First trimester exposure to paroxetine and risk of cardiac malformations in infants: the importance of dosage

BACKGROUND: Conflicting findings with regard to the teratogenic risks of first trimester use of paroxetine have prompted the FDA, Health Canada, and the manufacturer of the drug to issue warnings against its use during pregnancy. Given that untreated depression during pregnancy can lead to deleterious effect on the mother and her unborn fetus, data on the relationship between the dose and the range of malformations is warranted. This study attempts to quantify the association between first trimester exposure to paroxetine and congenital cardiac malformations, adjusting for possible confounders, and to quantify the dose-response relationship between paroxetine use and cardiac defects. METHODS: The Medication and Pregnancy registry was used. This population-based registry was built by linking three administrative databases (RAMQ, Med-Echo, and ISQ), and includes all pregnancies in Quebec between 01/01/1997 and 06/30/2003. Date of entry in the registry is the date of the first day of the last menstrual period. To be eligible for this study, women had to: 1) be 15-45 years of age at entry; 2) be covered by the RAMQ drug plan >or=12 months before and during pregnancy; 3) be using only one type of antidepressant during the first trimester; and 4) have a live birth. Two nested case-control studies were carried out comparing the prevalence of paroxetine use in the first trimester of pregnancy to the prevalence of other antidepressant exposures during the same time period. Cases were defined as: 1) any major malformations; or 2) any cardiac malformations diagnosed in the first year of life; controls were defined as no major or minor malformations. Multivariate logistic regression techniques were used to analyze data. RESULTS: Among the 1,403 women meeting inclusion criteria, 101 infants with major congenital malformations were identified; 24 had cardiac malformations. Adjusting for possible confounders, the use of paroxetine (odds ratio [OR] = 1.38, 95% confidence interval [CI] = 0.49-3.92), and the use of other SSRIs (OR = 0.89, 95% CI = 0.28-2.84) during the first trimester of pregnancy did not increase the risk of congenital cardiac malformations compared with the use of non-SSRI antidepressants. When considering the dose, however, a dose-response relationship was observed, thus women exposed to >25 mg/day of paroxetine during the first trimester of pregnancy were at increased risk of having an infant with major congenital malformations (adjusted [adj] OR = 2.23, 95% CI = 1.19, 4.17), or major cardiac malformations (adj OR = 3.07, 95% CI = 1.00, 9.42). CONCLUSIONS: Gestational exposure to paroxetine is associated with major congenital malformations and major cardiac malformations for only first trimester exposure above 25 mg/day.     

Congenital heart disease in the offspring and maternal habits and home exposures during pregnancy.

To test the effect of maternal habits and home exposures during early pregnancy on the occurrence of congenital heart disease in the offspring, 406 cases and 756 controls were studied. The cases included all cardiovascular malformations detected in Finland during 1982-1983, while the healthy controls were randomly selected from all babies born during the same period. Case and control mothers were interviewed after delivery using a structured and pre-tested questionnaire. Maternal overall drug consumption during the first trimester was as prevalent among case mothers (13.3%) as controls (14.6%). Neither was the risk of congenital heart disease associated with maternal use of contraceptive pills, salicylates, diazepam, or sweetening agents separately. Maternal exposures to disinfectants, dyes, lacquers, paints, pesticides, or glues at home were equally prevalent in case and control groups. Several earlier miscarriages was a predictor of an infant born with congenital heart disease (OR = 2.7, CI95 = 1.4-5.3). Maternal ultrasound examination was performed during the first 16 weeks of pregnancy more often among the case group (28.3%) than among the control group (22.0%). However, the association between ultrasound examination and the risk of congenital heart disease in the offspring was not statistically significant (OR = 1.2, 95% confidence interval 0.9-1.7) when adjusted for confounding factors such as the threat of miscarriage in logistic regression analysis. It is concluded that maternal ultrasound examination, intake of some common drugs, and exposure to a number of environmental factors at home during early pregnancy are probably not harmful for the developing fetal heart.     

