Association between body composition and mortality in patients requiring extracorporeal membrane oxygenation support

Abstract

To ascertain the association between body composition, including muscle mass and adiposity, and patient mortality in those requiring extracorporeal membrane oxygenation (ECMO) for acute respiratory failure. A retrospective study was undertaken of all patients with acute respiratory failure requiring veno-venous (VV) ECMO between January 2015 and December 2019. Automated image segmentation software was used to quantify the cross-sectional area and average radiodensity (in Hounsfield units) of different muscle and fat compartments at the L3 level of whole-body computed tomography (CT) images taken within 48 h of initiation of ECMO support. The primary endpoint was 30-day post-ECMO initiation all-cause mortality. Logistic regression was used to analyse the correlation between CT measurements, co-morbidities, and 30-day survival. The study included 189 patients (age=43.8±14.6, sex=42.3% female). There was no significant association between 30-day survival status and cross-sectional area of muscle or fat. Muscle attenuation (psoas, long spine, and abdominal muscles respectively) at the L3 level were significantly lower in those who died within 30 days of ECMO cannulation (p<0.05). On multivariable analysis including age, sex, and pre-existing respiratory comorbidities, psoas muscle attenuation was an independent predictor of survival at 30 days (OR 0.97; 95% CI 0.94 to 1.00; p=0.047). Reduced psoas muscle attenuation is associated with poorer survival outcomes at 30 days post-ECMO cannulation in patients who received VV ECMO support for respiratory failure. Cross-sectional areas of muscle and fat compartments did not correlate with survival outcomes at 30 days even when corrected for height and sex.