Abstract

The objective of this study was to examine the association between maternal asthma medication use during the periconceptional period and the risk of gastroschisis. In this case-control study, the authors used data on deliveries enrolled in the National Birth Defects Prevention Study (1997–2002) from eight collaborating centers. The cases included 381 infants with isolated gastroschisis, and the controls were 4,121 liveborn infants without malformations. The asthma medications used during the periconceptional period (1 month prepregnancy through the third pregnancy month) were divided into two groups, antiinflammatory and bronchodilator, and analyzed separately. Users of multiple asthma medications during the periconceptional period were also examined. Logistic regression was used to estimate odds ratios and 95% confidence intervals while controlling for maternal age, race/ethnicity, education, smoking, folic acid/vitamin use, and other vasoactive medications. Maternal bronchodilator use showed an elevated statistically significant risk of gastroschisis (adjusted odds ratio = 2.06, 95% confidence interval: 1.19, 3.59). No significant association was found between maternal use of asthma antiinflammatory medications and gastroschisis. Because information on maternal asthma status/severity was not available, the effects of disease on the risk of gastroschisis cannot be ruled out. Additional research is needed in determining whether a real risk exists and for guiding asthma treatment.

Asthma has been recognized as one of the most common obstructive pulmonary diseases and a serious medical complication in pregnancy. According to recent National Health Interview Survey data, 3.7–8.4 percent of pregnant women were affected by asthma, and there has been a twofold increase in asthma prevalence (1). Because of the potential direct and indirect consequences of uncontrolled asthma during pregnancy, benefit-risk comparisons or considerations often favor the use of asthma medications during pregnancy (2). Asthma medications are categorized into two general classes: long-term-controller and quick-relief medicines. Asthma medication use during pregnancy has been reported to be 2.83 percent (3). Animal studies demonstrated that approximately 10 percent of the circulating maternal albuterol (a bronchodilator) is transferred to the fetus, but the disposition in the fetal lungs is comparable to that in the maternal lungs (4). Pregnancy safety data for many asthma medications are still lacking because of ethical considerations in conducting randomized interventional studies in pregnant women. More importantly, as about half of pregnancies are unplanned, women with severe asthma or on asthma treatment who become pregnant could expose their fetuses to the effect of either medication or illness unknowingly.

Few studies have examined maternal asthma or asthma medication use and pregnancy outcomes, especially congenital malformations. A case study by Park et al. (5) and several epidemiologic studies (3, 6, 7) found signs of increased risks of cardiovascular defects, orofacial clefts, and anal atresia. However, other studies did not find associations between maternal asthma or asthma medication use and birth defects (8–10). Therefore, assessment of the safety of asthma medications during pregnancy and their impacts on adverse birth outcomes, including birth defects, is an important public health issue.

Gastroschisis, a paraumbilical defect of the abdominal wall resulting in the evisceration of abdominal contents into the amniotic cavity, is a rare birth defect (3.7 per 10,000 at birth) occurring before the 10th week of gestation (11). Studies have found that the prevalence has increased in many parts of the world in the last 2–3 decades (12, 13). Very rare familial clusters have been reported (11). Some epidemiologic studies suggest that environmental factors may play a role in the etiology of gastroschisis (14). Of the risk factors for gastroschisis, the most striking epidemiologic observation is young maternal age, especially those aged less than 20 years (11, 14). In addition, gastroschisis occurred more frequently among the offspring of smoking women, alcohol drinkers, recreational drug users, and women with low socioeconomic status (11, 14, 15). Werler et al. (16, 17) and Torfs et al. (18) found that gastroschisis was significantly associated with some vasoactive/vasoconstrictor medicines and decongestants used during the first trimester (16, 17). Although a potential biologic mechanism is unclear, a recent theory suggests that gastroschisis is a defect of body wall closure that could be affected by exposures in early pregnancy, including medications or immunologic events (19). Another theory suggests that gastroschisis may be due to a vascular disruption (14). It is known that asthma exacerbation and asthma medication can increase perinatal and fetal hypoxia or asphyxia (20), which may lead to vascular disruption because of the interrupted oxygen supply to the mother and fetus. Reduction of uterine blood flow and fetal hypoxia have been shown to be associated with malformations in animal experiments (21). The vasoconstrictive effects of some asthma medications, if used during early pregnancy, may increase the risk of vascular disruption and lead to gastroschisis (18).

Few studies have examined the impacts of maternal asthma medication use on congenital malformations, and of those that did, most did not examine the risks of different types of congenital malformations, especially gastroschisis. To address this important public health question, the current study assessed whether maternal use of asthma medications between 1 month preconception through the first trimester of pregnancy (periconceptional period) increased the risk of gastroschisis after controlling for other potential confounders.

