Objective  Based on ultrasound and clinical parameters, a nomogram model was constructed to explore the independent predictive value of non-invasive ultrasound examination on the total MEST-C score of IgA Nephropathy（IgAN） Oxford classification, providing guidance for the establishment of clinical treatment plans.

Methods 226 patients with primary IgAN who were diagnosed by renal biopsy at Nanchong Central Hospital Affiliated to North Sichuan Medical College from January 1, 2017 to April 30, 2022 were included. Ultrasound and clinical parameters were collected and patients were divided into two groups based on a MEST-C total score of ＞2 or ≤2, with a total score of ＞2 as the study outcome. Screen independent predictive factors and establish predictive models through single factor and multi factor logistic regression analysis.

Result  Kidney length（OR=0.938, 95%CI: 0.896−0.982, P=0.006）, urinary red blood cell count（OR=1.003, 95%CI:1.001−1.005, P=0.010）, serum creatinine（Scr）（OR=1.020, 95%CI:1.002−1.038, P=0.027）, and 24-hour urine protein quantification（OR=1.713, 95%CI:1.265-2.319, P=0.001） are independent predictors for evaluating the total MEST-C score of IgAN. Construct a nomogram model using independent predictive factors, with area under curve（AUC） of 0.821（95%CI:0.756−0.887）and 0.780（95%CI:0.656−0.904）for the training and validation sets. Based on the MEST-C total score ＞2 points as the research outcome, a clinical prediction model was established using clinical indicators（including 24-hour urine protein, Scr, and urinary red blood cell count） screened through multivariate logistic regression analysis. The model was compared with the nomogram model that included ultrasound parameters in this study. The results showed that the AUC of the clinical model is lower than that of the nomogram model:the AUC in the training set of the clinical model is 0.819 （95%CI:0.756−0.883）, while the AUC of the nomogram model is 0.821（95%CI:0.756−0.887）. The nomogram model combining clinical ultrasound parameters has better predictive performance in predicting MEST-C scores （＞2） compared to a simple clinical model.

Conclusion  This study constructs a nomogram model based on ultrasound and clinical parameters, which has good predictive performance and can accurately predict the total MEST-C score. It can be used for non-invasive assessment of renal pathological severity in clinical practice and provide guidance for the treatment of IgAN patients.