, thereby limiting their interpretation. Attaining a large sample size by “lumping” all types of cleft might increase the statistical power to detect an association. However, due to the heterogeneity of oral cleft subgroups, this “lumping” might not be a good idea. Furthermore, the lack of agreement between case-control and cohort studies in the combined analysis was most likely due to the lower number of cases in cohort studies.
If an association between use of benzodiazepines and oral clefts does exist, the increase in risk is likely weak and perhaps modified by other factors. A possible mechanism has been proposed based on the theory that these drugs may disrupt the neurotransmitter mechanisms that regulate embryonic development. There is evidence from mouse models that y-aminobutyric acid inhibits palatal shelf orientation. It is theorized that diazepam may mimic y-aminobutyric acid, leading to the development of oral clefts (Zimmerman, 1984). Although this theory is plausible, the evidence to date from human studies does not indicate a strong association between diazepam use and risk for oral clefts.
Corticosteroids
Corticosteroids, which are administered topically, systemically, or via inhalation, are mainly used to treat asthma, lupus, and rheumatoid disorders; and it has been proposed that they may promote fetal maturation (Czeizel and Rockenbauer, 1997). The potential association between oral clefts and corticosteroid use during pregnancy was first proposed from animal studies (Baxter and Fraser, 1950). A case-control study demonstrated an increased risk of oral clefts in infants of women who used corticosteroids during pregnancy (Carmichael and Shaw, 1999). In this study, mothers were interviewed by telephone and asked about various exposures, illnesses, and medication use during the period of 1 month prior to conception to the third month of gestation. The average interval between interview and delivery was similar for cases (3.7 years) and controls (3.8 years). A statistically significant increased risk was demonstrated for CL/P (OR = 4.3, 95% CI 1.1-17.2) as well as for CP (OR = 5.3, 95% CI 1.1-26.5). Similar increased risks were not observed for other congenital anomalies, such as neural tube defects and conotruncal defects, indicating that the effect of corticosteroid may be specific to oral cleft malformations.
In a case-control study in Spain, Rodriguez-Pinilla and Martinez-Frias (1998) demonstrated an increased risk of oral clefting in the offspring of mothers exposed to corticosteroids during pregnancy. This hospital-based case-control study reported the results of the Spanish Collaborative Study on Congenital Malformations, in which 1184 cases of nonsyndromic oral clefts were examined by physicians and the diagnoses were made within 3 days after birth. There were three sets of controls: (1) the next nonmalformed infant born in the same hospital, (2) infants born at the same hospital as the case infant within 45 days (before or after) of the date, of birth of the case infant, and (3) remaining malformed infants without oral clefts. Medication use was ascertained via physician interview and included month or week of intake, duration of treatment, and daily and total dosages. During the period April 1976 through December 1995, a total of 1,287,345 live-born infants were surveyed. There were a total of 24,038 (1.9%) malformed infants, 1184 oral clefts, and 23,517 controls. Exposure to any systemic corticosteroid during the first trimester was analyzed. Logistic regression was used to estimate ORs, with CL/P as the outcome and exposure to any systemic corticosteroid as the main exposure variable. The model was adjusted for smoking during the first trimester; first-degree relative with CL/P; maternal treatment with corticosteroids during the first trimester; and exposure to antiepileptic drugs, benzodiazepines, metronidazole, or sex hormones. Point estimates were similar when comparing all oral clefts to paired controls or controls born within 45 days of cases, 5.0 and 5.2, respectively. The increased risk was significant only for the second control group (95% CI 1.53-17.06). When the one CP case was excluded, the ORs for CL/P when compared to the second control group as well as to the malformed control group remained statistically significantly increased. No conclusion can be made regarding the risk for CP.
