Prior studies have demonstrated associations between body composition and clinical response to therapy as well as progression of Crohn's disease (CD). In patients with small bowel CD we sought to explore the impact of baseline body composition on radiological response and risk for small bowel surgery.Methods:
Patients with small bowel CD who underwent a baseline computed tomography enterography (CTE) prior to initiation or during maintenance of therapy and with subsequent serial enterography follow-ups were included. Small bowel lesions were scored for length, enhancement, mural thickness, dilated vasa recta, perienteric inflammation, and penetrating/stricturing disease. Patients were complete responders if all lesions improved, non-responders if existing lesion(s) worsened or new lesion(s) detected, and partial responders for other scenarios. Any small bowel CD-related surgery after first follow-up enterography was noted. Standard body morphomic parameters were quantified at mid-vertebral level over contiguous axial CT images from L1 to S1 on baseline CTE examination using semiautomated analytical software (NT, MATLAB 13.0, Math Works, Natick, MA). Measured parameters included area (cm2) and volume (cm3) of subcutaneous adipose tissue, visceral adipose tissue (VAT), total lean skeletal muscle, intramuscular adipose tissue (IMAT) and mean muscle attenuation (Hounsfield Units, HU). Right lower quadrant (RLQ) VAT was defined as between L3-S1 levels. Fat mass (0.042 x (L3 total adipose area) + 11.2) and skeletal muscle indices (SMI, L3 muscle area/height2) were calculated, with sarcopenia characterization according to current sex-specific consensus definitions (Male: SMI < 55 cm2/m2, Female: SMI < 39 cm2/m2). Baseline body composition was compared between response categories using Wilcoxon rank-sum test. Body composition as an independent risk factor for CD progression leading to surgery was explored in a logistic regression adjusting for smoking status, penetrating disease phenotype, and complete versus partial/non-response status using JMP 13 (SAS, Cary, NC).Results:
A total of 134 CD patients were included: 52 complete responders (39%), 34 partial responders (25%), and 48 non-responders (36%). At baseline CTE, a trend towards significance was seen for volumetric VAT (372.1 [IQR, 167.3–672.5] versus 550.1 [221.7–799.2], P = 0.09), RLQ VAT (131.7 [62.4–233.5] versus 180.2 [83.0–260.4], P = 0.09) and L3 VAT (65.6 [28.1–135.3] versus 106.4 [36.5–159.3], P = 0.07) comparing radiological non-responders and responders. At L3 level patients with increased fat mass (OR = 5.7, 95% CI, 5.5–5.9), increased IMAT (OR = 8.2, 95% CI, 7.9–8.5) and lower muscle attenuation (OR = 9.4, 95% CI, 9.1–9.7) demonstrated an independent increased risk of small bowel surgery adjusting for variables above including radiological response. Similar results were seen for volumetric IMAT and muscle attenuation. Body mass index and sarcopenia were not significantly associated with response or need for small bowel surgery.Conclusions:
In this cohort of patients, increased total fat mass, increased IMAT and decreased muscle attenuation (myosteatosis) were associated with an increased risk of small bowel surgery in Crohn's disease. These findings suggest a role for analysis of body fat and skeletal composition in risk stratification in treatment algorithms.