Objectives: Inflammatory bowel diseases (IBD) need long-term treatment, which can influence pregnancies in young women. Uncontrolled IBD is associated with poor pregnancy outcomes. Despite the labeling of Anti-tumor necrosis factor (TNF) antibodies (anti-TNFα) which indicates that their use is not recommended during pregnancy, anti-TNFα are increasingly being used during pregnancy and may expose women and their fetuses to treatment-related complications. Existing recommendations on the timing of treatment during pregnancy are inconsistent. We aimed to assess the safety of anti-TNFα treatment in pregnant women with IBD, and up to the first year of life for their children.
Methods: An exposed/non exposed retrospective cohort was conducted on the French national health system database SNIIRAM (Système National d'Information Inter-Régimes de l'Assurance Maladie). All IBD women who became pregnant between 2011 and 2014 were included. Women with concomitant diseases potentially treated with anti-TNFα were excluded. Anti-TNFα exposure (infliximab, adalimumab, golimumab or certolizumab pegol) during pregnancy was retrieved from the exhaustive prescription database in SNIIRAM. The main judgment criterion was a composite outcome of disease-, treatment- and pregnancy-related complications during pregnancy for the mother, and infections during the first year of life for children.
Results: We analyzed data from 11,275 pregnancies (8726 women with IBD), among which 1457 (12.9%) pregnancies were exposed to anti-TNFα, mainly infliximab or adalimumab, with 1313/7722 (17.0%) suffering from Crohn's disease and 144/3553 (4.1%) from ulcerative colitis. After adjusting for disease severity, steroid use, age, IBD type, and duration and concomitant 6-mercaptopurine use, anti-TNFα treatment was associated with a higher risk of overall maternal complications (adjusted Odds Ratio (aOR) = 1.49; 95% confidence interval (CI): 1.31-1.67) and infections (aOR = 1.31; 95% CI: 1.16-1.47). Maintaining anti-TNFα after 24 weeks did not increase the risk of maternal complication, but interrupting the anti-TNFα increased relapse risk. No increased risk for infection was found in children (aOR = 0.89; 95% CI: 0.76-1.05) born to mother exposed to anti-TNFα during pregnancy.
Conclusions: Anti-TNFα treatment during pregnancy increased the risk of maternal complications compared to unexposed; however, discontinuation before week 24 increased the risk of disease flare. There was no increased risk for children exposed to anti-TNFα up to 1 year of life.