Fig. 1

Association of muscle loss phenotype with clinical outcomes. A Proportion of patients with evidence of muscle loss phenotype (MLP) based on individual and combined criteria. Overall, 53.4% of subjects with chronic obstructive pulmonary disease (COPD) met at least one criterion for MLP. Proportion of patients with evidence of MLP was lowest using body mass index (BMI) criteria and highest using appendicular skeletal muscle index (ASMI) criteria. B Venn diagram of criteria. A Venn diagram using ellipses demonstrates all possible relationships between different definitions of MLP and provides information on the proportion of patients that met one or more criteria for MLP (see Supplementary Table 1 for MLP definitions). The greatest overlap between two definitions was FFMI-defined MLP and ASMI-defined MLP (n = 2911, or 10%). The greatest overlap between three definitions was FFMI-defined MLP, ASMI-defined MLP, and HGS-defined MLP (n = 1660, or 6%). C Hazard ratios from Cox proportional analysis of MLP and all-cause mortality. Fat free mass index (FFMI) and BMI were most associated with all-cause mortality. D Cox-regression analysis of MLP and survival (all-cause mortality). Evidence of MLP (defined by at least one criterion) was associated with increased risk for all-cause mortality over time. E Hazard ratios from Cox proportional analysis of MLP and COPD-related death. FFMI was most associated with COPD-related mortality. F Cox-regression analysis of MLP criteria and COPD-related death. Evidence of MLP was associated with increased COPD-related mortality over time. MLP (defined by at least one criteria) was associated with increased risk for COPD-related death over time