Unplanned hospital admission and emergency attendance following two-dose of BNT162B2 among patients with inflammatory bowel disease: A population-based cohort study

Abstract

Background: Real-world population-based safety data about the COVID-19 mRNA vaccine is lacking in patients with various immunocompromised conditions, including inflammatory bowel disease (IBD). Aim: To determine the incidence rates of unplanned IBD-related hospital admission and all-cause emergency attendance following BNT162B2 vaccination in IBD patients. Methods: Through the Government commissioned, territory-wide active COVID-19 safety surveillance, we linked population-level vaccination records and health outcome data, between March 10 (1st day of vaccination program) and September 30, 2021, to assess the association between two-dose of BNT162b2 and unplanned IBD-related hospitalization and all-cause emergency attendance. We used inverse probability treatment weighting-based cohort study design to balance the baseline characteristics between vaccinated and unvaccinated IBD patients. Poisson regression model was fitted to estimate the adjusted incidence rate ratio (IRR) of unplanned IBD hospital admission and 28-day emergency room attendance following the vaccination, using the unvaccinated group as the reference. Results: Among more than 4.1 million citizens with successful vaccine and health record-linkage, we identified 941 IBD patients (age: 46.0 ± 15.0 years, male: 64.2%) who completed two-dose of BNT162b2 and 1196 age-sex matched unvaccinated IBD patients as control (age: 49.3 ± 18.3 years, male: 58.9%). After inverse propensity weighting, all baseline demographic and clinical characteristics were well balanced (standard mean difference < 0.1; Table 1). During a median follow-up of 59-60 days (181.2 person-years for BNT162b2 group; 253.6 person-years for the unvaccinated group), there was no significant difference in the risk of unplanned IBD-related hospital admission [3.31 versus 5.13 per 100 person-years, IRR: 0.75 (0.38, 1.47)] and 28-day all-cause emergency room attendance [39.1 vs 47.5 per 100 person-years, IRR: 1.08 (0.76-1.53)] between BNT162b2 recipients and unvaccinated individuals. Series of stratified analyses, including patients with Crohn’s disease (N= 378) or ulcerative colitis (N=553), who received immunosuppressants (N=454) or biologics (N=192), all showed that receiving two-dose of BNT162b2 vaccine was not associated with a higher risk of unplanned IBD-admission and 28-day emergency attendance when compared to their counterparts without vaccination (Figure 1). Conclusion: Results from this population-based study showed no increase in risk of unplanned IBD-related hospitalization and all-cause emergency attendance following two-dose of BNT162b2 Covid-19 vaccination in patients with IBD. This observation potentially reassures the medium-term safety of mRNA vaccine in patients with IBD, although there is still possible self-selection bias in receiving the vaccine.