In a case-control study conducted in Hungary (Czeizel and Rockenbauer, 1997), cases were identified using the Hungarian Congenital Abnormality Registry. Controls were selected from the national birth registry and matched to cases according to sex, week of birth, and parents' residential district. A questionnaire was mailed to all eligible cases and controls. In addition, the prenatal care logbook (which included information on prescribed drugs and diseases during pregnancy recorded by the physician) and medical documents were requested of parents. Response rates were 82% for cases (n = 20,830) and 65% for controls 35,727). No significant increase in risk for CP was observed for topical or systemic corticosteroid use. An increase in risk for CL/P with topical steroid use was observed for use during the entire pregnancy (OR = 2.21, 95% CI 1.11-4.39) and use in the first month of gestation (OR = 4.19, 95% CI 1.47-11.97). However, the risk was not significant for use during the second or third month of gestation.
Possible mechanisms for the association between corticosteroid intake and oral clefts have been proposed. One theory is that the steroid acts directly on the fetus, resulting in disruption of glycosaminoglycan or collagen synthesis or both, weakened midline fusion, and loss of amniotic fluid (Greene and Kochhar, 1975). Further, receptors for glucocorticoids have been demonstrated to be more common in palatal mesenchymal cells, which may explain why palatal tissues are affected by this drug (Goldman et al., 1978; Pratt, 1985). Regardless of the underlying mechanism, the data suggest that there is an increased risk (ranging from threefold to ninefold) for CL/P associated with use of corticosteroids during pregnancy. The information regarding the risk for CP is less convincing.
Organic Solvents/Pesticides
Several investigators have examined the relationship between exposure to organic solvents and pesticides and risk of oral clefts. In one study, which investigated the association between organic solvents and oral clefts (Laumon et al., 1996), cases were live-born infants born between 1985 and 1989 in the Rhone-Alpes region of France with an oral cleft, ascertained through any of six maxillofacial surgery clinics. For each case, two controls were selected among newborns without congenital anomalies born in the same delivery unit and whose mothers had the same obstetrician as the case mother. Case mothers were interviewed at the time of their first visit to the surgeon. Controls were interviewed in the maternity unit a few days after birth. The interviewer and questionnaires were the same for cases and controls. Exposures were described by brand names of paints, dyes, glues, or other solvent-containing products. An occupational physician who translated them into trace or significant exposure categories blindly reviewed trade names. Solvents were divided into nine groups: halogenated aromatics, other aromatics, halogenated aliphatics, other aliphatics, alcohols, glycols, ketones, aldehydes, and esters. There were 200 cases of oral clefts and 400 controls. A significant increased risk was demonstrated only for halogenated aliphatic solutions and all oral clefts (OR = 4.4, 95% CI 1.41-16.15) and for CL/P (OR = 4.0, 95% CI 1.25-14.91). Since there was only one case with CP, the results are not clearly interpretable.
A cohort of men who were members of three printer's unions in Norway were studied and categorized into four exposure groups: lead only, solvents only, lead plus solvents, other exposure (Kristensen et al., 1993). Information on children was ascertained by linking parents' union membership files with the birth registry in Norway, which is a record of all live births and stillbirths of >16 weeks' gestation. The standardized morbidity ratio for CL/P was 1.6 (95% CI 0.97-2.5).
In a study in Finland, 388 mothers of infants with oral clefts born between December 1977 and May 1980 were identified from the Finnish Registry of Congenital Malformations (Holmberg et al., 1982). Mothers were questioned as to occupational and in-home exposures to organic solvents via personal interview. Nine cases of CP were born to exposed mothers compared to two cases in the nonexposed group; five CL/P cases were observed among exposed mothers compared to two cases in the nonexposed group. The total numbers of exposed and nonexposed subjects were not included; therefore, an OR could not be calculated. The authors reported a McNemar's test with p < 0.05. There are several limitations of this study, most notably, random and nonrandom misclassification bias.
