Establishment And Application Of Prediction Model And Nomogram For Uric Acid Stones

Research Background:Globally,with the continuous improvement of lifestyle and dietary habits,the incidence of uric acid calculi is increasing.According to the current research results,the pathogenesis of uric acid calculi is unclear.There are reports that diabetes caused by insulin resistance is an important risk factor for uric acid calculi formation and recurrence.However,insulin resistance is not only found in diabetic patients,all patients with abnormal glucose metabolism(including pre-diabetes and diabetes)have varying degrees of insulin resistance,and the relationship between abnormal glucose metabolism and uric acid calculi remains unclear.Therefore,we analyzed the data of 355 patients who visited the Urology Department of Qingdao Municipal Hospital in Shandong Province from January 2016 to December 2021 for related analysis.The purpose of this study is to clarify the correlation between abnormal glucose metabolism and uric acid calculi and to find the risk factors for uric acid calculi and to establish a prediction model for uric acid calculi,in order to provide relevant reference for clinical prevention,diagnosis and treatment of urinary uric acid calculi.Objective: The purpose of this study is to clarify the correlation between abnormal glucose metabolism and uric acid calculi and to find the risk factors for uric acid calculi and to establish a prediction model for uric acid calculi.Methods : A total of 267 patients who visited to Qingdao Municipal Hospital in Shandong Province from January 2016 to June 2018 with urinary calculi were selected as the training group,and the general information,past history,relevant laboratory tests and urinary calculi components of the patients were collected.The training group(267 cases)were divided into uric acid calculus group(observation group N=43)and other calculus group(control group N=224)according to whether uric acid calculi occurred.The differences of clinical data was analyzed in the two groups of patients.The chi-square test and Spearman correlation analysis were used to explore the relationship between glucose metabolism and uric acid stones.Univariate and multivariate Logistic regression analysis were used to analyze the risk factors and predictive factors of uric acid calculi,and establish a prediction model for uric acid calculi.In addition,88 patients who visited to Qingdao Municipal Hospital in Shandong Province from July 2018 to December 2021 with urinary calculi were selected as the verification group.And the prediction model was used to validation in the validation people.The discrimination and calibration were used to evaluate the model.The discriminative degree of the model was evaluated by the ROC-AUC.The calibration degree of the model was evaluated by the Hosmer-Lemeshow goodness-of-fit test.In this experiment,data analysis was performed using SPSS.24 software and R 4.1.2 software.Categorical data was analyzed by the chi-square test.Quantitative data was analyzed by the t-test or nonparametric test depending on distribution and variance.Results: In the comparison of clinical data,it was found that the differences in uric acidbase,serum uric acid,serum creatinine and whether they were associated with abnormal glucose metabolism were statistically significant(P<0.05).In the correlation analysis between glucose metabolism and uric acid calculi,the results showed that the probability of uric acid calculi in patients with abnormal glucose metabolism combined with urinary calculi was higher than that in patients with normal glucose metabolism combined with urinary calculi,and the difference was significant(29.89% vs.9.44% P<0.05),and the results of Spearman correlation analysis showed that the correlation coefficient was 0.261(P<0.05).Univariate Logistic regression analysis showed that uric acidity(OR=0.254,95%CI 0.131-0.492,P<0.05),serum creatinine(OR=1.028,95%CI 1.010-1.046,P<0.05),abnormal glucose metabolism(OR=4.087,95%CI 2.074-8.054,P<0.05),and serum uric acid(OR=1.008,95%CI 1.005-1.012,P<0.05)were the risk factors of uric acid calculi in patients.Multivariate Logistic regression analysis showed that uric acidity(OR=0.361,95%CI 0.176-0.741,P<0.05),blood uric acid(OR=1.007,95%CI 1.003-1.011,P<0.05),abnormal glucose metabolism(OR=3.078,95%CI 1.432-6.613,P<0.05),and serum creatinine(OR=1.025,95%CI 1.006-1.044,P<0.05)were the risk factors and predictors of uric acid calculi in patients.According to the above results,the prediction model for uric acid stones is P=1/1+EXP(-0.431 + 1.124* abnormal glucose metabolism-1.020*uric acid alkalinity + 0.025*serum creatinine + 0.007*blood uric acid).The ROC-AUC was 0.810(95% CI 0.734-0.885)in the training people,and after 200 Bootstrap replicates,the mean ROC-AUC was 0.769.The ROC-AUC was 0.637(95%CI 0.484-0.789)in the validation people.The Hosmer-Lemeshow test results of the training group and the validation group were χ~2=0.969(P>0.05)and χ~2=0.769(P>0.05).Results of clinical decision curve analysis showed that the model was higher in terms of net clinical benefit to patients.Conclusion: Abnormal glucose metabolism is closely related to the occurrence of uric acid stones.Abnormal glucose metabolism,blood uric acid,and serum creatinine,urine p H,have been identified as independent risk factors and predictive factors for uric acid calculi.The probability prediction model which was established based on the risk factors and predictive factors has good discrimination and calibration.It is helpful to find high-risk groups of uric acid stones and realize individualized prevention and treatment of uric acid calculi.