Smoking prevalence in inflammatory bowel disease and its effects on disease course and surgery

Abstract

Background: Smoking is an influential factor in the development and progression of inflammatory bowel diseases (IBD). It has a heightened prevalence in Crohn’s disease (CD) and detrimentally influences its natural history. Paradoxically, ulcerative colitis (UC) is predominantly a disease of non-smokers and ex-smokers, with varying reports of amelioration of disease in active smoking. Confirmation of the harmful effects of smoking in CD using local data may encourage patients to quit smoking. Aims: This study aims to determine the prevalence of smoking and its effects on disease progression and outcomes. Disease progression was defined by the need for surgical intervention and requirement to commence a thiopurine or biological agent. Methods: Ambulatory patients with IBD were recruited in Sydney from a metropolitan cohort diagnosed between 1942 and 2012. Case notes for each patient were reviewed to identify data on current smoking status, age at diagnosis, disease phenotype at diagnosis, thiopurine and biological agent use, need for surgery and number of hospital admissions. Chi square and the log-rank statistical tests were performed with cumulative surgery rate illustrated using the Kaplan-Meier method. Results: In total 803 IBD patients were recruited (374 CD, 393 UC and 36 IBD-undefined). Clear smoking status was available in 82.3%. The mean years of follow-up were 11.4 for CD patients and 12.2 for UC patients. Of the CD patients 12.6% were current smokers, 15.5% ex-smokers and 51.6% never smokers. There was no statistically significant difference in age or phenotype of disease at diagnosis between these three groups. Frequency of thiopurine and biological agent use and hospital admission between the groups was equivocal. Surgical management of disease was required more frequently in current and ex-smokers than in never smokers (P = 0.041) and ever smoking was associated with an increased cumulative probability of intestinal resection (Figure 1, P = 0.045).In UC 5.9% were current smokers, 13.2% ex-smokers and 65.4% never smokers. Ex-smokers had a later age of diagnosis than current smokers (P = 0.039) and never smokers (P < 0.001) with UC (45.7 vs. 37.0 vs. 34.9 years, respectively). Extent of disease at diagnosis was not influenced by smoking habit. Ever smoking was associated with a greater likelihood of thiopurine therapy induction requirement (P = 0.014). In UC there was no significant difference between groups in the need for biological agent use, colectomy or hospital admission.