MATERIALS AND METHODS

Design and study population

We analyzed data from the National Birth Defects Prevention Study, an ongoing, multicenter, population-based, case-control study in the United States, that investigates genetic and environmental risk factors of major birth defects. The methods used by the National Birth Defects Prevention Study have been described previously (22). All appropriate institutional review board approvals were obtained.

This study included cases of isolated gastroschisis and control births from eight participating states (Arkansas, California, Georgia, Iowa, Massachusetts, New Jersey, New York, and Texas, covering 482,000 births per year) with estimated dates of delivery from October 1, 1997, through December 31, 2002. Gastroschisis cases with multiple defects other than intestinal atresia were excluded, because the etiologies for these cases may be different from those for isolated cases. Cases included livebirths, stillbirths of 20 or more weeks' gestation or greater than 500 g, or elective terminations. The case infants were identified from each center's birth defects surveillance system, and they were reviewed by clinical geneticists using specific case criteria, including standardized definitions of defects, and required confirmatory diagnostic procedures (23). The controls consisted of nonmalformed liveborn infants, randomly selected from either birth certificates or from birth hospitals. The gastroschisis case:control ratio was 1:11. We excluded infants who were adopted or in foster care and those mothers who did not speak English or Spanish.

Data collection

Mothers were interviewed in English or Spanish by female staff interviewers using a computer-assisted telephone interview. A standard procedure was used for study recruitment (22). The information collected from maternal interviews included maternal health status, such as chronic diseases; medication used; pregnancy history and complications; maternal diet; vitamin use; caffeine, tobacco, alcohol, or illicit drug use; home and work exposures; and demographics. To improve recall and better define exposure, participants were asked for the specific names of prescribed medications and over-the-counter medications used from 3 months before pregnancy through the end of the pregnancy, as well as the health conditions and any drugs used due to that specific condition. A pregnancy calendar was used to allow mothers to respond to the questions regarding timing and frequency of exposure in the way that they remembered best: by date, month, or trimester of pregnancy. The interview was required to be completed after 6 weeks but before 2 years of the estimated date of delivery. This time window for interviews was chosen to respect mothers of newborns, especially those of children with defects, and to allow sufficient time to ascertain all cases with eligible defects not readily identified or diagnosed at birth.

Exposure variables

The Boston University Slone Center Drug Dictionary, a computerized database of prescription and nonprescription drugs that links drug products to their generic ingredients, was used to identify and classify asthma medication as a bronchodilator or an antiinflammatory agent. Maternal reports of medication use were matched to the active ingredient in the database to identify subjects who used asthma medications. A woman who reported using asthma medication at least once during the periconceptional period, that is, between 1 month prior to conception and the end of the third month of pregnancy, was defined as exposed in this study. In addition, women who used an asthma medication “as needed” were examined closely to detect any misclassification; that is, women who used asthma medication as needed but whose use period was unknown were excluded from the main analysis (0.87 percent). The category of asthma medication and the number of asthma medications used during the periconceptional period were used to further describe exposure. The method of administration (oral vs. inhaled) was not available for all participants; therefore, it was not considered. Analysis of the asthma medications used was divided into three main groups: antiinflammatories, bronchodilators, and both antiinflammatories and bronchodilators. Each group was considered separately as an exposure, because the mechanism of action of the drugs can be different. Mothers who started taking asthma medications during specific time windows, including the periconceptional period and 4th–9th months, were compared with nonusers. According to the distribution of the data, the number of maternal asthma medications used during the periconceptional period was also divided into three groups: no medication, one medication, and multiple medications. Multiple medication use and the timing of starting medications during the periconceptional period may indicate not only the possible intensity of exposure but also how well the asthma was controlled. These two variables might serve as indicators of asthma severity.

Statistical analysis and potential confounders

General crude associations between maternal characteristics and gastroschisis were evaluated by use of chi-square tests. The variables that were significantly associated with both gastroschisis and asthma medications in the crude analysis were considered as potential confounders and included in the multivariate analysis. Unconditional logistic regression was used to compute the adjusted odds ratios to assess the effect of maternal asthma medication use on gastroschisis while controlling for other confounding factors, including maternal age, ethnicity, educational level, smoking status, any other vasoactive medicines (i.e., aspirin, ibuprofen, acetaminophen, amoxicillin, pseudoephedrine, phenylpropanolamine, methylenedioxymethamphetamine), and folic acid use during the periconceptional period. Smoking, any vasoactive medications, and folic acid use are yes/no variables in the model. Separate logistic regression models were conducted for the bronchodilator and antiinflammatory categories. The relation between gastroschisis and number of maternal asthma medications used during the periconceptional period was examined to assess if there was any potential dose-response relation. Finally, the association between mothers who used asthma medications during different time windows, including the periconceptional period and beyond this period (4th–9th months of pregnancy), and gastroschisis was examined to assess if the effects found were biologically plausible.

