AKIN标准联合APACHEⅡ和SOFA评分对心脏术后急性肾损伤的预后评估  

Prognostic investigation of AKIN criteria combined with APACHE Ⅱ and SOFA scoring system in acute kidney injury after cardiac surgery

作　　者：晋玮[1] 滕杰[1] 方艺[1] 刘中华[1] 沈波[1] 徐雁[1] 衡艳艳[1] 杨兆华[2] 王春生[2] 丁小强[1] 

机构地区：[1]复旦大学附属中山医院肾内科,上海200032 [2]复旦大学附属中山医院心外科,上海200032

出　　处：《中华肾脏病杂志》2011年第3期170-175,共6页Chinese Journal of Nephrology

基　　金：教育部国家“211工程”重点学科建设项目（三期）（211XK20）;上海市医学发展基金重点研究课题（2003ZD001）;上海市科委重大项目（08dz1900602）

摘　　要：目的探讨急性肾损伤网络（AKIN）制定的急性肾损伤（AKI）诊断标准联合急性生理与慢性健康状况评分Ⅱ（APACHEⅡ）和序贯器官衰竭评估（SOFA）评分对心脏术后AKI的预后评估价值。方法前瞻性收集2009年4月至8月期间在本院行心脏手术患者的临床资料，采用AKIN标准对心脏术后患者进行AKI诊断和分期；根据患者术后第1个24h内的生理指标最差值进行APACHEⅡ和SOFA评分，并用受试者工作特征（ROC）曲线和Hosmer—Lemeshow拟合优度检验评价3项评估系统的分辨力和校准力。以Logistic多元回归分析它们对预后的影响。结果993例患者中309例术后出现AKI，发病率为31．1％。患者AKI诊断日和首次达AKIN最高分期日距手术的中位间隔时间分别为1d和2d。AKIN1、2、3期患者的APACHEⅡ及SOFA评分均高于非AKI患者（P〈0．01），且分值与AKIN分期呈正相关（APACHE11r=0．37，P〈0．01；SOFAr=0．42，P〈0．01）。病死率亦随AKIN分期升高而升高。非AKI组、AKIN1期患者根据APACHE11分值计算所得的校正预计病死率（PDR—A）明显高于实际病死率（P〈0．01），而AKIN3期PDR—A则低于实际病死率（P〈0．01）。APACHEⅡ、SOFA评分及AKIN分期的ROC曲线下面积（AUC）均〉0．8，且Hosmer—Lemeshow拟合优度检验提示模型拟合较好。Logistic多元回归分析显示APACHEⅡ≥19（OR=4．26）和AKIN3期（OR=76．15）是心脏术后患者院内死亡的独立预测指标。结论AKIN标准能在心脏术后早期对患者进行AKI诊断和分期，且在一定程度上发挥预后评估的作用。APACHEⅡ和SOFA在术后第1个24h内的评分能较好区分病情的严重程度。3者作为预测模型均显示了对于整体预后较好的分辨力和校准力，且APACHEⅡ≥19和AKIN3期是心脏术后患者院内死亡的独立预测指标。需注意APACHEⅡ计算所得的PDR—A与AKIN不同分期组实际病死率相比存在偏差，动态评分可Objective To explore the prognostic value of Acute Kidney Injury Network （AKIN） criteria combined with Acute Physiology and Chronic Health Evaluation Ⅱ （APACHEⅡ ） and Sequential Organ Failure Assessment （SOFA） scoring system in acute kidney injury （AKI） after cardiac surgery. Methods Clinical data of patients who underwent open-heart surgery in Zhongshan Hospital, Fudan University from April 2009 to August 2009 were prospectively collected. AKI after cardiac surgery was classified by AKIN staging system. APACHE Ⅱ and SOFA scores were evaluated according to the worst value of physiologic variables in the 1st 24 h after surgery. Discrimination and calibration of these three models were assessed by receiver operating characteristic （ROC） curve and Hosmer-Lemeshow goodness-of-fit test. Besides, their effects on in- hospital mortality were evaluated by muhivariate Logistic regression analysis. Results Of the 993 admissions, 309 patients developed AKI and the incidence was 31.1%. The median time that developed postoperative AKI and reached the Scr peak were 1 d and 2 d respectively. Either APACHE Ⅱ or SOFA scores, which was positively correlated with the severity of AKI （APACHE I r=0.37, P〈0.01; SOFA r=0.42, P〈0.01） was higher in AKI patients compared with that in non- AKI patients （P〈0.01）. The mortality rose corresponding to the severity of kidney injury. However, the predicted death rate-adjusted （PDR-A） calculated by APACHE Ⅱ scores was higher than the actual value in non-AKI patients and AKIN stage 1 （P〈0.01）, while it was lower in AKIN stage 3 （P〈0.01）. The areas under the ROC curve of APACHE Ⅱ, SOFA and AKIN criteria were all above 0.8 and the results of Hosmer-Lemeshow goodness-of-fit test indicated good calibration of three models. Multivariate analysis showed that APACHE Ⅱ ≥ 19 （OR=4.26） and AKIN stage 3 （OR= 76.15） were independent predictors of in-hospital mortality. Conclusions AKI can be classified by AKIN criteria in the early stag

关 键 词：肾功能不全 急性 心脏外科手术 预后 急性肾损伤网 急性生 理与慢性健康状况评分Ⅱ 序贯器官衰竭评估评分 

分 类 号：R654.2[医药卫生—外科学]