P139 Body composition and Crohn’s disease behavior: Is adiposity the main game changer?

Morão, B.(1);Velho, S.(2);Agostinho, L.(3);Torres, J.(1);Cravo, M.(4);

(1)Hospital Beatriz Ângelo, Gastroenterology, Lisbon, Portugal;(2)Hospital Beatriz Ângelo, Nutrition, Lisbon, Portugal;(3)Hospital Beatriz Ângelo, Radiology, Lisbon, Portugal;(4)Hospital da Luz, Gastroenterology, Lisboa, Portugal;

Background

Introduction:Visceral fat has been associated with unfavorable outcomes in Crohn’s disease (CD) patients. We investigated the association between body composition upon diagnosis and complicated phenotypes and time until surgery in patients with CD.

Methods

Methods:retrospective cohort study including patients with CD who had a computed tomography enterography (CT) or an entero magnetic resonance (MR) performed within 6 months of diagnosis. Skeletal muscle, visceral and subcutaneous adipose tissue cross-sectional areas were determined with CT or MR images at L3 vertebrae level, after being processed with Slice-o-Matic (Tomovison) and ABCS module (applicable only to CT scan images). Logistic regression was used to assess predictors of disease behaviourand time until abdominal urgery was modeled with proportional hazards Cox models.

Results

Results: We included 63 patients: 33(52%) men with median age of 35 years. Disease location and behaviour according to Montreal classification was as follows: L1=28(44%), L2=13(21%), L3=18(28%), L1+L4=1(2%), L3+L4=3(5%); B1=39(62%), B2=11(17%), B3=13(21%); 20(32%) patients had perianal disease. At diagnosis, 17(27%) patients were treated with corticosteroids. Anti-TNF agents were started in 8(13%) patients, azathioprine in 23(36%), combination therapy in 15(24%); 17(27%) patients did not start any medical therapy. Visceral obesity was present in 12(19%) patients at diagnosis. Visceral obesity was associated with higher age of CD onset (median 60 vs 34 years, p=0.002) and complicated disease behaviour(B2/B3) at diagnosis (66.7% vs 31.7%, p=0.021), with an approximately 4 times higher risk of complicated behaviour(OR=4.37, 95%IC=1.19-18.44, p=0.03).After adjusting for age and perianal disease, total adipose tissue (visceral and subcutaneous fat) remained statistically significant and was associated with a 4% increase in the odds of complicated behaviourat diagnosis per 10 cm2of total adipose tissue (OR 1.004,95%CI 1.00-1.008, p=0.043). Median follow-up time was 3.35 years, during which 15 (24%) of patients underwent abdominal surgery. Visceral obesity was associated with a 5.10 times higher risk of abdominal surgery during follow-up (95%CI 1.52-17.09, p=0.008); after adjusting to disease behaviour, visceral obesity maintained a near significant association with a 2.90 times higher risk of surgery (95%CI 0.83-10.08, p=0.09).

Conclusion

Conclusion: In our cohort of patients with newly diagnosed CD, total fat upon diagnosis seems to be associated with B2 and B3 disease phenotypes. Furthermore, visceral obesity was associated with a 5.10 times higher risk of abdominal surgery during follow-up and with a shorter time until surgery.