Background/Objectives:Delayed treatment failure occurs in a significant proportion of inflammatory bowel disease (IBD) patients treated with tumor necrosis factor-alpha (TNF) antagonists. Identification of predictors of loss of response (LOR) may help to optimize therapy. We sought to determine whether body composition parameters at the commencement of anti-TNF therapy were associated with earlier treatment failure.Subjects/Methods:A retrospective cohort study was performed on 68 patients who had undergone cross-sectional abdominal imaging coincident with the commencement of anti-TNF drugs. Analysis of the images at the third lumbar vertebra was performed using standard techniques to determine cross-sectional areas of skeletal muscle (SM), visceral adipose tissue, subcutaneous adipose tissue and intermuscular adipose tissue. Treatment failure was defined as: post-induction hospital admission or surgery for IBD, escalation of TNF dose or immunosuppressants for clinical LOR, emergence of a new fistula or Crohn’s Disease Activity Index (CDAI) >150.Results:Two-thirds of patients had myopenia. Patients with less than gender-specific median SM area had a median time to failure of 520 (s.d. 135) days compared to 1100 (s.d. 151) days for those with more than median SM area (P=0.036). No difference was found in disease duration, inflammatory markers or CDAI between quartiles of SM area. No relation between outcomes and measures of adipose tissue, weight or body mass index was observed.Conclusions:Identifying low muscle mass at anti-TNF induction as a risk factor for treatment failure may contribute to a more tailored approach to IBD therapy.European Journal of Clinical Nutrition advance online publication, 22 February 2017; doi:10.1038/ejcn.2017.10.