Predictive value of biochemical markers and clinical scoring systems for outcomes in nonagenarians and centenarians admitted to the ICU after cholecystectomy for acute cholecystitis: a binational multicenter analysis

초록

Introduction With the global rise in nonagenarian and centenarian populations, demand for surgical and critical care in this age group is increasing. Acute biliary disease, particularly cholecystitis, often necessitates surgery, yet outcomes following cholecystectomy in the oldest-old ICU patients remain poorly characterized. This study aimed to identify clinical and biochemical predictors of mortality and resource use in this cohort. Methods In this retrospective binational cohort study, we used the Australian and New Zealand Intensive Care Society Adult Patient Database (2010-2024) to identify patients aged >= 90 years admitted to ICU after cholecystectomy for gallbladder-related disease. The primary endpoint was 1-year all-cause mortality after ICU admission. Key secondary endpoints were ICU, in-hospital, 30-day, and 6-month mortality; the exploratory endpoint was mortality at final follow-up (censored 30 April 2025). Secondary non-mortality outcomes were ICU and hospital length of stay. Predefined admission predictors were APACHE III, SOFA, Glasgow Coma Scale, Clinical Frailty Scale, and peak 24-hour biochemical markers (urea, creatinine, albumin, bilirubin, white cell count). Discrimination was quantified using the area under the receiver-operating-characteristic curve, and associations with 1-year mortality were examined using multivariable logistic regression. Results Among 378 nonagenarian and centenarian ICU patients after cholecystectomy, 1-year mortality was 17.7%. ICU, in-hospital, 30-day, and 6-month mortality were 1.9%, 4.2%, 7.9%, and 14.0%, respectively, while mortality at final follow-up was 26.2%. Compared with 1-year survivors, non-survivors had higher APACHE III, SOFA, and frailty scores, and more deranged admission biochemistry. For 1-year mortality, APACHE III showed fair discrimination (AUROC 0.69), outperforming SOFA, frailty, and Glasgow Coma Scale. Serum urea provided the strongest discrimination among biochemical markers (AUROC 0.71). In adjusted models, higher urea and bilirubin remained independently associated with 1-year mortality. High-risk groups defined by ROC-derived thresholds for APACHE III and urea had substantially higher 1-year mortality than low-risk groups. Conclusion Routine admission data, particularly APACHE III and simple biochemical markers such as urea and bilirubin, provide clinically useful risk stratification for nonagenarians and centenarians after cholecystectomy and may support early decision-making and resource allocation in this high-risk, oldest-old surgical population.

키워드