Development and validation of a nomogram for predicting acute kidney injury after laparoscopic partial nephrectomy

Acute kidney injury (AKI) commonly occurs after laparoscopic partial nephrectomy (LPN). This study aimed to identify risk factors for postoperative AKI and develop a validated nomogram for individualized risk prediction. We retrospectively enrolled 266 LPN patients (training dataset) and 146 patients (external validation dataset). Lasso regression and AIC were used for feature selection. Independent risk factors were identified by multivariate logistic regression. A nomogram was constructed and its performance evaluated by ROC, calibration curves, DCA, as well as NRI and IDI. The incidence of AKI in training and external validation datasets was 29.3% and 29.5%, respectively. Gender, BMI, hypertension, duration of renal artery blockade, and cystatin C were independent risk factors. Incorporating the intraoperative feature (duration of renal artery blockade) into the model significantly improved its performance (IDI = 7.36%, P < 0.001; NRI = 16.2%, P = 0.008). The nomogram model developed based on the multivariable analysis results demonstrated an area under the curve (AUC) of 0.833 (95% CI: 0.78 ~ 0.89) in the training dataset and an AUC of 0.78 (95% CI: 0.70 ~ 0.86) in the external validation dataset. Both the calibration curves and DCA indicated favorable calibration and net clinical benefit. Based on five readily available independent risk factors, we have developed a nomogram model for the individualized prediction of AKI following LPN in patients with low-complexity renal tumors. The model has been validated to demonstrate robust performance, providing clinicians with a simple and intuitive tool for the precise and quantitative assessment of AKI risk in patients.