Search in the Abstract Database

Search Abstracts 2013

* = Presenting author

P614. Smoking prevalence and its effects on disease course and surgery in a cohort of inflammatory bowel disease patients from Sydney, Australia

P. Lunney1, V. Kariyawasam2, K. Middleton1, R. Wang1, T. Huang3, C. Selinger2, J. Andrews4, R. Leong2, 1The University of Sydney, Sydney Medical School, Sydney, Australia, 2Concord Repatriation General Hospital, Gastroenterology and Liver Services, Sydney, Australia, 3University of New South Wales, Australia, 4Royal Adelaide Hospital, Adelaide, Australia


Smoking influences inflammatory bowel diseases (IBD) development and progression, being detrimental in Crohn's disease (CD) and ameliorating in ulcerative colitis (UC). Confirmation of the harmful effects of smoking in CD using unprecedented Australian data may encourage patients to quit smoking.

This study aims to determine smoking prevalence and its effects on disease course and outcomes in Australian IBD patients.


Case notes were reviewed for demographic data, smoking status, Montreal classification of disease phenotype, medication use, surgery and hospitalisations and analysed using Chi square and Kruskal–Wallis statistical tests.


A total of 1,198 patients (624 CD, 574 UC: 52% female, median age 43 years) were recruited, affording 13,300 patient-years of follow-up with a median follow-up of 9 years.

In CD 19% were smokers and 25% ex-smokers without gender predominance. Disease location and behaviour were not significantly different between smokers, ex-smokers and nonsmokers. Smoking status did not modify the use of medical therapy in CD. However, smoking was associated with an increased need for surgery (P = 0.043, Figure 1) and hospitalisation (P = 0.005).

Smoking prevalence in UC was substantially lower at 10% (P < 0.001) and 18% were ex-smokers. There was no gender difference in smoking status. Ex-smokers demonstrated a significantly later age at diagnosis than current smokers and nonsmokers with UC (43 vs. 35 vs. 34, respectively, P = 0.001). Smoking status did not predict disease location or medication use. However, long-term steroid use in ex-smokers was significantly higher (44%) when compared to smokers (24%) and nonsmokers (36%, P = 0.048). Colectomy rates were equivocal between the three groups (8%) and hospitalisation rates did not differ significantly.

Figure 1. Proportion of patients requiring surgical intervention in Crohn's disease by smoking status.


Smoking was associated with an increased need for surgery and hospital admission in CD. Smoking poorly correlated to other markers of disease severity. Although UC is predominantly a disease of nonsmokers and ex-smokers this study did not find smoking to protect against colectomy or hospitalisation. These data should strongly encourage all smokers with IBD to quit smoking.