Association between admission diagnoses and intensive care unit mortality: A retrospective cohort study

Creative Commons License

YILMAZ M. A., Guney M. Z., Kivanc H. E., Uzun N., İbrahimbayli L., ÇATALKAYA H., ...Daha Fazla

Challenge Journal of Perioperative Medicine, cilt.4, sa.2, ss.42-52, 2026 (TRDizin)

Özet

Background: Mortality in the intensive care unit (ICU) is primarily driven by the se- verity of acute organ dysfunction; however, the prognostic contribution of admission clinical diagnoses beyond severity scores remains incompletely defined. This study aimed to evaluate the association between admission diagnoses and ICU mortality after adjustment for organ dysfunction severity. Methods: This retrospective cohort study included adult patients admitted to a ter- tiary ICU between January 2024 and December 2025. Patients with missing baseline data or ICU length of stay <24 hours were excluded. Demographic characteristics, comorbidities, admission diagnoses, and disease severity scores were recorded. ICU mortality was the primary outcome. Multivariable logistic regression analysis was performed using the Sequential Organ Failure Assessment (SOFA) score as the pri- mary severity adjustment variable. Results: A total of 1,248 patients were included; 763 (61.1%) died during the ICU stay. In the SOFA-adjusted multivariable model, age (adjusted odds ratio [aOR] 1.02; 95% CI 1.02–1.03), SOFA score (aOR 1.33 per point; 95% CI 1.25–1.41), post–cardiac arrest status (aOR 6.21; 95% CI 4.17–9.23), sepsis (aOR 1.73; 95% CI 1.15–2.59), and active malignancy (aOR 1.69; 95% CI 1.17–2.44) were independently associated with ICU mortality. Neuromuscular disease showed a trend toward increased mortality but did not reach statistical significance. Conclusion: Beyond organ dysfunction severity, selected admission diagnoses— particularly post–cardiac arrest status, sepsis, and active malignancy—provide inde- pendent prognostic information for ICU mortality. Incorporating diagnosis-based risk stratification alongside severity scores may improve early prognostic assess- ment in critically ill patients.