Objective  To explore the risk factors of secondary acute kidney injury (AKI) after gastrectomy and construct a practical model for predicting AKI and guiding decision-making in clinical practices.

Methods  From January 2020 to June 2022, the relevant clinical data were collected retrospectively from 1871 AKI patients after gastrectomy. Various independent risk factors of AKI after gastrectomy were determined by positive stepwise Logistic and multifactorial Logistic regressions and a nomogram prediction model was established. The capability of prediction model was evaluated and validated by C-index and calibration curve.

Results  A total of 1871 patients were finally recruited. One hundred subjects developed AKI after gastrectomy with an incidence of 5.34%. After stepwise regressions of albumin, blood sodium, estimated glomerular filtration rate and systolic blood pressure, a multifactorial Logistic regression model was constructed. Albumin <30 g/L, SBP>140 mmHg (1 mmHg=0.133 kPa), a previous history of urolithiasis and preoperative aspirin dosing and perioperative blood transfusion were independent risk factors for AKI after gastrectomy (all P<0.05). The predictive model could accurately predict the occurrence of AKI after gastrectomy. C-Index of initial model and internal validation were 0.805(95%CI: 0.758−0.852) and 0.816(95%CI:0.815−0.816) respectively. Calibration curve indicated that predicted probability was highly consistent with actual probability.

Conclusion  Based upon eGFR, albumin, blood sodium, SBP, perioperative blood transfusion, preoperative aspirin dosing and urolithiasis, the prediction model of AKI after gastrectomy offers an excellent potency. It helps clinicians to identify high-risk AKI patients after gastrectomy.