A study conducted in France, Italy, the United Kingdom, and The Netherlands using European Registration of Congenital Anomalies registries, included live-births, stillbirths (fetal death after 20 weeks' gestation), and induced abortions with a confirmed congenital anomaly diagnosed prenatally, at birth, or within the first week of life (Cordier et al., 1997). Oral clefts were divided into CL/P (n = 109) and CP (n = 52). Controls, which were the first infant born with no anomaly after the case, were selected from the same hospital as cases. A standardized questionnaire was used in all centers to obtain information regarding socioeconomic status, mother's medical and obstetric history, and occupational exposures. Occupational exposures were obtained by questioning the mother as to occupations before and during pregnancy. Occupations were coded following the International Standard Industrial Classification of all Economic Activities and the International Standard Classification of Occupations. Exposures were further categorized as to (1) route (inhalation, cutaneous, or both), (2) level (low, medium, or high), (3) frequency (<5% of work time or 5%-50% of work time), and (4) reliability of assessment (possible, probable, or certain exposure). Only mothers who worked during pregnancy were included. Potential exposure to glycol ethers was assessed by experts for a variety of occupations. The OR for CP was 1.68 (95% CI 0.75-3.76) and that for CL/P was 2.03 (95% CI 1.11-3.73). Several limitations hamper the interpretability of the results of this study, including small numbers, misclassification bias, and recall bias.
A study conducted in the United States evaluated fetuses exposed to agricultural chemicals in Iowa and Michigan. Data were obtained from the National Center for Health Statistics and county and city data book computerized files based on the 1970 U.S. Census, the 1974 U.S. Census of Agriculture, and the 1976 update of Area Health Education Centers (Gordon and Shy, 1981). Iowa and Michigan were selected because they best met the criteria for this study, which included good birth defect monitoring programs, adequate numbers of births per year, participation in a community pesticide study, and being agricultural states. The study population included live-born singletons and controls, which were selected randomly by sampling two per center of live births in each state with a 5:1 controhcase ratio. Exposures were classified as crops only, all pesticidal chemicals and fertilizers, and suspect chemicals. Each county was rated as “high” or “low,” using midpoints of each of the three categories. A score was calculated for each exposure category within each county. Results revealed an increased risk of oral clefts for males and females in Iowa, with ORs of 4.25 (95% CI 1.08-16.67) and 7.13 (95% CI 1.20-42.49), respectively. In Michigan, the ORs for oral clefts were not statistically significantly increased (for males, OR = 1.46, 95% CI 0.59-3.62; for females, OR = 1.00, CI 0.33-2.99). Subgroup analyses were not included; thus, the interpretation of the results is limited. Caution must be used in drawing conclusions based on this study because of the likelihood of bias due to misclassification of exposure, as well as the lumping of heterogeneous cases of oral clefts.
In another multicenter case-control study (Lorente et al., 2000a), data were obtained from the EUROCAT study and included live births, still births, and therapeutic abortions with a major congenital malformation diagnosed prior to birth or during the perinatal period. Controls were recruited for each case. A standardized questionnaire was used to obtain information from cases and controls for the periods before conception and during each trimester of pregnancy regarding socioeconomic status, age, residence, country of origin, mother's medical and obstetric history, alcohol use, tobacco use, drug use, occupation, and hobbies. Information on occupations before and during pregnancy was collected, including a description of tasks, products handled, frequency and timing of use (1 month before conception and first, second, and third trimesters), and typical job categories. Of eligible cases with oral clefts, 63% were interviewed. Controls were matched to oral cleft or to other cases in the parent study. Analysis of occupational exposures was conducted for those women who worked during pregnancy. The ORs were adjusted for center, mother's socio-economic status, urbanization, and country of origin. The ORs for oral clefts were estimated for each type of cleft and each type of occupation. The only statistically significant ORs were reported for CP among housekeepers (OR = 2.8, 95% CI 1.08-7.24) and hairdressers (OR = 5.1, 95% CI 1.10-25.9). The two most likely limitations of this study include misclassification of exposure and the substantial likelihood of spurious associations due to the high number of analyses. Chance is a likely explanation for the findings for CP.
