SHA.LIN、S.T.O.N.E.评分系统与腔道泌尿外科协会临床研究办公室列线图对预测经皮肾镜取石术后结石清除率及围手术期情况的对比

目的 评估和对比SHA.LIN评分系统、S.T.O.N.E.评分系统与腔道泌尿外科协会临床研究办公室(CROES)列线图在预测经皮肾镜取石术(PCNL)后结石清除率及围手术期情况的准确性。方法 回顾性分析90例在2015年1月至2018年3月于中日友好医院泌尿外科收治的行PCNL的肾结石患者的临床数据。收集患者的一般状况、病史情况、结石特征、术中及围手术期情况等。利用术前CT影像,根据SHA.LIN、S.T.O.N.E.评分系统与CROES列线图的标准进行结石评分。术后结石清除状态根据术后1个月的腹部平片进行评估。分析3种结石评分系统与结石清除率、术后并发症、手术时间、估计出血量、住院天数、血红蛋白下降量等临床数据的相关性,并绘制受试者工作特征曲线(ROC),比较3种评分系统的预测准确性。结果 90例患者结石清除率为72.2%(65/90),33例(36.7%)出现术后并发症。手术时间平均(103.1±39.6)min,估计出血量平均(46.1±53.0)ml,总住院天数平均(15.3±5.2)d,术后住院天数平均(8.5±3.4)d,术后血红蛋白下降量(16.1±10.2)g/L。结石清除组和残留组的SHA.LIN评分分别为8.23和10.36分,S.T.O.N.E.评分分别为7.05和8.16分,CROES列线图分别为188.50和143.89分,差异均有统计学意义(P均=0.000)。3项评分均与术后并发症的发生无相关性(P均>0.05)。SHA.LIN评分与估计失血量(P=0.028)、手术时间(P=0.006)、术后住院天数(P=0.001)、血红蛋白下降量(P=0.014)相关;S.T.O.N.E.评分与估计失血量(P=0.047)、手术时间(P=0.012)、术后住院天数(P=0.005)、血红蛋白下降量(P=0.011)相关;CROES列线图与估计失血量(P=0.045)、手术时间(P=0.040)、血红蛋白下降量(P=0.013)相关。在Logistic回归模型中,SHA.LIN评分(OR=2.491)、S.T.O.N.E.评分(OR=3.030)及CROES列线图(OR=0.973)均与结石清除状态相关。SHA.LIN、S.T.O.N.E.评分及CROES列线图的ROC曲线下面积(AUC)分别为0.808(95% CI=0.711~0.905)、0.748(95% CI= 0.632~0.864)和0.770(95% CI=0.664~0.877)。SHA.LIN评分与S.T.O.N.E.评分的AUC差异有统计学意义(P=0.047),CROES与另外两种评分的AUC差异无统计学意义(P>0.05)。结论 SHA.LIN评分、S.T.O.N.E.评分及CROES列线图均能有效预测PCNL术后清石率,SHA.LIN评分对术后清石率的预测能力比S.T.O.N.E.评分更强。但是3项评分均不能预测术后并发症的发生。

Evaluation and Comparison of SHA.LIN,S.T.O.N.E.Nephrolithometry Scoring System,and Clinical Research Office of the Endourological Society Nephrolithometry Nomogram for Predicting Stone Free Rate and Postoperative Outcomes after Percutaneous Nephrolithotomy

Objective To compare the accuracy of SHA.LIN,S.T.O.N.E.nephrolithometry scoring system,and Clinical Research Office of the Endourological Society(CROES)nephrolithometry nomogram in predicting percutaneous nephrolithotomy(PCNL)outcomes including stone free rate(SFR)and perioperative status. Methods The clinical data of 90 patients with nephrolithiasis undergoing PCNL in department of urology,China-Japan Friendship Hospital from January 2015 to March 2018 were retrospectively analyzed.The general data,stone characteristics,operation approaches,and perioperative variables were recorded.SHA.LIN,S.T.O.N.E.score,and CROES nomogram were assigned according to the computed tomography(CT)findings before surgery.Stone free status was evaluated by kidney-ureter-bladder one month after PCNL.The relationships of SHA.LIN score,S.T.O.N.E.score,and CROES score with SFR,postoperative complications,operation time(OT),length of hospital stay(LOS),estimated blood loss(EBL),and decrease of hemoglobin was evaluated.Receiver operating characteristic(ROC)curves were used to analyze the predictive accuracy. Results The SFR was 72.2%(65/90)and postoperative complications occurred in 33 cases(36.7%).The mean OT was(103.1±39.6)min,the mean EBL was(46.1±53.0)ml,the mean LOS was(15.3±5.2)d,the mean postoperative LOS was(8.5±3.4)d,and the mean decrease of hemoglobin was(16.1±10.2)g/L.Stone-free patients had significantly lower SHA.LIN score(8.23 vs. 10.36,P=0.000)and S.T.O.N.E.score(7.05 vs.8.16,P=0.000)and significantly higher CROES score(188.50 vs. 143.89,P=0.000)compared to patients with residual fragments.All these scores were not significantly associated with complications(P>0.05).On the other hand,all these scores were significantly correlated with OT,EBL,and decrease of hemoglobin(SHA.LIN:P_OT=0.006,P_EBL=0.028,P_{decrease of hemoglobin}=0.014;S.T.O.N.E.:P_OT=0.012,P_EBL=0.047,P_{decrease of hemoglobin}=0.011;and CROES:P_OT=0.040,P_EBL=0.045,P_{decrease of hemoglobin}=0.013).SHA.LIN(P=0.001)and S.T.O.N.E.(P=0.005)scores were associated with LOS.Logistic regression analysis revealed that SHA.LIN(OR=2.491),S.T.O.N.E.(OR=3.030),and CROES(OR=0.973)scores were significantly associated with stone-free status.ROC curves in predicting SFR showed that there was significant difference in the areas under the curves(AUC)for the SHA.LIN vs. S.T.O.N.E.score 0.808(95% CI=0.711-0.905)vs. 0.748(95% CI=0.632-0.864),P=0.047].AUC for the CROES score 0.770(95% CI=0.664-0.877)] showed no significantly different for the SHA.LIN score or the S.T.O.N.E.score(P>0.05). Conclusions All these three scoring systems have good predictive accuracy for SFR.SHA.LIN is more precise than S.T.O.N.E.in predicting SFR.However,they can not predict postoperative complications.