Ronald G. Munger
Inadequate maternal nutrition during pregnancy has been suspected as a possible cause of oral clefts in humans since at least the early 1900s (Strauss, 1914). Evidence for this view has accumulated from several areas of research, including animal experiments, observational studies of human populations, and in some limited cases human experimental studies, yet many gaps remain in our understanding of the etiology of oral clefts. Each of these areas of research will be reviewed in this chapter.
Oral clefts comprise one of the most common groups of birth defects in the world (World Health Organization, 1998), and there is considerable geographic variation in occurrence. Asian populations in Asia and Native Americans (of Asian genetic ancestry) have higher rates of oral clefts than other major groups (Chung and Kau, 1985; Croen et al., 1998). The burden of oral cleft birth defects is great because affected children require substantial medical care and speech training. In areas of the developing world where corrective surgery is not available, affected children face having a serious disability for their entire lives and are often shunned by others. Hence, there is a strong public health imperative for a better understanding of the causes and prevention of oral clefts in all populations, and these insights may provide clues to the etiology of other birth defects.
Poverty has been linked to the occurrence of oral clefts. In Glasgow, oral clefts were found to be more common in socioeconomically deprived areas than elsewhere in the city (Womersley and Stone, 1987). Studies in metropolitan Manila found evidence of a social class gradient in oral cleft risk: the prevalence was higher among births in a hospital serving a more impoverished group of patients at the Philippine General Hospital than among births to wealthier patients in the Makati financial district a few kilometers away (Lasa and Manalo, 1989). Social class may be a determinant of maternal nutritional status, but it may also be related to other environmental exposures that have a bearing on the risk of oral clefts.
Several environmentally induced causes of clefting have been identified. Maternal cigarette smoking is perhaps the best-studied environmental risk factor for oral clefts. In a meta-analysis of 11 published studies, smoking was associated with an increased risk for both cleft lip with or without cleft palate (CL/P) and cleft palate alone (CP) (Wyszynski and Beaty, 1996). The association between maternal smoking and risk of clefting may be due in part to confounding in studies that do not account for maternal diet or nutritional status. Maternal alcohol use has been associated with an increased risk of oral clefts in some studies (Werler et al., 1991; Munger et al., 1996; Romitti et al., 1998; Shaw and Lammer, 1999; Lorente et al., 2000) but not all (Natsume et al., 2000). Examination of the social and dietary context in which alcohol consumption takes place may help to clarify its relation to risk of oral clefts. For example, the risk from alcohol consumed while drinking beer at a pub is not likely to be equivalent to that from the same amount of alcohol consumed while drinking wine with a nutritious meal. Other environmental exposures linked to increased risks for oral clefts include agricultural chemicals, solvents, and medications (reviewed in detail in Chapter 13).
Genetic and developmental studies have provided evidence that the etiology of cleft lip (CL) with or without cleft of the primary (hard) palate is different com pared to clefts affecting only the secondary (soft) palate (Fraser, 1955). Oral clefts are often classified into four groups: isolated CL/P, isolated CP, CL/P in association with other major birth defects, and CP in association with other major birth defects. In isolated cases, affected individuals have no other physical or developmental anomalies (Murray, 1995). Most studies suggest that about 60% to 70% of cases are isolated and, thus, not associated with any syndrome of malformations (Jones, 1988). Most of the remaining 30% to 40% of cases occur with a pattern of multiple malformations and either are classified as recognized syndromes related to chromosomal abnormalities, one of several hundred known Mendelian disorders, or known teratogen exposures (e.g., phenytoin or alcohol) or become candidates for newly defined syndromes. Some studies of maternal nutrition and oral clefts have used subgroups of clefts in their analyses, and this approach is clearly needed in all future studies. More information on the clinical classification of oral clefts may be found in Chapters 5 and 6 of this volume.
Neural Tube Defects and Maternal Nutrition: Relevance for Oral Clefts
The literature on nutritional causes of neural tube defects (NTDs) is extensive and relevant to oral clefts because some structures of the craniofacial region are derived from cephalic neural crest cells; thus, these two conditions may share some developmental origins (Been and Lieuw Kie Song, 1978; Lammer et al., 1985). Malnutrition was suspected as a cause of NTDs as early as the 1960s, when it was noted that women in the United Kingdom from the poorer social classes had an elevated risk of NTDs in their pregnancies (Hibbard, 1964; Hibbard and Smithells, 1965). Attention was later focused on the role of folic acid, but early attempts at supplementation trials provoked controversy because they were not fully controlled by randomization and there was no masking of the treatment status (Smithells et al., 1976; Wald and Polani, 1984). The Medical Research Council Vitamin Study was a randomized, controlled trial of folic acid and multivitamin supplementation that provided strong evidence of the role of folic acid supplementation in reducing the risk of NTDs in the pregnancies of women who had previously had a pregnancy with an NTD (MRC Vitamin Study Research Group, 1991). Folic acid supplementation has also been shown to reduce the primary occurrence of NTDs in Hungary (Czeizel and Hirschberg, 1997) and China (Berry et al., 1999).
The mechanism for the protective effect of folic acid against NTDs remains unknown, and several features of folate-homocysteine metabolic pathways have been under investigation (Daly et al., 1997). Red cell folate is a more accurate indicator of long-term folate status compared to plasma folate (Herbert, 1989). Research has focused on understanding the shape of the doseresponse curve between the maternal red cell folate level, the risk of NTDs, and the levels of folate intake required to result in a desired red cell folate level (Daly et al., 1997; Wald et al., 1998). The risk of NTDs decreases in a continuous dose-response relationship with increasing maternal level of red cell folate. A high level of plasma total homocysteine is an indicator of impaired folate and vitamin B12 metabolism has been observed in mothers of children with NTDs compared to unaffected controls (Steegers-Theunissen et al., 1994; Mills et al., 1995). The major catabolic pathway for homocysteine involves cystathione β-synthase with vitamin B6 as a co-factor. Low levels of vitamin B12 have also been implicated as a cause of NTDs and, thus, may be relevant for the discussion of oral clefts (Kirke et al., 1993)
An irony of the folic acid-NTD success story is that other nutrients, including those identified years ago in the animal experiments described below, have been largely overlooked in recent epidemiologic studies of human populations.
Experimental Animal Studies on Maternal Nutrition and Oral Clefts
The role of maternal nutrition in the formation of oral clefts has been studied extensively in experimental animal models over the past 70 years (Table 14.1). A surprising number of specific nutritional deficiencies have been found to cause oral clefts, and in most cases multiple birth defects and excessive fetal loss also have been observed. In light of the history of animal experimentation, it is remarkable that recent investigations in human populations have been so narrowly focused, neglecting many nutrients that may be important causes of oral clefts.
The first reported observations from animal studies were anecdoctal, yet with scant evidence authors were eager to promote nutritional interventions for the prevention of oral clefts in humans. As early as 1914, maternal diet was thought to influence the occurrence of CP among lions at the London Zoological Gardens (Pickerill, 1914). Cleft palate was reported among nearly all of the lion cubs born during a period when the lionesses were fed a diet of meat alone. When the diet was changed to whole small animals, which were consumed entirely, the occurrence of oral clefts among subsequent births dropped immediately and considerably.
|
TABLE 14.1. Nutritional Factors Found to Cause Oral Clefts in Experimental Animal Models |
||||||||||||||||||||||||
|
Pickerell attributed the reduction to the added consumption of bones and suggested that women with a family history of frequent oral clefts should be given a diet during early pregnancy with added powdered fresh bone or lime phosphate. In the same year, experiments with cats in the United States were reported, in which oral clefts were easily induced by unspecified manipulations of diet and environmental conditions (Strauss, 1914). Strauss noted that in humans the occurrence of clefts seemed to be more frequent among the poor and suggested that a “hypoplastic condition of the blood” due to “faulty metabolism” was responsible even if the mother generally seemed to be wellnourished and that these nutritional factors acted in concert with “hereditary tendencies” to result in oral clefts. Nearly 100 years later, the interaction between maternal nutritional status and genes is still invoked as a cause of oral clefts, yet the specific components remain unclear.
The identification and isolation of vitamins led to animal experiments with great specificity, and the first vitamins characterized were included in the earliest teratologic studies. Hale, of the Texas Agricultural Experiment Station, reported in 1933 that sows raised on rations deficient in vitamin A gave birth to pigs without eyes and in 1935 that cleft lip and palate (CLP) and ear and kidney malformations also occurred in these animals. Hale cited further breeding experiments as evidence against genetic causes of the defects in his experimental animals. When control animals were bred on the same rations but with added cod-liver oil or were allowed to graze on green pastures, no birth defects were produced. Impressed with the protective effect of green pastures, Hale (1935) wrote that “perhaps we have been forcing our spinach on the wrong victims; it ought to be administered to the mothers instead of the children.”
Warkany and the Development of Animal Models of Teratogenesis
Josef Warkany began his pioneering work on nutrition and birth defects in 1938 after immigrating to Cincinnati from Vienna (Kalter 1993). Warkany was inspired by Hale's previous work and set out to challenge the prevailing view that nearly all congenital malformations were hereditary (Warkany, 1954). At that time, it was commonly asserted that maternal nutrition could not have an important effect on the development of the embryo, which was viewed as a small parasite that could easily extract the small amount of needed nutrients from the unlimited stores of the mother.
Warkany had developed a broad view of nutrition and health from his early childhood experiences. As a boy foraging for food for his family during wartime in the Austrian countryside, Warkany was struck by the sight of physically and mentally disabled cretins and developed a lifelong interest in the role of maternal iodine nutrition in this condition. Likewise, he became interested in the role of nutrition in rickets, then common in the European mountain villages he frequented (Kalter, 1993). Warkany often cited these as examples of evidence of important nutritional causes of birth defects.
Warkany and his first co-worker in the United States, Rose Nelson, developed techniques for breeding Sprague-Dawley rats in states of nutritional deficiency. Female rats fed on rations made from cornmeal, wheat gluten, calcium carbonate, and sodium chloride gave birth to young with multiple skeletal and craniofacial anomalies, including CP. The addition of 2% dried pig liver to the rations completely prevented the anomalies, and these investigators set out to discover the specific vitamins responsible. B-complex vitamins, including thiamine, riboflavin, niacin, pyridoxine, and pantothenic acid, had recently become available in crystalline form, and Warkany found that riboflavin alone prevented oral clefts and all of the other malformations among offspring of female rats bred while on the deficient diet. In further studies of the timing of deficiencies during gestation, Warkany found that riboflavin supplementation before day 13 prevented the malformations but that later supplementation did not. Other investigators confirmed that riboflavin deficiency caused malformations in rats (Noback and Kupperman, 1944; Giroud and Boisselot, 1947, 1951; Leimbach, 1949; Piccioni and Bologna, 1949) and fowl (Lepkovsky et al., 1938; Romanoff and Bauernfeind, 1942). The mechanism by which riboflavin deficiency caused oral clefts and other malformations was not clear, and subsequent studies in the late 1940s of riboflavin deficiency in mice did not replicate the cleft findings in the rat studies (Fraser and Fainstat 195la). A subsequent study by Kalter and Warkany (1957) reported malformations in mice caused by riboflavin deficiency, but little if any work has been done on the teratogenic effects of riboflavin deficiency since then.
Vitamin A
The role of vitamin A deficiency in the growth and reproduction of rats was further studied by Warkany, who reported patterns of malformations including defects of the skeleton, eyes, kidney, urogenital tract, and aortic arch; no oral clefts were mentioned however (Warkany and Shraffenberger, 1944). The possible role of vitamin A in craniofacial abnormalities did not appear in the literature until many years later, when reports on the role of excess exposure to vitamin A and related compounds in oral clefts and other birth defects in animal studies were published. Retinoic acid and other retinoids were discovered to cause oral clefts and other malformations in several animal species, including mice, rats, rabbits, and primates; and the specific effects were found to be dependent on the precise timing of exposure during early fetal development (Kochhar, 1967, 1973; Fantel et al., 1977; Kochhar et al., 1984; Soprano and Soprano, 1995). After gestational day 10 in the mouse, retinoic acid-induced oral clefts result from abnormally small palatal shelves that do not contact one another and medial epithelial cells of the shelves differentiate into oral-like epithelial tissue; after day 12, shelves of normal size form but fail to fuse and medial cells proliferate and abnormally differentiate into nasal-like epithelium (Abbott et al., 1988a,b; Abbott and Pratt, 1988a,b). Abbott and Birnbaum (1990) found that RA exposure disrupts the specific expression pattern of growth factors, including transforming growth factors α, β1, and β2, during this period of development and suggested that the timing of this disruption determines whether the cleft is due to the formation of small palatal shelves that never contact one another or to abnormal differentiation of medial epithelial cells and failure of fusion of the palatal shelves. Other important variables found to influence the expression of oral clefts after retinoid exposure in animal experiments include form of retinoid, dose, species studied, rate of metabolism (Ross, 1999), interaction with ethanol and other exposures including vitamin A (Whitby et al., 1994), and genetic susceptibility (Tyan and Tyan, 1993).
