Search in the Abstract Database

Abstracts Search 2014

* = Presenting author

P114. The influence of obesity on clinical course and behaviour of Crohn's disease

S.V. Venkatachalapathy1, V.E. Sayer1, N. Grasso1, A. Salerno2, R. Saleem1, P. Dubois1, G. Chung-Faye1, B. Hayee1, 1King's College Hospital NHS Foundation Trust, Gastroenterology, London, United Kingdom, 2King's College Hospital NHS Foundation Trust, Colorectal Surgery, London, United Kingdom


Despite the widespread perception of Crohn's disease (CD) as a wasting disorder there is a sub-population of obese patients with the disease [1]. Notwithstanding the possibility of either a pathogenic link to CD, or even of a distinct enteropathy [2], obesity may influence the severity and phenotype of CD [3].


We conducted a retrospective analysis of our Institution's electronic patient record to stratify and compare CD patients by body mass index (BMI) at diagnosis, with regard to their disease characteristics, medication use and outcomes. Smoking status recorded on the hospital database was validated by survey with general practitioners. All statistical analysis was conducted using GraphPad Prism v4.0.


Reliable data was available for 282 patients with CD (130 female, age 43.6±14.3 y). At diagnosis, BMI >30 was recorded in n = 63 (22.3%), 20–29 in 181 (64.2%), and <19 in 38 (13.5%). The proportion of smokers did not differ significantly in each group. Patients with BMI >30 were older (p = 0.0001), with more colonic (p = 0.0003) and less penetrating disease (p = 0.01) compared to all other groups. After adjustment for age, colonic disease remained significantly different (OR 3.20; 95% CI: 1.74 to 5.91, p = 0.002). Rates of operative intervention for intestinal disease were not statistically different: 0.076 vs 0.063 vs 0.74 per patient year for the above BMI categories, respectively (p = 0.36). Neither the courses of steroids (per patient year), proportion of patients receiving more than 3 courses of steroids in their lifetime, or the proportion of patients requiring immunosuppressant or biologic medication, differed significantly between the two groups The median time-to-first-use of anti-TNF (in months) was 14 (range 3–120), 36 (3–504) and 24 (4–600) in the above BMI categories, respectively (p < 0.05).


Obesity in CD is a common phenomenon in this cohort, with 22.3% having BMI >30 - nearing the national average of 26% (source: Office of National Statistics 2012). Obesity is associated with colonic disease location, less penetrating disease, but no difference in the rates of stricturing disease or of operative intervention. Our data also imply a more aggressive inflammatory phenotype in terms of requiring biologic therapy significantly earlier in the natural history of disease. The mechanism for these effects is unclear, but environmental and luminal factors could be critical.

1. Ungar B, (2013), et al., Digestion, 26–32, 88.

2. Mendall MA, (2010), et al, Gut, A60, 59.

3. Nascimento ATM, (2012), et al, JCC, 1041, 6.