The prediction of in-hospital mortality in elderly patients with sepsis-associated acute kidney injury utilizing machine learning models

Jie Tang; Jian Huang; Xin He; Sijue Zou; Li Gong; Qiongjing Yuan; Zhangzhe Peng

doi:10.1016/j.heliyon.2024.e26570

The prediction of in-hospital mortality in elderly patients with sepsis-associated acute kidney injury utilizing machine learning models

Heliyon. 2024 Feb 16;10(4):e26570. doi: 10.1016/j.heliyon.2024.e26570. eCollection 2024 Feb 29.

Authors

Jie Tang^{1

2}, Jian Huang³, Xin He^{2

4}, Sijue Zou^{2

4}, Li Gong^{2

4}, Qiongjing Yuan^{2

4

5

6}, Zhangzhe Peng^{2

4

5}

Affiliations

¹ Department of Gastroenterology, Xiangya Hospital, Central South University, Changsha, Hunan, China.
² Organ Fibrosis Key Lab of Hunan Province, Central South University, Changsha, Hunan, China.
³ Department of Diagnostic Ultrasound and Echocardiography, Sir Run Run Shaw Hospital, Zhejiang University College of Medicine, Hangzhou, Zhejiang, China.
⁴ Department of Nephrology, Xiangya Hospital, Central South University, Changsha, Hunan, China.
⁵ National International Joint Research Center for Medical Metabolomices, Xiangya Hospital, Central South University, Changsha, Hunan, China.
⁶ National Clinical Medical Research Center for Geriatric Diseases, Xiangya Hospital of Central South University, Changsha, Hunan, China.

Abstract

Background: Sepsis-associated acute kidney injury (SA-AKI) is a severe complication associated with poorer prognosis and increased mortality, particularly in elderly patients. Currently, there is a lack of accurate mortality risk prediction models for these patients in clinic.

Objectives: This study aimed to develop and validate machine learning models for predicting in-hospital mortality risk in elderly patients with SA-AKI.

Methods: Machine learning models were developed and validated using the public, high-quality Medical Information Mart for Intensive Care (MIMIC)-IV critically ill database. The recursive feature elimination (RFE) algorithm was employed for key feature selection. Eleven predictive models were compared, with the best one selected for further validation. Shapley Additive Explanations (SHAP) values were used for visualization and interpretation, making the machine learning models clinically interpretable.

Results: There were 16,154 patients with SA-AKI in the MIMIC-IV database, and 8426 SA-AKI patients were included in this study (median age: 77.0 years; female: 45%). 7728 patients excluded based on these criteria. They were randomly divided into a training cohort (5,934, 70%) and a validation cohort (2,492, 30%). Nine key features were selected by the RFE algorithm. The CatBoost model achieved the best performance, with an AUC of 0.844 in the training cohort and 0.804 in the validation cohort. SHAP values revealed that AKI stage, PaO₂, and lactate were the top three most important features contributing to the CatBoost model.

Conclusion: We developed a model capable of predicting the risk of in-hospital mortality in elderly patients with SA-AKI.

Keywords: Acute kidney injury; Elderly; Machine learning; Model interpretation; Sepsis.