Journal of Gastrointestinal Surgery · Mar 15, 2017

The Association of Computed Tomography-Assessed Body Composition with Mortality in Patients with Necrotizing Pancreatitis

By van Grinsven J, van Vugt JLA, Gharbharan A et al.

BACKGROUND: Identification of patients with necrotizing pancreatitis at high risk for a complicated course could facilitate clinical decision-making. In multiple diseases, several parameters of body composition are associated with impaired outcome, but studies in necrotizing pancreatitis are lacking. METHODS: A post hoc analysis was performed in a national prospective cohort of 639 patients with necrotizing pancreatitis. Skeletal muscle mass, skeletal muscle density, and visceral adipose tissue were measured at the third lumbar vertebra level (L3) on contrast-enhanced computed tomography (CT) within 10 days after initial admission and 1 month thereafter. RESULTS: In total, 496 of 639 patients (78%) were included. Overall mortality rate was 14.5%. Skeletal muscle mass and density and visceral adipose tissue on first CT were not independently associated with in-hospital mortality. However, low skeletal muscle density was independently associated with increased mortality in patients ≥65 years (OR 2.54 (95%CI 1.12-5.84, P = 0.028). Skeletal muscle mass and density significantly decreased within 1 month, for both males and females, with a median relative loss of muscle mass of 12.9 and 10.2% (both P < 0.001), respectively. Skeletal muscle density decreased with 7.2 and 7.5% (both P < 0.001) for males and females, respectively. A skeletal muscle density decrease of ≥10% in 1 month was independently associated with in-hospital mortality: OR 5.87 (95%CI 2.09-16.50, P = 0.001). CONCLUSION: First CT-assessed body composition parameters do not correlate with in-hospital mortality in patients with necrotizing pancreatitis. Loss of skeletal muscle density ≥10% within the first month after initial admission, however, is significantly associated with increased mortality in these patients.

doi.org/10.1007/s11605-016-3352-3Open accessPubMed 28299618Biomarkers