Systematic review with meta-analysis: The effect of tobacco smoking on the natural history of ulcerative colitis

Abstract

Background Tobacco smoking is associated with a reduced risk of developing ulcerative colitis (UC). A high proportion of UC patients perceive a benefit in disease outcomes secondary to smoking. However, the effects of smoking on the natural history of UC are uncertain. Aim To conduct a systematic review and meta-analysis of the effects of tobacco smoking on the natural history of UC. Methods A search of MEDLINE, EMBASE and EMBASE classic was carried out (up to December 2015) to identify observational studies reporting data on smoking and rates of colectomy, flare of disease activity, proximal disease extension, and development of pouchitis following panproctocolectomy and ileal pouch-anal anastomosis in patients with UC. Dichotomous data were pooled to obtain odds ratios (ORs), with 95% confidence intervals (CIs). Results The search identified 16 eligible studies: five (2615 patients) studying colectomy; four (620 patients) reporting on flare of disease activity; four (687 patients) examining proximal disease extension and three (355 patients) assessing development of pouchitis. Compared with nonsmokers, the odds of colectomy (OR = 0.89; 95% CI 0.62-1.26), flare of disease activity (OR = 1.26; 95% CI 0.65-2.44), proximal extension of disease (OR = 0.57; 95% CI 0.20-1.66) or the development of pouchitis (OR = 0.57; 95% CI 0.21-1.53) were not significantly lower in smokers. Conclusions Smoking may not improve the natural history of ulcerative colitis. Given the health benefits of smoking cessation and the lack of clear benefit in ulcerative colitis, smoking cessation advice should be incorporated into guidance on the management of ulcerative colitis.