RESULTS

After excluding 34 cases (8.2 percent) with multiple defects, one case who used terbutaline (a medication also prescribed to prevent preterm labor), and five participants who did not complete all the medication questions, we identified 381 isolated gastroschisis cases. The control group consisted of 4,121 nonmalformed liveborn infants after exclusion of one participant who used terbutaline and 21 participants with incomplete medication information. The response rate among cases (68 percent) was similar to that among controls (65 percent). In addition, there was no significant difference in the response rate between young women aged less than 20 years (64 percent) and the rest of the women (66 percent). Table 1 shows the results of a crude analysis of the association between maternal characteristics and gastroschisis. The results found that young maternal age at delivery (<20 years of age), low maternal body mass index (<18.5), primigravid women, Hispanic ethnicity, low maternal education, maternal smoking, cocaine or crack use, and vasoactive medication use were statistically positively associated with gastroschisis (all: p < 0.05). On the other hand, maternal folic acid use during pregnancy had a statistically significant protective effect (p < 0.001). Among these risk factors, young maternal age is the strongest risk factor.

TABLE 1.

Demographic and maternal factors among controls and gastroschisis cases, National Birth Defects Prevention Study, 1997–2002

 Total* Cases Controls 
 No. No. 
Maternal age (years) at delivery†      
    <20 612 149 39.11 463 11.23 
    ≥20 3,890 232 60.89 3,658 88.76 
Maternal body mass index†      
    <18.5 271 37 9.95 234 5.91 
    ≥18.5 4,058 335 90.05 3,723 94.09 
Parity (no. of children)†      
    0 1,893 243 63.78 1,650 40.07 
    ≥1 2,606 138 36.22 2,468 59.93 
Maternal ethnicity†      
    Black non-Hispanic 519 26 6.84 493 12.00 
    Hispanic 1,068 132 34.74 936 22.77 
    Others 238 25 6.58 213 5.18 
    White non-Hispanic 2,665 197 51.84 2,468 60.05 
Maternal level of education (years)†      
    ≤12 1,977 266 70.56 1,711 41.73 
    >12 2,500 111 29.44 2,389 58.27 
Maternal drinking habits‡      
    Yes 1,757 157 41.53 1,600 39.05 
    No 2,718 221 58.47 2,497 60.95 
Child's gender      
    Male 2,255 191 50.26 2,064 50.15 
    Female 2,241 189 49.74 2,052 49.85 
Maternal smoking habits†‡      
    Yes 934 134 35.17 800 19.45 
    No 3,561 247 64.83 3,314 80.55 
Folic acid-containing vitamin†‡      
    Yes 2,182 134 35.17 2,048 49.70 
    No 2,320 247 64.83 2,073 50.30 
Cocaine or crack use†‡      
    Yes 39 10 2.62 29 0.71 
    No 4,453 371 97.38 4,082 99.29 
Other vasoactive medications†‡§      
    Yes 1,100 118 30.97 982 23.83 
    No 3,402 263 69.03 3,139 76.17 
Fever in first trimester      
    Yes 532 45 11.81 487 11.82 
    No 3,970 336 88.19 3,634 88.18 
 Total* Cases Controls 
 No. No. 
Maternal age (years) at delivery†      
    <20 612 149 39.11 463 11.23 
    ≥20 3,890 232 60.89 3,658 88.76 
Maternal body mass index†      
    <18.5 271 37 9.95 234 5.91 
    ≥18.5 4,058 335 90.05 3,723 94.09 
Parity (no. of children)†      
    0 1,893 243 63.78 1,650 40.07 
    ≥1 2,606 138 36.22 2,468 59.93 
Maternal ethnicity†      
    Black non-Hispanic 519 26 6.84 493 12.00 
    Hispanic 1,068 132 34.74 936 22.77 
    Others 238 25 6.58 213 5.18 
    White non-Hispanic 2,665 197 51.84 2,468 60.05 
Maternal level of education (years)†      
    ≤12 1,977 266 70.56 1,711 41.73 
    >12 2,500 111 29.44 2,389 58.27 
Maternal drinking habits‡      
    Yes 1,757 157 41.53 1,600 39.05 
    No 2,718 221 58.47 2,497 60.95 
Child's gender      
    Male 2,255 191 50.26 2,064 50.15 
    Female 2,241 189 49.74 2,052 49.85 
Maternal smoking habits†‡      
    Yes 934 134 35.17 800 19.45 
    No 3,561 247 64.83 3,314 80.55 
Folic acid-containing vitamin†‡      
    Yes 2,182 134 35.17 2,048 49.70 
    No 2,320 247 64.83 2,073 50.30 
Cocaine or crack use†‡      
    Yes 39 10 2.62 29 0.71 
    No 4,453 371 97.38 4,082 99.29 
Other vasoactive medications†‡§      
    Yes 1,100 118 30.97 982 23.83 
    No 3,402 263 69.03 3,139 76.17 
Fever in first trimester      
    Yes 532 45 11.81 487 11.82 
    No 3,970 336 88.19 3,634 88.18 
*

Total may not add up because of missing observations.

