Comparative analysis of clinical scores in predicting ICU and hospital mortality in nonagenarians and centenarians after in-hospital cardiac arrest: A retrospective observational study using the Australian and New Zealand Intensive Care Society (ANZICS) Adult Patient Database (2010-2024)

Background Nonagenarians and centenarians admitted to intensive care units (ICUs) following in-hospital cardiac arrest (IHCA) represent a growing yet understudied population. Clinicians require accurate prognostic tools to inform early goals of care discussions and resource allocation. This study evaluated the predictive performance of commonly used clinical scores in this unique cohort. Methods We conducted a retrospective binational cohort study of nonagenarian and centenarian patients admitted to ICUs in Australia and New Zealand between 2010 and 2024 after IHCA, using data from the ANZICS Adult Patient Database. We assessed the prognostic accuracy of four clinical scores: Acute Physiology and Chronic Health Evaluation III (APACHE III), Sequential Organ Failure Assessment (SOFA), Clinical Frailty Scale (CFS) and Glasgow Coma Scale, in predicting ICU and hospital mortality. Discrimination was measured using area under the receiver operating characteristic curve (AUROC). Multivariable Cox regression and Fine-Gray competing risk models were used to examine associations with mortality and discharge outcomes. Results A total of 219 patients (median age 91.6 years; 44% female) were included. ICU and hospital mortality were 45.2% and 55.7%, respectively. The APACHE III score showed the highest discriminatory ability (ICU mortality AUROC=0.850; hospital mortality AUROC=0.842), followed by the SOFA score (AUROCs=0.758 and 0.761, respectively). The CFS showed poor prognostic performance (AUROCs close to 0.5). In adjusted Cox models, both APACHE III and SOFA scores were independently associated with mortality. SOFA scores were associated with longer ICU length of stay, while higher APACHE III scores were associated with shorter hospital stay, likely reflecting early mortality. Conclusions In the oldest critically ill patients following IHCA, physiologic severity scores, particularly APACHE III and SOFA, outperform frailty in predicting short-term mortality and resource use. These findings support the integration of validated scoring systems into early clinical decision-making to improve care precision and guide resource allocation in ageing ICU populations.