Folate
The role of folate in malformations began to be studied in the late 1940s by several groups of investigators using folate antagonists in animal experiments. By this time, Nelson had moved to Berkeley from Cincinnati, further developed the experimental methods she had established earlier with Warkany, and in 1947 demonstrated that folic acid deficiency caused the resorption of 26% of rat fetuses using a purified diet with the addition of succinyl-sulfathiazole (Nelson and Evans, 1947). Folic acid deficiency induced by folate antagonists was shown to cause beak malformations in chickens (Karnofsky et al., 1949). Oral clefts and other malformations were produced in rats by folic acid deficiency in combination with sulfonamide treatment to suppress folic acid production by maternal intestinal bacteria (Giroud and Boisselot, 1951; Giroud and Lefebvres, 1951). The folate antagonist 4-aminofolic acid was shown to be sufficient to cause resorption of embryos (Thiersch and Philips, 1950), and Nelson found that addition of the folic acid antagonist Xmethyl-pteroylglutamic acid to her previous protocol further increased adverse reproductive effects in rats and that the timing of the induced folate deficiency was more important than the dose of folate antagonist in producing malformations and fetal deaths (Evans et al., 1951; Nelson et al., 1952). Folate deficiency induced between days 1 and 9 of gestation resulted in 100% fetal resorptions and that induced between days 10 and 11 resulted in 100% stillbirths, in which 94% had CP; when the induction of folate deficiency was delayed until day 15, only 6% were stillborn and none had cleft palates. Nelson et al. (1952) noted that the malformations induced by folic acid deficiency were similar to but more severe than those produced by deficiencies of riboflavin (Warkany and Nelson, 1941), vitamin A (Wilson and Warkany, 1948), pantothenic acid (Boisselot, 1949, 1951), and vitamin B12 (O'Dell et al., 1951) and concluded that similar patterns of malformations result from interference with any one of a number of metabolic pathways during critical developmental periods.
Folate antagonists were subsequently found to be highly toxic to human embryos (Meltzer, 1956; Thiersch, 1956; Warkany et al., 1959; Emerson, 1962; Goetsch, 1962; Milunsky et al., 1968; Shaw and Steinbach, 1968; Powell and Ekert, 1971). Methotrexate, a folate antagonist used in cancer therapy, was found to cause oral clefts, craniofacial anomalies, and other malformations in rats and rabbits, although rabbits required a dose level 60 times greater than rats to produce similar effects (Jordan et al., 1977). The A/WySn strain of mice is genetically predisposed to oral clefts and has been used to examine the interaction of genetic susceptibility to oral clefts with folate status. The occurrence of oral clefts in the offspring of A/WySn dams was increased after they had been fed a low-folate diet (Lidral et al., 1991) and reduced when dams were treated with folinic acid, the most stable intermediate of folic acid metabolism, via continuous delivery by way of an osmotic minipump (Paros and Beck, 1999).
A high incidence of CP was observed over a 30-year period in a breeding line of Boston Terrier dogs, and in 1982 animals from this group were supplemented with a daily dose of 5 mg folic acid for the purpose of reducing aggressiveness in the mother toward the pups after birth. Cleft prevention was apparently not the original reason for the supplementation (Elwood and Colquhoun, 1997). The occurrence of oral clefts in the period after supplementation began (1982–1997) was significantly less than in the presupplementation period (1974–1981) [4.2% of 191 pups from supplemented pregnancies were affected vs. 17.6% of 51 unsupplemented pups; odds ratio (OR) = 0.20, 95% confidence interval (CI) 0.07–0.62]. Studies in several animal models have thus shown that folic acid plays an important role in fetal viability and normal development of the craniofacial region and other structures. The mechanisms by which folate deficiency causes oral clefts and other malformations remain unclear.
Vitamin B6
Vitamin B6 has been shown to protect against teratogen-induced oral clefts in many animal studies. Vitamin B6 is the generic term for 3-hydroxy-2-methylpyridine derivatives that have the biologic activity of pyridoxine. Vitamin B6 plays many vital roles in amino acid metabolism, including transamination and decarboxylation reactions, acting as the coenzyme of glycogen phosphorylase, steroid hormone action, and acting as a coenzyme in the degradation of homocysteine; thus, there are many potential pathways in which vitamin B6 protects against oral clefts. Corticosteroid exposure has been used as a model for teratogen-induced oral clefts in mice since the 1950s (Fraser and Fainstat, 1951a,1951b; Kalter, 1957; Bonner and Slavkin, 1975; Melnick et al., 1981), and Peer et al. (1958a,1958b) demonstrated that vitamin B6 supplementation reduced the occurrence of corticosteroid-induced oral clefts in mice. Vitamin B6 deficiency alone was demonstrated to cause CP and other birth defects in mice (Davis et al., 1970). Miller (1972) demonstrated that dietary vitamin B6 deprivation in mice resulted in isolated oral clefts 20% of the offspring, and the prevalence increased to 61% to 100% when deoxypyridine, a vitamin B6 antagonist, or cortisone was administered to the pyridoxinedeficient mothers. Later studies demonstrated that vitamin B6 also prevented the induction of oral clefts by vitamin A excess (Yamaguchi, 1968), cyclophosphamide (Dostal and Schubert, 1990), β-aminoproprionitrile (Jacobsson and Granstrom, 1997); hence, the role of vitamin B6 in cleft prevention may be complex and involve several different mechanisms.
The susceptibility to corticosteroid-induced oral clefts in mice is strain-dependent, raising the possibility that the protective effects of vitamin B6 may depend on genotype (Marazita et al., 1988). Vitamin B6appears to regulate the activity of hormones, including androgens, estrogens, progesterone, retinol, retinoic acid, thyroid hormones, calcitriol, and gluco-corticoids, by binding to nuclear receptor proteins and influencing transcription and gene expression (Jacobsson and Granstrom, 1997). Cytoplasmic levels of glucocorticoid receptors in the developing secondary palate of mice are associated with the susceptibility to glucocorticoidinduced oral clefts (Salomon et al., 1979; Salomon and Pratt, 1979). These authors suggested that glucocorticoids inhibit the growth of palatal mesenchyme cells at a critical point in palatal formation. Yoneda and Pratt (1982a,b) found that vitamin B6 inhibited the specific binding of labeled glucocorticoid to cytosolic receptors from cultured mouse palatal mesenchyme and suggested that it reduces the occurrence of cortisone-induced oral clefts by altering the binding of glucocorticoids to their cytoplasmic receptors and, consequently, nuclear acceptors.
Plant Toxins
Naturally occurring toxins in plants cause malformations, including oral clefts in sheep, cattle, and other grazing animals. Extensive studies of the role of plant toxins have been conducted since the 1950s in the arid mountainous regions of the western United States, where many poisonous plant species have caused epidemics of birth defects with significant morbidity and mortality in livestock. The production and sequestering of toxins by plants appear to have evolved as defense mechanisms against herbivores, especially in arid regions where plant growth is quite limited. Many of the plants with potent toxins are in the legume family of flowering plants, and alkaloids are often found to be the active compound. Exposure of pregnant ewes to Veratrum californicum (false hellebore) on day 14 of gestation resulted in cyclopic craniofacial anomalies and CLP in the lambs (James, 1999). The teratogens isolated from V. californicum include the steroidal alkaloids cyclopamine, jervine, and cycloposine (Keeler, 1968,1990). A potential mechanism for the teratogenic action of the Veratrum alkaloids includes disruption of signal transduction mediated by Sonic hedgehog proteins, responsible for developmental patterning of the mammalian craniofacial region (Cooper et al., 1998; Incardona et al., 1998). Oral clefts and skeletal defects are also caused in cattle that graze on Lupine species (James, 1999). Humans are exposed to alkaloids from a variety of plant sources, including Nicotiana (tobacco), Lobelia (medications), Punica (pomegranate), Duboisia (medications), Carica (papaya), and Prosopis (mesquite); but the relevance of these alkaloid exposures to human birth defects is largely unexplored.
Human Observational Studies of Maternal Nutrition and Oral Clefts
Peer et al. (1958b) were among the first to publish data on a series of mothers of cleft patients that included vitamin use. The subtitle of their 1958 paper on the results of 400 human pregnancies is “protective effect of folic acid and vitamin B6 therapy,” yet no evidence from human studies was presented to support these conclusions. Recognizing that existing medical records were inadequate, the authors sent questionnaires to 1000 mothers of cleft-affected children who were surgical patients at St. Barnabus Medical Center in Newark, New Jersey. Four hundred mothers responded. The racial composition of the sample was not reported, although it was the impression of the authors that oral clefts were less common among African-Americans than in other groups. Because 77% of the oral clefts occurred in families without a previous history, the authors believed that environmental causes were important. Five pairs of identical twins with oral clefts were described, three of whom had one affected and one unaffected twin. Douglas (1958) had cited similar discordant sets of identical twins in which the cleft-affected twin had a lower birth weight and length than the unaffected twin as evidence that differences in the fetal environment, perhaps related to unequal maternal blood supply between twins, may give rise to a cleft in the more compromised fetus. Peer et al. (1958b) noted that 90% of the mothers in their series either had not used vitamin supplements or had used them late in pregnancy or irregularly; no data from a control group were cited, making the significance of this observation uncertain.
Montreal Case-Control Study
Fraser and Warburton (1964) collected family history and medical information from mothers of 187 children with CL/P and 59 with CP in the Montreal area and compared these data to reports from 90 mothers of children with genetically determined diseases that did not include oral clefts. This study thus represents the first of a growing list of case-control studies of oral clefts (Table 14.2). The authors were primarily interested in whether stressful life events were associated with the risk of oral clefts via excessive adrenal gland secretion of glucocorticoids. Both case and control mothers reported more stressful events during the pregnancy of their affected children than during the pregnancies of their other unaffected children, and the authors attributed this finding to maternal recall bias. Prescribed nutritional supplements were taken during pregnancy by 67% of the CL/P mothers, 77% of the CP mothers, and 71% of the control mothers; the authors concluded that use of prescribed dietary supplements was not associated with the occurrence of CL/P or CP. The strengths of this early case-control study include collection of data from a structured questionnaire, use of a control group with children affected with known genetic diseases, and consideration of the role of maternal recall bias by comparing the results of reports from affected pregnancies to those of unaffected pregnancies from the same mothers. No information was provided on the demographic characteristics of the sample, type of nutritional supplements, or diets of the mothers.
Finland Case-Control Study
The Finnish Register of Congenital Malformations was used by Saxen (1975) in a study of medication and vitamin supplement use during pregnancy in relation to oral clefts. The register covered 98% of Finnish births in 1967; thus, this report was the most representative population-based sample at that time. Mothers of 232 CL/P children, 232 CP children, and 232 control children were contacted by midwives after their deliveries while working with the Maternity Welfare Centres and interviewed with a standard questionnaire. Controls were unaffected children born in the same district just prior to a case child; thus, they were matched on place of residence and time of pregnancy. Use of iron, vitamin supplements, or both was tabulated by trimester of pregnancy; during the first trimester, use was greater for CL/P case mothers compared to controls (63.8% vs. 54.4%) and similar for CP case mothers and controls (46.6% vs. 47.8%). Use of supplements increased to 82% for all mothers in the third trimester, with no significant differences between groups. Several medications were used more often among cases vs. controls in the first trimester, including salicylates, other antipyretic analgesics, opiates, tetracyclines, chloramphenicol, and antineurotics. The strengths of the Finnish study included being among the first population-based case-control studies of nutritional supplements and cleft risk, an adequate sample size, and assessment of exposure by trimester of pregnancy. The limitations of the study include the simple exposure variable of all supplement use combined (most of which may have been iron alone), lack of dietary data, possible overmatching of cases and controls by place of residence, and lack of information on possible confounding factors. The medication data indicated that case mothers had more illnesses during pregnancy than controls; thus, it is possible that these illnesses compromised the nutritional status of case mothers more often than controls.