In a case-control study in 15 maternity hospitals in France (Cordier et al., 1992), cases included all products of conception with major congenital malformations born to mothers attending one of the study hospitals, including all live births, still births, and therapeutic abortions with major defects detected prenatally or during prenatal period. For each case, one control who was normal at birth and born immediately after the case in the same maternity ward was selected. Mothers of cases and controls were interviewed in the hospital regarding age; place of residence; family, medical, and obstetric history; use of alcohol or drugs during pregnancy; and specific occupations. Occupational histories of mothers were reviewed blindly by an industrial hygienist, who assessed the presence of any chemical exposure for each work period and estimated frequency of exposure for 10% of the work time, 10% to 50% of the work time, and >50% of the work time. Twenty-nine cases of oral clefts were included. A statistically significant increased risk for all oral clefts was demonstrated for solvents with high frequency (OR = 7.9, 95% CI 1.8-4.9). These results are difficult to interpret since all oral clefts were evaluated as one case group. The limitations of misclassification bias and spurious associations cannot be ruled out as possible explanations for the findings of these studies.
A population-based case-control study (Shaw et al., 1991) included all live-born infants and fetal deaths (≥20 weeks' gestation) diagnosed with orofacial clefts, neural tube defects, conotruncal heart defects, and limb anomalies from most California counties between January 1987 and December 1988 (n = 344,214). Orofacial clefts were ascertained by the California Birth Defects Monitoring Program staff, reviewed by medical geneticists, confirmed by surgical or autopsy report. Controls (n = 972) were randomly selected from all infants born alive (n = 548,844) in the same geographic region and time period who had no major congenital anomaly diagnosed before the first birthday. Telephone interviews were conducted, and the information collected included pesticide exposure and other covariates. The periconceptional window was defined as 1 month prior to and 3 months following conception. Of cases with oral clefts, 85% completed interviews, as did 78% of controls. Pesticide information was collected from the following sources: (1) occupation (reviewed by an industrial hygienist); (2) home use, including weed killer treatments for insects, tree diseases, treatment of pests inside and outside of the home, and treatment of fleas; and (3) residential proximity to agricultural crops. The majority of exposed women were employed in agriculture, in housekeeping, as florists, or as animal handlers. The covariates included in the analysis were periconceptional vitamin use, cigarette smoking, education level, race, and ethnicity. The only statistically significant increased risk was found for CP associated with multiple anomalies and use of insect fogger (OR = 2.0, 95% CI 1.0-4.4). The limitations of this study include misclassification of exposure, the likelihood of spurious associations due to multiple testing, and low power for stratified analysis.
Other Potential Risk Factors
In addition to the studies discussed above, there have been numerous investigations regarding other potential risk factors for oral clefts (Wyszynski and Beaty, 1996). Reports have been published on chemical combustion (ten Tusscher et al., 2000), bed-heating devices (Shaw et al., 1999b; Dlugosz et al., 1992), proximity to hazardous waste sites (Croen et al., 1997), and stress (Carmichael and Shaw, 2000). The findings from these studies are variable, and most are limited by significant methodologic challenges encountered in capturing these exposures accurately. Little information is available to draw any conclusions on the role of these exposures and the risk of oral cleft formation.
Summary
In summary, some of the exposures discussed in this chapter have demonstrated associations with increased risk of oral clefts, most notably use of alcohol, corticosteroids, or anticonvulsant medications and seizure disorders. For the most part, these association are weak or modest for CL/P but not for CP. Perhaps this indicates that CL/P has a stronger environmental component to its etiology than CP. None of the environmental risk factors studied to date stands out as a strong and consistent risk factor for oral clefts. There is a tremendous amount of heterogeneity in the study designs and assessment of exposures. It is likely that the mystery surrounding the etiology of oral clefts will be unlocked as technology advances and the ability to examine gene-environment interactions increases.