Congenital malformations and maternal smoking during pregnancy

The relationship between smoking during pregnancy and congenital malformations was studied in prospective studies of 33,434 live births in the Kaiser-Permanente Birth Defects Study and 53,512 live births in the Collaborative Perinatal Project (CPP). In the Kaiser study, statistically significant positive associations (P less than .05) were observed for ventral hernias, omphaloceles, and "other major gut abnormalities," but, for each comparison the estimates were based on one or two unexposed cases. Statistically significant negative associations were found for ventricular septal defects (odds ratio, 0.5 [95% confidence interval, 0.2-0.96]), hydroceles (0.7 [0.6-0.9]), clubfoot (0.7 [0.6-0.9]), pigmented nevi (0.7 [0.6-0.9]), hemangiomas (0.8[0.7-0.98]) and Down syndrome (0.2 [0.1-0.9]). To determine if the findings noted above were an artifact of multiple comparisons, seven of these nine malformations were analyzed by smoking status for women in the CPP. All but one of the associations were not confirmed in the CPP. Previously described associations between smoking and specific congenital malformations were also tested using data from the CPP. We conclude that smoking is unlikely to be responsible for a large increase in malformations at birth.     

Inflammatory Bowel Disease and Risk of Adverse Pregnancy Outcomes

Background and Objectives

Existing data on pregnancy complications in inflammatory bowel disease (IBD) are inconsistent. To address these inconsistencies, we investigated potential associations between IBD, IBD-related medication use during pregnancy, and pregnancy loss, pre-eclampsia, preterm delivery, Apgar score, and congenital abnormalities.

Methods

We conducted a cohort study in >85,000 Danish National Birth Cohort women who were pregnant in the period 1996-2002 and had information on IBD, IBD-related medication use (systemic or local corticosteroids, 5-aminosalicylates), pregnancy outcomes and potential confounders. We evaluated associations between IBD and adverse pregnancy/birth outcomes using Cox regression and log-linear binomial regression.

Results

IBD was strongly and significantly associated with severe pre-eclampsia, preterm premature rupture of membranes and medically indicated preterm delivery in women using systemic corticosteroids during pregnancy (hazard ratios [HRs] >7). IBD was also associated with premature preterm rupture of membranes in women using local corticosteroid medications (HR 3.30, 95% confidence interval [CI] 1.33-8.20) and with medically indicated preterm delivery (HR 1.91, 95% CI 0.99-3.68) in non-medicated women. Furthermore, IBD was associated with low 5-minute Apgar score in term infants (risk ratio [RR] 2.19, 95% CI 1.03-4.66). Finally, Crohn’s disease (but not ulcerative colitis) was associated with major congenital abnormalities in the offspring (RR 1.85, 95% CI 1.06-3.21). No child with a congenital abnormality born to a woman with IBD was exposed to systemic corticosteroids in utero.

Conclusion

Women with IBD are at increased risk of severe pre-eclampsia, medically indicated preterm delivery, preterm premature rupture of membranes, and delivering infants with low Apgar score and major congenital malformations. These associations are only partly explained by severe disease as reflected by systemic corticosteroid use.     

Combined effect of prenatal solvent exposure and GSTT1 or GSTM1 polymorphisms in the risk of birth defects

Exposure to solvents during pregnancy has long been suspected to increase the risk of congenital malformations. Glutathione S-transferases (GSTs) are enzymes essential for the detoxification of various chemicals. Our objective here was to assess whether GST polymorphisms might modify the association between maternal solvent exposure and the risk of birth defects. A prospective cohort included 3421 pregnant women in Brittany, France (2002-2006). Occupational exposure to solvents was assessed from a job-exposure matrix. Congenital malformations were diagnosed among livebirths, stillbirths, and medical pregnancy terminations. Using a nested case-control design, 32 babies with major birth defects were compared to 348 normal births for babies' cord blood genotypes (at GSTT1 and GSTM1) and maternal occupational solvent exposure. Logistic models were used to adjust for potential confounders. The risk of major defects increased significantly in women with solvent exposure (20% of controls and 34% of cases). Frequencies of the null genotype of both the GSTT1 and GSTM1 genes were similar among controls and cases. There was a significantly increased risk of birth defects in GSTM1 not-null cord-blood genotype in pregnancies exposed to solvents (odds ratio [OR], 1.0 for not-null, not-exposed; OR, 4.0 for not-null, exposed; 95% confidence interval [CI], 1.4-11.2; OR, 1.6 for null, not-exposed; 95% CI, 0.6-3.9; OR, 1.0 for null, exposed; 95% CI, 0.2-4.7; p = 0.05). This nested case-control study suggests that the child's GSTM1 genotype modifies the risk of major birth defects among offspring of solvent-exposed women. Replication and additional investigations are necessary to confirm and elucidate these findings.     