Baltimore Case-Control Study of Maternal Metabolic Factors
Maternal metabolic factors were studied in a case-control study by Niebyl et al. (1985). Cases were 59 mothers of children with CL/P recruited through surgical clinics and advertisements in the Baltimore area. Controls were a convenience sample of 56 mothers of unaffected children drawn from the authors' private patients or hospital employees. The sample size was thus small, was not population-based, and yielded controls of higher socioeconomic status (40% college graduates vs. 24% among cases); thus, the results may be confounded by these differences. No information on diet or supplement use was presented. The authors collected blood specimens and conducted a broad spectrum of biochemical studies, including folate status and phenytoin pharmacokinetics. In cases compared to controls, mean red cell folate was higher (143 vs. 138 ng/ml) and serum folate was lower (6.0 vs. 6.3 ng/ml), but these differences were not statistically significant. In a more detailed substudy of 10 cases and 10 controls, the mean baseline red cell folate was higher in cases vs. controls (150 vs. 118 ng/ml) but the mean peak serum folate after a 1.0 mg dose of pteroylmonoglutamate was lower in cases vs. controls (63.5 vs. 83.3 ng/ml); these differences were not statistically significant, but this approach may be useful in characterizing differences in folate metabolism in larger samples of cleft case mothers vs. controls. After a 1-week period of phenytoin treatment (300 mg/day), the mean peak serum folate values of cases and controls were similar (56.4 vs. 52.9 ng/ml), as were the phenytoin elimination kinetics.
|
TABLE 14.2. Case-Control Studies of Maternal Nutritional Status in Pregnancy and Risk of Oral Cleftsa |
||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
|
||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
Hill's Case-Control Study in England
Medication use during pregnancy was the focus of a study of 676 mothers cleft-affected children and an equal number of mothers of unaffected children born during 1983-1984 in England (Hill et al., 1988). Cases were ascertained by the congenital malformation surveillance system of the Office of Population Censuses and Surveys. Data were not presented by type of cleft. Each case child was matched with the next unaffected child born in the same general medical practice. The family medical practitioners of case and control children were identified from the District Health Authorities and visited by a medical officer of the study, who abstracted information on maternal medical and family history from the written records of the medical practice. No demographic data were presented for cases or controls, and no data were collected directly from mothers. Medical records provided very limited information on supplement use during the 3-month period before the last menstrual period: vitamin use was indicated in only 1.9% of cases and 1.3% controls, folic acid supplementation was indicated in only 1.2% of cases and 0.9% of controls. More case than control mothers reported use of one or more of 13 categories of medications. Thus, this study design was useful for evaluation of exposure to medications but inadequate to capture data on use of nutritional supplements. Prescription of nutritional supplements by physicians for mothers in this study seemed nearly nonexistent, and no data were provided on the use of supplements obtained outside of the medical system.
Metropolitan Atlanta Congenital Defects Program
The Atlanta Birth Defect Case-Control Study investigated a wide variety of risks for birth defects, including oral clefts, in the Atlanta metropolitan area (Khoury et al., 1989). Cleft cases were ascertained during 1968-1980 via multiple sources as part of the routine surveillance of the Metropolitan Atlanta Congenital Defects Program and included 238 with CL/P and 107 with CP; data were reported on these separate groups, but each group included isolated, multiple, and syndromic birth defects. Controls were mothers selected among 323,421 live births to women in the same population and frequency-matched to cases by birth period, race, and hospital of birth The focus of the report was on maternal cigarette smoking, but data on vitamin use were presented, although not controlled for potential confounding factors. Mothers were interviewed by telephone, and the participation rate was 70%. Mothers were asked whether they used vitamins during the periconceptional period, defined as 3 months prior to conception through 3 months after pregnancy began. No further information was provided regarding what constituted vitamin use, i.e., frequency, regularity, and type of vitamins used; and no information was provided on maternal diet. The percent of mothers reporting vitamin use was 53.3% for CL/P, 58.9% for CP, and 60.7% for controls; the OR for risk of oral clefts associated with vitamin use was 0.74 for CL/P (95% CI 0.56-0.98) and 0.93 for CP (95% CI 0.62-1.40).
Western Australia
A case-control study in Western Australia, in which an inverse association was found between maternal dietary intake of folate and risk of NTDs (Bower, Stanley, 1989), was extended to examine midline birth defects, some of which included oral clefts. (Bower and Stanley, 1992). In the original study, all cases of NTDs in infants born during mid-1982 through 1984 and ascertained via the Western Australia Birth Defects Registry were eligible. Two control groups were drawn from the same population, one with major malformations other than NTDs and the other including unaffected live-born infants. Midline defects were identified in 59 infants of the malformed control group in the original study, and these were designated as cases in the study of midline defects. Infants with chromosomal anomalies or known Mendelian syndromes were excluded. The controls used in the midline study were 115 unaffected infants used as controls in the original study of NTDs. Mothers were interviewed by telephone and completed a self-administered food-frequency questionnaire regarding their diet for the period 3 months before to 9 months after their last menstrual period before pregnancy. The questionnaire was limited to food items that contained substantial amounts of folate and included questions on cooking methods, dietary changes, and use of vitamin supplements. Intake of free folic acid and total folate was calculated using food-composition tables and adjusted for cooking practices (Paul and Southgate, 1979; Truesdale, 1984). Folate intake was divided into quartiles, and the risk of midline defects was evaluated with the lowest quartile as the reference level in multivariate models that included several potential confounding factors. No consistent relationship was observed between risk of midline defects and either free folic acid or total folate intake. The power of this study was limited by the small sample size, and the heterogeneity of the sample further compounded this problem. Only 13 of the 59 cases with midline defects had oral clefts. While there remains a strong and plausible argument for shared causes in this diverse group of birth defects, the most promising approach in identifying these causes is to focus on cleft defects per se and to use homogeneous sub-groups such as isolated CL/P and CP in case-control analyses.
California Birth Defects Monitoring Program
The California Birth Defects Monitoring Program provided a large population-based sample of oral clefts for the case-control study reported by Shaw et al. (1995). Cases were infant births or fetal deaths during 1987-1989 with oral clefts ascertained via multiple sources throughout California. All cases were reviewed by a medical geneticist and classified as isolated CL/P or CP either in isolation or with multiple birth defects. Known monogenic syndromes were excluded. Controls were randomly sampled from all unaffected infants born during the same period and matched to cases by county of residence. Mothers of 731 cases and 734 controls were interviewed by telephone and asked about use of vitamin and mineral supplements, intake of cold breakfast cereals, and other traits and demographic factors. The periconceptional exposure period of interest for vitamin use was defined as 1 month before conception through 2 months afterward, and mothers who started vitamin use in the third month after conception were excluded from the analyses. Doses of folic acid were imputed as 0.8 mg for reports of prenatal vitamin use and 0.4 mg for all other vitamin supplements that contained folic acid. Fewer case than control mothers reported any use of multivitamins containing folic acid during the periconceptional period (57.2% vs. 69.1%) and during the 1-month period before conception (14.3% vs. 18.9%). The ORs indicating risk of specific types of oral cleft with any use of multivitamins with folic acid in the periconceptional period were 0.50 for isolated CL/P, 0.61 for CL/P with multiple birth defects, 0.73 for isolated CP, and 0.64 for CP with multiple birth defects. The 95% CI for isolated CL/P excluded 1.0 and was narrow (95% CI 0.36-0.68), based on the substantial sample size of 348 cases in this subgroup; the CIs for the other subgroups included 1.0 but were based on much smaller sample sizes. The risk of oral clefts by subgroup was evaluated over four levels of average daily intake of folic acid from multivitamins, with the lowest (none) as the reference level. The highest level of intake (≥1.0 mg/day) was uninformative because only four cases of isolated CL/P, one case of multiple CL/P, and no CP cases were in this level and the number of controls in this level was not presented. The ORs for isolated CL/P were 0.47 (95% CI 0.33-0.67) for 0.4 mg or less folic acid per day and 0.52 (95% CI 0.36-0.76) for 0.4 to 0.9 mg/day, indicating that the risk of this type of cleft was reduced by half with any use of multivitamins containing folic acid, and no dose-response relationship was discernible. The results appeared similar for the other subgroups of oral clefts but more variable, perhaps because of smaller sample sizes in each. Vitamin use may be associated with other healthy lifestyle characteristics, so Shaw et al. (1995) rigorously controlled for potential confounding factors, including race and ethnicity, education, age, gravidity, smoking, alcohol use, family history of oral clefts, maternal medical history, and use of folate-antagonist medications; the reduced risk with multivitamin use was not influenced by these factors. Throughout this report, the relevant exposure was described as use of “multivitamins containing folic acid,” yet this placed an artificial emphasis on folic acid because most of the multivitamin preparations contained folic acid and a large number of other vitamins and minerals. The authors noted that of the 954 case and control mothers using vitamins, only two reported using a supplement with folic acid only and only 27 used a multivitamin supplement without folic acid. Shaw et al. (1995)extended their analyses by examining the risk of oral clefts in mothers who consumed cold cereals, an unspecified proportion of which were fortified with vitamins and minerals. Among the mothers who used no multivitamins in the periconceptional period, the risk of clefting was reduced among those with daily cereal consumption (OR = 0.41, 95% CI 0.17-0.98) for isolated CL/P and suggestive for other cleft types, although the small number of case mothers with these characteristics (22 case and 32 control mothers) limited the analyses. The Californian study, one of the most exhaustive to date, provides compelling evidence that maternal multivitamin use in the periconceptional period is associated with a reduced risk of oral clefts, but the specific role of folic acid remains uncertain.
Hungarian Case-Control Surveillance of Congenital Abnormalities
The Hungarian Case-Control Surveillance of Congenital Abnormalities has collected data since 1980 on cases of birth defects identified among infants in the first 3 months of life and in fetal deaths ascertained by the Hungarian Congenital Abnormality Registry. Controls were selected from unaffected births via the national birth registry and matched to cases by sex, birth week, and district of parental residence. Data on cleft cases were reported by Czeizel et al. (1996) and later updated (Czeizel et al., 1999); the latest analyses were based on 1246 matched pairs of CL/P cases and controls and 537 matched pairs of CP and cases controls. Data were collected from mothers on use of vitamin supplements, use of medications, and medical and pregnancy history by mailed questionnaire, review of antenatal logbooks and medical records, and (for nonresponding cases) interviews by visiting nurses. Folic acid supplementation was evaluated from these records, and according to the authors, the commonly available form during the period of study was a 3.0 mg tablet prescribed by obstetricians in the amount of 1 to 3 tablets/day; the usual dose was assumed to be 6.0 mg/day. Use of folk acid alone was uncommon (14% of controls and 13% of cases) and not further evaluated; among the remaining majority of participants, other vitamins, minerals, and medicines were taken; but these details were not reported. The onset of folic acid supplement use was reported by month of pregnancy for cases and controls, and the critical period of exposure was defined as months 1 to 2 for CL/P and 1 to 3 for CP. The ORs describing risk of oral clefts for mothers using folic acid supplements in the critical period vs. all others were 0.72 (95% CI 0.55-0.92) for CL/P and 0.86 (95% CI 0.66-1.13) for CP. Shaw et al. (1995) excluded from their analyses mothers in California who began use of supplements after 3 months; thus, the remaining comparison was mothers using supplements in the periconceptional period vs. nonusers. Applying this approach to the 1999 Hungarian data yields an OR of 0.62 (95% CI 0.46-0.83) for the risk of CL/P among mothers using folic acid supplements in the critical period vs. nonusers. When late users (beyond the second month of pregnancy) were compared to nonusers, the OR for risk of CL/P was 0.77 (95% CI 0.62-0.96); because this exposure period is after cleft formation, these results indicate that the association between supplement use and reduced risk of oral clefts may be confounded by other personal characteristics associated with supplement use and risk of oral clefts. A similar analysis of the CP data yields an OR of 0.71 (95% CI 0.52-0.98) for use in the critical period (months 1-3) vs. nonuse, and the contrast of late supplement users (beyond the third month) and nonusers yields an OR of 0.79 (95% CI 0.54-1.15). The Hungarian findings, like the Californian findings, reveal associations between the use of nutritional supplements and reduced risk of clefting, but the independent role of folic acid per se also remains uncertain. Some of the reduced risk associated with maternal vitamin use appears to be due to confounding factors because mothers who initiated supplement use after the period of cleft formation also appeared to have a reduced risk of clefting compared to mothers who did not use supplements.
Strasbourg Case-Control Study
Plasma samples were routinely collected during the first antenatal visit of women in the Strasbourg area for subsequent case-control comparisons of mothers of children affected and unaffected with malformations (Stoll et al., 1999). The bank of samples represented births occurring between 1985 and 1994, and each was stored at -20°C until the assays were completed. Biochemical studies of maternal plasma trace elements and vitamin B12, vitamin A, and folic acid were evaluated in a nested case-control study. Mothers of 170 children with congenital malformations ascertained by the local registry of congenital malformations were included as cases, and two controls per case were matched for sex, age of mother, hospital of delivery, obstetric history, and social class. The sample included only 14 cases of CL/P. The other major groups and numbers of malformations included heart (n = 72), vesicorenal reflux (n = 8), limb reduction defects (n = 12), spina bifida (n = 6), and hydrocephaly (n = 6). There were no signficant differences between all malformations and controls or between specific subgroups of defects in mean plasma zinc, copper, magnesium, manganese, folate, vitamin B12, or vitamin A. The sample size was adequate to detect a 1.0 standard deviation difference in mean trace elements at α = 0.05 and β = 0.10 between the congenital heart defects (n = 72) and controls but lacked power for comparisons with the other subgroups including oral clefts. The close matching of cases and controls on a number of characteristics including social class and hospital of delivery may have resulted in an artificial similarity between cases and controls.
