Development and external validation of a prediction model for sepsis-associated acute kidney injury: a two-centre retrospective cohort study

Sepsis-associated acute kidney injury (SA-AKI) is common and is associated with poor outcomes. This study aimed to develop a parsimonious model based on routinely available clinical variables and to evaluate its performance in an independent external cohort. This two-centre retrospective cohort study included 1,874 adults with sepsis admitted between 2012 and 2023. The development cohort comprised 1,472 patients from the Second Affiliated Hospital of Dalian Medical University, and the external validation cohort comprised 402 patients from the Central Hospital Affiliated to Dalian University of Technology. Thirty-six candidate predictors obtained within 48 h of admission were assessed. Least absolute shrinkage and selection operator (LASSO) regression was used for feature selection, followed by multivariable logistic regression. Model discrimination, calibration, and clinical utility were evaluated using the area under the receiver operating characteristic curve (AUC), calibration analyses, and decision curve analysis (DCA), respectively. Seven predictors were retained in the final model: sex, arrhythmia, abdominal infection, D-dimer, the aspartate aminotransferase-to-alanine aminotransferase ratio (AST/ALT), total protein, and albumin. The AUC was 0.711 (95% confidence interval [CI], 0.681–0.741) in the development cohort and 0.675 (95% CI, 0.614–0.737) in the external validation cohort. At the selected cut-off values, sensitivity and specificity were 0.613 and 0.723 in the development cohort and 0.668 and 0.628 in the validation cohort, respectively. The corresponding positive and negative predictive values were 0.365 and 0.877 in the development cohort and 0.402 and 0.835 in the validation cohort. The Hosmer-Lemeshow P values were 0.362 and 0.053, respectively. DCA suggested a net benefit across a range of clinically relevant threshold probabilities. A seven-predictor model based on routinely available clinical variables showed moderate discrimination for SA-AKI and retained predictive ability in an external cohort. The model may support early risk stratification, but prospective multicentre validation is required before clinical implementation.

Authors

Institutions

Publication Details

Journal
BMC Infectious Diseases
Published
2026-11-14
DOI
https://doi.org/10.1186/s12879-026-14825-y
Primary Topic
Acute Kidney Injury Research
Type
article
Field-Weighted Citation Impact
0.00
Controls
|||
ALL TIME
JAN
FEB
MAR
APR
MAY
JUN
JUL
AUG
SEP
OCT
NOV
article

Development and external validation of a prediction model for sepsis-associated acute kidney injury: a two-centre retrospective cohort study

Yingzhuo Dong, Zhuang Ge, Xuhan Liu, Gong Gu et al.
BMC Infectious Diseases
Acute Kidney Injury Research
article

Development and external validation of a prediction model for sepsis-associated acute kidney injury: a two-centre retrospective cohort study

Yingzhuo Dong, Zhuang Ge, Xuhan Liu, Gong Gu, Feili Liu, Cencen Liu, Chuwei Yang
article en

Abstract

Sepsis-associated acute kidney injury (SA-AKI) is common and is associated with poor outcomes. This study aimed to develop a parsimonious model based on routinely available clinical variables and to evaluate its performance in an independent external cohort. This two-centre retrospective cohort study included 1,874 adults with sepsis admitted between 2012 and 2023. The development cohort comprised 1,472 patients from the Second Affiliated Hospital of Dalian Medical University, and the external validation cohort comprised 402 patients from the Central Hospital Affiliated to Dalian University of Technology. Thirty-six candidate predictors obtained within 48 h of admission were assessed. Least absolute shrinkage and selection operator (LASSO) regression was used for feature selection, followed by multivariable logistic regression. Model discrimination, calibration, and clinical utility were evaluated using the area under the receiver operating characteristic curve (AUC), calibration analyses, and decision curve analysis (DCA), respectively. Seven predictors were retained in the final model: sex, arrhythmia, abdominal infection, D-dimer, the aspartate aminotransferase-to-alanine aminotransferase ratio (AST/ALT), total protein, and albumin. The AUC was 0.711 (95% confidence interval [CI], 0.681–0.741) in the development cohort and 0.675 (95% CI, 0.614–0.737) in the external validation cohort. At the selected cut-off values, sensitivity and specificity were 0.613 and 0.723 in the development cohort and 0.668 and 0.628 in the validation cohort, respectively. The corresponding positive and negative predictive values were 0.365 and 0.877 in the development cohort and 0.402 and 0.835 in the validation cohort. The Hosmer-Lemeshow P values were 0.362 and 0.053, respectively. DCA suggested a net benefit across a range of clinically relevant threshold probabilities. A seven-predictor model based on routinely available clinical variables showed moderate discrimination for SA-AKI and retained predictive ability in an external cohort. The model may support early risk stratification, but prospective multicentre validation is required before clinical implementation.

BMC Infectious Diseases
Dalian Medical University (CN), Dalian University of Technology (CN), Soochow University (CN), Second Affiliated Hospital of Dalian Medical University (CN), Dalian Municipal Central Hospital (CN), Zhangjiagang First People's Hospital (CN)
Acute Kidney Injury Research
AI Navigator

Ask Laika to Summarize, Analyze, and Connect papers live on the map.

Summarize Papers & Methodologies

Extract key findings, datasets, and comparative methods across publications.

Benchmark Rankings & Visual Analytics

Rank top research institutions, authors, funders, topics, and journals by Field-Weighted Citation Impact (FWCI) and paper volume with instant charts.

Connect Distant Disciplines

Bridge topological clusters on the map to find hidden collaborative intersections.

Laika access coming soon