Background: 59-81% of patients given infliximab for Crohn's disease will respond. Although now in widespread use, little consensus exists regarding the optimal place in patient care. Recently developed guidelines have identified need for markers that predict response.
Aims: We aimed to identify markers of response to infliximab given for Crohn's disease.
Methods: Markers of response (defined at 4 weeks) were prospectively assessed in 74 infliximab-treated patients with Crohn's disease. Patients were followed-up to 1 year.
Results: Fifty-four of 74 (73%) patients responded. Univariate analysis identified that smokers were less likely to respond than non-smokers [P = 0.005, odds ratio (OR) 0.22]. Patients established on immunosuppression (P = 0.034, OR 7.31) and with isolated colonic disease (P = 0.042, OR 3.83) were more likely to respond. Multiple logistic regression confirmed smoking (P = 0.035, OR 0.24) and colonic disease (P = 0.035, OR 4.87) as independent markers of response. One-year relapse rates differed significantly between smokers and non-smokers (100% vs. 39.6%, P = 0.0026, relative risk 3.2) and between patients established on immunomodulators or not (58.0% vs. 92.8%, P = 0.0054, relative risk 2.6).
Conclusions: Smoking has a strong adverse effect on the response rates and maintenance of response to infliximab. Patients on immunomodulators have a more favourable short- and long-term response. These results have important implications for clinical practice.