American Journal of Medical Genetics 43:1016-1022 (1992)
Can Maternal Risk Factors Influence the Presence of Major Birth Defects in Infants With Down Syndrome? Muin J. Khoury and J. David Erickson Birth Defects and Genetic Diseases Branch, Division of Birth Defects and Developmental Disabilities, Center for Environmental Health and Znjury Control, Centers for Disease Control, Atlanta, Georgia 30333 Although the manifestations of Down syndrome (DS) are well known, certain major birth defects such as duodenal atresia and endocardia1 cushion defects are present in some infants but not others, suggesting the possible role of other genetic or environmental factors interacting with the trisomy genotype. To explore the possible role of maternal factors in the presence of major defects among DS infants, we examined data from an epidemiologic study of DS conducted in metropolitan Atlanta. Of 219 DS infants born between 1968and 1980,50had recorded cardiac defects, 9 had selected gastrointestinal atresias and 4 had oral clefts. We evaluated the association of these defects with several maternal factors including age, race, first trimester cigarette smoking, alcohol use, and fever. We found that different maternal factors were associated with several defects: (1)mother’s race with cardiac defects (40% in blacks vs. 17% in whites, P < 0.01), (2) mother’s age with oral clefts (6%for < 25 years, 1%for 25-34, and 0% for > 34, P< 0.051, and (3) maternal first trimester fever with gastrointestinal defects (15% in infants with history of fever and 3% in infants without a history of fever, P < 0.01). We also observed an inverse relationship between maternal alcohol use and the presence of ventricular septal defect. These findings suggest that maternal risk factors may influence the clinical manifestations of DS. In addition to searching for a genetic basis for the DS phenotype, we suggest that the role of environmental factors and maternal exposures be
Received for publication September 9, 1991; revision received December 27, 1991. Address reprint requests to Muin J. Khoury, M.D., Ph.D., Birth Defects and Genetic Diseases Branch, Centers for Disease Control (MailStop F45), 1600 Clifton Road, Atlanta, GA 30333.
0 1992 Wiley-Liss, Inc.
specifically explored in clarifying the genesis of various birth defects in Down syndrome.0 1992 Wiley-Lisa, Inc.
KEY
WORDS: Congenital abnormalities, Down syndrome, maternal factors
INTRODUCTION Although the clinical manifestations of Down syndrome (DS) are well recognized, certain major birth defects are present in only a proportion of cases [Jones, 19881. Defects most commonly associated with DS are cardiac malformations (up to 40% of cases), notably endocardial cushion defects, and include, among others, ventricular septal defects, atrial septal defects and patent ductus arteriosus [Greenwood and Nadas, 1976; Rehder, 1981; Spicer, 1984;Ferencz et al., 19891. In addition, gastrointestinal defects are also seen in some DS infants. These include duodenal atresia, anorectal atresia, Hirschsprung disease, annular pancreas, and tracheo-esophageal fistula [Safra et al., 1976; Gilbert and Opitz, 1982; Buchin et al., 19861. Although most cardiac, gastrointestinal, and other defects occur not only in DS, but also with other chromosome abnormalities, they occur in DS infants much more than expected by chance alone [Opitz and Gilbert, 1982;Khoury and Cordero, 19891. The variability of the presence of major birth defects in DS has not been fully explored. In addition to direct dosage effects associated with genes located on chromosome 21 [Korenberg et al., 19901, adverse regulatory effects of gene products on other chromosomes [Weil and Epstein, 19791, chromosome imprinting [Cattanach, 1986; Nicholls et al., 19891 and simple stochastic processes [Kurnit et al., 19851,it is also possible that maternal risk factors, such as pregnancy events and exposures, may interact with an already unbalanced and susceptible genotype leading to the presence of certain defects in some cases but not others. Opitz and Gilbert [19821 suggested that in aneuploidy, there could be a state of “hyperreactivity to teratogens through lack of
Maternal Factors and Birth Defects in Down Syndrome protective developmental buffering,” as a result of generalized instability in growth and development, as well as disruption of homeostatic mechanisms [Waddington, 1957; Shapiro, 1975, 19831. Nevertheless, the association between maternal exposures and various phenotypic manifestations has not been adequately evaluated in epidemiologic studies of DS because such studies usually focus on prior maternal events that may lead to nondisjunction [Janerich and Bracken, 19861. Although maternal pregnancy events and other risk factors are irrelevant to the etiology of nondisjunction, they could theoretically be important in the development of individual birth defects in some DS infants but not others. In this report, we present an exploratory analysis of the association between selected major birth defects in infants with DS (cardiac, gastrointestinal, and oral clefts) and several maternal demographic factors (age and race), and pregnancy exposures (smoking, alcohol use, and fever).Results of this population-based epidemiologic study suggest interesting associations between several maternal factors and different types of defects in DS. METHODS The Metropolitan Atlanta Congenital Defects Program Cases of DS in this study come from the Metropolitan Atlanta Congenital Defects Program (MACDP),a population-based birth defects surveillance system in operation since 1968. The system ascertains all stillborn and live-born infants diagnosed with serious birth defects in the first year of life and whose mothers are residents of 5 counties in metropolitan Atlanta. Multiple sources of ascertainment are used, including hospital records, vital records, and specialty clinics. Results of chromosome diagnoses are recorded from various cytogenetic laboratories in the area. For each case, staff from the Centers for Disease Control record all defects on a special abstraction form. Birth defect diagnoses are periodically updated in the system as more information (e.g., surgery, cardiac evaluation, etc.) on cases become available. Thus, for each DS infant in the registry, information is recorded on individual categories of defects (e.g., cardiac, gastrointestinal), in addition to the overall diagnosis. The system has been described in detail elsewhere [Edmonds et al., 19811.