References
Abrishamchian, AR, Khoury, MJ, Calle, EE (1994). The contribution of maternal epilepsy and its treatment to the etiology of oral clefts: a population based case-control study. Genet Epidemiol 11: 343–351.
Baxter, HD, Fraser, FC (1950). Production of congenital defects in the offspring of female mice treated with cortisone. McGill Med J 19: 245–249.
Beaty, TH, Maestri, NE, Hetmanski, JB, et al. (1997). Testing for interaction between maternal smoking and TGFA genotype among oral cleft cases born in Maryland 1992–1996. Cleft Palate Craniofac J 34: 447–454.
Bergman, U, Roas, FW, Baum, C, et al. (1992). Effects of exposure to benzodiazepine during fetal life. Lancet 340: 694–696.
Carmichael, SL, Shaw, GM (1999). Maternal corticosteroid use and risk of selected congenital anomalies. Am J Med Genet 86: 242–244.
Carmichael, SL, Shaw, GM (2000). Maternal life event stress and congenital anomalies. Epidemiology 11: 30–35.
Christensen, K, Olsen, J, Norgaard-Pedersem, B, et al. (1999). Oral clefts: transforming growth factor alpha gene variants, and maternal smoking: a population-based case-control study in Denmark, 1991–1994. Am J Epidemiol 149: 248–255.
Chung, KG, Kowalski, CP, Kim, HM, Buchman, SR (2000). Maternal cigarette smoking during pregnancy and the risk of having a child with cleft lip/palate. Plast Reconstr Surg 205: 485-491
Cordier, S, Bergeret, A, Goujard, J, et al. (1997). Congenital malformations and maternal occupational exposure to glycol ethers. Epidemiology 8: 355–363.
Cordier, S, Ha, MC, Ayme, S, Goujard, J (1992). Maternal occupational exposure and congenital malformations. Scand J Work Environ Health 18: 11–17.
Croen, LA, Shaw, GM, Sanbonmatsu, L, et al. (1997). Maternal residential proximity to hazardous waste sites and risk for selected congenital malformations. Epidemiology 8: 347-354
Czeizel, A. (1987). Lack of evidence of teratogenicity of benzodiazepine drugs in Hungary. Reprod Toxicol 1: 183–188.
Czeizel, AE, Rockenbauer, M (1997). Population based case control study of teratogenic potential of corticosteroids. Teratology 56: 335–340.
Dansky, LV, Finnell, RH (1991). Parental epilepsy, anticonvulsant drugs, and reproductive outcome: epidemiologic and experimental findings spanning three decades. 2: Human studies. Reprod Toxicol 5: 301–355.
Dansky, LV, Rosenblatt, DS, Andermann, E (1992). Mechanisms of teratogenesis. Neurology 42(Suppl 5): 32–42.
Dlugosz, L, Vena, J, Byers, T, et al. (1992). Congenital defects and electric bed heating in New York state: a register-based case control study. Am J Epidemiol 135: 1000–1011.
Dolovich, LR, Addis, A, Vaillancourt, JMR, et al. (1998). Benzodiazepine use in pregnancy and major malformations or oral cleft: meta-analysis of case control and cohort studies. BMJ 327: 839–843.
Dronamraju, KR (1970). Epilepsy and cleft lip and palate. Lancet 2: 876–877.
Durner, M, Greenberg, DA, Delgado-Escueta, V (1992). Is there a genetic relationship between epilepsy and birth defects. Neurology 42(Suppl 5): 63–67.
Ericson, A, Kallen, B, Welesterholm, P (1979). Cigarette smoking as an etiologic factor in cleft lip and palate. Am J Obstet Gynecol 235: 348–351.
Evans, DR, Newcombe, RG, Campbell, H (1979). Maternal smoking habits and congenital malformations: a population study. BMJ 2: 171–173.
Friis, ML (1979). Epilepsy among parents of children with facial clefts. Epilepsia 20: 69–76.
Friis, ML, Broeng-Nielsen, B, Sindrup, EH, et al. (1981). Facial clefts among epileptic patients. Arch Neurol 38: 227–229.
