Cachexia as an inpatient vulnerability phenotype in lung cancer: A National Inpatient analysis.

Abstract

e23225 Background: Cancer cachexia is a multisystem metabolic syndrome characterized by involuntary tissue catabolism and systemic inflammation. In pivotal randomized interventional trials, cachexia has been defined as a composite syndrome incorporating nutritional, inflammatory, and tissue-loss domains. This trial-anchored framework was applied to a national inpatient cohort to evaluate associations with acute outcomes among hospitalized lung cancer patients. Methods: A survey-weighted analysis of the National Inpatient Sample (2016–2022) was conducted among adult lung cancer hospitalizations identified using ICD-10-CM code C34*, excluding childbirth- and delivery-related admissions. Cachexia was defined using a trial-anchored, multi-domain syndromic framework requiring ≥1 ICD-10–mapped domain: nutritional depletion (E43, E44, E46, R62.7, R63.4), skeletal muscle loss (M62.84, M62.5*), systemic inflammatory or metabolic burden (E88.09, R77.0), or clinician-recognized cachexia (R64); burden was categorized as 0, 1, or ≥2 domains. The primary outcome was in-hospital mortality; secondary outcomes included sepsis (A40/A41, R65.20, R65.21), acute kidney injury (N17*), ICU-level care proxied by invasive mechanical ventilation (5A1935Z, 5A1945Z, 5A1955Z), length of stay, and CCR-adjusted cost. Results: The weighted cohort comprised 3,197,184 hospitalizations (unweighted n = 639,437). Cachexia prevalence was 25.3%, with domain prevalences of nutritional depletion 22.2%, skeletal muscle loss 0.16%, inflammatory/metabolic burden 1.69%, and clinician-recognized cachexia 5.45%. Unadjusted outcomes were worse among hospitalizations with cachexia versus without: mortality 13.35% vs 7.40%, sepsis 20.17% vs 12.14%, acute kidney injury 21.73% vs 16.03%, and invasive mechanical ventilation 7.15% vs 4.80%. Mean length of stay was 7.96 vs 5.50 days, mean cost $24,662 vs $20,336, and mean charges $91,313 vs $74,016. After adjustment, cachexia was independently associated with higher in-hospital mortality (aOR 1.79, 95% CI 1.75–1.83), with a graded dose–response by domain burden (1 domain aOR 1.68, 95% CI 1.64–1.72; ≥2 domains aOR 2.45, 95% CI 2.34–2.55). Cachexia was also associated with sepsis (aOR 1.57, 95% CI 1.54–1.60), acute kidney injury (aOR 1.26, 95% CI 1.24–1.28), invasive mechanical ventilation (aOR 1.40, 95% CI 1.36–1.44), longer length of stay (+2.26 days), and higher hospitalization cost (+$5,132). Conclusions: A trial-anchored, syndromic cachexia phenotype is common among hospitalized lung cancer patients and identifies a graded, dose-dependent inpatient vulnerability signature associated with increased mortality, complications, and resource utilization. This framework translates cachexia trial concepts into real-world inpatient oncology and offers a scalable phenotype for risk stratification and evaluation of emerging anti-cachexia therapies.