Objective: In sepsis, serum cortisol generally tends to increase as part of the systemic stress response. This study aimed to evaluate the value of serum cortisol for predicting in-hospital mortality in elderly patients with sepsis at Thong Nhat Hospital.
Subjects and Methods: A prospective cohort study was conducted on 132 elderly patients (≥60 years) diagnosed with sepsis at Thong Nhat Hospital between October 2024 and August 2025. Demographic characteristics and clinical data were collected, including age, sex, comorbidities, Sequential Organ Failure Assessment (SOFA) score, Acute Physiology and Chronic Health Evaluation II (APACHE II) score and serum cortisol levels measured at the time of sepsis diagnosis. Multivariable logistic regression analysis was performed to assess the independent prognostic value of serum cortisol for mortality. The prognostic performance of serum cortisol for predicting in-hospital mortality was evaluated using the receiver operating characteristic (ROC) curve.
Results: The mean age of patients was 76.0 ± 10.1 years, and 57.6% were female. Most patients had comorbidities (92.4%), mainly hypertension. Septic shock occurred in 75.8% of cases. The in-hospital mortality rate was 58.3%. Multivariable logistic regression analysis showed that SOFA score and serum cortisol levels were significantly associated with in-hospital mortality. The AUC of serum cortisol was 0.711, indicating moderate predictive value. The optimal cut-off value was 22.9 µg/dL, with sensitivity of 51.9% and specificity of 85.5%.
Conclusions: Serum cortisol with a cut-off value of 22.9 µg/dL showed moderate predictive value with AUC of 0.711 for predicting in-hospital mortality in elderly patients with sepsis. Therefore, it should be combined with other indicators to improve mortality prediction.