The Atlanta Birth Defects Case-Control Study Infants born between 1968 and 1980 ascertained via MACDP were enrolled in a case-control study conducted during 1982 and 1983. The objective of the study was to evaluate the association of a variety of parental risk factors with various defects [Erickson et al., 1984; Erickson, 19911. Maternal and paternal interviews were conducted on about 4,000 babies with birth defects and 3,000 live-born infants without birth defects, and frequency-matched to case infants by period of birth, race, and hospital of birth. A total of 219 infants with DS were part of this study and are the study population in this analysis. Five had translocation abnormalities and 3 were mosaic. Information on parental medical history, demographic information, lifestyle factors, family his-
1017
tory, occupational data, as well as pregnancy illnesses and exposures were collected using a structured interview instrument. This study does not include DS cases diagnosed prenatally. The impact of prenatal diagnosis on the rate of DS at birth was negligible during this time period (1968-1980). Methods of Analysis We first evaluated the association between several maternal factors and the presence or absence of specific major defects in infants with DS. Maternal factors examined included: race (white and black), age in years (34), cigarette smoking (from - 1 to + 3 months of pregnancy: nonsmokers, 3 drinks per week) and reported maternal fever (from - 1 to + 3 months of pregnancy: no fever, and any fever). Because of the relatively small sample size of this series, we could not adequately examine other maternal illnesses, exposures and occupational information. We focused on 3 relatively frequent categories of major defects in these infants: (1) cardiac defects (excluding patent ductus arteriosus in infants born
Can Maternal Risk Factors Influence the Presence of Major Birth Defects in Infants With Down Syndrome? Muin J. Khoury and J. David Erickson Birth Defects and Genetic Diseases Branch, Division of Birth Defects and Developmental Disabilities, Center for Environmental Health and Znjury Control, Centers for Disease Control, Atlanta, Georgia 30333 Although the manifestations of Down syndrome (DS) are well known, certain major birth defects such as duodenal atresia and endocardia1 cushion defects are present in some infants but not others, suggesting the possible role of other genetic or environmental factors interacting with the trisomy genotype. To explore the possible role of maternal factors in the presence of major defects among DS infants, we examined data from an epidemiologic study of DS conducted in metropolitan Atlanta. Of 219 DS infants born between 1968and 1980,50had recorded cardiac defects, 9 had selected gastrointestinal atresias and 4 had oral clefts. We evaluated the association of these defects with several maternal factors including age, race, first trimester cigarette smoking, alcohol use, and fever. We found that different maternal factors were associated with several defects: (1)mother’s race with cardiac defects (40% in blacks vs. 17% in whites, P < 0.01), (2) mother’s age with oral clefts (6%for < 25 years, 1%for 25-34, and 0% for > 34, P< 0.051, and (3) maternal first trimester fever with gastrointestinal defects (15% in infants with history of fever and 3% in infants without a history of fever, P < 0.01). We also observed an inverse relationship between maternal alcohol use and the presence of ventricular septal defect. These findings suggest that maternal risk factors may influence the clinical manifestations of DS. In addition to searching for a genetic basis for the DS phenotype, we suggest that the role of environmental factors and maternal exposures be
Received for publication September 9, 1991; revision received December 27, 1991. Address reprint requests to Muin J. Khoury, M.D., Ph.D., Birth Defects and Genetic Diseases Branch, Centers for Disease Control (MailStop F45), 1600 Clifton Road, Atlanta, GA 30333.
0 1992 Wiley-Liss, Inc.
specifically explored in clarifying the genesis of various birth defects in Down syndrome.0 1992 Wiley-Lisa, Inc.
