Abstract
BACKGROUND: Sepsis is a major health threat in the elderly, with high morbidity and mortality. Emerging evidence suggests an "obesity paradox," where higher body mass index (BMI) may improve survival. However, the association between BMI and outcomes in elderly sepsis patients remains unclear. METHODS: Using data from the MIMIC-IV v3.1 database (2008-2022), we conducted a retrospective cohort study of sepsis patients aged ≥ 60 years. After 1:1 propensity score matching based on demographics and severity scores, 9,602 patients (4,801 obese; 4,801 non-obese) were included. Kaplan-Meier curves, Cox regression, and restricted cubic spline (RCS) analyses were used to evaluate mortality at 28, 90, and 180 days. Secondary outcomes included hospital and ICU length of stay (LOS), and mechanical ventilation (MV) duration. RESULTS: Obese patients had significantly lower mortality at 28 (HR = 0.83), 90 (HR = 0.79), and 180 days (HR = 0.76; all p < 0.001). RCS showed an L-shaped curve, with lowest mortality at BMI ≈ 36 kg/m². However, obesity was associated with longer hospital (p = 0.021), ICU stays (p = 0.0005), and MV duration (p = 0.0002). Subgroup analysis confirmed that overweight and moderately obese patients had the best survival, while underweight patients fared worst. CONCLUSION: Obesity is linked to improved survival in elderly sepsis patients, supporting the obesity paradox. Yet, it also entails greater healthcare utilization. Moderate obesity (BMI 30-35 kg/m²) appears most beneficial. Further studies are needed to clarify mechanisms and guide personalized sepsis care.