Nomogram Predicts Survival in ICU Patients with AKI: Study Reveals A New Approach to Individualized Drug Therapy

"By integrating multiple factors, the model achieved high accuracy, with AUC values of 0.827, 0.799, and 0.777 for 1-, 3-, and 5-year survival, respectively, in the training dataset. The survival outcomes were also influenced by medication use, indicating that the nomogram could support clinical decision-making," the researchers reported in the European Journal of Medical Research.

The researchers note that acute kidney injury is a common and life-threatening condition in ICU patients, characterized by a sudden decline in the estimated glomerular filtration rate (eGFR). It is associated with higher mortality and prolonged hospital stays, making it a significant challenge for healthcare providers to accurately predict survival outcomes. These outcomes can vary considerably depending on factors such as the patient's underlying health, the severity of the injury, and the treatments administered. Despite its importance, the role of medication history in influencing the survival of AKI patients has received limited attention, highlighting the need to explore its potential as a predictor of survival outcomes in ICU settings.

Against the above background, Hao Su, Department of Urology, The Second Xiangya Hospital of Central South University, Changsha, Hunan, China, and colleagues aimed to develop a model that predicts the survival of ICU patients with AKI and examines the relationship between their medication history and survival outcomes. By utilizing machine learning techniques, this study aims to enhance the understanding of AKI survival and support the clinical application of such predictive models in intensive care settings.

For this purpose, the researchers sourced critically ill AKI patients from the MIMIC-IV database. Univariate Cox analysis and stepwise Cox regression were performed to ascertain significant, drug-related, independent predictors of survival. A nomogram was developed based on the identified predictors to estimate the individualized survival probability for AKI patients. A propensity score matching procedure was employed to account for potential confounders related to medication use. Subsequently, a comparative analysis was conducted to explore the prognostic differences among the patient subgroups identified through this approach.

The investigation uncovered the following findings:

• The study included 1,208 patients and developed a nomogram that incorporated oxygen flow rate, respiratory frequency, continuous venovenous hemodiafiltration status, age, and medication use (including ibuprofen, epinephrine, cefazolin, warfarin, and vasopressin).
• The predictive model demonstrated strong diagnostic accuracy, with AUC values for 1-year, 3-year, and 5-year survival among AKI patients of 0.827, 0.799, and 0.777, respectively, in the training dataset.
• In the internal validation dataset, the AUC values for 1-year, 3-year, and 5-year survival were 0.760, 0.743, and 0.740, respectively.
• Kaplan-Meier survival analyses revealed significant differences in survival outcomes among AKI patients based on their exposure to different medications.

The findings revealed that the survival of AKI patients in the ICU is shaped by multiple factors, including oxygen flow, respiratory rate, continuous venovenous hemodiafiltration (CVVHDF) status, age, and medication history (ibuprofen, epinephrine, cefazolin, warfarin, and vasopressin).

"The nomogram developed using these nine key factors demonstrates impressive predictive accuracy and clinical relevance. This model can assist in the early identification of AKI patients at high risk of mortality, ultimately contributing to improved patient outcomes," the researchers wrote.

Reference:

Yang, R., Su, X., Liu, Z. et al. Individualized drug therapy and survival prediction in ICU patients with acute kidney injury: construction and validation of a nomogram. Eur J Med Res 30, 65 (2025). https://doi.org/10.1186/s40001-025-02300-4