Relevant infections in inflammatory bowel disease, their relationship with immunosuppressive therapy and their effects on disease mortality.

Background and Aims: There is controversy as to whether the risk of relevant infection in IBD is related to immunosuppressants or the disease itself. The aims of this study were to evaluate: [1] the life-long prevalence and types of relevant infections in patients with IBD related to immunosuppressive treatment, and [2] the relationship of both infection and patient comorbidity to mortality. Methods: Observational multicentre retrospective study of IBD patients that presented a relevant infection. For each case, four periods of infection exposure were analysed: P1: pre-IBD diagnosis, P2: from IBD diagnosis to immunosuppressant initiation, P3: during immunosuppressant therapy, and P4: after treatment withdrawal. Results: The life-long prevalence of relevant infection in the total cohort of patients [6914] was 3%, and 5% in immunosuppressed patients [4202]. 366 relevant infections were found in 212 patients [P1: 9, P2: 17, P3: 334, and P4: 6]. Differences between periods were significant [p < 0.0001]. The most frequent types of infection were respiratory, intestinal and urinary. The most frequent opportunistic infections were tuberculosis [prevalence: 2.6/1000] and herpes zoster [prevalence: 3.9/1000]. Herpes zoster infection was associated with thiopurines alone or in combination with anti-TNF in 75% of the cases, whereas tuberculosis was associated with anti-TNF in 94% of patients. The overall mortality was 4.2%. Infection-related mortality was 2.8% and it was not influenced by comorbidity. Conclusions: Relevant infections in IBD patients are rare and appear to be related to immunosuppression. Relevant infection is a major cause of death in IBD.