Preclinical Kidney Stones Prediction Model Based On Metabolism Matter In The Body For Study

Objective:The clinical data of patients with kidney stones were retrospectively analyzed to explore the related risk factors of the occurrence of kidney stones,and the binary Logistic regression prediction model was established according to the obtained results.Finally,the model was visualized to better predict the risk of stone occurrence.Methods:A retrospective analysis was performed on 140 patients and 65 patients who underwent physical examination or suffered from other kidney diseases during the same period.The data of 140 patients with kidney stones(stone group)and 65 patients with physical examination or other kidney diseases(control group)were analyzed by univariate and Logistic multivariate analysis,then independent risk factors of kidney stones were screened out.The Logistic regression prediction model was established by stepwise regression method.C index and consistency curve were used to evaluate the effectiveness of the prediction model,and then R language software was used to draw the kidney stone nomogram to predict the risk of kidney stone.Results:Univariate analysis found that EO#,MCHC,TBA,UREA,CREA,UA,Lp(a),CI,SG,RH,WH,UNCX,urine phosphorus,ERY and LEU levels,RBC,HCT,TBIL,CHOL,HDL,LDL,K,urinary potassium,urinary chlorine,urinary creatinine,urinary uric acid,citric acid and OPN were different between the two groups,and the difference was statistically significant(P<0.05).It was found that urine potassium/blood potassium,urine chlorine/blood chlorine,urine creatinine/blood creatinine and urinary uric acid/blood uric acid were also statistically significant between the two groups(P<0.05).Finally,multivariate Logistic regression analysis showed that TBA,RH,Lp(a),MCHC,HDL and urinary uric acid /blood uric acid were independent risk factors for the occurrence of kidney stones.The above independent risk factors were included in the regression equation.The model is P=-21.424+0.376* TBA(μmol/L)+2.235* RH+0.006*Lp(a)(mg/L)+0.067* MCHC(g/L)-2.243*HDL(mmol/L)-0.155* urinary uric acid /blood uric acid.C index was 0.932,indicating that the model was highly differentiated.The consistency curve shows a good consistency between the prediction ability of the model and the actual results.Conclusion:1.TBA,RH,Lp(a),MCHC,HDL and urinary uric acid /blood uric acid were independent predictors of kidney stones;2.The multivariate Logistic regression prediction model was P=-21.424+0.376* TBA(μmol/L)+2.235* RH +0.006*Lp(a)(mg/L)+0.067* MCHC(g/L)-2.243*HDL(mmol/L)-0.155* urinary uric acid /blood uric acid;3.The risk of kidney stones can be predicted easily,quickly and efficiently according to the nomogram.