Chi-square test: p < 0.05.

Yes/no during the critical period.

§

Use of any of the following vasoactive medications: aspirin, ibuprofen, acetaminophen, amoxicillin, pseudoephedrine, phenylpropanolamine, and methylenedioxymethamphetamine.

Among all pregnant women, 1.7 percent used antiinflammatories, and 3.3 percent used bronchodilators during the periconceptional period (data not shown). The 10 specific antiinflammatory drugs and seven bronchodilators used in the National Birth Defects Prevention Study during the periconceptional period of pregnancy are listed in table 2. The most commonly used antiinflammatories are fluticasone, prednisone, beclomethasone, and triamcicolone. Among bronchodilator users, about 71 percent of the cases and 89 percent of the controls used albuterol, followed by salmeterol during the periconceptional period.

TABLE 2.

Individual asthma medication reported during the periconceptional period of pregnancy among 381 cases and 4,121 controls, National Birth Defects Prevention Study, 1997–2002

Medication Cases Controls 
No. %* No. %* 
Antiinflammatories (N = 70)     
    Fluticasone 0.79 21 0.51 
    Prednisone 0.52 11 0.27 
    Beclomethasone 0.26 10 0.24 
    Triamcinolone 0.52 0.22 
    Cromolyn sodium 0.17 
    Budesonide 0.10 
    Montelukast 0.05 
    Zafirlukast 0.05 
    Prednisolone 0.02 
    Methylprednisolone 0.02 
Bronchodilators (N = 113)     
    Albuterol 12 3.17 85 2.08 
    Salmeterol 0.52 11 0.27 
    Pirbuterol 0.26 0.07 
    Ipratropium bromide 0.07 
    Ephedrine† 0.52 0.05 
    Epinephrine† 0.26 0.02 
    Theophylline† 0.26 0.02 
Medication Cases Controls 
No. %* No. %* 
Antiinflammatories (N = 70)     
    Fluticasone 0.79 21 0.51 
    Prednisone 0.52 11 0.27 
    Beclomethasone 0.26 10 0.24 
    Triamcinolone 0.52 0.22 
    Cromolyn sodium 0.17 
    Budesonide 0.10 
    Montelukast 0.05 
    Zafirlukast 0.05 
    Prednisolone 0.02 
    Methylprednisolone 0.02 
Bronchodilators (N = 113)     
    Albuterol 12 3.17 85 2.08 
    Salmeterol 0.52 11 0.27 
    Pirbuterol 0.26 0.07 
    Ipratropium bromide 0.07 
    Ephedrine† 0.52 0.05 
    Epinephrine† 0.26 0.02 
    Theophylline† 0.26 0.02 
*

Total percentage is not equal to 100% because some people took more than one medication.

Statistical difference between the cases and controls (chi-square test).

The adjusted risks of gastroschisis for maternal bronchodilator and antiinflammatory use are reported in table 3. We found that maternal bronchodilator use had a statistically significant elevated risk of gastroschisis (adjusted odds ratio (OR) = 2.06, 95 percent confidence interval (CI): 1.19, 3.59) after controlling for maternal age, ethnicity, education, smoking, folic acid use, and other vasoactive medication use. On the other hand, antiinflammatory use showed an increased but not statistically significant association with gastroschisis (adjusted OR = 2.00, 95 percent CI: 0.88, 4.51). Those women who used both antiinflammatories and bronchodilators had the highest risk, although not statistically significant (adjusted OR = 2.69, 95 percent CI: 0.87, 8.28). When mothers who used asthma medications outside the periconceptional period only were added to the reference group, we found similar results. These women might have asthma but did not take any medication during the periconceptional period because of less severity. When the analysis was stratified by maternal age (<20 years and ≥20 years), the odds ratios were similar in both strata (OR = 2.02 and 1.99, respectively). To estimate recall bias due to time to interview, we compared the risks of gastroschisis in various strata of time to interview and found that they did not differ greatly (0–<6 months: OR = 1.71, 95 percent CI: 0.60, 4.90; 6–<12 months: OR = 1.57, 95 percent CI: 0.81, 3.07; 12–<24 months: OR = 1.98, 95 percent CI: 1.17, 3.36).