Goldman, AS, Shapiro, BH, Katsumata, M (1978). Human foetal palatal corticoid receptors and teratogens for cleft palate. Nature 272: 464–466.
Gordon, JE, Shy, CM (1981). Agricultural chemical use and congenital cleft lip and/or palate. Arch Environ Health 36: 213–221.
Greene, RM, Kochhar, DM (1975). Some aspects of corticosteroidinduced cleft palate: a review. Teratology 11: 47–56.
Hecht, JT, Annegers, JF (1990). Familial aggregation of epilepsy and clefting disorders: a review of the literature. Epilepsia 32: 574–577.
Hecht, JT, Annegers, JT, Kutland, LT (1989). Epilepsy and clefting disorders: lack of evidence of familial association. Am J Med Genet 33: 244–247.
Holmberg, PC, Hernberg, S, Kurppa, K, et al. (1982). Oral clefts and organic solvent exposure during pregnancy. Int Arch Occup Environ Health 50: 371–376.
Hwang, SJ, Beaty, TH, Panny, SR, et al. (1995). Association study of transforming growth factor alpha (TGFA) Taql polymorphism and oral clefts: indication of gene-environment interaction in a population-based sample of infants with birth defects. Am J Epidemiol 242: 629–636.
Kallen, K (1997). Maternal smoking and orofacial clefts. Cleft Palate Craniofac J 34: 11–16.
Kelly, TE, Rein, M, Edwards, P (1984). Teratogenicity of anticonvulsant drugs: the association of clefting and epilepsy. Am J Med Genet 29: 451–458.
Kelsey, JL, Dwyer, T, Holford, TR, Bracken, MB (1978). Maternal smoking and congenital malformations: an epidemiological study. J Epidemiol Community Health 32: 102–107.
Khoury, MJ, Gomez-Frias, M, Mulinare, J (1989). Does maternal cigarette smoking during pregnancy cause cleft lip and palate in offspring? Am J Dis Child 243: 333–337.
Kristensen, P, Irgens, LM, Daltveit, AK, Andersen, A (1993). Perinatal outcome among children or men exposed to lead and organic solvents in the printing industry. Am J Epidemiol 237: 134–144.
Laegreid, L, Olegard, R, Conradi, N, et al. (1990). Congenital malformations and maternal consumption of benzodiazepines: a case control study. Dev Med Child Neurol 32: 432–441.
Laumon, B, Martin, JL, Bertucat, I, et al. (1996). Exposure to organic solvents during pregnancy and oral clefts: a case control study. Reprod Toxicol 20: 15–19.
Lieff, S, Olshan, AF, Werler, MM, et al. (1999). Maternal cigarette smoking during pregnancy and risk of oral clefts in newborns. Am J Epidemiol 150: 683–94.
Lorente, C, Cordier, S, Bergeret, A, et al. (2000a). Maternal occupational risk factors for oral clefts. Scand J Work Environ Health 26: 137–145.
Lorente, C, Cordier, S, Goujard, J, et al. (2000b). Tobacco and alcohol use during pregnancy and risk of oral clefts. Am J Public Health 90: 415–419.
Meadow, SR (1970). Congenital abnormalities and anticonvulsant drugs. Proc R Soc Med 63: 48–49.
Millar, JHD, Nevin, NC (1973). Congenital malformations and anticonvulsant drugs. Lancet 10: 328.
Monson, R, Rosenberg, L, Hartz, SC, et al. (1973). Diphenylhydantoin and selected congenital malformations. N Engl J Med 289: 1049–1052.
Munger, RG, Romitti, PA, Daack-Hirsch, S, et al. (1996). Maternal alcohol use and risk of orofacial cleft birth defects. Teratology 54: 27–33.
Niswander, JD, Wertelecki, W (1973). Congenital malformation among offspring of epileptic women. Lancet 1: 1062.
