Procedure-specific mortality risk in octogenarians undergoing curative GI cancer resection: A contemporary U.S. analysis.

Abstract

737 Background: Octogenarians increasingly undergo curative GI cancer surgery, yet most risk tools apply a uniform “age penalty” across procedures. Whether age ≥80 confers similar operative mortality across resections is unclear. We examined if the mortality impact of advanced age differs between pancreatic resection and gastrectomy. Methods: We used a nationally representative U.S. inpatient dataset (2020–2022) to identify adults with pancreatic or gastric cancer who underwent the corresponding resection. The primary outcome was in-hospital mortality; secondary outcomes were length of stay (LOS), and hospital charges. Multivariable models adjusted for sex, hospital teaching status, and year. An age (≥80 vs <80) × operation interaction tested whether the age effect varied by procedure. A sensitivity analysis excluded records with metastatic disease codes. Results: We identified 15,821 discharges (weighted ≈79,095), including 9,924 pancreatic resections and 5,897 gastrectomies; 16.1% were aged ≥80 years. In pooled models, adjusted in-hospital mortality was 3.54% for patients ≥80 versus 2.50% for those <80 (absolute difference +1.04 percentage points; aOR 1.43; p=0.003). The age effect differed by procedure (interaction p=0.024). For pancreatic resection, mortality was 3.34% in patients ≥80 versus 1.77% in those <80 (difference +1.56 points; aOR 1.91; 95% CI 1.36–2.68). For gastrectomy, mortality was 4.09% in patients ≥80 versus 3.74% in those <80 (difference +0.35 points; aOR 1.10; 95% CI 0.78–1.54), indicating no significant age-related increase. Length of stay was similar by age (8.41 vs 8.75 days; p=0.06), while hospital charges were lower among patients ≥80 ($133k vs $154k; p<0.001). Conclusions: In contemporary U.S. practice, the excess operative mortality associated with age ≥80 is concentrated in pancreatic resections and is not significant for gastrectomy after adjustment. Preoperative counseling and risk stratification for very elderly patients should be procedure-specific rather than applying a uniform “age penalty,” with particular caution for octogenarians considered for pancreatic resection. In-hospital mortality by age (≥80 vs <80) for curative GI cancer resections, U.S. inpatient 2020–2022. In-hospital mortality (%) ≥80 <80 Absolute difference (pp) Adjusted OR (95% CI) Pooled 3.54 2.50 +1.04 1.43 (—) Pancreatic resection 3.34 1.77 +1.56 1.91 (1.36–2.68) Gastrectomy 4.09 3.74 +0.35 1.10 (0.78–1.54) Mortality models adjusted for sex, hospital teaching status, and year. Pooled mortality aOR p=0.003; age×procedure interaction p=0.024.