Long-term clinical outcomes in critically ill patients with sepsis and pre-existing low muscle mass: a retrospective cohort study

Abstract

Abstract Purpose Critically ill patients with sepsis account for significant disease morbidity and healthcare costs. Low muscle mass has been proposed as an independent risk factor for poor short-term outcomes, although its effect on long-term outcomes remains unclear. Methods Retrospective cohort analysis of patients treated at a quaternary care medical center over 6 years (09/2014 - 12/2020). Critically ill patients meeting Sepsis-3 criteria were included, with low muscle mass defined by $$\le$$ ≤ 5th percentile skeletal muscle index, measured at the L3 lumbar level (L3SMI) on Computed-Tomography (CT) scan ($$\le$$ ≤ 41.6 cm2/m2 for males and $$\le$$ ≤ 32.0 cm2/m2 for females). L3SMI was calculated by normalizing the CT-measured skeletal muscle area to the square of the patient’s height (in meters). Measurements were taken from abdominal/pelvic CT scan obtained within 7 days of sepsis onset. The prevalence of low muscle mass and its association with clinical outcomes, including in-hospital and one-year mortality, and post-hospitalization discharge disposition in survivors, was analyzed. Unfavorable post-hospitalization disposition was defined as discharge to a location other than the patient’s home. Results Low muscle mass was present in 34 (23%) of 150 patients, with mean skeletal muscle indices of 28.0 ± 2.9 cm2/m2 and 36.8 ± 3.3 cm2/m2 in females and males, respectively. While low muscle mass was not a significant risk factor for in-hospital mortality (hazard ratio 1.33; 95% CI 0.64 – 2.76; p = 0.437), it significantly increased one-year mortality after adjusting for age and illness severity using Cox multivariate regression (hazard ratio 1.9; 95% CI 1.1 – 3.2; p = 0.014). Unfavorable post-hospitalization discharge disposition was not associated with low muscle mass, after adjusting for age and illness severity in a single, multivariate model. Conclusion Low muscle mass independently predicts one-year mortality but is not associated with in-hospital mortality or unfavorable hospital discharge disposition in critically ill patients with sepsis.