Integrative evaluation of shear wave elastography and renal function biomarkers for predicting renal fibrosis in chronic kidney disease patients

Objective We hypothesize that combining point shear wave elastography (PSWE) with clinical risk factors enables accurate renal fibrosis assessment. This retrospective study integrates PSWE with serum creatinine (Scr) and estimated glomerular filtration rate (eGFR) to develop and validate a nomogram for personalized renal fibrosis evaluation in chronic kidney disease (CKD) patients. Methods A total of 157 patients underwent renal PSWE and kidney biopsy. PSWE measured cortical stiffness in the mid-portion of the right kidney. Feature importance was selected using elastic net regression, XGBoost, and random forest, with PSWE, Scr, and eGFR identified as key variables. Three models were established: Model 1 (PSWE + Scr + eGFR), Model 2 (Scr + eGFR), and Model 3 (PSWE). Diagnostic performance was evaluated using receiver operating characteristic (ROC) curves and area under the curve (AUC) values. A nomogram based on PSWE, Scr, and eGFR was developed for precise fibrosis risk assessment. The Hosmer-Lemeshow test and K-fold cross-validation were used to evaluate the nomogram's generalizability. Results Model 1 achieved an AUC of 0.928, outperforming Model 2 (AUC = 0.878) and Model 3 (AUC = 0.824). The Hosmer-Lemeshow test yielded a P -value of .7969, and K-fold cross-validation showed an accuracy of 0.8419 and a Kappa value of 0.6780. Conclusion PSWE combined with Scr and eGFR enhances diagnostic accuracy in differentiating renal fibrosis severity in CKD patients, aiding clinicians in making precise clinical decisions. The PSWE-based nomogram demonstrates excellent performance in predicting renal fibrosis severity.