Key points are not available for this paper at this time.
Background: Liver transplantation serves as an essential therapeutic intervention for patients with end-stage liver disease. However, the occurrence of postoperative acute kidney injury (AKI) can markedly affect the clinical prognosis of these patients. Existing models to predict AKI after liver transplantation have limitations in specificity and accuracy, necessitating an updated model. Methods: We conducted a study adhering to the TRIPOD guidelines, including patients who underwent liver transplantation at West China Hospital from 2016 to 2020. Clinical data encompassing demographics, comorbidities, and intraoperative variables were collected. The LASSO regression was used to identify optimal predictors of AKI, leading to the development of a predictive nomogram. The model’s discrimination and calibration were assessed using AUC and calibration curves, respectively. Results: The nomogram, developed from 296 patients in the development cohort and validated on 142 patients, identified surgery duration, intraoperative blood loss, and preoperative serum creatinine as predictors of AKI. It demonstrated good discrimination with AUCs of 0.720 and 0.725 for the development and validation cohorts, respectively. The calibration curve confirmed the model’s accuracy in predicting AKI probabilities. Conclusion: The developed nomogram offers a novel model for predicting AKI risk after liver transplantation, with robust discrimination and calibration. Further multicenter validation and potential integration of genetic and molecular biomarkers for improved accuracy are needed.
Siying Wang (Fri,) studied this question.