KEY
WORDS: Congenital abnormalities, Down syndrome, maternal factors
INTRODUCTION Although the clinical manifestations of Down syndrome (DS) are well recognized, certain major birth defects are present in only a proportion of cases [Jones, 19881. Defects most commonly associated with DS are cardiac malformations (up to 40% of cases), notably endocardial cushion defects, and include, among others, ventricular septal defects, atrial septal defects and patent ductus arteriosus [Greenwood and Nadas, 1976; Rehder, 1981; Spicer, 1984;Ferencz et al., 19891. In addition, gastrointestinal defects are also seen in some DS infants. These include duodenal atresia, anorectal atresia, Hirschsprung disease, annular pancreas, and tracheo-esophageal fistula [Safra et al., 1976; Gilbert and Opitz, 1982; Buchin et al., 19861. Although most cardiac, gastrointestinal, and other defects occur not only in DS, but also with other chromosome abnormalities, they occur in DS infants much more than expected by chance alone [Opitz and Gilbert, 1982;Khoury and Cordero, 19891. The variability of the presence of major birth defects in DS has not been fully explored. In addition to direct dosage effects associated with genes located on chromosome 21 [Korenberg et al., 19901, adverse regulatory effects of gene products on other chromosomes [Weil and Epstein, 19791, chromosome imprinting [Cattanach, 1986; Nicholls et al., 19891 and simple stochastic processes [Kurnit et al., 19851,it is also possible that maternal risk factors, such as pregnancy events and exposures, may interact with an already unbalanced and susceptible genotype leading to the presence of certain defects in some cases but not others. Opitz and Gilbert [19821 suggested that in aneuploidy, there could be a state of “hyperreactivity to teratogens through lack of
Maternal Factors and Birth Defects in Down Syndrome protective developmental buffering,” as a result of generalized instability in growth and development, as well as disruption of homeostatic mechanisms [Waddington, 1957; Shapiro, 1975, 19831. Nevertheless, the association between maternal exposures and various phenotypic manifestations has not been adequately evaluated in epidemiologic studies of DS because such studies usually focus on prior maternal events that may lead to nondisjunction [Janerich and Bracken, 19861. Although maternal pregnancy events and other risk factors are irrelevant to the etiology of nondisjunction, they could theoretically be important in the development of individual birth defects in some DS infants but not others. In this report, we present an exploratory analysis of the association between selected major birth defects in infants with DS (cardiac, gastrointestinal, and oral clefts) and several maternal demographic factors (age and race), and pregnancy exposures (smoking, alcohol use, and fever).Results of this population-based epidemiologic study suggest interesting associations between several maternal factors and different types of defects in DS. METHODS The Metropolitan Atlanta Congenital Defects Program Cases of DS in this study come from the Metropolitan Atlanta Congenital Defects Program (MACDP),a population-based birth defects surveillance system in operation since 1968. The system ascertains all stillborn and live-born infants diagnosed with serious birth defects in the first year of life and whose mothers are residents of 5 counties in metropolitan Atlanta. Multiple sources of ascertainment are used, including hospital records, vital records, and specialty clinics. Results of chromosome diagnoses are recorded from various cytogenetic laboratories in the area. For each case, staff from the Centers for Disease Control record all defects on a special abstraction form. Birth defect diagnoses are periodically updated in the system as more information (e.g., surgery, cardiac evaluation, etc.) on cases become available. Thus, for each DS infant in the registry, information is recorded on individual categories of defects (e.g., cardiac, gastrointestinal), in addition to the overall diagnosis. The system has been described in detail elsewhere [Edmonds et al., 19811.
The Atlanta Birth Defects Case-Control Study Infants born between 1968 and 1980 ascertained via MACDP were enrolled in a case-control study conducted during 1982 and 1983. The objective of the study was to evaluate the association of a variety of parental risk factors with various defects [Erickson et al., 1984; Erickson, 19911. Maternal and paternal interviews were conducted on about 4,000 babies with birth defects and 3,000 live-born infants without birth defects, and frequency-matched to case infants by period of birth, race, and hospital of birth. A total of 219 infants with DS were part of this study and are the study population in this analysis. Five had translocation abnormalities and 3 were mosaic. Information on parental medical history, demographic information, lifestyle factors, family his-
1017
tory, occupational data, as well as pregnancy illnesses and exposures were collected using a structured interview instrument. This study does not include DS cases diagnosed prenatally. The impact of prenatal diagnosis on the rate of DS at birth was negligible during this time period (1968-1980). Methods of Analysis We first evaluated the association between several maternal factors and the presence or absence of specific major defects in infants with DS. Maternal factors examined included: race (white and black), age in years (34), cigarette smoking (from - 1 to + 3 months of pregnancy: nonsmokers, 3 drinks per week) and reported maternal fever (from - 1 to + 3 months of pregnancy: no fever, and any fever). Because of the relatively small sample size of this series, we could not adequately examine other maternal illnesses, exposures and occupational information. We focused on 3 relatively frequent categories of major defects in these infants: (1) cardiac defects (excluding patent ductus arteriosus in infants born