Pratt, RM (1985). Receptor-dependent mechanism of glucocorticoid and dioxin-induced cleft palate. Environ Health Perspect 61: 35–40.
Rodriguez-Pinilla, E, Martinez-Frias, ML (1998). Corticosteroids during pregnancy and oral clefts: a case control study. Teratology 58: 2–5.
Romitti, PA, Lidral, AC, Munger, RG, et al. (1999). 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 59: 39–50.
Rosenberg, L, Mitchell, A, Parsells, JL, et al. (1983). Lack of relation of oral clefts to diazepam use during pregnancy. N Engl J Med 309: 1282–1285.
Rosenberg, L, Mitchell, A, Shapiro, S, Slone, D (1982). Selected birth defects in relation to caffeine-containing beverages. JAMA 247: 1429–1432.
Safra, MJ, Oakley, GP (1975). Association between cleft lip with or without cleft palate and prenatal exposure to diazepam. Lancet 13: 478–480.
Saxen, I (1974). Cleft lip and palate in Finland: parental histories, course of pregnancy and selected environmental factors. Int J Epidemiol 3: 263–270.
Schwaninger, M, Ringleb, P, Winter, R, et al. (1999). Elevated plasma concentrations of homocysteine in antiepileptic drug treatment. Epilepsia 40: 345–350.
Seidman, DS, Ever Hadani, P, Gale, R (1990). Effect of maternal smoking and age on congenital anomalies. Obstet Gynecol 76: 1046–1050.
Shaw, GM, Lammer, EJ (1999). Maternal periconceptional alcohol consumption and risk for orofacial clefts. J Pediatr 134: 298–303.
Shaw, GM, Nelson, V, Todoroff, K, et al. (1999b). Maternal periconceptional use of electric bed-heating devices and risk for neural tube defects and orofacial clefts. Teratology 60: 124–129.
Shaw, GM, Wasserman, CR, Lammer, EJ, et al. (1996). Orofacial clefts, parental cigarette smoking, and transforming growth factor alpha gene variants. Am J Hum Genet 58: 551–561.
Shaw, GM, Wasserman, CR, O'Malley, CD, et al. (1999a) Maternal pesticide exposure from multiple sources and selected congenital anomalies. Epidemiology 10: 60–66.
Shiono, P, Mills, J (1984). Oral clefts and diazepam use during pregnancy. N Engl J Med 311: 919–920.
Shiono, PH, Klebanoff, MA, Berendes, HW (1986). Congenital malformations and maternal smoking during pregnancy. Teratology 34: 65–71.
Speidel BD, Meadow SR (1972). Maternal epilepsy and abnormalities of the fetus and newborn. Lancet 2: 839–843.
ten Tusscher, GW, Stam, G, Koppe, J (2000). Open chemical combustions resulting in a local increased incidence of orofacial clefts. Chemosphere 40: 1263–1270.
Van den Eeden, SK, Karagas, MR, Daling, JR, Vaughan, TL (1990). A case-control study of maternal smoking and congenital malformations. Paediatr Perinat Epidemiol 4: 147–155.
Werler, MM, Lammer, EJ, Rosenberg, L, Mitchell, AA (1990). Maternal cigarette smoking during pregnancy in relation to oral clefts. Am J Epidemiol 132: 926–932.
Werler, MM, Lammer, EJ, Rosenberg, L, Mitchell, AA (1991). Maternal alcohol use in relation to selected birth defects. Am J Epidemiol 134: 691–697.
Wyszynski, DF, Beaty, TH (1996). The role of potential teratogens in the origin of human non-syndromic oral clefts. Teratology 53: 309–317.
Wyszynski, DF, Duffy, DL, Beaty, TH (1997). Maternal cigarette smoking and oral clefts: a meta-analysis. Cleft Palate Craniofac J 34: 206–210.
Zimmerman, EF (1984). Neuropharmacologic teratogenesis and neurotransmitter regulation of palate development. Am J Ment Defic 88: 548–558.