Boston-Philadelphia-Toronto Case-Control Studies
The Slone Epidemiology Unit has combined surveillance of birth defects and medication use with case-control studies since 1976 in participating hospitals in the Boston, Philadelphia, and Toronto metropolitan areas. Hayes et al. (1996) reported results of a case-control study of oral clefts that, like the Western Australia Study, was derived from a previous investigation of NTDs. Cases included mothers of infants live- or stillborn during 1988-1991 with CL/P (n = 195) or CP (n = 108). Mendelian conditions were excluded. Controls were 1167 mothers of live-born children with congenital anomalies other than NTDs or midline defects, and the larger groups were renal anomalies (n = 177), gastrointestinal anomalies (n = 156), limb defects 139), craniosynostosis (n = 115), and chromosomal anomalies (n = 90). Case and control mothers were interviewed by a trained nurse within 6 months of birth regarding demographic background, medical and family history, dietary practices, and use of nutritional supplements and medications. Dates of initiation and cessation of supplement use were determined with the aid of a calendar, highlighting the exposure periods of interest. The relevant period of periconceptional exposure was defined as 1 month prior to the last menstrual period though the fourth month of pregnancy (16 weeks after last menstrual period). Medication and supplement bottles were examined for content and dose information. Supplementation was defined as use of a folic acid-containing multivitamin or a folic acid supplement alone; 95% of the folic acid supplementation, however, was in the form of a multivitamin, so the effects of folic acid could not be separated from those of other nutrients as in the Californian and Hungarian studies. Reported use of supplements at any time during the periconceptional period was 68% among controls, 70% among CL/P cases, and 62% among CP cases. When mothers using supplements daily throughout the periconceptional period were compared with mothers not using them in the same period, the OR estimate of risk was 1.2 (95% CI 0.7-2.0) for CL/P and 0.9 (95% CI 0.5-1.7) for CP; thus, no evidence was found of a protective association between maternal use of multivitamin and mineral supplements and risk of oral clefts.
A second study of maternal nutrition and oral clefts in the same populations was reported from the Slone group by Werler et al. (1999), in which case and control mothers for birth years 1993-1996 were recruited using similar procedures. A second control group was included of mothers of children without birth defects selected from the same birth hospitals in the surveillance program. The 822 case mothers represented eight major categories of birth defects, and of these, 114 were CL/P and 46 were CP. Data-collection procedures were the same as in the earlier study, and participation rates were nearly the same in cases, malformed controls, and normal controls (66% overall). Multivitamin supplementation was defined more broadly than in the earlier study and included daily use of a supplement with two or more water-soluble vitamins and two or more fat-soluble vitamins. Reported use of supplements at any time during months 1 through 4 was 76% among both malformed and non-malformed control mothers, 75% among CL/P case mothers, and 64% among CP case mothers; each represented slight increases over the previous study. Supplementation was categorized by the month of pregnancy of first use, and use in the 1-month period before the last menstrual period was not a criterion for exposure as in the earlier study. The critical exposure period was defined as months 1 through 3 for CL/P and 1 through 4 for CP because of the later period of palatal closure. The reference group included mothers with no supplement use or use that began after the critical period. Maternal characteristics associated with supplement use included age, education, race, planned pregnancy, nausea and vomiting in first month, and geographic center; these were included as covariates in multivariate models. Using the non-malformed control group, the OR indicating risk of CL/P with any use of supplements in the critical period was 0.8 (95% CI 0.5-1.3); the CL/P results using the malformed control group were similar (OR = 0.7, 95% CI 0.4-1.1). A reduced risk of CP was significantly associated with supplement use in the comparisons using nonmalformed controls (OR = 0.4, 95% CI 0.2-0.8) and malformed controls (OR = 0.4, 95% CI 0.2-0.9). Both studies used similar methods of ascertainment and the same rigorous data-collection methods; supplement use increased only slightly between the earlier and later periods of study. The data from these studies do not provide evidence for a protective effect of multivitamin supplements against CL/P. The difference between the null association for CP in the earlier study and the positive association in the later study may be partially explained by the use in the earlier study of some groups of malformed controls excluded in the later one, i.e., groups including limb defects and urinary tracts defects. The parallel use of malformed and non-malformed controls and the similarity of results provide strong evidence that maternal recall of supplement use is not biased between mothers of affected and unaffected children. The possibility remains, despite the best efforts of investigators, that residual confounding factors related to maternal vitamin use and risk of oral clefts may affect the findings of the Slone studies as in other case-control studies.
The Netherlands Case-Control Study of Maternal Metabolic Factors
The University of Nijmegen in The Netherlands has been a center of innovative studies of homocysteine metabolism in relation to NTDs (Steegers-Theunissen et al., 1991, 1994, 1995; Wong, 1999). Exposure to high levels of homocysteine alters the migration of neural crest cells and causes malformations in experimental avian models (Rosenquist et al., 1996). Wong et al. (1999) demonstrated the usefulness of investigating biochemical markers of maternal nutritional status in case-control studies of oral clefts. Cases of nonsyndromic oral clefts, both CL/P and CP, were ascertained by the Nijmegen Cleft Palate Team from patients treated between 1992 and 1997, and 71 eligible mothers of case children between ages 1 and 5 years were identified. Of these, three could not be contacted, 17 declined participation, and others were excluded because of current vitamin use (n = 6) or pregnancy (n = 10); 35 case mothers completed the study. Control mothers were selected from the nearby population (methods not specified), and 56 completed the study; 1 was excluded because of recent vitamin use. Mean ages of mothers were 32.0 years for cases and 34.8 years for controls. Venous blood samples were collected for measurement of serum and red cell folate, plasma homocysteine, serum vitamin B12, and vitamin B6 (as pyridoxal-5-phosphate) in whole blood. Oral methionineloading tests (Steegers-Theunissen et al., 1992) were then performed, and plasma homocysteine was measured again after 6 h. Case mothers had higher mean plasma homocysteine levels than controls for both the fasting (12 vs. 9 µmol/1, p < 0.01) and methionine afterload (35 vs. 31 µtmol/l, p < 0.05) samples. Hyperhomocysteinemia, defined as fasting or afterload values above the 97.5 percentile, was found in 15.6% of case mothers and 3.6% of control mothers (OR = 5.5, 95% CI 1.1-24.2). Unexpectedly, case mothers compared to controls had higher mean levels of serum folate (16 vs. 13 nmol/1, p < 0.01) and red cell folate (550 vs. 490 nmol/1, p < 0.05). Vitamin B12 levels were not significantly different between groups, but case mothers had lower levels of whole-blood pyridoxal-S-phosphate compared to controls (41 vs. 52 nmol/1, p < 0.05). The mild hyperhomocysteinemia of case mothers thus was not explained by low folate values but may have been related to the poorer vitamin B6 status of case mothers compared to controls. The advent of biochemical studies of maternal nutritional status, even years after the affected pregnancies, appears to be a significant methodologic advance in case-control studies of oral clefts. This method, however, is still subject to many of the pitfalls of case-control studies, including biased selection of participants. The Nijmegen results are based on a small number of case mothers from a single clinical center and had different rates of participation and exclusion between cases and controls. These studies should be extended to a larger sample in The Netherlands and other populations.
Human Experimental Trials of Vitamin Supplementation for the Prevention of Oral Clefts
Soon after the demonstration that vitamin deficiencies produced oral clefts in animal models in the 1940s, physicians became interested in whether vitamin supplementation could prevent the recurrence of oral clefts. The prevailing view at that time was that gene mutations were the most important cause of birth defects. In contrast, Douglas (1958) presented clinical observations of identical twins that were discordant for clefting in which the affected twin weighed less than the normal twin as evidence that variations in maternal blood supply between twins may adversely affect fetal nutritional status and risk of clefting. After the report by Warkany (1942) that riboflavin deficiency induced skeletal abnormalities in rats, Douglas began adding riboflavin to the diets of pregnant women in Tennessee who had previously given birth to a child with a cleft. Douglas (1958) justified the intervention with the statement that “riboflavin and other vitamins in proper amounts as a supplement could not possibly harm the mother or the fetus in any way; further when one considers that it could only improve the mother's general welfare and might lower the risk of her having a second deformed child.” Douglas reported no further details of efforts to reduce the recurrence of oral clefts other than the fact that not a single treated mother had another affected child; she acknowledged that the expected prevalence of oral clefts at birth in the area at that time was 0.6 to 1.0/1000 births and that the number of patients in the study was too small to be significant.
The next reported experimental trial to prevent the recurrence of oral clefts in humans was described by Conway (1958) just months after the report by Douglas appeared. Aspects and results of the trial by Conway (1958) and subsequent ones are listed in Table 14.3. Conway (1958) cited the work of Fogh-Andersen (1942) on the genetic determination of oral clefts in humans and the induction of oral clefts in animal models and concluded that “the occurrence of induced deformities is influenced by the genetic background of the experimental animals and that alteration of environmental factors is important in bringing out the deformities.” Conway (1958) was also influenced by the work of Warkany and Kalter (1957) and cited their admonition that “it would be premature and dangerous to infer that parallels exist between the etiologies of the experimental and human abnormalities.” Conway, working at the Cornell Medical College during 1946-1957, provided vitamin supplements for mothers of cleft-affected children during the periconceptional period of their subsequent pregnancies. Multivitamin capsules were given daily and included vitamin A (12,500 USP units), vitamin D (1000 USP units), vitamin C (100 mg), vitamin B1 (5 mg), vitamin B2 (5 mg), vitamin B6 (2 mg), vitamin B12 (4 µg), calcium pantothenate (10 mg), nicotinamide (30 mg), and folk acid (0.5 mg). In addition, an intramuscular injection of vitamins was given every other day, including vitamin B1 (5 mg), vitamin B2 (2 mg), vitamin B6 (5 mg) vitamin B12 (2.5 µg), nicotinamide (75 mg), and sodium pantothenate (2.5 mg). No details were provided on the method of assignment of treatment; thus, it is unclear what selective factors and possible biases were involved. Participants included private, semiprivate, and “pavillion” (presumably indigent) patients; but the distributions of these patient types in the treated and untreated groups was not reported. Of the 196 mothers under observation, 87 subsequently become pregnant; of these, 39 were given vitamin therapy and 48 received no vitamin therapy. The 39 mothers receiving vitamin therapy had a total of 59 subsequent pregnancies, and none of the infants had an oral cleft. Among the 48 mothers receiving no vitamin therapy, there were 78 subsequent pregnancies and four infants with oral clefts, including one with congenital heart defects. No statistical analyses were reported; application of Fisher's exact test to these data revealed that the difference between groups was not statistically significant (p = 0.13).
|
TABLE 14.3. Trials of Maternal Vitamin Supplementation for the Prevention of Recurrence of Oral Clefts in the Pregnancies of High-Risk Women |
||||||||||||||||||||||||||||||||||||
|
||||||||||||||||||||||||||||||||||||
Peer et al. (1958a) reported that because animal studies showed protective effects of vitamin B6 and folic acid against cleft induction in cortisone-treated mice, he routinely provided these vitamins to mothers of children with oral clefts during the early weeks of their subsequent pregnancy. Appeals were made to other plastic surgeons to do the same with their patients and contribute their data on pregnancy outcome to a pooled analysis. Peer et al. (1963, 1964) reported on patients in Newark, New Jersey, and 51 patients referred by colleagues at other locations. No details of the patient population were provided other than that each mother had a previous child with a cleft. Treatment consisted of prenatal vitamin capsules and an additional 5 mg folic acid and 10 mg vitamin B6, each taken daily during the first trimester. The prenatal vitamin capsule also contained vitamin B6 (2 mg); thus, the total daily dose of vitamin B6 was 12 mg. Other vitamins in the prenatal capsule included vitamin B1 (10 mg), vitamin B2 (10 mg), vitamin B12 (4 µg), vitamin C (300 mg), niacinamide (100 mg), and calcium pantothenate (20 mg). Briggs (1976) provided an updated account of Peer's work and reported on the results of births to 228 treated mothers and 417 untreated control mothers. No details were provided on the methods of assigning women to the treatment or control group. Among the 228 subsequent births of the treated mothers, seven had oral clefts (3.1%); among the 417 subsequent births to the untreated women, 20 had oral clefts (4.8%). No statistical analyses were provided in the original publication; the difference between groups was not statistically significant (Fisher's exact (p = 0.41). Briggs (1976) further stratified the data according to the type of cleft of the index child of each mother. Among mothers of children with CL/P, recurrence was 1.9% among the 161 pregnancies of treated mothers and 5.5% among the 275 pregnancies of untreated mothers (p = 0.08). Among mothers of index children with CP, recurrence was 6.0% in the treated group and 3.5% in the untreated group (p = 0.47). Briggs (1976) concluded, without statistical support for his claim, that the supplementation was most helpful for mothers of CL/P index children and that supplementation might be “specifically detrimental when evaluating the incidence of cleft palate alone.” He attempted to extend the sample size by including additional births of treated mothers who had already delivered one child while in the trial, but no comparable mothers in the untreated group were included; this may have introduced confounding related to differences in parity and social class between treated and untreated mothers.