Early pregnancy azathioprine use and pregnancy outcomes

BACKGROUND: Azathioprine (AZA) is used during pregnancy by women with inflammatory bowel disease (IBD), other autoimmune disorders, malignancy, and organ transplantation. Previous studies have demonstrated potential risks. METHODS: The Swedish Medical Birth Register was used to identify 476 women who reported the use of AZA in early pregnancy. The effect of AZA exposure on pregnancy outcomes was studied after adjustment for maternal characteristics that could act as confounders. RESULTS: The most common indication for AZA use was IBD. The rate of congenital malformations was 6.2% in the AZA group and 4.7% among all infants born (adjusted OR: 1.41, 95% CI: 0.98–2.04). An association between early pregnancy AZA exposure and ventricular/atrial septal defects was found (adjusted OR: 3.18, 95% CI: 1.45–6.04). Exposed infants were also more likely to be preterm, to weigh <2500 gm, and to be small for gestational age compared to all infants born. This effect remained for preterm birth and low birth weight when infants of women with IBD but without AZA exposure were used as a comparison group. A trend toward an increased risk of congenital malformations was found among infants of women with IBD using AZA compared to women with IBD not using AZA (adjusted OR: 1.42, 95% CI: 0.93–2.18). CONCLUSIONS: Infants exposed to AZA in early pregnancy may be at a moderately increased risk of congenital malformations, specifically ventricular/atrial septal defects. There is also an increased risk of growth restriction and preterm delivery. These associations may be confounded by the severity of maternal illness. Birth Defects Research (Part A), 2009. © 2009 Wiley‐Liss, Inc.     

Infants with single ventricle: a population-based epidemiological study

BACKGROUND: Single ventricle, a rare congenital cardiac defect, often occurs as part of a complex group of cardiovascular abnormalities. Little is known of its epidemiologic associations. METHODS: Using data from the Baltimore-Washington Infant Study [BWIS], (1981-89), a population based case-control study of cardiovascular malformations, infants with single ventricle were evaluated with respect to infant and family characteristics and maternal and paternal exposures. The cases were analyzed according to presence/absence of abnormal cardio-visceral situs. Controls were 3,572 infants without heart defects randomly selected from the regional cohort of live births. Odds ratios and 95% confidence intervals were used as measures of association. RESULTS: Single ventricle occurred in 1.25% of infants with congenital cardiovascular defects in the BWIS. Fifty-five infants had single ventricle. In 48 families (87.3%) the parents were interviewed. Thirty-three infants had normal situs and 15 had abnormal situs. Paternal alcohol consumption (OR = 2.0, 95% CI 1.1-3.9) and paternal cigarette smoking (OR = 2.4, 95% CI 1.1-5.1) were associated with all cases of single ventricle. These associations were even stronger in the subset of infants with abnormal situs. Maternal history of a previous induced abortion was also associated with infants born with abnormal situs (OR = 3.2, 95% CI 1.1-11.5). Paternal marijuana use was associated with cases of single ventricle in normal situs (OR = 2.2, 95% CI 1.0-5.2). CONCLUSIONS: Potential risk factors included paternal smoking and alcohol consumption, highlighting the need for future studies to consider environmental factors in the pathogenesis of this cardiac defect.     

Outcomes of pregnancy in insulin dependent diabetic women: results of a five year population cohort study.

OBJECTIVE: To monitor pregnancies in women with pre-existent insulin dependent diabetes for pregnancy loss, congenital malformations, and fetal growth in a geographically defined area of north west England. DESIGN: Population cohort study. SETTING: 10 maternity units in Cheshire, Lancashire, and Merseyside which had no regional guidelines for the management of pregnancy in diabetic women. SUBJECTS: 462 pregnancies in 355 women with insulin dependent diabetes from the 10 centres over five years (1990-4 inclusive). MAIN OUTCOME MEASURES: Numbers and rates of miscarriages, stillbirths, and neonatal and postneonatal deaths; prevalence of congenital malformations; birth weight in relation to gestational age. RESULTS: Among 462 pregnancies, 351 (76%) resulted in a liveborn infant, 78 (17%) aborted spontaneously, nine (2%) resulted in stillbirth, and 24 (5%) were terminated. Of the terminations, nine were for congenital malformation. The stillbirth rate was 25.0/1000 total births (95% confidence interval 8.9 to 41.1) compared with a population rate of 5.0/1000, and infant mortality was 19.9/1000 live births (5.3 to 34.6) compared with 6.8/1000. The prevalence of congenital malformations was 94.0/1000 live births (63.5 to 124.5) compared with 9.7/1000 in the general population. When corrected for gestational age, mean birth weight in the sample was 1.3 standard deviations greater than that of infants of non-diabetic mothers. Infants with congenital malformations weighed less than those without. CONCLUSION: In an unselected population the infants of women with pre-existent insulin dependent diabetes mellitus have a 10-fold greater risk of a congenital malformation and a fivefold greater risk of being stillborn than infants in the general population. Further improvements in the management of pregnancy in diabetic women are needed if target of the St Vincent declaration of 1989 is to be met.     

