BACKGROUND: Four studies investigating the association between inhaled corticosteroid (ICS) use during pregnancy and perinatal mortality reported no significantly increased risk. These studies must be interpreted with caution because they have insufficient statistical power and a lack of adjustment for potential confounders. OBJECTIVES: We sought to evaluate whether asthmatic women exposed to ICSs during pregnancy are more at risk of perinatal mortality than asthmatic women not exposed. We also sought to estimate the risk of perinatal mortality as a function of the daily ICS dose taken during pregnancy. METHODS: From the linkage of 3 administrative databases from Quebec, a cohort including 13,004 single pregnancies from asthmatic women was constructed. We used a 2-stage sampling cohort design to obtain information on cigarette smoking from the medical charts of 487 mothers. The final estimates of the odds ratios (ORs) of perinatal mortality were estimated with a logistic regression model. RESULTS: The cohort was formed of 4,140 women who used greater than 0 to 250 μg/d ICS, 1,140 women who used greater than 250 μg/d ICS, and 7,724 nonusers of ICSs during pregnancy. Women exposed to ICSs (any dose) had a nonsignificant increased risk of perinatal mortality (OR, 1.07; 95% CI, 0.70-1.61). The use of greater than 250 μg/d ICS was associated with a nonsignificant 52% increased risk of perinatal mortality (OR, 1.52; 95% CI, 0.62-3.76). CONCLUSION: The risk of perinatal mortality was not found to be significantly associated with ICS use during pregnancy. The result associated with higher doses of ICSs is limited due to a lack of statistical power and a possibility of residual confounding by asthma severity and control.
BACKGROUND: Four studies investigating the association between inhaled corticosteroid (ICS) use during pregnancy and perinatal mortality reported no significantly increased risk. These studies must be interpreted with caution because they have insufficient statistical power and a lack of adjustment for potential confounders. OBJECTIVES: We sought to evaluate whether asthmatic women exposed to ICSs during pregnancy are more at risk of perinatal mortality than asthmatic women not exposed. We also sought to estimate the risk of perinatal mortality as a function of the daily ICS dose taken during pregnancy. METHODS: From the linkage of 3 administrative databases from Quebec, a cohort including 13,004 single pregnancies from asthmatic women was constructed. We used a 2-stage sampling cohort design to obtain information on cigarette smoking from the medical charts of 487 mothers. The final estimates of the odds ratios (ORs) of perinatal mortality were estimated with a logistic regression model. RESULTS: The cohort was formed of 4,140 women who used greater than 0 to 250 μg/d ICS, 1,140 women who used greater than 250 μg/d ICS, and 7,724 nonusers of ICSs during pregnancy. Women exposed to ICSs (any dose) had a nonsignificant increased risk of perinatal mortality (OR, 1.07; 95% CI, 0.70-1.61). The use of greater than 250 μg/d ICS was associated with a nonsignificant 52% increased risk of perinatal mortality (OR, 1.52; 95% CI, 0.62-3.76). CONCLUSION: The risk of perinatal mortality was not found to be significantly associated with ICS use during pregnancy. The result associated with higher doses of ICSs is limited due to a lack of statistical power and a possibility of residual confounding by asthma severity and control.