Additional studies of the effect of vitamin supplementation on the reduction of the recurrence of oral clefts were reported by von Krebig and Stoeckenius (1978), Gabka (1981), and Schubert et al. (1990). These authors claimed success in reducing the risk of oral clefts with vitamin supplements, yet each of these studies either was too small or had insufficient data to allow a detailed evaluation of the results.
Czech Cleft Prevention Trial
A collection of pedigrees of 8250 cleft patients born between 1886 and 1982 in Bohemia provided an opportunity for Tolarova (1986, 1987a,b) at the Czechoslovak Academy of Sciences in Prague to explore the occurrence, genetics, and prevention of oral clefts. The Bohemian data include careful clinical classification of cleft type and genetic studies of the more recent patients and their families. In 1976, Tolorova began providing vitamin supplements to mothers deemed at high risk of cleft recurrence because they had either a cleft themselves or a child with a cleft. Mothers were instructed to take three tablets of the “Spofavit” multivitamin preparation each day beginning 3 months before conception and continuing until the end of the first trimester of pregnancy. Each tablet included vitamin A (2000 IU), vitamin B1 (1 mg), vitamin B2 (1 mg), vitamin B6 (1 mg), vitamin C (50 mg), vitamin D (100 IU), vitamin E (2 mg), nicotinamide (10 mg), and calcium pantothenate (1 mg); in addition, mothers were given 10 mg of folic acid per day. The treatment group consisted of mothers who accepted the vitamin supplements when offered, and controls were mothers who declined the vitamin supplements or failed to comply with the treatment regimen. Tolarova (1982) reported that 1 of 85 supplemented pregnancies and 10 of 212 unsupplemented pregnancies were affected with oral clefts. The results were updated in 1987 (Tolarova, 1987a,b) and 1995 (Tolarova and Harris, 1995), and the most recent results revealed that 3 of 211 supplemented pregnancies and 77 of 1824 unsupplemented pregnancies were affected with oral clefts. A Fisher exact p value of 0.03 from a one-sided test was provided; the p value for the two-sided test was 0.058.
The exclusion of noncompliant participants in a clinical trial may seriously bias the results, even if the trial began with random assignment of treatments; this is the basis for “intention-to-treat” analyses in the design of modern clinical trials (Meinert, 1986). A serious limitation of the Czech study was the lack of random assignment of mothers to treatment or no treatment. Mothers in the treatment group were acceptors of the offer of supplementation and compliant, while the control group consisted of mothers who rejected the offer of supplementation or failed to adequately comply with the treatment regimen. This may have resulted in important confounding differences between groups related to lifestyle factors and risk of clefting. The authors noted that mothers in the supplemented group received additional interventions, including advice to conceive in the late spring and summer months because of the greater availability of fresh fruit and green vegetables and a lesser risk of respiratory tract infections; this advice was not given to the control group. Because the Czech trial did not include randomization, was a comparison of compliers and noncompliers by design, included additional lifestyle interventions for the supplemented group but not for the control group, and included an intervention with multiple vitamins, the results are uninterpretable with regard to the role of any specific nutrient in the prevention of oral clefts.
Hungarian Birth Defects Prevention Trial
The Hungarian Family Planning Program (HFPP) is a comprehensive program of medical care and social services devoted to improving reproductive health. This setting was used for a clinical trial of the efficacy of periconceptional multivitamin supplementation in the prevention of birth defects and other complications of pregnancy (Meinert, 1986; Czeizel and Dudas, 1992; Czeizel 1993a,b; Czeizel et al., 1994; Czeizel and Hirschberg, 1997; Czeizel, 1998; Czeizel et al., 1999). The HFPP included 26 centers coordinated by the Department of Human Genetics and Teratology at the Hungarian National Institute of Hygiene and is a World Health Organization Collaborating Center for the Community Control of Hereditary Diseases. Participants were offered extensive medical examinations, genetic counseling, and health education, including recommendations on diet and the avoidance of tobacco, alcohol, and other hazardous substances. Women attending the HFPP were recruited for a multivitamin supplementation trial and randomly assigned to take either a multivitamin or a trace-element tablet daily for the period 1 month before conception until the third month of gestation. The trial was double-blind. The multivitamin (Elevit Prenatal; Hoffman-La Roche, Nutley, NJ) contained vitamin A (6000 IU until 1989 and 4000 IU thereafter); vitamin B1 (1.6 mg); vitamin B2 (1.8 mg); vitamin B6 (2.6 mg); vitamin B12 (4 µg); vitamin C (100 mg); vitamin D (500 IU); vitamin E (15 mg); folic acid (15 mg); nicotinamide (19 mg); calcium pantothenate (10 mg); biotin (0.2 mg); four minerals including calcium (125 mg), phosphorus (125 mg), magnesium (100 mg), and iron (60 mg); and three trace elements including copper (1 mg), manganese (1 mg); and zinc (7.5 mg). The trace-element control group took a tablet with the same amounts of copper, manganese, and zinc with the addition of vitamin C (7.3 mg) and lactose (736 mg). All live births and other pregnancy outcomes were evaluated for the presence of congenital abnormalities through participating obstetrical clinics. The compliance of participants in pill taking was carefully assessed, and participants in 29% of the multivitamin group and 27% in the trace-element group were classified as taking no supplements or only a partial course. The outcome of pregnancy was assessed in 99% of participants, and the final results (Czeizel, 1998; Czeizel et al., 1999), based on an intention-to-treat analysis of all participants, revealed a significant reduction in NTDs (0 in 2471 vitaminsupplemented pregnancies vs. 6 in 2391 trace elementtreated pregnancies; p = 0.02), but no significant differences were observed in the occurrence of a small number of oral clefts between treatment groups (4 among the vitamin-supplemented and 5 in the trace element-supplemented pregnancies; p = 0.57). The multivitamin group had higher rates of conception (71.3% vs. 67.9%), multiple births (3.7% vs 2.7%), and fetal deaths (13.4% vs. 11.4%) compared to the trace-element group.
Nutrient-Gene Interactions in Oral Clefts
A major role for genes in the development of normal craniofacial structures is evident from the common observation that monozygotic twins are nearly identical in appearance. Patterns of genetic inheritance of clefting were first described by Fogh-Andersen (1942) and confirmed by segregation analysis (Marazita et al., 1986; Mitchell and Risch, 1992; Mitchell, Christensen, 1996). The processes involved in the construction of the face have begun to be established through studies of animal models, human mutations, and recognized teratogens. Facial development is a complex process dependent on a broad spectrum of signaling molecules, homeobox genes, and growth factors (Murray, 1995). Development is initiated by migrating neural crest cells, which combine with mesodermal cells to establish the facial primordia.
A variation in the methylenetetrahydrofolate reductase gene (MTHFR, C677T allele) is the first known genetic risk factor for NTDs and may be relevant to oral clefts. Persons homozygous for the MTHFR T allele have a more thermolabile form of the MTHFR enzyme with reduced activity, which in the presence of low plasma folate (presumably low folate intake) results in higher plasma homocysteine levels (Jacques et al., 1996; Christensen et al., 1997). Not all studies, however, have shown a significant increase in risk of NTDs associated with the MTHFR T allele (Mornet et al., 1997). Shaw et al. (1998b) found a modest association between the MTHFR genotype of infants and risk of NTDs but no evidence of interaction between infant MTHFR genotype and maternal intake of folic acidcontaining supplements.
The MTHFR-NTD association is an example of a possible nutrient-gene interaction in which the combination of these factors explains more than genotype or nutritional status alone. Christensen et al. (1999) extended the analysis of nutrient-gene interactions to include the genotype of both mother and child. The combination of the MTHFR TT genotype for mother and child yields an OR estimate of NTD risk of 6.0 (95% CI 1.26-28.53). The OR for mothers alone with the TT genotype was 1.29, and that for TT children alone was 2.0 (neither was statistically significant). When the combination of mother's TT genotype and low red cell folate was considered, the OR was 3.28 (95% CI 0.84-12.85). The combination of child's TT genotype and low maternal red cell folate yielded an OR of 13.43 (95% CI 2.49-72.33). This example illustrates both the potentials and the pitfalls of analyses of nutrient-gene interactions: uncovering strong interactions yields far more insight than studying single genotypes or nutrients alone, yet large sample sizes are needed to accurately estimate the effect because the combination of relatively uncommon genotypes with the probability of being in the lowest part of the distribution of an indicator of nutritional status results in a comparison group that is a small fraction of the total original sample. This is presumably why Christensen et al. (1999) were unable to report on the risk of NTDs among mother-child pairs in which both mother and child had the TT genotype and the mother had low red cell folate. Substantially larger sample sizes will be necessary to probe the extent of this and other complex interactions. As more data become available on multiple genes related to vitamin B6 and folate metabolism, there will be more opportunities to explore the interaction between genes. Botto and Mastroiacovo (1998) analyzed the risk of NTDs in the presence of variations in both MTHFR and cystathionine β3-synthase and found an OR for gene-gene interaction of 5.2 (95% CI 0.8-2.1); this represents the estimate of the factor by which the OR for persons with both mutations is different from the multiplied effect of each mutation alone.
The MTHFR-NTD associations provide an interesting model, which at first glance may seem useful in understanding the causes of oral clefts. Early evidence, however, indicates that the association between MTHFR genotype and cleft risk may be different from that of NTDs. In a California study of isolated CL/P, Shaw and colleagues (1998a) found no increase in cleft risk among infants with the C677T allele (T); in the largest ethnic subgroup, non-Hispanic whites, the T allele was associated with a reduced risk of CL/P. In a follow-up report on isolated CP in the same study, a reduced risk of clefting was also found; the risk of CP, relative to infants with the CC genotype, was 0.6 (95% CI 0.3-1.3) for TT infants and 0.8 (95% CI 0.5-1.2) for CT infants (Shaw et al., 1999).
Further understanding of the role of MTHFR polymorphisms and cleft risk will require determinations of biochemical markers of folate and other nutrients, and insights might be gained from study of other MTHPJR-disease associations. The C677T allele has also been associated with a reduced risk of colon cancer in the Health Professionals Follow-up Study (Chen et al., 1999), the Physicians' Health Study (Ma et al., 1997), and a case-control study conducted in Utah and Minnesota (Slattery et al., 1999). The mechanism by which the C677T allele contributes to the reduction in colon cancer risk has been hypothesized to include either imbalanced DNA methylation or altered synthesis of nucleic acids (Ma et al., 1996, 1997, 1999; Bagley and Selhub, 1998; Chen et al., 1998; Chen et al., 1998). Presence of the C677T allele was shown by Bagley and Selhub (1998) to be associated with an accumulation of formylated folates in red blood cells relative to the CC genotype; this shift in the distribution of folate forms may favor DNA synthesis and repair because of the dependence of these pathways on nonmethylated forms of folate.
Discussion
Maternal nutrition is likely an important environmental factor responsible for the occurrence of oral clefts in humans. Evidence for this view has accumulated from studies over most of the past century from experimental animals, observational studies of human populations, and some limited human experimental studies. The history of experimental studies of oral clefts in experimental animals reveals several nutrients that might be used to prevent oral clefts in susceptible populations. The leading candidate nutrients include folic acid, vitamin B6, and vitamin A. A lesser body of evidence implicates riboflavin, vitamin B12, zinc, pantothenic acid, and biotin.
Josef Warkany was a pioneer in the early experimental studies of teratogenesis but remained cautious about the relevance of animal studies for congenital anomalies in humans: “experimental cleft palate has been going on for 40 years… to my knowledge the brilliant studies on mechanisms involved in cleft palate formation have not led to the prevention of human palate defects” (Kalter, 1993). Warkany died at the age of 90 about the time that the results from the MRC trial (MRC Vitamin Study Research Group, 1991) provided proof that maternal supplementation with folic acid substantially reduced the recurrence of NTDs and a source of optimism that improved maternal nutritional status might reduce the burden of other birth defects.
