Objective To identify the risk factors associated with uric acid stones,construct a nomogram model for predicting the occurrence of the disease,and evaluate its predictive performance.Methods A retrospective analysis was conducted on the general and clinical data of 876 patients who underwent surgical treatment for stones at the Department of Urology,the First Affiliated Hospital of Soochow University,during Jan.2020 and Dec.2022.Based on the analysis results of stone composition,the patients were divided into the uric acid stone group(n=82)and non-uric acid stone group(n=794).All patients were then randomly split into the training group(n=526)and validation group(n=350)in a ratio of 6∶4.The training group underwent LASSO regression,univariate,and multivariate logistic regression analyses to identify predictive factors associated with the occurrence of uric acid stones.Based on the factors,a nomogram model was constructed.The performance of the model was evaluated using the validation group data by comparing it with models from other research centers.Results LASSO regression,univariate,and multivariate logistic regression analyses revealed that cystatin C,urine pH,and stone CT values were predictive factors for uric acid stones.The area under the receiver operating characteristic curve(AUC)of the model was 0.968 for the training group and 0.956 for the validation group.Compared to other models,this model showed better predictive performance.The integrated discrimination improvement(IDI)and net reclassification index(NRI)in the training group were 0.420 0(95％CI:0.328 2-0.511 8),P＜0.001,and 0.484 2(95％CI:0.321 3-0.647 2),P＜0.001,respectively.In the validation group,the IDI and NRI were 0.405 9(95％CI:0.330 7-0.481 1),P＜0.001,and 0.365 3(95％CI:0.211 6-0.519 0),P＜0.001,respectively.Conclusion The nomogram model based on cystatin C,urine pH,and stone CT values can predict the occurrence of uric acid stones more accurately than other models,and can serve as a clinical supportive tool for the treatment and prevention of stone recurrence.
维普
万方数据