Results
Descriptive Data
There were 197 060 eligible children, of whom 101 069 (51.3%) were boys. A total of 2238 (1.1%) children had been exposed prenatally to PPIs, including 1238 (55.3%) exposed during the first trimester. Advanced maternal age at delivery, maternal asthma and high pregravid BMI were more frequent among PPI-exposed children than among unexposed children. So were maternal use of antibiotics during pregnancy (Table 1).
The maximum follow-up time was 14 years, with a median follow-up of 6.8 years. By the end of follow-up, 24 506 (12.4%) children were identified as having asthma. The median age at asthma diagnosis was similar among unexposed and exposed to PPIs: 1.5 and 1.6 years respectively. The prevalence rate of maternal PPI prescriptions over time increased continuously from 1996 to 2008 and no substantial change was observed in 2006 or 2007 (during which some PPIs became over-the-counter drugs).
Risk Estimates
Among the children without prenatal exposure to PPIs, the 2-year risk of asthma was 7.5% and the 10-year risk was 14.4%. The corresponding estimates were 12.2% and 21.1% among the children prenatally exposed to PPIs (Figure 1). We found similar estimates among those exposed prenatally to H2RAs (results not shown).
(Enlarge Image)
Figure 1.
Cumulative incidence of asthma in children according to prenatal exposure to proton pump inhibitors (PPIs).
After adjustment for potential risk factors, risk of asthma among children exposed prenatally to PPIs was 41% higher than among unexposed children (aIRR = 1.41, 95% CI: 1.27–1.56). Risk of asthma was also elevated among children exposed to more than 28 pills compared with 28 or fewer PPI pills (Table 2). When examining the risk of asthma according to trimester of exposure the aIRR was 1.46 (95% CI: 1.27–1.67) when exposed in first trimester and 1.34 (95% CI: 1.15–1.56) when exposed in second and/or third trimester.
In the analysis restricted to children aged 5 years and older (N = 129 888), 4966 children had asthma (3.8%) and the aIRR was 1.38 (95% CI: 1.00–1.89). The analyses restricted to children with a hospitalisation with asthma [10 632 children with asthma (43.4% of all asthma cases)] also yielded similar associations for PPI exposure as the main results (aIRR = 1.19, 95% CI: 1.01–1.40).
After adjusting for maternal BMI in the subset of women with available BMI data (N = 78 671), the crude IRR was 1.41 (95% CI: 1.23–1.62) and the aIRR was 1.25 (95% CI: 1.09–1.44) when exposed during gestation.
The estimates did not change substantially after varying the earliest PPI exposure between 0 and 60 days before pregnancy (results not shown).
Among the 197 060 children, 1605 (0.8%) had been exposed prenatally to H2RAs, as measured by maternal prescription dispensations. We found that 315 (19.6%) of the exposed children and 24 191 (12.4%) of the unexposed children had asthma, a result which was similar to the main results for PPIs (Table 2). The association did not vary according to the cumulative dose as measured by the number of pills.
Maternal use of PPIs in the year after giving birth, but not during pregnancy [2515 (1.3%) women], was associated with increased risk of asthma among offspring, with estimates similar to those in the main analysis. Corresponding results were observed for maternal postnatal use of H2RAs [695 (0.4%) women] Table 3.