The relevance of folic acid supplementation for the prevention of human oral clefts remains unclear despite a number of studies that have addressed this hypothesis. Folate was an early focus in animal experiments on oral clefts, long before it was implicated in NTDs. Warkany's colleague Rose Nelson established methods for inducing dietary folate deficiency and exacerbating it with folate antagonists to produce fetal deaths, oral clefts, and other malformations. This work was replicated by Giroud and Boisselot (1951) in Paris, and a detailed description of the importance of the timing of folate deficiency during gestation emerged. Folate deficiency may even have relevance in dogs: a daily dose of 5 mg folic acid prevented oral clefts in a line of Boston Terriers with a genetic predisposition to this disorder.
Case-control studies of folate nutrition and oral clefts in humans have been limited by the facts that folate intake is difficult to estimate, folates have a wide range of bioavailability, and folic acid supplements are usually taken with other vitamins, minerals, and trace elements that may also have protective effects against oral clefts. Studies of medications that disrupt folate metabolism may shed more light on the role of folate nutrition in human oral clefts. Another useful approach is the study of biochemical indicators of maternal folate status (and other nutrients) even long after the fact of the affected pregnancy. The approach of examining biochemical markers of maternal nutritional status has rarely been employed in case-control studies but makes sense because mothers tend to resume their prepregnancy dietary habits and each mother, of course, has the same genotype at each time point in the study, which is important in determining tissue levels of nutrients. Maternal red cell folate levels were determined 1 year after delivery in a study in north London and found to be reasonably correlated with values determined early in pregnancy (Leek et al., 1983).
Studies of nutrient-genotype interactions may clarify the role of folate and other nutrients in oral clefts. The early indications from MTHFR studies are that the oral cleft results are quite different from expectations based on studies of NTDs. These findings underscore the importance of developing new methods for folate analyses that will characterize the distribution of various forms of folate so that the metabolic consequences of genetic variation may be examined in large-scale epidemiologic studies.
The human experimental trials conducted to date have done little to clarify the causal role of maternal folate nutrition in human oral clefts. These trials either have had an inadequate number of participants to provide sufficient statistical power, have not included study of folic acid apart from other nutrients, have not randomly assigned participants to treatment or control groups, or have failed to use intention-to-treat analyses.
Despite the extensive investigation of the role of vitamin B6 in animal models of oral clefts since the 1950s, there is little information on the relevance of vitamin B6 to oral clefts in humans, and this remains an important gap in our present knowledge. The use of antinausea medications was associated with a reduced risk of congenital heart defects in the Atlanta Birth Defects Case-Control Study (Erickson, 1991), and vitamin B6 may have a role in this story. Further analyses of these findings revealed that maternal nausea during the first 2 months of pregnancy and use of the antinausea combination of doxylamine, dicyclomine and vitamin B6 (Bendectin) (Merrell Dow Pharmaceuticals, Midland, MI) was significantly associated with a reduced risk of congenital heart defects (Boneva et al., 1999). The authors pointed out that Bendectin was composed of doxylamine succinate (10 mg), dicyclomine (10 mg, dropped from the formulation in 1976), and vitamin B6 (pyridoxine, 10 mg) and that vitamin B6 was often used alone for treatment of nausea in pregnancy. Portions of the developing heart are linked to cellular migration from the cranial neural crest; thus, it is plausible that vitamin B6 may be related to several types of birth defect, including oral clefts, with common developmental origins in the neural tube and crest.
Biochemical analyses of maternal vitamin B6 status in case-control studies also hold great promise, as shown in the Nijmegen study in The Netherlands (Wong et al., 1999). Adequate maternal vitamin B6nutrition may be important in lowering the risk of oral clefts by reducing plasma homocysteine levels. Animal experiments have provided evidence of other mechanisms for the protective effects, including a role for vitamin B6 in glucocorticoid metabolism. The prevalence of vitamin B6 deficiency is not well described worldwide, and little attention has been given to vitamin B6 in public health nutrition policy in the United States and elsewhere. There are no regulations mandating the fortification of food with vitamin B6 in the United States as there are with thiamine (vitamin B1), riboflavin (vitamin B2), niacin, and folic acid. Vitamin B6 deficiency results from the use of medications including isoniazid treatment of tuberculosis and oral contraceptives (Sauberlich et al., 1972); thus, these factors should be considered in future studies of vitamin B6 status.
Both abnormally low and high levels of vitamin A intake by mothers early in pregnancy may increase the risk of oral clefts and other malformations in their offspring. Despite Male's early discovery that vitamin A deficiency caused ocular defects, oral clefts, and other malformations in pigs (Hale, 1933, 1935), little research has been subsequently conducted on the role of inadequate dietary vitamin A intake in experimental animal or human oral clefts. One recent exception has been the work of Natsume and colleagues (1999) in Japan, who found that intake of vegetables rich in β-carotene (and folate) was associated with a reduced risk of CL/P. Vitamin A deficiency is widespread, especially in developing countries around the world (West et al., 1999). Much work has been done on the effects of vitamin A deficiency on maternal and child health, especially related to morbidity and mortality due to increased susceptibility to infectious diseases, impaired vision, blindness, and other eye problems. Congenital anomalies may occur in the setting of vitamin A deficiency but may go unnoticed because of the larger burden of other health problems.
In a prospective study of over 22,000 pregnancies of American women, Rothman et al. (1995) found that malformations of structures derived from the embryonic cranial neural crest were more common among women who consumed more than 10,000 IU of vitamin A in the periconceptional period. Humans and other primates are thought to be more susceptible to retinoid-induced birth defects than rodents and rabbits, although the types of malformation are similar between species and include the craniofacial and other malformations mentioned above. The greater susceptibility to retinoid exposure among humans compared to other animals may be due to a higher rate of placental transfer, longer plasma half-life, and differences in metabolite formation (Ross, 1999).
Our understanding of the mechanisms involved in teratogenesis requires narrowly focused experiments in human and nonhuman animal studies and well-targeted hypotheses in observational epidemiologic studies. The flip side of this coin is that reductionism may obscure our recognition of other nutrients that may play important roles in the prevention of birth defects. Many researchers and public health advocates seem to have lost sight of the fact that a balanced diet based on wholesome foods is critical for providing the vitamins, minerals, macronutrients, and energy needed to sustain a healthy pregnancy.
A new generation of epidemiologic studies may help to establish a broader and more balanced view of the role of maternal nutrition in oral clefts. A good starting point would be studies of the co-occurrence of NTDs and oral clefts (or the lack of it). While it is possible that NTDs and clefts share developmental origins, it may be useful to try to understand how the nutritional and other causes of NTDs and oral clefts are different. Well-designed case-control studies with detailed assessment of biochemical markers of maternal nutritional status are needed to fill the gaps in our understanding of the role of maternal nutrition in oral clefts. Genetic studies of the underlying susceptibility to nutritional deficiency will clarify our understanding further. International collaboration will be essential to uncover nutrition-related and other important causes of oral clefts and to discover feasible public health interventions, to reduce their occurrence. Studies in diverse populations around the globe will likely uncover multiple nutritional, environmental, and genetic causes of oral clefts. Consistency of research methods across studies is highly desirable, and considerable efforts are now under way to organize a global network of cleft researchers with common case-control study protocols that will enhance the comparison and combination of the results of individual studies (Wyszynski and Mitchell, 1999). As the results from the next wave of enhanced case-control studies become available, rigorously designed intervention trials of vitamin supplementation will be the next logical step toward the prevention of oral clefts.
Acknowledgements
Supported by grant RO1-HD39061 from the U.S. National Institute of Child Health and Human Development and the National Institute for Dental and Craniofacial Research.
References
Abbott, BD, Adamson, ED, Pratt, RM (1988). Retinoic acid alters EGF receptor expression during palatogenesis. Development 202: 853–867.
Abbott, BD, Birnbaum, LS (1990). Retinoic acid-induced alterations in the expression of growth factors in embryonic mouse palatal shelves. Teratology 42: 597–610.
Abbott, BD, Harris, MW, Birnbaum, LS (1988b). Etiology of retinoic acid-induced cleft palate varies with the embryonic stage. Teratology 40: 533–553.
Abbott, BD, Pratt, RM (1988a). EGF receptor expression in the developing tooth is altered by exogenous retinoic acid and EGF. Dev Biol 128: 300–304.
Abbott, BD, Pratt, RM (1988b). Influence of retinoids and EGF on growth of embryonic mouse palatal epithelia in culture. In Vitro Cell Dev Biol 24: 343–352.
Bagley, PJ, Selhub, J (1998). A common mutation in the methylenetetrahydrofolate reductase gene is associated with an accumulation of formylated tetrahydrofolates in red blood cells. Proc Natl Acad Sci USA 95: 13217–13220.
Been, W, Lieuw Kie Song, SH (1978). Harelip and cleft palate conditions in chick embryos following local destruction of the cephalic neural crest. A preliminary note. Acta Morphol Neerl Scand 16: 245–255.
Berry, RJ, Li, Z, Erickson, JD (1999). Prevention of neural-tube defects with folk acid in China. China-U.S. Collaborative Project for Neural Tube Defect Prevention. N Engl J Med 341: 1485–1490.
Boisselot, J (1949). Malformations foetales par insuffisance en acide pantothenique. Arch Pediatr 6: 1.
Boisselot, J (1951). Role tertogene de la deficience en acide pantothenique chez le rat. Ann Med 52: 225.
Boneva, RS, Moore, CA, Botto, L, et al. (1999). Nausea during pregnancy and congenital heart defects: a population-based case-control study. Am J Epidemiology 149: 717–725.
Bonner, J, Slavkin, H (1975). Cleft palate susceptibility linked to histocompatibility-2 (H-2) in the mouse. Immunogenetics 2: 213–218.
Botto, LD, Mastroiacovo, P (1998). Exploring gene-gene interactions in the etiology of neural tube defects. Clin Genet 53: 456–459.
Bower, C, Stanley, FJ (1989). Dietary folate as a risk factor for neuraltube defects: evidence from a case-control study in Western Australia. Med J Aust 150: 613–619.
Bower, C, Stanley, FJ (1992). Dietary folate and nonneural midline birth defects: no evidence of an association from a case-control study in Western Australia [see comments]. Am J Med Genet 44: 647–650.
Briggs, RM (1976). Vitamin supplementation as a possible factor in the incidence of cleft lip/palate deformities in humans. Clin Plast Surg 3: 647–652.
Chen, B, Wang, X, Yu, J, et al. (1998). Relationship between methylenetetrahydrofolate reductase gene polymorphism and coronary heart disease [in Chinese]. Chung Hua I Hsueh I Chuan Hsueh Tsa Chih 15: 300–302.
Chen, J, Giovannucci, E, Hankinson, SE, et al. (1998). A prospective study of methylenetetrahydrofolate reductase and methionine synthase gene polymorphisms, and risk of colorectal adenoma. Carcinogenesis 19: 2129–2132.
Chen, J, Giovannucci, EL, Hunter, DJ (1999). MTHFR polymorphism, methyl-replete diets and the risk of colorectal carcinoma and adenoma among U.S. men and women: an example of gene-environment interactions in colorectal tumorigenesis. J Nutr U9(Suppl 2): 560S-564S.
Christensen, B, Arbour, L, Tran, P, et al. (1999). Genetic polymorphisms in methylenetetrahydrofolate reductase and methionine synthase, folate levels in red blood cells, and risk of neural tube defects. Am J Med Genet 84: 151–157.
Christensen, B, Frosst, P, Lussier-Cacan, S, et al. (1997). Correlation of a common mutation in the methylenetetrahydrofolate reductase gene with plasma homocysteine in patients with premature coronary artery disease. Arterioscler Thromb Vase Biol 17: 569–573.
Chung, CS, Kau, MC (1985). Racial differences in cephalometric measurements and incidence of cleft lip with or without cleft palate. J Craniofac Genet Dev Biol 5: 341–349.
Conway, H (1958). Effect of supplemental vitamin therapy on the limitation of incidence of cleft lip and cleft palate in humans. Plast Reconstr Surg 22: 450–453.
Cooper, MK, Porter, JA, Young, KE, et al. (1998). Teratogenmediated inhibition of target tissue response to Shh signaling [see comments]. Science 280: 1603–1607.
Croen, LA, Shaw, GM, Wasserman, CR, et al. (1998). Racial and ethnic variations in the prevalence of orofacial clefts in California, 1983–1992. Am J Med Genet 79: 42–47.
Czeizel, AE (1993a). Controlled studies of multivitamin supplementation on pregnancy outcomes. Ann N Y Acad Sci 678: 266–275.
Czeizel, AE (1993b). Prevention of congenital abnormalities by periconceptional multivitamin supplementation. BMJ 306: 1645–1648.
