Prognostic Utility and Exploratory Cut-off Threshold of the First 24-Hour APACHE II Score for In-Hospital Mortality After Thoracotomy: A Single-Center Retrospective Cohort Study

Abstract

Background: Open thoracotomy imposes severe physiological stress, so objective early risk stratification is needed at General Intensive Care Unit (GICU) admission.

Objective: To evaluate how well the first 24-hour Acute Physiology and Chronic Health Evaluation II (APACHE II) score predicts in-hospital mortality after thoracotomy, to set an exploratory cut-off, and to examine differences by surgical indication and organ component.

Methods: This retrospective cohort study analyzed 33 consecutive adults admitted directly to the GICU of Dr. Mohammad Hoesin General Hospital, Palembang, Indonesia, after open thoracotomy (January–December 2024). Analyses included Mann-Whitney U, Spearman's ρ, logistic regression, and Receiver Operating Characteristic (ROC) analysis with Youden's J index.

Results: In-hospital mortality was 27.3% (9/33). Non-survivors had higher median APACHE II scores than survivors (23.0 [IQR: 19.0–27.0] vs. 8.0 [IQR: 6.75–12.0]; U = 9.5, p < 0.001). The score correlated with mortality (ρ = 0.706, p < 0.001) and remained significant on multivariable regression (aOR = 1.74; 95% CI: 1.13–2.69; p = 0.0127). The AUC was 0.956 (95% CI: 0.891–1.000). A cut-off of ≥12.5 gave 88.9% sensitivity, 87.5% specificity, 72.7% positive predictive value, and 95.5% negative predictive value. Mortality was 32.0% in malignancy and 12.5% in non-malignancy cases (p = 0.394). Glasgow Coma Scale (6.33 vs. 1.83 points), serum creatinine (3.44 vs. 0.96), and arterial pH (1.89 vs. 0.33) components differed between non-survivors and survivors (all p < 0.001).

Conclusion: The initial 24-hour APACHE II score is a hypothesis-generating predictor of post-thoracotomy mortality at a candidate threshold of ≥12.5.