TABLE 3.

Maternal use of bronchodilators, antiinflammatories, and both bronchodilators and antiinflammatories and the risk of gastroschisis, National Birth Defects Prevention Study, 1997–2002

 Cases Controls Crude odds ratio 95% confidence interval Adjusted odds ratio* 95% confidence interval 
 No. No. 
Bronchodilators         
    Yes 17 4.53 96 2.38 1.94 1.14, 3.29 2.06 1.19, 3.59 
    No 358 95.47 3,932 97.62 1.00 Referent 1.00 Referent 
Antiinflammatories         
    Yes 1.85 63 1.55 1.20 0.55, 2.64 2.00 0.88, 4.51 
    No 371 98.15 4,003 98.45 1.00 Referent 1.00 Referent 
Bronchodilators and antiinflammatories         
    Yes 1.11 19 0.49 2.30 0.78, 6.79 2.69 0.87, 8.28 
    No 355 98.89 3,873 99.52 1.00 Referent 1.00 Referent 
 Cases Controls Crude odds ratio 95% confidence interval Adjusted odds ratio* 95% confidence interval 
 No. No. 
Bronchodilators         
    Yes 17 4.53 96 2.38 1.94 1.14, 3.29 2.06 1.19, 3.59 
    No 358 95.47 3,932 97.62 1.00 Referent 1.00 Referent 
Antiinflammatories         
    Yes 1.85 63 1.55 1.20 0.55, 2.64 2.00 0.88, 4.51 
    No 371 98.15 4,003 98.45 1.00 Referent 1.00 Referent 
Bronchodilators and antiinflammatories         
    Yes 1.11 19 0.49 2.30 0.78, 6.79 2.69 0.87, 8.28 
    No 355 98.89 3,873 99.52 1.00 Referent 1.00 Referent 
*

Adjusted for maternal age, ethnicity, educational level, smoking, folic acid, and any of the following vasoactive medications: aspirin, ibuprofen, acetaminophen, amoxicillin, pseudoephedrine, phenylpropanolamine, and methylenedioxymethamphetamine.

Use of both antiinflammatories and bronchodilators is a clinical recommendation for severe or persistent asthma cases. Because we observed a positive association for bronchodilators, we continued to examine the effects of maternal bronchodilator use patterns by examining the effects on gastroschisis of the number of bronchodilators used and the specific gestational periods when bronchodilators were started. As demonstrated in table 4, women who had a child with gastroschisis were about two times more likely to use one or multiple bronchodilators (>70 percent are concurrent users) during the periconceptional period compared with their controls. The concurrent users of multibronchodilators had a higher risk (OR = 4.12) than did the sequential users (OR = 1.93), although it was not statistically significant. In addition, mothers of cases had a significantly higher rate of starting bronchodilators during the periconceptional period than did control mothers, and 77 percent of these women used these medications for 4 months or longer. On the other hand, there were no differences in terms of maternal bronchodilator use starting in the 4th–9th months of the pregnancy between cases and controls.

TABLE 4.

Evaluation of the number of bronchodilators used during the periconceptional period and early/late bronchodilator use on the risk of gastroschisis, National Birth Defects Prevention Study, 1997–2002

 Cases Controls Crude odds ratio 95% confidence interval Adjusted odds ratio* 95% confidence interval 
 No. No. 
No. of bronchodilators used during the periconceptional period (medications)         
    ≥2 0.53 10 0.25 2.20 0.48, 10.07 2.33 0.47, 11.46 
    1 15 4.00 86 2.14 1.92 1.10, 3.35 2.03 1.13, 3.66 
    0 358 95.47 3,932 97.62 1.00 Referent 1.00 Referent 
Period of starting bronchodilators         
    1 month before–3rd month 17 4.49 96 2.35 1.95 1.15, 3.31 2.07 1.19, 3.59 
    4th–9th months 1.06 44 1.08 1.00 0.36, 2.81 1.01 0.35, 2.93 
    Nonusers 358 94.46 3,946 96.57 1.00 Referent 1.00 Referent 
 Cases Controls Crude odds ratio 95% confidence interval Adjusted odds ratio* 95% confidence interval 
 No. No. 
No. of bronchodilators used during the periconceptional period (medications)         
    ≥2 0.53 10 0.25 2.20 0.48, 10.07 2.33 0.47, 11.46 
    1 15 4.00 86 2.14 1.92 1.10, 3.35 2.03 1.13, 3.66 
    0 358 95.47 3,932 97.62 1.00 Referent 1.00 Referent 
Period of starting bronchodilators         
    1 month before–3rd month 17 4.49 96 2.35 1.95 1.15, 3.31 2.07 1.19, 3.59 
    4th–9th months 1.06 44 1.08 1.00 0.36, 2.81 1.01 0.35, 2.93 
    Nonusers 358 94.46 3,946 96.57 1.00 Referent 1.00 Referent 
*

Adjusted for age, ethnicity, educational level, smoking, folic acid, and any of the following vasoactive medications: aspirin, ibuprofen, acetaminophen, amoxicillin, pseudoephedrine, phenylpropanolamine, methylenedioxymethamphetamine.