Czeizel, AE (1998). Periconceptional folic acid containing multivitamin supplementation. Eur J Obstet Gynecol Reprod Biol 78: 151–161.
Czeizel, AE, Dudas, I (1992). Prevention of the first occurrence of neural-tube defects by periconceptional vitamin supplementation [see comments]. N Engl J Med 327: 1832–1835.
Czeizel, AE, Dudas, I, Metneki, J, et al. (1994). Pregnancy outcomes in a randomised controlled trial of periconceptional multivitamin supplementation. Final report. Arch Gynecol Obstet 255: 131–139.
Czeizel, AE, Hirschberg, J (1997). Orofacial clefts in Hungary. Epidemiological and genetic data, primary prevention. Folia Phoniatr Logop 49: 111–116.
Czeizel, AE, Timar, L, Sarkozi, A, et al. (1999). Dose-dependent effect of folic acid on the prevention of orofacial clefts. Pediatrics 104: e66.
Czeizel, AE, Toth, M, Rockenbauer, et al. (1996). Populationbased case control study of folic acid supplementation during pregnancy. Teratology 53: 345–351.
Daly, S, Mills, JL, et al. (1997). Minimum effective dose of folic acid for food fortification to prevent neural-tube defects [see comments]. Lancet 350: 1666–1669.
Davis, SD, Nelson, T, Shepard, T, et al. (1970). Teratogenicity of vitamin B6 deficiency: omphalocele, skeletal and neural defects, and splenic hypoplasia. Science 169: 1329–1330.
Dostal, M, Schubert, J (1990). Further studies on protective effects of vitamins in cyclophosphamide-induced cleft palate. Int J Oral Maxillofac Surg 19: 308–311.
Douglas, B (1958). The role of environmental factors in the etiology of so-called congenital malformations. I. Deductions from the presence of cleft lip and palate in one of identical twins, from embryology and from animal experiments. Plast Reconstr Surg 22: 94–108.
Elwood, JM, Colquhoun, TA (1997). Observations on the prevention of cleft palate in dogs by folic acid and potential relevance to humans. N Z V J 45: 254–256.
Emerson, DJ (1962). Congenital malformation due to attempted abortion with aminopterin. Am J Obstet Gynecol 84: 356–357.
Erickson, JD (1991). Risk factors for birth defects: data from the Atlanta Birth Defects Case-Control Study. Teratology 43: 41–51.
Evans, HM, Nelson, MM, Asling, CW (1951). Multiple congenital abnormalities resulting from folic acid deficiency during gestation. Science 114: 479.
Fantel, AG, Shepard, TH, Newell-Morris L, Moffett, BC (1977). Teratogenic effects of retinoic acid in pigtail monkeys (Macaca nemestrina). I. General features. Teratology 15: 65–71.
Fogh-Andersen, P (1942). Inheritance of Harelip and Cleft Palate. Copenhagen: Arnold Busck.
Fraser, FC (1955). Thoughts on the etiology of clefts of the palate and lip. Acta Genet 5: 358–369.
Fraser, FC, Fainstat, TD (1951a). Causes of congenital defects. Am Med Assoc J Dis Child 32: 593–603.
Fraser, FC, Fainstat, TD (1951b). Production of congenital defects in the offspring of pregnant mice treated with cortisone. Pediatrics 8: 527–533.
Fraser, FC, Warburton, D (1964). No association of emotional stress or vitamin supplement during pregnancy to cleft lip or palate in man. Plast Reconstr Surg 33: 395–399.
Gabbka, J (1981). Verhutung von lippen-kiefer-gaumenspalten: klinisch erfahrungen. Munch Med Wschr 123: 1139–1141.
Giroud, A, Boisselot, J (1947). Repercussions de I'avitaminose B2 sur I'embryon du rat. Arch Pediatr 4(4): 317–327.
Giroud, A, Boisselot, J (1951). Influence teratogene de la carence en acide folique. C R Soc Biol 145: 526–527.
Giroud, A, Lefebvres, J (1951). Anomalies provoquées chez le foetus en I'absence d'acide folique. Arch Pediatr 8: 648–656.
Giroud, A, Levy, G, Lefebvres, J, et al. (1951). Variations du taux de la riboflavine chez la mere et le foetus, et leurs repercussions. Bull Soc Chim Biol 33(9): 1214–1222.
Goetsch, C (1962). An evaluation of aminopterin as an abortifacient. Am J Obstet Gynecol 83: 1474–1477.
Hale, F (1933). Pigs born without eyeballs. J Hered 24: 105–106.
Hale, F (1935). The relation of vitamin A to anophthalmos in pigs. Am J Opthalmol 18: 1087–1092.
Hayes, C, Werler, MM, Willett WC, Mitchell AA (1996). Case-control study of periconceptional folic acid supplementation and oral clefts. Am J Epidemiol 143: 1229–1234.
Herbert, V (1989). Development of human folate deficiency. In: Folic Acid Metabolism in Health and Disease, edited by MF Picciano, ELR Stokstad, and JFI Gregory. New York: Wiley-Liss, pp. 195–200.
Hibbard, B (1964). The role of folic acid in pregnancy with particular reference to anaemia, abruption and abortion. J Obstet Gynaecol Br Commonw 71: 529–542.
Hibbard, B, Smithells, R (1965). Folic acid metabolism and human embryopathy. Lancet i: 1254.
Hill, L, Murphy, M, McDowall Paul AH (1988). Maternal drug histories and congenital malformations: limb reduction defects and oral clefts. J Epidemiol Community Health 42: 1–7.
Incardona, JP, Gaffield, W, Kapur RP, Roelink H (1998). The teratogenic Veratrum alkaloid cyclopamine inhibits sonic hedgehog signal transduction. Development 225: 3553–3562.
Jacobsson, C, Granstrom, G (1997). Effects of vitamin B6 on beta-aminoproprionitrile-induced palatal cleft formation in the rat. Cleft Palate Craniofac J 34: 95–100.
Jacques, PF, Bostom, AG, Williams, RR, et al. (1996). Relation between folate status, a common mutation in methylenetetrahydrofolate reductase, and plasma homocysteine concentrations [see comments]. Circulation 93: 7–9.
James, LF (1999). Teratological research at the USDA-ARS poisonous plant research laboratory. J Nat Toxins 8: 63–80.
Jones, MC (1988). Etiology of facial clefts: prospective evaluation of 428 patients. Cleft Palate J 25: 16–20.
Jordan, RL, Wilson, JG, Schumacher HJ, et al. (1977). Embryotoxicity of the folate antagonist methotrexate in rats and rabbits. Teratology 15: 73–80.
Kalter, H (1957). Factors influencing the frequency of cortisone-induced cleft palate in mice. J Exp Zool 234: 449–467.
Kalter, H (1993). Josef Warkany, 1902–1992. Teratology 48: 1–3.
Kalter, H, Warkany, J (1957). Congenital malformations in inbred strains of mice induced by riboflavin-deficient, galactoflavincontaining diets. J Exp Zool 136: 531–536.
Karnofsky, DA, Patterson, PA, Ridgway, LR, et al. (1949). Effect of folic acid, “4-amino”-folic acids, and related substances on growth of chick embryo. Proc Soc Exp Biol Med 71: 447.
Keeler, RF (1968). Teratogenic compounds of Veratrum californicum (Durand). IV, First isolation of veratramine and alkaloid Q and a reliable method for isolation of cyclopamine. Phytochemistry 7: 303–306.
Keeler, RF (1990). Early embryonic death in lambs induced by Veratrum californicum. Cornell Vet. 80: 203–207.
Khoury, MJ, Gomez-Farias, M, Mulinare, J, et al. (1989). Does maternal cigarette smoking during pregnancy cause cleft lip and palate in offspring? Am J Dis Child 143: 333–337.
Kirke, PN, Molloy, AM, Daly, LE, et al. (1993). Maternal plasma folate and vitamin BH are independent risk factors for neural tube defects. Q J Med 86: 703–708.
Kochhar, DM (1967). Teratogenic activity of retinoic acid. Acta Pathol Microbiol Scand 70: 398–404.
Kochhar, DM (1973). Limb development in mouse embryos. I. Analysis of teratogenic effects of retinoic acid. Teratology 7: 289–298.
Kochhar, DM, Penner, JD, Tellone, CI (1984). Comparative teratogenic activities of two retinoids: effects on palate and limb development. Teratog Carcinog Mutagen 4: 377–387.
Lammer, EJ, Chen, DT, Hoar, RM, et al. (1985). Retinoic acid embryopathy. N Engl J Med 313: 837–841.
Lasa, C, Manalo, P (1989). Update on the occurrence rate of cleft lip and palate. Philippine Journal of Surgery and Surgical Specialities 44: 109–111.
Leek, I, lies, CA, Sharman, IM, et al. (1983). Maternal diet and nutrition during early pregnancy and after delivery in north London. In: Prevention of Spina Bifida and Other Neural Tube Defects, edited by J Dobbin. London: Academic Press. Appendix I. pages 197–218.
Leimbach, DG (1949). Rev Int Vitaminol 21: 222.
Lepkovsky, S, Taylor, LW, Jukes, TH, Almquist, HJ (1938). Hilgardia 11: 559–591.
Lidral, AC, Johnston, MC, Switzer, R (1991). The relationship between vitamins and the prevalence of cleft lip in mice. J Dent Res 68(Suppl): 524.
Lorente, C, Cordier, S, Goujard, J, et al. (2000). Tobacco and alcohol use during pregnancy and risk of oral clefts. Occupational Exposure and Congenital Malformation Working Group. Am J Public Health 90: 415–419.
Ma, J, Stampfer, MJ, Christensen, B, et al. (1999). A polymorphism of the methionine synthase gene: association with plasma folate, vitamin B12, homocyst(e)ine, and colorectal cancer risk. Cancer Epidemiol Biomarkers Prev 8: 825–829.
Ma, J, Stampfer, MJ, Giovannucci, E, et al. (1997). Methylene-tetrahydrofolate reductase polymorphism, dietary interactions, and risk of colorectal cancer. Cancer Res 57: 1098–1102.
Ma, J, Stampfer, MJ, Hennekens, CH, et al. (1996). Methylene-tetrahydrofolate reductase polymorphism, plasma folate, homocysteine, and risk of myocardial infarction in US physicians [see comments]. Circulation 94: 2410–2416.
Marazita, ML, Goldstein, AM, Smalley, SL, Spence, MA (1986). Cleft lip with or without cleft palate: reanalysis of a three-generation family study from England. Genet Epidemiol 3: 335–342.
Marazita, ML, Jaskoll, T, Melnick, M (1988). Corticosteroidinduced cleft in short-ear mice. J Craniofac Genet Dev Biol 8: 47–51.
Meinert, CL (1986). Clinical Trials. Design, Conduct, and Analysis. New York: Oxford University Press.
Melnick, M, Jaskoll, T, Slavkin, HC (1981). Corticosteroid-induced cleft palate in mice and H-2 haplotype: maternal and embryonic effects. Immunogenetics 13: 443–450.
Meltzer, HJ (1956). Congenital anomalies due to attempted abortion with 4-amino-pteroylglutamic acid. JAMA 161: 1253.
Miller, TJ (1972). Cleft palate formation: a role for pyridoxine in the closure of the secondary palate in mice. Teratology 6: 351–356.
Mills, JL, McPartlin, JM, Kirke, PN, et al. (1995). Homocysteine metabolism in pregnancies complicated by neural-tube defects [see comments]. Lancet 345: 149–151.
Milunsky, A, Graef, JW, Gaynor, ME (1968). Methotrexate-induced congenital malformations. J Pediatr 72: 790–795.
Mitchell, LE, Christensen, K (1996). Analysis of the recurrence patterns for nonsyndromic cleft lip with or without cleft palate in the families of 3,073 Danish probands. Am J Med Genet 61: 371–376.
Mitchell, LE, Risch, N (1992). Mode of inheritance of nonsyndromic cleft lip with or without cleft palate: a reanalysis. Am J Hum Genet 51: 323–332.
Mornet, E, Muller, F, Lenvoise-Furet A, et al. (1997). Screening of the C677T mutation on the methylenetetrahydrofolate reductase gene in French patients with neural tube defects. Hum Genet 100: 512–514.
MRC Vitamin Study Research Group (1991). Prevention of neural tube defects: results of the Medical Research Council Vitamin Study. MRC Vitamin Study Research Group [see comments]. Lancet 338: 131–137.
Munger, RG, Romitti, PA, Daack-Hirsch, S, et al. (1996). Maternal alcohol use and risk of orofacial cleft birth defects. Teratology 54: 27–33.
Murray, JC (1995). Face facts: genes, environment, and clefts [editorial; comment]. Am J Hum Genet 57: 227–232.
Natsume, N, Kawai, T, Ogi, N, et al. (2000). Maternal risk factors in cleft lip and palate: case control study. Br J Oral Maxillofac Surg 38: 23–25.
