Abstract Background Corticosteroids have been fundamental in the management of ulcerative colitis (UC) flares, yet many patients do not respond or become corticosteroid dependent. The evolution of this clinical response across successive treatment courses remains underexplored. In this study we aimed to analyze the dynamics of the corticosteroid response over time and identify associated factors. Methods We conducted a retrospective cohort study including adult patients with UC diagnosed between 1975 and 2023 and treated with ≥2 corticosteroid courses. The treatment response was defined by Partial Mayo Score (PMS) criteria. Transitions between response states across courses were modeled using a Markov approach to estimate probabilities and identify factors associated with response. Results Of 571 patients with UC, 201 (35.2%) had received ≥2 corticosteroid courses and were included in the study. Over a median follow-up of 9.8 years (IQR, 6.4-21.3 years) there were 899 corticosteroid courses (708 78.8% with prednisone). During follow-up, 89 patients (44.3%) experienced nonresponse at some point and 84 (41.8%) developed steroid dependence. The probability of maintaining a “non-responsive” status through corticosteroid courses was 37.8% (95% CI, 29.6%-46.8%), while complete response persistence was 79.5% (95% CI, 75.5%-82.9%). Intercurrent enteric infections were identified in 23 (11.4%) patients and were associated with corticosteroid non-response within that flare. Beclomethasone use was associated with non-response in the first cycle compared to prednisone (odds ratio OR, 8.70; 95% CI, 3.65-20.71). The presence of extraintestinal manifestations (OR, 5.34; 95% CI, 1.39-20.45) and greater disease extension (OR, 1.57; 95% CI, 1.05-2.35) were predictors of complete response to corticosteroids through corticosteroid courses. Conclusions Corticosteroid response in UC is a dynamic phenomenon. Over a third of non-responders remain unresponsive in subsequent courses. Extraintestinal manifestations, corticosteroid type, and greater disease extension are associated with an increased likelihood of clinical response to corticosteroids.
Gros et al. (Mon,) studied this question.