The Association Between Smoking and Microscopic Colitis: A Meta-Analysis

Introduction: Microscopic colitis (MC) is a chronic inflammatory disease of the colon which typically manifests clinically with watery diarrhea. MC is usually characterized by a normal gross endoscopic examination with abnormal histological findings and thereby can be divided into two subtypes; collagenous colitis (CC) and lymphocytic colitis (LC). There are several well-known risk factors for MC including multiple medications and celiac disease. It has been suggested that smoking may be associated with MC in a few smaller studies; however, there are conflicting results in the current literature and many of these studies have significant limitations. Our study aims to offer the first meta-analysis analyzing the association between MC and smoking. Methods: A systemic review was conducted in PUBMED, Embase, Pubmed Central and ScienceDirect databases from inception through April 2018 to identify the studies that evaluated the association between smoking and MC or either of its subtypes. Effect estimates from the individual studies were extracted and combined using the random-effect, generic inverse variance method of DerSimonian and Laird and a pooled odds ratio (OR) was calculated. Forest plots were generated, and publication bias was assessed for using conventional techniques. Results: Eight observation studies with a total of 1461 patients with MC were included in this study, 383 of whom were active smokers (26.2%). Current smoking was significantly associated with MC (OR 3.58, 95% CI, 2.51-5.11, Figure 1A), LC (OR 3.64, 95% CI, 2.46-5.38, Figure 1B), and CC (OR 4.43, 95% CI, 2.68-7.32, Figure 1C). Subgroup analysis was performed for both male and female sexes and a significant association with smoking was found for CC in men (OR 4.53, 95% CI, 1.59-12.85), CC in women (OR 3.27, 95% CI, 2.35-4.54), LC in women (OR 2.27, 95% CI, 1.27-4.06) and MC in women (OR 2.93, 95% CI, 2.09-4.10). We found no publication bias as assessed by the funnel plots and Egger’s regression asymmetry test.131 Figure 1. Forest plots: A. Smoking and microscopic colitis (all subtypes). B. Smoking and lymphocytic colitis. C. Smoking and collagenous colitis.Conclusion: Our meta-analysis found a statistically significant association between smoking and MC. Significant associated between smoking and the subtypes of LC and CC was also observed. Smoking may represent a modifiable risk factor for MC and appropriate clinical counselling for smoking cessation should be employed in these patients. Future high-quality prospective studies are needed to analyze whether this association is causal.