DISCUSSION

In the current study, women treated with bronchodilators during the periconceptional period had a risk of gastroschisis in their offspring approximately two times that of women not using these medicines. This positive association persisted when women taking these medications outside the periconceptional period were added to the reference group. Furthermore, the positive association found in this study was observed only for women who started use of asthma medication during the periconceptional period when gastroschisis occurs (11), but not in those who started during late pregnancy, which is biologically plausible. Our finding is consistent with those of some previous studies (3–7). In an animal study, albuterol with doses equal to the maximum recommended daily inhalation dose for adults was found to have teratogenic effects (4). Park et al. (5) reported three cases of cardiovascular anomalies after the mothers took asthma medication (e.g., theophylline, albuterol) during the first trimester of the pregnancy. Demissie et al. (6) found that maternal asthma was associated with an increase in congenital anomalies. Kallen and Olausson (3) followed a cohort of pregnant women in Sweden and found a weak risk between antiasthmatic drugs during pregnancy and all congenital malformations. In contrast, three other studies found no significant associations between major congenital malformations and first trimester or any use of beta-agonists, theophylline, cromolyn, corticosteroids, antihistamines, or decongestants (8–10). None of the prior studies had sufficient power to examine specific types of birth defects.

The current study found that the most frequently used bronchodilator was albuterol (3.17 percent for cases and 2.08 percent for controls), which is consistent with the Kaiser Permanente and Michigan Medicaid studies (3, 6). Albuterol is a long-acting beta-agonist, which is the only beta-agonist currently available as a metered dose inhaler for which there are published human data on use during pregnancy. For persistent asthma, an aggressive asthma management plan, including use of long-term controller medications, has been recommended in different position statements from the National Institutes of Health since budesonide, an inhaled corticosteroid, was recently changed from pregnancy category C to B by the Federal Drug Administration (24). As such, risk-benefit considerations may favor salmeterol use during pregnancy in those women whose asthma is not controlled by medium-dose inhaled steroids (25).

This study found no significant association between maternal antiinflammatory (mainly corticosteroid) use and gastroschisis, although the odds ratio was similar to that found for bronchodilators. Inhaled corticosteroids are the most effective antiinflammatory medications to reduce airway inflammation and are recommended as the main controller therapy for all patients with persistent asthma. No studies to date, including studies of large birth registries, have reported that inhaled corticosteroid use was related to any increase in congenital malformations (26), which is consistent with our findings.

One of the critical questions raised by the results of this study is whether maternal asthma rather than maternal use of bronchodilators is associated with the increased risk of gastroschisis in this analysis. Because information on maternal asthma status was not available in the current study, we could not distinguish between these two important factors. However, this study found that women with some indicators of severe/persistent asthma, such as women who used both bronchodilators and antiinflammatories, had a higher risk of gastroschisis compared with the women who used only one type of asthma medication. We also found that those women with indicators of uncontrolled asthma, such as use of multiple bronchodilators during the periconceptional period to relieve acute symptoms, were at higher risk of gastroschisis in their offspring than were those who used one medication. This hypothesis is further supported by our findings that the positive association persisted when we compared the women taking asthma medications during the periconceptional period versus nonusers plus those who took medicines outside this period, which may indicate less severe asthma. Another important finding is that, among the women who used bronchodilators during pregnancy, 77 percent of them used bronchodilators throughout the entire periconceptional period. The requirement for regular use of reliever treatment with short-acting beta-agonists implies that an activated inflammatory process exists and may require an inhaled corticosteroid or an increased dose of one (25). It also indicates that more than half of the asthmatic women in the current study may not have controlled their asthma well during their pregnancies. Our findings suggest that poorly controlled or severe asthma may increase the risk of gastroschisis. Asthma frequently complicates pregnancy, and the course of asthma may also be adversely altered by gestation, which may place the mother and/or fetus at risk. To distinguish whether asthma symptoms or asthma therapy influenced reproductive outcomes, several studies (8, 26) found that increasing severity and symptoms of asthma rather than medication type were associated with fetal growth restriction. These studies suggest that poor asthma management or control is an important remedial risk factor of fetal growth. Potential mechanisms postulated to explain the adverse neonatal effect of maternal asthma or asthma medication use include hypoxia or other physiologic consequences of poorly controlled asthma.

