Background: The impact of smoking at diagnosis and subsequent smoking cessation on clinical outcomes in Crohn's disease (CD) has not been evaluated in a population-based cohort.
Methods: Using a nationally representative clinical research database, we identified incident cases of CD between 2005 and 2014. We compared the following outcomes: overall corticosteroid (CS) use; flares requiring CS; CS dependency and intestinal surgery between smokers and non-smokers at time of CD diagnosis. Differences in these outcomes were also compared between persistent smokers and smokers who quit within 2 years of diagnosis.
Results: We identified 3553 patients with a new CD diagnosis over the study period of whom 1121 (32%) were smokers. Smokers at CD diagnosis had significantly higher CS-use (56 versus 47%, p < 0.0001), proportionally more CS flares (>1 CS flare/year: 9 versus 6%, p < 0.0001), and higher CS dependency (27 versus 21%, p < 0.0001) than non-smokers. Regression analysis identified smoking at diagnosis to be associated with a higher risk of intestinal surgery (HR 1.64, 95% CI 1.16-2.52). There was a significantly higher proportion of 'quitters' who remained steroid-free through follow-up in comparison to 'persistent smokers' (45.4 versus 37.5%, respectively, p = 0.02). 'Quitters' also had lower rates of CS dependency compared to 'persistent smokers' (24 versus 33%, p = 0.008).
Conclusions: Smokers at CD diagnosis have higher CS-use, CS dependency and higher risk of intestinal surgery. Quitting smoking appears to have beneficial effects on disease related outcomes, including reducing CS dependency highlighting the importance of offering early smoking cessation support.