Interrupted aortic arch: an epidemiologic study

BACKGROUND: Interruption of the aortic arch (IAA) is a rare but severe anomaly associated with major intracardiac defects and with multisystem noncardiac malformations, recently linked to chromosome deletion of 22q11.2. METHODS: The Baltimore-Washington Infant Study (1981-1989), a population-based epidemiologic study of cardiovascular malformations, evaluated 53 infants with IAA in comparison with 3,572 controls. Risk factors for the anatomic subtypes were evaluated in 14 cases of IAA type A and 32 cases of IAA type B, but no molecular genetic tests were available. The distribution of associated cardiac defects was similar for both types. RESULTS: DiGeorge syndrome (DGS) occurred more frequently in IAA type B. Case-control comparisons demonstrated that infants in both groups were growth retarded at birth. A family history of noncardiac defects occurred only in IAA type B cases and included relatives with cleft lip and/or cleft palate. Candidate risk factors were associated only in type B cases and differed for those with (n = 10) and for those without (n = 19) DGS: a family history of noncardiac defects (odds ratio [OR] = 7.2, 95% confidence interval [CI] = 1.5-39.2) and maternal use of aspirin during the critical period (OR = 4.8, 95% CI = 1.3-25.4) occurred with DGS, while previous stillbirth (OR = 9.4, 95% CI = 1.3-53.1), bleeding during pregnancy (OR = 3.7, 95% CI = 1.4-11.4), and maternal exposure to arts/crafts paints (OR = 4.8, 95% CI = 1.3-17.4) were associated in those without DGS. CONCLUSIONS: These findings confirm the heterogeneity of IAA and of the type B subtype. Risk factors specific for cases with DGS may open a window to further investigations of the etiology of IAA and of the associated molecular genetic abnormalities.     

Exogenous sex hormone exposure and the risk for VACTERL association

In several studies investigators have suggested that maternal use of exogenous sex hormones during early pregnancy may be associated with various congenital malformations. A group of malformations, the VACTERL (vertebral, anal, cardiac, trachea, esophageal, renal, limb-acronym) association, has been statistically associated with maternal exposure to exogenous sex hormones during the first trimester of pregnancy. The VACTERL association is a nonrandom group of major malformations that occur together more often than would be expected on the basis of chance. To assess this association, we conducted a case-control study of first-trimester exposure to sex hormones among mothers of 34 infants with the VACTERL association and of 1,024 comparison infants with one or more of ten major malformations or Down syndrome. The study subjects were malformed infants born between July 1970 and June 1979 and registered in a population-based birth defects registry. Information concerning the use of exogenous sex hormones during pregnancy was obtained by systematically interviewing the mothers of the malformed infants. Most of the mothers were interviewed within 6 months of their children's births. Each mother was interviewed within a year of her child's birth. We found an odds ratio of 0.98 (90% confidence limits 0.40, 2.38) for the relationship between VACTERL association and use of any sex hormone in the first trimester of pregnancy. Our study had adequate statistical power to detect a true relative risk of 2.8 or greater.     

First-trimester hemoglobin A1 and risk for major malformation and spontaneous abortion in diabetic pregnancy

The relationship between the level of hemoglobin A1 (Hb A1) in the first trimester and major malformations and spontaneous abortions was examined in 303 insulin-requiring diabetic gravidas. During the study period, all patients with insulin-requiring diabetes mellitus antedating pregnancy who registered for prenatal care prior to 12 weeks' gestation and who had a known outcome were included. Thirty-five percent of the patients entered with a first-trimester Hb A1 of greater than 11.0% of total hemoglobin (9 standard deviations above the mean for a nondiabetic population). A broad spectrum of glycemic control was therefore represented. The risk of spontaneous abortion was 12.4% with first-trimester Hb A1 less than or equal to 9.3% and 37.5% with Hb A1 greater than 14.4% (risk ratio 3.0; 95% confidence interval 1.3-7.0). The risk for major malformation was 3.0% with Hb A1 less than or equal to 9.3% and 40% with Hb A1 greater than 14.4% (risk ratio 13.2; 95% confidence interval 4.3-40.4). Although the risks for both adverse outcomes were markedly elevated following a first trimester in very poor metabolic control, there was a broad range of control over which the risks were not substantially elevated. To keep malformations and spontaneous abortions to a minimum among diabetic women does not require "excellent" control; there seems to be a fairly broad range of "acceptable" control.     