This is the first study to evaluate the risk of gastroschisis among maternal asthma medication users. Because most teratogenic agents tend to affect the development of specific types of birth defects rather than birth defects overall, it is important to assess the use of specific asthma medications in relation to specific defects (27). Unlike other studies that have looked at birth defects combined, the National Birth Defects Prevention Study has a large sample size and systematic case review to possibly investigate the effect of major classes of asthma medications on gastroschisis. We were also able to control for multiple potential confounders due to a detailed interview. Population-based selection of cases and controls provided the potential of reducing selection bias. Particularly, maternal characteristics were similar while comparing control participants with all livebirths at each study center. Also, the participation rate of cases was similar to that of controls. Because gastroschisis is strongly related to young maternal age, a possible selection bias may have been created if young mothers were more mobile and less likely to respond than older mothers were. However, similar response rates by maternal age suggest that any bias due to this reason is less likely. Our analysis was restricted to isolated gastroschisis, since gastroschisis with multiple defects or syndromes may have different etiologies from those of isolated cases. However, information regarding asthma medication use relied on mothers' ability to recall exposures. Therefore, recall bias may exist since mothers of affected children may be more likely to report their exposures than mothers of controls. We did, however, include questions about specific types of medicines, as well as the timing and dosage of these exposures by using a pregnancy calendar to minimize such bias. Moreover, the risk of gastroschisis in various strata of time to interview was similar, and therefore recall bias due to length of time to interview was less likely. Because of the rarity of gastroschisis in older mothers and potential uneven distributions of cases and controls by age, we also performed stratified analysis by age. Similar risks between younger and older women minimized the possibility of a bias due to maternal age on our findings. Another critical limitation is lack of information on asthma status, which makes it difficult to distinguish whether the potential effect was due to asthma medication or disease status/severity.

In conclusion, the present study suggests that maternal use of bronchodilators during the periconceptional period of pregnancy may be associated with gastroschisis. However, we cannot rule out the effect of asthma severity or uncontrolled asthma during pregnancy on the risk of gastroschisis among the offspring of pregnant asthmatics. Further studies with more complete information regarding asthma status, asthma severity, and asthma treatment/management are required to clarify whether this adverse effect is due to maternal asthma severity or use of bronchodilators.

Abbreviations

    Abbreviations
  • CI

    confidence interval

  • OR

    odds ratio

This research was supported by a cooperative agreement from the Centers for Disease Control and Prevention (U50CCU213244). Coding of drug information in the study utilized the Slone Epidemiology Center at Boston University.

The authors acknowledge Dr. Marilyn Browne, Dr. Kimberly Campbell, and other National Birth Defects Prevention Study investigators for their valuable contributions.

The findings and conclusions of this report are those of the authors and do not necessarily represent the views of the Centers for Disease Control and Prevention.

Conflict of interest: none declared.

References

1.
Kwon
HL
Belanger
K
Bracken
MB
Asthma prevalence among pregnant and childbearing-aged women in the United States: estimates from national health surveys
Ann Epidemiol
 , 
2003
, vol. 
13
 (pg. 
317
-
24
)
2.
NAEPP Expert Panel report
Managing asthma during pregnancy: recommendations for pharmacologic treatment— 2004 update
J Allergy Clin Immunol
 , 
2005
, vol. 
115
 (pg. 
34
-
46
)
3.
Kallen
B
Olausson
PO
Use of anti-asthmatic drugs during pregnancy. 3. Congenital malformations in the infants
Eur J Clin Pharmacol
 , 
2007
, vol. 
63
 (pg. 
383
-
8
)
4.
Proventil, brand of albuterol, USP
Physicians' desk reference
 , 
2006
53rd ed
Montvale, NJ
Medical Economics Company
(pg. 
3067
-
9
)
5.
Park
JM
Schmer
V
Myers
TL
Cardiovascular anomalies associated with prenatal exposure to theophylline
South Med J
 , 
1990
, vol. 
83
 (pg. 
1487
-
8
)
6.
Demissie
K
Breckenridge
MB
Rhoads
GG
Infant and maternal outcomes in the pregnancies of asthmatic women
Am J Respir Crit Care Med
 , 
1998
, vol. 
158
 (pg. 
1091
-
5
)
7.
Rosa
F
Databases in the assessment of the effects of drugs during pregnancy
J Allergy Clin Immunol
 , 
1999
, vol. 
103
 