Natsume, N, Sugimoto, S, Yoshida, K (1999). Influence of maternal anaemia during early pregnancy on the development of cleft palate [letter; comment]. Br J Oral Maxillofac Surg 37: 330–331.
Nelson, M, Asling, C, Evans, HM (1952). Production of multiple congenital abnormalities in young by maternal pteroylglutamic acid deficiency during gestation. J Nutr 48: 61–66.
Nelson, MM, Evans, HM (1947). Reproduction in the rat on purified diets containing succinylsulfathiazole. Proc Soc Exp Biol Med 66: 289.
Niebyl, JR, Blake, DA, Rocco, LE, et al. (1985). Lack of maternal metabolic, endocrine, and environmental influences in the etiology of cleft lip with or without cleft palate. Cleft Palate J 22: 20–28.
Noback, CR, Kupperman, HS (1944). Proc Soc Exp Biol Med 57: 183.
O'Dell, BL, Whitley, JR, Hogan, AG (1951). Vitamin B12, a factor in prevention of hydrocephalus in infant rats. Proc Soc Exp Biol Med 76: 349.
Paros, A, Beck, SL (1999). Folinic acid reduces cleft lip [CL(P)] in A/WySn mice. Teratology 60: 344–347.
Paul, AA, Southgate, DA (1979). McCance and Widdowson's “The composition of foods”: dietary fibre in egg, meat and fish dishes. J Hum Nutr 33: 335–336.
Peer, L, Bryan, W, Strean, LP, et al. (1958a). Induction of cleft palate in mice by cortisone and its reduction by vitamins. J Int Coll Surg 30: 249–254.
Peer, L, Gordon, H, Bernhard, WG (1964). Effect of vitamins on human teratology. Plast Reconstr Surg 34: 358–363.
Peer, L, Strean, L, Walker, JC, et al. (1958b). Study of 400 pregnancies with birth of cleft lip-palate infants: protective effect of folic acid and vitamin B6 therapy. Plast Reconstr Surg 22: 442–449.
Peer, LA, Gordon, HW, Bernhard, WG (1963). Experimental production of congenital deformities and their possible prevention in man. J Int Coll Surg 39:
Piccioni, V, Bologna, U (1949). Clin Obstet 51: 173.
Pickerill, HP (1914). The anatomy and physiology of cleft palate and a new method of treatment. In: Transactions of the Sixth International Dental Congress, London, pp. 453–469.
Powell, HR, Ekert, H (1971). Methotrexate induced congenital malformations. Med J Aust 58: 1076–1077. Romanoff, AL, Bauernfeind, JC (1942). Anat Rec 82: 11.
Romitti, PA, Munger, RG, Murray, JC, et al. (1998). The effect of follow-up on limiting non-participation bias in genetic epidemiologic investigations. Eur J Epidemiol 14: 129–138.
Rosenquist, TH, Ratashak, SA, Selhub, J (1996). Homocysteine induces congenital defects of the heart and neural tube: effect of folic acid. Proc Natl Acad Sci USA 93: 15227–15232.
Ross, CA (1999). Vitamin A and retinoids. In: Modern Nutrition in Health and Disease, edited by ME Shils, JA Olson, M Shike and AC Ross. Baltimore: Williams and Wilkins.
Rothman, KJ, Moore, LL, Singer, MR, et al. (1995). Teratogenicity of high vitamin A intake [see comments]. N Engl J Med 333: 1369–1373.
Salomon, DS, Gift, VD, Pratt, RM (1979). Corticosterone levels during midgestation in the maternal plasma and fetus of cleft palatesensitive and -resistant mice. Endocrinology 104: 154–156.
Salomon, DS, Pratt, RM (1979). Involvement of glucocortkoids in the development of the secondary palate. Differentiation 13: 141–154.
Sauberlich, HE, Canham, JE, Baker, EM, et al. (1972). Biochemical assessment of the nutritional status of vitamin B6 in human. Am J Clin Nutr 25: 629–642.
Saxen, I (1975). Associations between oral clefts and drugs taken during pregnancy. Int J Epidemiol 4: 37–44.
Schubert, J, Schmidt, R, Raupach, HW (1990). New findings explaining the mode of action in prevention of facial clefting and first clinical experience. J Craniomaxillofac Surg 18: 343–347.
Shaw, EB, Steinbach, HL (1968). Aminopterin-induced fetal malformations. Am J Dis Child 115: 477–482.
Shaw, GM, Lammer, EJ (1999). Maternal perkonceptional alcohol consumption and risk for orofacial clefts. J Pediatr 134: 298–303.
Shaw, GM, Lammer, EJ, Wasserman, CR, et al. (1995). Risks of orofacial clefts in children born to women using multivitamins containing folk acid perkonceptionally [see comments]. Lancet 346: 393–396.
Shaw, GM, Rozen, R, Finnell, RH, et al. (1998). Infant C677T mutation in MTHFR, maternal perkonceptional vitamin use, and cleft lip [see comments]. Am J Med Genet 80: 196–198.
Shaw GM, Rozen, R, Finnell, RH, et al. (1998b). Maternal vitamin use, genetic variation of infant methylenetetrahydrofolate reductase, and risk for spina bifida [see comments]. Am J Epidemiol 148: 30–37.
Shaw, GM, Todoroff, K, Finnell, RH, et al. (1999). Maternal vitamin use, infant C677T mutation in MTHFR, and isolated cleft palate risk [letter]. Am J Med Genet 85: 84–85.
Slattery, ML, Potter, JD, Samowitz, W, et al. (1999). Methylenetetrahydrofolate reductase, diet, and risk of colon cancer. Cancer Epidemiol Biomarkers Prev 8: 513–518.
Smithells, RW, Sheppard, S, Schorah, CJ (1976). Vitamin dificiencies and neural tube defects. Arch Dis Child 51: 944–950.
Soprano, DR, Soprano, KJ (1995). Retinoids as teratogens. Annu Rev Nutr 15: 111–132.
Steegers-Theunissen, RP, Boers, GH, Blom, HJ, et al. (1995). Neural tube defects and elevated homocysteine levels in amniotic fluid. Am J Obstet Gynecol 172: 1436–1441.
Steegers-Theunissen, RP, Boers, GH, Steegers, EA, et al. (1992). Effects of sub-50 oral contraceptives on homocysteine metabolism: a preliminary study. Contraception 45: 129–139.
Steegers-Theunissen, RP, Boers, GH, Trijbels, FJ, Eskes, TK, (1991). Neural-tube defects and derangement of homocysteine metabolism [letter]. N Engl J Med 324: 199–200.
Steegers-Theunissen, RP, Boers, GH, Trijbels, FJ, et al. (1994). Maternal hyperhomocysteinemia: a risk factor for neural-tube defects? Metabolism 43: 1475–1480.
Stoll, C, Dott, B, Alembik, Y, et al. (1999). Maternal trace elements, vitamin B12, vitamin A, folic acid, and fetal malformations. Re-prod Toxicol 13: 53–57.
Strauss, OA (1914). Predisposing causes of cleft palate and harelip. In: Transactions of the Sixth International Dental Congress, London, pp. 470–471.
Sunde, ML, Cravens, WW, Elvehjem, CA, et al. (1950). The effect of folic acid on embryonic development of the domestic fowl. Poultry Sci 29: 696.
Thiersch, JB (1956). The control of reproduction in rats with the aid of antimetabolites, and early experiments with antimetabolites as abortifacient agents in man. Acta Endocr 28(Suppl): 37–45.
Thiersch, JB, Philips, FS (1950). Effect of 4-amino-pteroylglutamic acid (aminopterin) on early pregnancy. Proc Soc Exp Biol Med 74: 204–208.
Tolarova, M (1982). Periconceptional supplementation with vitamins and folic acid to prevent recurrence of cleft lip [letter]. Lancet 2: 217.
Tolarova, M (1986). Incidence of orofacial clefts in Bohemia 1975-1982 [in Czech]. Cesk Pediatr 41: 131–135.
Tolarova, M (1987a). Orofacial clefts in Czechoslovakia. Incidence, genetics and prevention of cleft lip and palate over a 19-year period. Scand J Plast Reconstr Surg Hand Surg 21: 19–25.
Tolarova, M (1987b). A study of the incidence, sex-ratio, laterality and clinical severity in 3,660 probands with facial clefts in Czechoslovakia. Acta Chir Plast 29: 77–87.
Tolarova, M, Harris, J (1995). Reduced recurrence of orofacial clefts after periconceptional supplementation with high-dose folic acid and multivitamins. Teratology 51: 71–78.
Truesdale, K (1984). Nutrition to meet the needs of the elderly. Can Dent Hyg 18: 40–41.
Tyan, ML, Tyan, DB (1993). Vitamin A-enhanced cleft palate susceptibility gene maps between C4 and B144 within the H-2 complex. Proc Soc Exp Biol Med 202: 482–486.
von Krebig, VT, Stoeckenius, M (1978). Fehlbildungen beim menschen: lippen-kiefer-gaumenspalten. Med Monatsschr Pharm 1: 243–249.
Wald, NJ, Law, M, Jordan, R (1998). Folic acid food fortification to prevent neural tube defects [letter; comment]. Lancet 351: 834–835.
Wald, NJ, Polani, PE (1984). Neural-tube defects and vitamins: the need for a randomized clinical trial. Br J Obstet Gynaecol 91: 516–523.
Warkany, J (1954). Congenital malformations induced by maternal dietary deficiency: experiments and their interpretation. In: The Harvey Lectures. 1952–1953. New York: Academic Press, pp. 89–109.
Warkany, J, Beaudry, PJ, Hornstein, S (1959). Attempted abortion with aminopterin (4-amino-pteroylglutamic acid). Am J Dis Child 97: 274–281.
Warkany, J, Kalter, H (1957). Experimental cleft palate: interpretations and misinterpretations. Bull Am Assoc Cleft Palate Rehabil 7: 9.
Warkany, J, Nelson, RC (1940). Appearance of skeletal abnormalities in the offspring of rats reared on a deficient diet. Science 92: 383–384.
Warkany, J, Nelson, RC (1941). Skeletal abnormalities in the offspring of rats reared on deficient diets. Anat Rec 79: 83.
Warkany, J, Nelson, RC, Schraffenberger, E (1942). Congenital malformations induced in rats by maternal nutritional deficiency. II. Use of varied diets and of different strains of rats. Am J Dis Child 64: 860.
Warkany, J, Shraffenberger, E (1944). J Nutr 27: 477.
Werler, MM, Hayes, C, Louik, C, et al. (1999). Multivitamin supplementation and risk of birth defects. Am J Epidemiol 150: 675–682.
Werler, MM, Lammer, EJ, Rosenberg, L, Mitchell, AA (1991). Maternal alcohol use in relation to selected birth defects. Am J Epidemiol 134: 691–698.
West, KP, Jr, Katz, J, Khatry, SK, et al. (1999). Double blind, cluster randomised trial of low dose supplementation with vitamin A or beta carotene on mortality related to pregnancy in Nepal. The NNIPS-2 Study Group [see comments]. BMJ 318: 570–575.
Whitby, KE, Collins, TF, Welsh, JJ, et al. (1994). Developmental effects of combined exposure to ethanol and vitamin A. Food Chem Toxicol 32: 305–320.
Wilson, JG, Warkany, J (1948). Malformations in the genito-urinary tract induced by maternal vitamin A deficiency in the rat. Am J Anat 83: 357.
Womersley, J, Stone, DH (1987). Epidemiology of facial clefts. Arch Dis Child 62: 717–720.
Wong, WY, Eskes, TK, Kuijpers-Jagtman, AM (1999). Nonsyndromic orofacial clefts: association with maternal hyperhomocysteinemia. Teratology 60: 253–257.
World Health Organization (1998). World Atlas of Birth Defects. Geneva: World Health Organization.
Wyszynski, DF, Beaty, TH (1996). Review of the role of potential teratogens in the origin of human nonsyndromic oral clefts. Teratology 53: 309–317.
Wyszynski, DF, Mitchell, LE (1999). Report of the newly formed International Consortium for Oral Clefts Genetics. Cleft Palate Craniofac J 36: 174–178.
Yamaguchi, T (1968). Effects of riboflavin, pridoxine, and folic acid on the incidence of malformations in the rat caused by hypervitaminosis A. Congenit Anom 8: 175–182.
Yoneda, T, Pratt, RM (1982a). Vitamin B6 reduces cortisone-induced cleft palate in the mouse. Teratology 26: 255–258.
Yoneda, T, Pratt, RM (1982b). Glucocorticoid receptors in palatal mesenchymal cells from the human embryo: relevance to human cleft palate formation. J Craniofac Genet Dev Biol 1: 411–423.