Search in the Abstract Database

Abstracts Search 2019

P438 Lupus-like reactions in patients with inflammatory bowel disease treated with anti-TNFs are rare but insidious adverse events: data from a large single-centre cohort

F. S. Macaluso*1, C. Sapienza1, M. Ventimiglia1, M. Cottone1, A. Orlando1

1IBD Unit, ‘Villa Sofia-Cervello’ Hospital, Palermo, Italy


The occurrence of lupus-like reactions (LLRs) may complicate the management of patients with inflammatory bowel disease (IBD) treated with anti-TNFs. However, very few data on the incidence, predictors, and clinical outcomes of LLRs have been reported. We aimed to describe all these features in a large cohort of IBD patients treated with anti-TNF drugs


All records of consecutive patients who started a treatment with an anti-TNF from January 2006 to June 2018 were retrospectively reviewed. Patients were defined as having LLR by the presence of immunologic abnormalities (positivity for ANA and/or anti-ds-DNA), along with clinical features that included at least two of the following: arthralgia, fatigue, fever, cutaneous manifestations, or serositis, which had a clear temporal association with exposure to the anti-TNFs, and resolved without recurrence once the drug was discontinued. Univariable and multiple Cox proportional hazard models were used to estimate the association between all variables at baseline and occurrence of LLRs.


In total, 760 patients (1059 total treatments with anti-TNFs) were included. Participants contributed a total of 2863.5 person-years of follow-up, during which 16 cases of LLRs (2.1% of patients) were reported, with an incidence rate of 5.6 per 1000 person-years. Female gender and being former smokers were more prevalent in the LLR group (75.0% vs. 44.1%, p = 0.02; and 18.8% vs. 5.4%, p = 0.037, respectively), with a hazard ratio of 3.86 (95% CI: 1.21–12.38; p = 0.023) and 4.42 (95% CI: 1.20–16.24; p = 0.025), respectively, at Cox regression analysis adjusted for possible confounders. LLRs occurred after a mean of 12.0 ± 9.7 months of therapy with anti-TNFs. Antinuclear antibodies were universally positive, and 10 out 16 (62.5%) patients had also anti-ds-DNA. Arthropathy was the most frequent symptom (87.5%), followed by fatigue (81.2%), and fever (31.2%). Three cases presented with a concomitant autoimmune hepatitis-like syndrome. The diagnosis of LLR was further confirmed by a re-challenge with the culprit agent in half of the cases. All LLRs resolved following discontinuation of the drug after a mean of 8.1 ± 4.2 weeks, even if 10 patients required corticosteroids for the control of symptoms. Five patients (31.2%) were switched to a second anti-TNFs, and one of them developed a second LLR.


In this very large cohort of patients treated with anti-TNFs, LLRs were rare adverse events, more common in women and former smokers. Clinical features are non-specific and insidious. All LLRs resolved following discontinuation of the drug, but the use of corticosteroids was required in most of the cases.