Maternal influenza during pregnancy and risk of congenital abnormalities in offspring

BACKGROUND: The teratogenic effect of influenza viruses is currently being debated, and we examined the large population-based data set of the Hungarian Case-Control Surveillance of Congenital Abnormalities (HCCSCA) to study the possible association between maternal influenza and various congenital abnormalities (CAs). METHODS: The 1980-1996 HCCSCA includes 22,843 newborns or fetuses with CAs, and 38,151 matched controls (newborn infants without any abnormalities). RESULTS: In the case group, 1328 (5.8%) mothers had influenza at some time during their pregnancies compared to 1838 (4.8%) mothers in the control group (adjusted prevalence odds ratios [PORs], 1.3; 95% confidence interval [CI], 1.2-1.4). In the calculation of the adjusted PORs, the use of antifever drugs and maternal employment status were considered. When cases and their matched controls were compared, there was a higher prevalence of maternal influenza during the second and/or third month of pregnancy for the group of newborns with cleft lip +/- palate (adjusted POR, 3.2; 95% CI, 2.0-5.3), neural-tube defects (adjusted POR, 1.9; 95% CI, 1.1-3.3), and cardiovascular malformations (adjusted POR, 1.7; 95% CI, 1.3-2.3). However, a direct teratogenic effect from influenza viruses appears to be unlikely, and we suggest that the higher prevalence of the CAs indicated above can be explained mainly by fever, because this risk was reduced by the use of antifever drugs. Periconceptional folic acid supplementation also showed some preventive effect for these CAs. CONCLUSIONS: The indirect teratogenic effect of maternal influenza during pregnancy may be restricted by appropriate medical treatment (e.g., antifever drugs) and periconceptional folic acid supplementation.     

Risk of complications of pregnancy in women with type 1 diabetes: nationwide prospective study in the Netherlands

Objective To investigate maternal, perinatal, and neonatal outcomes of pregnancies in women with type 1 diabetes in the Netherlands.Design Nationwide prospective cohort study.Setting All 118 hospitals in the Netherlands.Participants 323 women with type 1 diabetes who became pregnant between 1 April 1999 and 1 April 2000.Main outcome measures Maternal, perinatal, and neonatal outcomes of pregnancy.Results 84% (n = 271) of the pregnancies were planned. Glycaemic control early in pregnancy was good in most women (HbA_1c ≤ 7.0% in 75% (n = 212) of the population), and folic acid supplementation was adequate in 70% (n = 226). 314 pregnancies that went beyond 24 weeks'' gestation resulted in 324 infants. The rates of pre-eclampsia (40; 12.7%), preterm delivery (101; 32.2%), caesarean section (139; 44.3%), maternal mortality (2; 0.6%), congenital malformations (29; 8.8%), perinatal mortality (9; 2.8%), and macrosomia (146; 45.1%) were considerably higher than in the general population. Neonatal morbidity (one or more complications) was extremely high (260; 80.2%). The incidence of major congenital malformations was significantly lower in planned pregnancies than in unplanned pregnancies (4.2% (n = 11) v 12.2% (n = 6); relative risk 0.34, 95% confidence interval 0.13 to 0.88).Conclusion Despite a high frequency of planned pregnancies, resulting in overall good glycaemic control (early) in pregnancy and a high rate of adequate use of folic acid, maternal and perinatal complications were still increased in women with type 1 diabetes. Neonatal morbidity, especially hypoglycaemia, was also extremely high. Near optimal maternal glycaemic control (HbA_1c ≤ 7.0%) apparently is not good enough.     

Prospective population based survey of outcome of pregnancy in diabetic women: results of the Northern Diabetic Pregnancy Audit, 1994.

