BACKGROUND: Budesonide is accepted as first-choice therapy for microscopic colitis (MC); however, symptoms often recur and some patients may be dependent, intolerant, or even fail budesonide. We performed a systematic review and meta-analysis to determine the effectiveness of non-budesonide therapies (thiopurines, bismuth subsalicylate [BSS], bile acid sequestrants [BAS], loperamide and biologics) for MC suggested by international guidelines.

METHODS: We searched the CENTRAL, MEDLINE, and EMBASE databases from their inception to 18 April 2023 for the above-mentioned therapeutics in MC. We pooled the response and remission rates by medication using a random-effects model.

RESULTS: Twenty-five studies comprising 1475 patients were included in the meta-analysis. Treatment with BSS showed the highest response rate of 75% (95% confidence interval [CI] 0.65-0.83; I2 = 70.12%), with 50% achieving remission of symptoms (95% CI 0.35-0.65; I2 = 71.06%). Treatment with tumor necrosis factor (TNF) inhibitors (infliximab and adalimumab) demonstrated a response rate of 73% (95% CI 0.63-0.83; I2 = 0.00%), with a remission rate of 44% (95% CI 0.32-0.56; I2 = 0.00%). The response rate for those treated with vedolizumab was similar; 73% responded to treatment (95% CI 0.57-0.87; I2 = 35.93%), with a remission rate of 56% (95% CI 0.36-0.75; I2 = 46.30%). Loperamide was associated with response and remission rates of 62% (95% CI 0.43-0.80; I2 = 92.99%) and 14% (95% CI 0.07-0.25), respectively, whereas BAS use was associated with response and remission rates of 60% (95% CI 0.51-0.68; I2 = 61.65%) and 29% (95% CI 0.12-0.55), respectively. Finally, the outcomes for thiopurine use were 49% (95% CI 0.27-0.71; I2 = 81.45%) and 38% (95% CI 0.23-0.54; I2 = 50.05%), respectively DISCUSSION: The present systematic review and meta-analysis provides rates of effectiveness of non-budesonide therapies for MC based on available data in the field. Studies in the meta-analysis showed a large amount of heterogeneity due to the variability in assessing the clinical effects of intervention between the studies caused by differences in the definitions of response or remission rates between the studies included. This may likely result in overestimating the benefit of a treatment. Furthermore, the number of participants and drug dosages varied, and only a few studies applied disease-specific activity indices. Only one randomized controlled trial (RCT) was identified. All other 24 included studies were either case series or (retrospective) cohort studies, which complicated efforts to perform further sensitivity analyses to adjust for potential confounders and risk of bias. In addition, the overall evidence on the effect of these treatment options was judged as low, mostly due to comparability bias and the observational nature of the available studies, which limited statistically robust comparisons of rates of effectiveness of the different non-budesonide agents ranked against each other. However, our observational findings may inform clinicians regarding the most rational selection of non-budesonide therapies to patients with MC.

CLINICAL TRIALS REGISTRATION: PROSPERO protocol #CRD42020218649.