Opportunistic Infections With Anti-tumor Necrosis Factor-α Therapy in Inflammatory Bowel Disease

Meta-analysis of Randomized Controlled Trials

Alexander C Ford MBChB, MD, FRCP; Laurent Peyrin-Biroulet MD, PhD


Am J Gastroenterol. 2013;108(8):1268-1276. 

In This Article

Abstract and Introduction


Objectives: Several anti-tumor necrosis factor-α (TNFα) antibodies have demonstrated efficacy in Crohn's disease (CD) and ulcerative colitis (UC). These drugs carry the theoretical risk of opportunistic infection, but no systematic review and meta-analysis has examined this issue specifically.

Methods: MEDLINE, EMBASE, and the Cochrane central register of controlled trials were searched (through to November 2012). Randomized controlled trials (RCTs) recruiting adults with active or quiescent CD or UC comparing anti-TNFα therapy with placebo were eligible. Dichotomous data were pooled to obtain a relative risk (RR) of opportunistic infection, with a 95% confidence interval (CI). The number needed to harm (NNH) was estimated from the reciprocal of the risk difference from the meta-analysis.

Results: The search strategy identified 20,563 citations, 21 of which were eligible, reporting 22 separate RCTs with between 2 and 56 weeks of follow-up. In total, there were 39 (0.9%) opportunistic infections among 4,135 patients allocated to anti-TNFα therapy, compared with 9 (0.3%) among 2,919 assigned to placebo. Among patients receiving active therapy these included eight cases of Mycobacterium tuberculosis, eight cases of herpes simplex infection, six cases of oral or esophageal candidiasis, six cases of herpes zoster infection, two cases of varicella-zoster virus infection, two cases of cytomegalovirus or Epstein–Barr virus infection, and one case of Nocardia infection. The RR of developing an opportunistic infection was significantly higher with anti-TNFα therapy (2.05; 95% CI 1.10–3.85, NNH=500; 95% CI 200–1,567). The RR of tuberculosis infection was 2.52 (95% CI 0.62–10.21).

Conclusions: Anti-TNF therapy doubles the risk of opportunistic infections in inflammatory bowel disease patients. This underlines the importance of adherence to guidelines for their prevention and management.


Crohn's disease (CD) and ulcerative colitis (UC) are chronic inflammatory disorders of the gastrointestinal tract of unknown etiology. A proposed mediator of inflammation in inflammatory bowel disease (IBD) is the pro-inflammatory cytokine tumor necrosis factor-α (TNFα). Concentrations of TNFα are elevated in the stool,[1] mucosa,[2,3] and blood of patients with IBD.[4] Monoclonal antibodies to TNFα were developed in the 1990s, and pre-clinical studies demonstrated that they could treat the spontaneous colitis observed in the cotton-top tamarin effectively.[5] Chimeric, partly humanized, or fully humanized monoclonal antibodies, or antibody fragments, have been compared with placebo in randomized controlled trials (RCTs), with meta-analyses consistently demonstrating their efficacy in both CD and UC.[6–14]

Although anti-TNFα therapies have revolutionalized the treatment of IBD, there are theoretical, but serious, safety concerns, which include infections, such as tuberculosis, development of hematologic malignancy, including hepatosplenic T-cell lymphoma, and death.[15–17] However, a meta-analysis found no increased risk in overall infections in CD patients treated with anti-TNFα agents in placebo-controlled trials.[9] Similarly, two pooled analyses of infliximab- and adalimumab-treated patients from available RCTs also failed to identify any safety signal in terms of an increased risk of infection.[18,19] More recently, in a report of data from the TREAT™ registry, a 1.5-fold increase in the rates of serious infection was observed among patients treated with infliximab.[20] As TNFα plays a role in host defense, there is thought to be an increased risk of opportunistic infection with therapy,[21] but this was not specifically evaluated in any of these studies.

The latter type of adverse events are very rare, and individual RCTs of anti-TNFα therapies have not been powered with this end point in mind. As only case reports or case series are available at the present time, a precise estimate of the risk of such an event occurring in an IBD patient receiving treatment with anti-TNFα agents is difficult. However, given that there have now been numerous RCTs of anti-TNFα therapies in patients with IBD published, pooling data from all available placebo-controlled studies to estimate the likelihood of experiencing opportunistic infection with active therapy may resolve this uncertainty. We have therefore performed a systematic review and meta-analysis to examine this issue.