Low muscle mass determined by psoas muscle area does not correlate with dual-energy x-ray absorptiometry or total lumbar muscle mass scores: A prospective cohort study of patients undergoing vascular surgery

Abstract

Background/Objective: Low muscle mass and sarcopenia have been explored as risk factors for poor outcomes following vascular surgery. The findings have been variable. The use of a diverse range of techniques to identify low muscle mass is a confounder in establishing the true relationship between low muscle mass, sarcopenia and outcomes. Our aim was to establish if different scoring methods identified the same patients as sarcopenic. We also explored which method best predicted outcomes. Method: 70 patients undergoing vascular surgery were prospectively assessed for sarcopenia using dual-energy x-ray absorptiometry (DEXA) scan, grip strength and gait speed. These patients underwent abdominal CT imaging as routine care. The muscle mass of each patient was determined using DEXA and by both psoas muscle and total skeletal muscle area on CT, normalised for patient height (PMI and CT-SMI, respectively). Low muscle mass was defined by published age- and sex-specific cut-offs. Grip strength data was combined with muscle mass to define sarcopenic patients. One- and 3-year mortality and time to readmission was recorded. Conclusion and Results: 10–22% of patients had low muscle mass and 4–10% of patients were sarcopenic, depending on the method employed. PMI did not correlate with DEXA or CT-SMI for low muscle mass, but CT-SMI correlated with DEXA (p = 0.0007). For sarcopenia, CT-SMI and DEXA scoring correlated (p = 0.002); PMI correlated with CT-SMI (p = 0.0006) but not DEXA. Low muscle mass by PMI predicted 1-year mortality (p = 0.02, X2 = 5.34, Effect size = 1.04) but the applicability of this finding is limited by the diverse pathologies explored. No other method predicted 1- or 3-year mortality or readmissions in this heterogenous cohort. The psoas area did not correlate with muscle mass defined by DEXA or total lumbar skeletal muscle area. Low psoas muscle index may be an independent marker of poor outcome, unrelated to generalised sarcopenia, and this warrants investigation in specific pathologies. A lower total number of patients were sarcopenic than had been expected, emphasising the need to use population-based pre-defined cut-offs.