尿pH值、血尿酸及相关临床指标对尿路感染性结石的诊断价值及其列线图模型的构建

The diagnostic value of urine pH,serum uric acid,and related clinical indicators for urinary infection stones and the construction of a nomogram model

目的探讨尿pH值、血尿酸及相关临床指标在尿路感染性结石诊断中的价值,并构建预测感染性结石的列线图模型。方法回顾性分析徐州市中心医院2018年8月至2023年11月收治的432例尿路结石患者的临床资料。男289例,女143例;年龄(52.72±13.46)岁;高血压病98例;糖尿病67例;复发性结石100例;肾结石152例,输尿管结石242例,膀胱结石38例;尿细菌培养阳性97例。根据术后结石成分分析结果分为感染性结石组与非感染性结石组,比较两组一般资料的差异。采用单因素及多因素logistic回归分析评估尿pH值、血尿酸及相关临床指标对尿路感染性结石的诊断价值。应用受试者工作特征(ROC)曲线和曲线下面积(AUC)评估尿pH值、血尿酸值及联合指标对术前诊断尿路感染性结石的临床意义,并构建列线图预测模型。结果432例中感染性结石127例,非感染性结石305例。感染性结石患者较非感染性结石患者的尿pH值更高[7.0(6.5,7.5)与6.0(5.5,6.5),P<0.001]、血尿酸值更低[(301.38±70.12)μmol/L与(358.88±88.99)μmol/L,P<0.001]、女性比例更高[55.1%(70/127)与23.9%(73/305),P<0.001]、年龄更低[(48.36±14.83)岁与(53.12±12.61)岁,P<0.001]、复发性结石比例更高[34.6%(44/127)与18.4%(56/305),P<0.001]、尿细菌培养阳性比例更高[29.9%(38/127)与19.3%(59/305),P=0.016]、尿亚硝酸盐阳性比例更高[18.9%(24/127)与6.3%(19/305),P<0.001]。单因素logistic回归分析结果显示,尿pH值、血尿酸、性别、年龄、复发性结石、尿细菌培养、尿亚硝酸盐与尿路感染性结石相关(P<0.05)。多因素logistic回归分析结果显示,尿pH值(OR=4.836,95%CI3.342~6.997)、女性(OR=2.320,95%CI1.286~4.186)、复发性结石(0R=2.225,95%CI1.208~4.101)、尿细菌培养阳性(0R=2.061,95%CI1.094~3.883)、血尿酸(0R=0.992,95%CI0.949~0.989)、年龄(0R=0.969,95%CI0.949~0.990)是尿路感染性结石的独立危险因素(均P<0.05)。上述6个指标联合诊断价值最高(AUC=0.874,95%CI0.837~0.911),其次为尿pH值(AUC=0.818,95%CI 0.778~0.858)和血尿酸(AUC=0.704,95%CI0.652~0.756)。将上述6个指标构建列线图模型,Bootstrap法重抽样1000次后R0C曲线的平均AUC为0.864(95%CI0.828~0.900),校准曲线显示预测曲线与理想曲线贴合良好,平均绝对误差为0.005,Hosmer-Lemeshow检验P>0.05。临床决策曲线分析显示,当模型的阈概率值0.01时,模型有更高的临床净获益。结论尿pH值、血尿酸与尿路感染性结石密切相关,二者与性别、年龄、复发性结石、尿细菌培养结果共同构建的列线图模型可预测尿路感染性结石的发生概率,有较高的临床应用价值。

Objective To investigate the diagnostic value of urine pH,serum uric acid and related clinical indicators in the diagnosis of urinary infection stones,and to construct a prediction nomogram.Methods The clinical data of 432 patients with urinary calculi admitted to Xuzhou Central Hospital from August 2018 to November 2023 were retrospectively analyzed.The study included 289 males and 143 females,with an average age of(52.72±13.46)years old.Among the patients,there were 98 cases of hypertension,67 cases of diabetes,and 100 cases of recurrent calculi.Kidney stones were present in 152 cases,ureteral stones in 242 cases,and bladder stones in 38 cases.Urine bacterial culture yielded positive results in 97 cases.According to the results of postoperative stone composition analysis,the two groups were categorized as infection and no-infection stone groups,and the differences in general data between the two groups were compared.Univariate and multivariate logistic regression analysis were conducted to assess the diagnostic value of urine pH,serum uric acid,and related clinical indicators for urinary infection stones.Receiver operating characteristic(ROC)curve and area under curve(AUC)were utilized to evaluate the clinical significance of urine pH,serum uric acid,and combined indexes in preoperatively diagnosing urinary infection stones,as well as constructing a nomogram prediction model.Results There were 127 cases of infection stones and 305 cases of no-infection stones.The infection stone group exhibited a higher urine pH value[7.0(6.5,7.5)vs.6.0(5.5,6.5),P<0.001],lower serum uric acid levels[(301.38±70.12)vs.(358.88±88.99)μmol/L,P<0.001],a higher proportion of females[55.1%(70/127)vs.23.9%(73/305),P<0.001],younger age[(48.36±14.83)vs.(53.12±12.61)years old,P<0.001],a higher proportion of recurrence stones[34.6%(44/127)vs.18.4%(56/305),P<0.001],and a higher rate of positive urine bacteria culture[29.9%(38/127)vs.19.3%(59/305),P=0.016]and nitrite test results[18.9%(24/127)vs.6.3%(19/305),P<0.001].Univariate logistic regression analysis revealed that urine pH value,serum uric acid levels,gender,age,recurrent stones,urine bacterial culture,and urine nitrite were associated with urinary infection stones(P<0.05).Multivariate logistic regression analysis revealed that urine pH value(OR=4.836,95%CI 3.342-6.997),female gender(OR=2.320,95%CI 1.286-4.186),recurrent stones(0R=2.225,95%CI 1.208-4.101),positive urine bacterial culture(0R=2.061,95%CI 1.094-3.883),serum uric acid(0R=0.992,95%CI 0.949-0.989),age(OR=0.969,95%CI 0.949-0.990)were independent risk factors for urinary infection stones(P<0.05).The combined diagnostic value of six indicators was the highest,with an AUC of 0.874(95%CI 0.837-0.911).Following this,urine pH exhibited an AUC of 0.818(95%CI 0.778-0.858),while serum uric acid had an AUC of 0.704(95%CI 0.652-0.756).The nomogram model was successfully constructed based on the six indicators.The mean AUC of the ROC curve after 1000 resamples of the Bootstrap method was 0.864(95%CI 0.828-0.900),and the calibration curve showed that the predicted curve fit the ideal curve well,with a mean absolute error of 0.005 and a Hosmer-Lemeshow test of P>0.05.Clinical decision curve analysis(DCA)showed that the model had a higher net clinical benefit when the model had a threshold probability value≥0.01.Conclusions Urine pH and serum uric acid are closely related to urinary infection stones.A nomogram model combining these factors with gender,age,recurrent stones,and urine culture results can effectively predict the probability of infection-related stones,providing significant clinical value.