Related Experiment Video
Updated: Oct 15, 2026
Point-of-Care Ultrasound for Peripheral Veno-Arterial Extracorporeal Membrane Oxygenation Without Left Ventricular Venting
Published on: January 17, 2025
Computed Tomography-Derived Psoas Muscle Quantity, Attenuation, and 1 Year Mortality in Venovenous Extracorporeal
Jin Ho Jang1, Eunjeong Choi, Woo Hyun Cho
1From the Division of Pulmonary, Allergy and Critical Care Medicine, Department of Internal Medicine, Pusan National University School of Medicine, Transplantation Research Center, Research Institute for Convergence of Biomedical Science and Technology, Pusan National University Yangsan Hospital, Yangsan, Republic of Korea.
Abstract:
Computed tomography (CT)-derived skeletal muscle characteristics may provide prognostic information in patients receiving venovenous extracorporeal membrane oxygenation (VV ECMO), but the relative contributions of muscle quantity and attenuation remain uncertain. We retrospectively studied adults who received VV ECMO for severe respiratory failure between 2014 and 2023 and underwent abdominal CT within 48 h before ECMO initiation. Psoas muscle index (PMI) and attenuation were measured using an artificial intelligence-based pipeline, and combined PMI-attenuation phenotypes were evaluated for 1 year mortality. Among 92 patients, 53 (57.6%) died within 1 year. In Cox proportional hazards models adjusted for age, sex, and initial Sequential Organ Failure Assessment score and stratified by bridge-to-transplantation status, higher PMI (adjusted hazard ratio [aHR] per standard deviation increase, 0.58; 95% confidence interval [CI], 0.41-0.82) and higher attenuation (aHR per 5 Hounsfield unit increase, 0.87; 95% CI, 0.77-0.97) were associated with lower 1 year mortality. When modeled simultaneously, PMI remained significant, whereas attenuation did not. The low PMI/low attenuation phenotype had the highest mortality risk (aHR, 5.30; 95% CI, 2.10-13.34). PMI showed a more robust association with 1 year mortality, while combined PMI-attenuation phenotyping may identify a particularly high-risk subgroup.