Erector spinae muscle characteristics predict 90-day survival in elderly pneumonia patients

Pneumonia in elderly patients represent a critical clinical challenge with substantial mortality risks, where muscle deterioration has emerged as a potential prognostic biomarker. However, the specific morphological characteristics of the erector spinae muscle (ESM) in this vulnerable population and their predictive value for short-term outcomes remain poorly understood. This prospective observational cohort study enrolled 189 elderly patients (age 84.7 ± 4.1 years) diagnosed with pneumonia at Lishui Traditional Chinese Medicine Hospital. A pre-specified, retrospective analysis of baseline chest CT images was conducted to quantify erector spinae muscle (ESM) parameters, including cross-sectional area (ESMCSA), thickness (ESMT), fat infiltration rate (FI), and CT attenuation values. Clinical variables, inflammatory markers, and severity scores were prospectively collected. LASSO regression was applied for feature selection, followed by univariable and multivariable Cox proportional hazards analyses to determine factors associated with 90-day mortality. Among 48 non-survivors (25.4%), the death group exhibited significantly lower ESM CSA (adjusted for body surface area), ESMT, and albumin levels, alongside elevated FI, APACHE II/SOFA scores, CRP, and diabetes prevalence (all P < 0.05). Kaplan-Meier analysis revealed a 9.6-fold higher mortality risk (P < 0.001) in patients with lower ESMT/BSA. LASSO regression identified CRP, albumin, FI, ESMT, and diabetes as key predictors. Although univariable analysis associated lower ESMCSA/BSA, ESMT/BSA, and elevated FI with mortality (P < 0.05), multivariable Cox regression identified only higher albumin levels (HR = 0.882, 95% CI: 0.827-0.941) and ESMT/BSA (HR = 0.553, 95% CI: 0.449-0.683) as independent protective factors. Elevated CRP (HR = 1.098, 95% CI: 1.051-1.147) and FI (HR = 1.044, 95% CI: 1.006-1.084) independently predicted mortality. In patients with pneumonia, elevated CRP levels and the degree of erector spinae muscle fat infiltration were identified as independent risk factors for mortality, while higher albumin levels and the ratio of erector spinae muscle thickness to body surface area served as independent protective factors.