OBJECTIVE: To determine whether the St Vincent declaration (1989) target of diabetic pregnancy outcome approximating non-diabetic pregnancy outcome in near to being achieved. DESIGN: Prospective collection of population based information on pregnancies in women with diabetes from all participating hospitals. SETTING: District general and teaching hospitals of the former Northern region. SUBJECTS: 111 diabetic women booking with pregnancy during 1 January to 31 December 1994. MAIN OUTCOME MEASURES: Diabetic control, perinatal mortality rate, fetal abnormality rate. RESULTS: The perinatal mortality rate was 48/1000 for diabetic pregnancies compared with 8.9/1000 for the background population (odds ratio 5.38; 95% confidence interval 2.27 to 12.70) and the neonatal mortality rate was 59/1000 compared with 3.9/1000 (15.0; 6.77 to 33.10). Two late neonatal deaths were due to congenital heart defects. Six per cent of all fetal losses (6/109 cases) were due to major malformations. The congenital malformation rate was 83/1000 compared with 21.3/1000 (3.76; 2.00 to 7.06) in the background population. CONCLUSIONS: Diabetic pregnancy remains a high risk state with perinatal mortality and fetal malformation rates much higher than in the background population.     

Medications used to treat nausea and vomiting of pregnancy and the risk of selected birth defects

BACKGROUND Nausea and vomiting of pregnancy (NVP) occurs in up to 80% of pregnant women, but its association with birth outcomes is not clear. Several medications are used for the treatment of NVP; however, data are limited on their possible associations with birth defects. METHODS Using data from the National Birth Defects Prevention Study (NBDPS)—a multi‐site, population‐based, case‐control study—we examined whether NVP or its treatment was associated with the most common noncardiac defects in the NBDPS (nonsyndromic cleft lip with or without cleft palate [CL/P], cleft palate alone [CP], neural tube defects, and hypospadias) compared with randomly selected nonmalformed live births. RESULTS Among the 4524 cases and 5859 controls included in this study, 67.1% reported first‐trimester NVP, and 15.4% of them reported using at least one agent for NVP. Nausea and vomiting of pregnancy was not associated with CP or neural tube defects, but modest risk reductions were observed for CL/P (adjusted odds ratio [aOR] = 0.87; 95% confidence interval [CI], 0.77–0.98) and hypospadias (aOR = 0.84; 95% CI, 0.72–0.98). Regarding treatments for NVP in the first trimester, the following adjusted associations were observed with an increased risk: proton pump inhibitors and hypospadias (aOR = 4.36; 95% CI, 1.21–15.81), steroids and hypospadias (aOR = 2.87; 95% CI, 1.03–7.97), and ondansetron and CP (aOR = 2.37; 95% CI, 1.18–4.76), whereas antacids were associated with a reduced risk for CL/P (aOR = 0.58; 95% CI, 0.38–0.89). CONCLUSIONS NVP was not observed to be associated with an increased risk of birth defects; however, possible risks related to three treatments (i.e., proton pump inhibitors, steroids and ondansetron), which could be chance findings, warrant further investigation. Birth Defects Research (Part A) 2012. © 2011 Wiley Periodicals, Inc.     

Birth defects after maternal exposure to corticosteroids: prospective cohort study and meta-analysis of epidemiological studies

BACKGROUND: Corticosteroids are first-line drugs for the treatment of a variety of conditions in women of childbearing age. Information regarding human pregnancy outcome with corticosteroids is limited. METHODS: We collected prospectively and followed up 184 women exposed to prednisone in pregnancy and 188 pregnant women who were counseled by Motherisk for nonteratogenic exposure. The primary outcome was the rate of major birth defects. A meta-analysis of all epidemiological studies was conducted. The Mantel-Haenszel summary odds ratio was calculated for the pooled studies with 95% confidence intervals. A cumulative summary odds ratio was also calculated by combining studies in chronological order. Chi-squared for homogeneity was determined to establish the comparability of the studies. RESULTS: In our prospective study, there was no statistical difference in the rate of major anomalies between the corticosteroid-exposed and control groups. In the meta-analysis, the Mantel-Haenszel summary odds ratio for major malformations with all cohort studies was 1.45 [95% CI 0.80, 2.60] and 3.03 [95% CI 1.08, 8. 54] when Heinonen et al. ('77) was removed. This suggests a marginally increased risk of major malformations after first-trimester exposure to corticosteroids. In addition, summary odds ratio for case-control studies examining oral clefts was significant (3.35 [95% CI 1.97, 5.69]). CONCLUSIONS: Although prednisone does not represent a major teratogenic risk in humans at therapeutic doses, it does increase by an order of 3.4-fold the risk of oral cleft, which is consistent with the existing animal studies.