Elsevier

Joint Bone Spine

Volume 79, Issue 5, October 2012, Pages 457-463
Joint Bone Spine

Original article
Development of inflammatory bowel disease during anti-TNF-α therapy for inflammatory rheumatic disease. A nationwide series

https://doi.org/10.1016/j.jbspin.2011.10.001Get rights and content

Abstract

Objectives

To describe cases of new onset of inflammatory bowel disease (IBD) in patients with inflammatory rheumatic disease (IRD) receiving anti-TNF-α therapy.

Methods

A call for observations of such cases was sent to members of the French “Club rhumatismes et inflammation”. Only patients without intestinal symptoms before introduction of anti TNF-α agents were included.

Results

During a 2-year period, 16 patients were declared: nine men and seven women, mean age 41.5 ± 17.4 years, 12 patients with ankylosing spondylitis, one with rheumatoid arthritis, one with psoriatic arthritis and two juvenile idiopathic arthritis with enthesitis related arthritis. Overall, 14 patients received etanercept and two had infliximab. The meantime frame between onsets of anti-TNF--α drugs and development of IBD was 29.3 ± 20.1 months. According to endoscopic and histological findings, IBD was classified as typical Crohn's disease in eight cases, Crohn's-like disease in six cases, indeterminate in one case and definite ulcerative colitis in one case. For all cases, each TNF-α blocking agent was discontinued and replaced by another monoclonal anti TNF-α antibody. After a mean follow up period of 23.4 ± 19.5 months, outcome was favorable without recurrent or flaring IBD.

Conclusions

Paradoxical IBD may occur during anti TNF-α therapy for inflammatory rheumatic disease, mostly in patients with spondylarthropathies while receiving etanercept, at a frequency estimated to 0.15% in the French patients with spondylarthropathies exposed to TNF-α antagonists. The IBD mainly corresponded to Crohn's or Crohn's-like disease. On the contrary, new onset IBD is less frequently observed in other cases of IRD and with other TNF--α blockers.

Introduction

TNF-α antagonists have been shown to be highly effective in different inflammatory conditions, namely RA, AS, PsA, JIA, as well as psoriasis. Different anti-TNF-α drugs are currently available: infliximab and adalimumab are two anti-TNF-α monoclonal antibodies while etanercept is a p75 TNF-α receptor fusion protein. Another product is certolizumab pegol, which has been available for a few months in France and is a construct of anti-TNF-α Fab’ fragment linked to pegol residue. Anti-TNF-α monoclonal antibodies have also demonstrated their efficacy in IBD, and particularly in CD. In Europe and the USA, both infliximab and adalimumab are licensed for the treatment of CD, whereas etanercept was not shown to be effective in CD [1]. Infliximab is also effective in the treatment of UC.

Due to their high efficacy, TNF-α antagonists are used increasingly in IRD, psoriasis and IBD. Various side effects have been described, mainly infections. However, some unexpected and rare events that were not observed during the clinical trials of these drugs have emerged over the last few years: they are qualified as paradoxical because they occur after the initiation of the TNF-α blocking agent which is normally used to treat them [2]. These paradoxical events include new onset or exacerbation of psoriasis, development of aseptic granulomatous disease such as sarcoidosis [3], and appearance of uveitis or IBD [2].

This study was undertaken in order to better describe the clinical features of IBD occurring after the introduction of any TNF-α blocking agent in the French patient-population exposed to these drugs. We report here the largest series of patients under treatment with TNF-α antagonist developing new onset of IBD.

Section snippets

Methods

Between January 2009 and December 2010, all members of the “Club Rhumatismes and Inflammation (CRI)”, a section of the French Society of Rheumatology (Société française de rhumatologie [SFR]), including 2006 rheumatologists, internal medicine practitioners and pediatric rheumatologists, were regularly contacted via the CRI website system and by mail in order to collect information on all cases of patients with an IRD (RA, AS, PsA or JIA) who developed IBD while being treated with an anti-TNF-α

Patient demographics

Sixteen patients were reported, including nine men and seven women with a mean age of 41.5 ± (SD) 17.4 years (median 40.0). Three of these patients have been previously reported but individual details were not given [2]. The underlying IRD was AS in 12 cases, RA in one case, PsA in one case and JIA in two cases (including two patients with ERA). The mean disease duration was 13.7 ± 7.6 years (median: 12). Eight patients had extra-articular manifestations: uveitis in five cases and psoriasis in five

Discussion

We observed 16 cases of IBD after TNF-α blocker initiation, which represents the largest series nationwide. This series mainly included patients with AS or related spondylarthropathies (PsA or AS with juvenile onset or ERA) and only one patient with RA. There was a predominance of CD (50%) or CLD (43.7%) while UC was more rarely observed (6.25%). According to endoscopic findings, the extent of gastrointestinal involvement and the observance of pancolitis in combination with the occurrence of

Disclosure of interest

The authors declare that they have no conflicts of interest concerning this article.

Acknowledgements

The authors are indebted to Mrs. Frances Sheppard (CIC Biotherapy 506, Besançon, France) for help in preparing the manuscript and to Pr Jean Sibilia (Strasbourg, France), president of Le CRI, for critically reading the manuscript.

References (25)

  • E. Tsochatzis et al.

    First appearance of Crohn's disease following infliximab treatment for spondylarthropathy

    Scand J Gastrenterol

    (2007)
  • V. Yazisiz et al.

    Developemnt of Crohn's disease following anti-tumour necrosis factor therapy (etanercept)

    Colorectal Dis

    (2008)
  • Cited by (0)

    1

    Eric Toussirot is the main investigator who conceived and designed the study and drafted the manuscript. He contributed to the recruitment of the patients involved in the study and analyzed the results. Éric Houvenagel, Vincent Goëb, Damien Fouache, Antoine Martin, Philippe Le Dantec, Emmanuelle Dernis, Daniel Wendling, Thiphaine Ansemant, Jean-Marie Berthelot and Brigitte Bader-Meunier each contributed to the work as a clinical investigator by declaring one or more patient. Bernadette Kantelip provided histological results of two patients.

    2

    Le CRI: Club Rhumatismes and Inflammation, University Hospital St Antoine Paris, France.

    View full text