摘要目的:探讨快速序贯器官衰竭评分(qSOFA)对经皮肾镜取石术(PCNL)后脓毒症休克的预测价值。方法:回顾性分析2018年5月至2019年10月苏州大学附属第一医院收治的309例肾结石行PCNL患者的临床资料。男192例，女117例；年龄(51.4±12.8)岁；伴高血压病82例，糖尿病23例；78例既往有泌尿系手术史；88例术前尿细菌培养阳性。102例经腹部CT和腹部X线平片(KUB)诊断为鹿角形结石。309例术中8例(2.6%)采用多通道。术后24 h对所有患者进行qSOFA和全身炎症反应综合征(SIRS)评分，将qSOFA≥2分和SIRS评分≥2分作为诊断脓毒症休克的阳性标准。绘制受试者工作特征(ROC)曲线，比较qSOFA、SIRS评分对脓毒症休克的诊断价值。应用单因素logistic回归筛选出PCNL术后发生脓毒症的相关危险因素，并纳入多因素logistic回归进行校正，应用校正后的多因素logistic回归分析两种评分标准对PCNL术后相关临床结局指标(术后住院时间、术后输血、术后再干预、结石残留、30 d内计划再入院例数、30 d内非计划再入院例数)的预测价值。结果:本研究309例的中位手术时间为85(56，115)min。309例中，23例(7.4%)术后符合qSOFA阳性，84例(27.2%)符合SIRS评分阳性。7例(2.3%)术后进入ICU并诊断为脓毒症休克，其中6例qSOFA、SIRS评分均阳性，1例qSOFA、SIRS评分均为阴性。术后结石成分分析结果显示64例(20.7%)为感染性结石。术后复查KUB显示179例(57.9%)存在结石残留。术后中位住院时间为7(6，9)d。10例(3.2%)术后行输血治疗。9例(2.9%)术后进行再干预治疗。30 d内计划、非计划再入院例数分别为41例(13.3%)和16例(5.2%)。ROC曲线显示qSOFA和SIRS评分的曲线下面积(AUC)分别为0.900和0.799。qSOFA诊断脓毒症休克的特异性、阳性似然比、阳性预测值均明显高于SIRS评分[94.4%(285/302)与74.2%(224/302)，15.23与3.32，26.1%(6/23)与7.1%(6/84)，均 P<0.001]，差异有统计学意义。qSOFA和SIRS评分的敏感性均为85.7%(6/7)，阴性似然比分别为0.15和0.19，阴性预测值分别为99.7%(285/286)和99.6%(224/225)，差异均无统计学意义( P>0.05)。单因素logistic回归分析结果显示，qSOFA阳性和SIRS评分阳性均与尿细菌培养阳性、结石大小、鹿角形结石、感染结石、既往泌尿系手术史、手术时间密切相关(均 P<0.05)。多因素logistic回归分析结果显示，术后住院时间( OR＝1.237，95% CI 1.048～1.459， P＝0.012)和术后输血( OR＝8.265，95% CI 1.409～48.481， P＝0.019)是qSOFA阳性的危险因素，而术后临床结局指标与SIRS评分阳性均无明显相关性(均 P>0.05)。 结论:与SIRS评分比较，qSOFA能够更好地预测PCNL术后脓毒症休克的发生。

abstractsObjective:To evaluate the predictive value of the quick sequential organ failure assessment(qSOFA) score in septic shock after percutaneous nephrolithotomy(PCNL).Methods:309 patients who underwent PCNL at the First Affiliated Hospital of Soochow University between May 2018 and October 2019 were retrospectively reviewed. Among them, there were 192 men and 117 women, whose mean age was (51.4±12.8)years (range from 20 to 79 years). There were 82 cases(26.5%) of hypertension and 23 cases(7.4%) of diabetes. There were 88 patients(28.5%) with positive preoperative urine culture.102 patients(33.0%) were diagnosed with staghorn calculi by abdominal CT and urinary tract abdominal plain film(KUB).78 patients(25.2%) had a history of urinary surgery. The qSOFA and SIRS were evaluated to all patients within 24 h after PCNL and the best diagnostic criteria was considered as qSOFA≥2 and SIRS≥2. Receiver operating characteristic(ROC) curves were constructed and the areas under the curve(AUC) were calculated to compare the discriminatory ability of qSOFA and SIRS with the post-PCNL septic shock. A univariate logistic regression analysis was used to identify the covariates associated with post-PCNL sepsis. Then adjusted multivariate analysis was used to identify the predictive value of positive qSOFA and SIRS for the postoperative clinical outcomes including postoperative hospitalization days, postoperative blood transfusion, postoperative re-intervention, residual stone, planned readmission within 30 days and unplanned readmission within 30 days.Results:Among the 309 patients who underwent PCNL, 23 patients(7.4%) met the positive qSOFA criterion while 84 patients(27.2%) developed to SIRS. 7 patients(2.3%) were admitted to ICU after operation and were eventually diagnosed as septic shock, among which 6 patients met the criteria of qSOFA and SIRS. 8 patients(2.6%) underwent multi-channel operation. The median operative time of 309 patients was 85(56, 115) min. Postoperative calculus composition analysis showed that 64 patients(20.7%) were infectious calculi. Postoperative KUB showed residual calculi in 179 patients (57.9%). The median postoperative hospital stay was 7(6, 9) days. 10 patients(3.2%) received blood transfusion. 9 patients(2.9%) received re-intervention after surgery. There were 41 patients (13.3%) of planned readmissions and 16 cases (5.2%) of unplanned readmissions within 30 days. The AUC of qSOFA and SIRS was 0.900 and 0.799 respectively. The qSOFA had a higher specificity, positive likelihood ratio and positive predictive value(94.4%, 15.23, 26.1%)than that of SIRS(74.2%, 3.32, 7.1%)for septic shock. In univariate logistic regression analysis significant associations were observed between positive urine culture, stone size, staghorn stones, struvite stones, surgery history, operation time and sepsis after PCNL. Multivariate logistic regression analysis revealed that postoperative length of stay( OR=1.237, 95% CI 1.048-1.459, P=0.012) and postoperative transfusion( OR=8.265, 95% CI 1.409-48.481, P=0.019) were closely associated with qSOFA after adjusting for covariates shown to be related to post-PCNL sepsis mentioned above. Conclusions:The qSOFA could be superior to SIRS in predicting septic shock after PCNL.