(2 pt 2)
(pg. 
S360
-
1
)
8.
Namazy
JA
Schatz
M
Update in the treatment of asthma during pregnancy
Clin Rev Allergy Immunol
 , 
2004
, vol. 
26
 (pg. 
139
-
47
)
9.
Briggs
GG
Freeman
RA
Yaffe
SJ
Michigan Medicaid study. Drugs in pregnancy and lactation. 4th ed
1994
Baltimore, MD
Williams and Wilkins
10.
Liu
S
Wen
SW
Demissie
K
, et al.  . 
Maternal asthma and pregnancy outcomes: a retrospective cohort study
Am J Obstet Gynecol
 , 
2001
, vol. 
184
 (pg. 
90
-
6
)
11.
Hunter
A
Soothill
P
Gastroschisis—an overview
Prenat Diagn
 , 
2002
, vol. 
22
 (pg. 
869
-
73
)
12.
Rankin
J
Dillon
E
Wright
C
Congenital anterior abdominal wall defects in the north of England, 1986 –1996: occurrence and outcome
Prenat Diagn
 , 
1996
, vol. 
19
 (pg. 
662
-
8
)
13.
Suita
S
Okamatsu
T
Yamamoto
T
, et al.  . 
Changing profile of abdominal wall defects in Japan: results of a national survey
J Pediatr Surg
 , 
2000
, vol. 
35
 (pg. 
66
-
71
)
14.
Werler
MM
Mitchell
AA
Shapiro
S
Demographic, reproductive, medical, and environmental factors in relation to gastroschisis
Teratology
 , 
1992
, vol. 
45
 (pg. 
353
-
60
)
15.
Goldbaum
G
Daling
J
Milham
S
Risk factors for gastroschisis
Teratology
 , 
1990
, vol. 
42
 (pg. 
397
-
403
)
16.
Werler
MM
Mitchell
AA
Shapiro
S
First trimester maternal medication use in relation to gastroschisis
Teratology
 , 
1992
, vol. 
45
 (pg. 
361
-
7
)
17.
Werler
MM
Sheehan
JE
Mitchell
AA
Association of vasoconstrictive exposures with risks of gastroschisis and small intestinal atresia
Epidemiology
 , 
2003
, vol. 
14
 (pg. 
349
-
54
)
18.
Torfs
CP
Katz
EA
Bateson
TF
, et al.  . 
Maternal medications and environmental exposures as risk factors for gastroschisis
Teratology
 , 
1996
, vol. 
54
 (pg. 
84
-
92
)
19.
Feldkamp
ML
Carey
JC
Sadler
TW
Development of gastroschisis: review of hypotheses, a novel hypothesis, and implications for research
Am J Med Genet A
 , 
2007
, vol. 
143
 (pg. 
639
-
52
)
20.
Johnson
TRB
Niebyl
JR
Gabbe
SG
Niebyl
JR
Simpson
JL
Preconception and prenatal care: part of the continuum
Obstetrics: normal and problem pregnancies
 , 
2002
Philadelphia, PA
Churchill Livingstone
(pg. 
1038
-
46
)
21.
Danielsson
BR
Skold
AC
Johansson
A
, et al.  . 
Teratogenicity by the hERG potassium channel blocking drug almokalant: use of hypoxia marker gives evidence for a hypoxia-related mechanism mediated via embryonic arrhythmia
Toxicol Appl Pharmacol
 , 
2003
, vol. 
193
 (pg. 
168
-
76
)
22.
Yoon
PW
Rasmussen
SA
Lynberg
MC
, et al.  . 
The National Birth Defects Prevention Study
Public Health Rep
 , 
2001
, vol. 
116
 
suppl 1
(pg. 
32
-
40
)
23.
Rasmussen
SA
Olney
RS
Holmes
LB
, et al.  . 
Guidelines for case classification for the National Birth Defects Prevention Study
Birth Defects Res A Clin Mol Teratol
 , 
2003
, vol. 
67
 (pg. 
193
-
201
)
24.
Osur
SL
The management of asthma and rhinitis during pregnancy
J Womens Health (Larchmt)
 , 
2005
, vol. 
14
 (pg. 
263
-
76
)
25.
Lipworth
BJ
Fortnightly review: modern drug treatment of chronic asthma
BMJ
 , 
1999
, vol. 
318
 (pg. 
380
-
4
)
26.
Bracken
MB
Triche
EW
Belanger
K
, et al.  . 
Asthma symptoms, severity, and drug therapy: a prospective study of effects on 2205 pregnancies
Obstet Gynecol
 , 
2003
, vol. 
102
 (pg. 
739
-
52
)
27.
Mitchell
AA
Systematic identification of drugs that cause birth defects: a new opportunity
N Engl J Med
 , 
2003
, vol. 
349
 (pg. 
2556
-
9
)