RIFLE，AKIN和KDIGO三种急性肾损伤诊断标准在百草枯中毒患者中的应用

发布时间：2018-06-12 15:28

本文选题：急性肾损伤 + 百草枯中毒　；参考：《安徽医科大学》2016年硕士论文

【摘要】：目的探讨RIFLE(Risk,Injury,Failure,Loss of kidney function and End-stage renal failure),急性肾损伤网络(Acute Kidney Injury Network,AKIN)和全球改善肾脏疾病预后(Kidney Disease:Improving Global Outcomes,KDIGO)三种急性肾损伤(acute kidney injury,AKI)诊断标准对百草枯(paraquat,PQ)中毒患者AKI的诊断效率及对患者死亡的预测能力。方法选取2010年11月至2015年4月在上海市第十人民医院肾脏科就诊的PQ中毒患者112例。根据改良的MDRD(Modification of Diet in Renal Disease)公式计算基础血肌酐值。依据三种AKI诊断标准,在每种标准下均将患者分为非AKI组与AKI组;并依据各诊断标准中的分级标准,进一步将各AKI组分为Risk期、Injury期和Failure期(RIFLE标准),或1期、2期和3期(AKIN和KDIGO标准)。各患者均随访中毒后90天,观察肾功能进展及患者存活情况。用卡方检验比较不同AKI诊断标准的诊断效率以及各组间的死亡率,用Kaplan-Meier法进行生存曲线分析,用Logistic回归评估三种诊断标准各组及其分期之间的患者的死亡风险,用受试者工作特征曲线(ROC)下面积(AUC)比较三种诊断标准对患者死亡的预测能力。P0.05为差异有统计学意义。结果在112例PQ中毒患者中,应用AKIN诊断标准诊断AKI的发生率(31.3%,35/112)最低,与RIFLE(47.3%,53/112)和KDIGO(54.5%,61/112)标准相比,差异有统计学意义(P0.05);而RIFLE和KDIGO两标准之间,差异无统计学意义(P=0.285)。与生存组相比,死亡组患者的AKI(依据三种诊断标准分别诊断)发生率均显著升高(P0.001);与非AKI组相比,AKI组患者死亡率均显著升高(P0.001)。生存曲线分析显示,三种诊断标准下AKI组患者的生存率均较非AKI组显著降低(P0.001)。Logistic回归显示:依据RIFLE和KDIGO标准进行分级,可能与患者的死亡关系更密切;而根据两标准的AKI分级标准,风险期(Risk期)/1期患者的死亡风险与非AKI组患者相比无显著增加。AUC分析显示:AKIN诊断标准(AUC=0.692)对患者的死亡预测能力最低,与RIFLE(AUC=0.840)和KDIGO(AUC=0.861)标准相比,差异有统计学意义(P0.05);而RIFLE和KDIGO标准相比,差异无统计学意义(P=0.700)。结论RIFLE和KDIGO诊断标准均适用于PQ中毒患者AKI的诊断及对患者死亡的预测,而AKIN标准的诊断效率及对死亡的预测能力均较差。不同病因所致的AKI可能需要依据不同的AKI诊断标准。在PQ中毒中,AKI可以作为患者预后的一个预测指标;但AKI分级后,风险期(RIFLE标准)和1期(KDIGO标准)不能作为PQ中毒预后的预测因子。
[Abstract]:Objective to study the diagnostic efficiency of acute of kidney function and injury-induced renal (AKI) in patients with paraquat PQs (acute renal injury network) and three diagnostic criteria for acute renal injury (acute kidney injuryAkei), acute Kidney injury network (ARN) and Kidney DiseaseImproving Global improvement (KDIGO) in the Global improvement of Renal Disease prognosis (KDIGO) in patients with acute renal injury (RIFLELE): loss of kidney function and end stage renal failure, acute kidney injury network (ARN) and Kidney DiseaseImproving Global improvement (KDIGO). The ability to predict death. Methods 112 patients with PQ poisoning were selected from November 2010 to April 2015 in Renal Department of Shanghai Tenth people's Hospital. The basic creatinine value was calculated according to the modified MDRD Modification of Diet in Renal Diseaseformula. According to the three AKI diagnostic criteria, patients were divided into non-AKI group and AKI group under each criteria, and according to the classification criteria of each diagnostic criteria, The AKI components were further divided into risk stage injury and failure stage and RIFLE standard, or stage 1, stage 2 and stage 3, AKIN and KDIGO standards. All patients were followed up 90 days after poisoning to observe the progression of renal function and survival of the patients. The diagnostic efficiency and mortality of different AKI diagnostic criteria were compared by chi-square test. Kaplan-Meier method was used to analyze the survival curve. Logistic regression was used to evaluate the risk of death among the three diagnostic criteria. The area under the operating characteristic curve (ROC) was used to compare the predictive ability of the three diagnostic criteria for the death of patients. Results among 112 patients with PQ poisoning, the incidence of diagnosis of AKI by AKIN diagnostic criteria was 31.3% / 112), which was lower than that of RIFLER 47.3R 53 / 112 and KDIGO 54.55.There was no significant difference between RIFLE and KDIGO (P 0. 285) and between RIFLE and KDIGO (P 0. 285). Compared with the survival group, the incidence of AKI (diagnosed according to the three diagnostic criteria) in the death group was significantly higher than that in the non-AKI group, and the mortality rate in the AKI group was significantly higher than that in the non-AKI group. Survival curve analysis showed that the survival rate of AKI group was significantly lower than that of non-AKI group under the three diagnostic criteria (P 0.001). Logistic regression analysis showed that according to RIFLE and KDIGO criteria, the survival rate of AKI patients was significantly lower than that of non-AKI group. According to the two AKI classification criteria, the risk of death of patients in risk period and risk phase 1 was not significantly increased compared with those in non-AKI group. AUC analysis showed that the death prediction ability of patients was the lowest according to the diagnostic criteria of: AUC 0.692), compared with the criteria of RIFLELEA AUC 0.840) and KDIGOAUC 0.861), AUC analysis showed that the mortality prediction ability of the patients was the lowest, compared with that of RIFLEX AUC 0.840 and KDIGOAUC 0.861. There was significant difference between RIFLE and KDIGO, but there was no significant difference between RIFLE and KDIGO. Conclusion both RIFLE and KDIGO diagnostic criteria are applicable to the diagnosis of AKI and the prediction of death in patients with PQ poisoning. AKI caused by different etiology may need to be based on different AKI diagnostic criteria. In PQ poisoning, AKI can be used as a predictor of prognosis, but after AKI classification, RIFLE criteria and KDIGO criteria can not be used as prognostic factors of PQ poisoning.
【学位授予单位】：安徽医科大学
【学位级别】：硕士
【学位授予年份】：2016
【分类号】：R595.4;R692

【相似文献】