目的

探讨重症监护病房（ICU）合并急性肾损伤（AKI）患者的预后及相关危险因素。

方法

回顾性分析2006年9月至2011年9月入住大连医科大学附属第一医院ICU的862例患者临床资料，根据急性肾损伤网络（AKIN）标准诊断分期，收集人口学资料、肾功能、尿量、血清钾、APACHEⅡ评分评估脏器衰竭程度及血液净化治疗方式等临床资料，应用logistic回归和COX回归分析影响患者生存和预后的相关危险因素。

结果

ICU中有26.8%（231/862）患者出现AKI，其首位基础病因是重症感染，占28.5%（66/231）。862例ICU患者中共有419例死亡，其中合并AKI的患者病死率为71.0%（164/231），明显高于无AKI患者（40.4%，255/631）。患者病死率与血清肌酐（Scr）水平、治疗前后Scr差值、AKI初发时间、AKI分期、衰竭脏器数量及APACHEⅡ评分均呈正相关。连续性肾脏替代治疗（CRRT）后患者的APACHEⅡ评分、平均动脉压（MAP）、Scr、血尿素氮（BUN）、尿量及血钾水平均较治疗前有明显改善，但与患者病死率无显著相关性。多因素回归分析显示治疗后高血钾（OR = 4.282，95%CI 1.519 ～ 12.070）、高APACHEⅡ评分（OR = 1.318，95%CI 1.192 ～ 1.457）是AKI患者死亡的独立危险因素，而高MAP（OR = 0.972，95%CI 0.946 ～0.999）则是保护因素。COX多因素回归分析表明治疗后Scr、衰竭脏器数量及APACHEⅡ评分均是影响AKI患者生存时间的独立影响因素，而CRRT则是保护因素。

结论

ICU患者的AKI患病率高，发生AKI后患者病死率增高，治疗后高血钾、高APACHEⅡ评分和低MAP是AKI患者死亡的独立危险因素。CRRT能够延长ICU患者住院期间的生存时间，但与AKI患者的病死率无显著相关性。

Objective

To assess the prognosis and risk factors of patients with acute kidney injury (AKI) in intensive care unit (ICU).

Methods

We performed a retrospective study of 862 patients in the ICU of the First Affiliated Hospital of Dalian Medical University from September 2006 to September 2011. AKI were defined by Acute Kidney Injury Network (AKIN) criteria. The clinical data were collected including demography, renal function, urine output, serum potassium, APACHEⅡscore, and continuous renal replacement therapy (CRRT) treatment scheme. Logistic regression and COX regression were used to analyze the risk factors relevant to prognosis and survival of the patients.

Results

26.8%of patients (231/862) developed AKI, the leading cause of which was severe infection (28.5%, 66/231).The all-cause mortality of ICU patients was 48.6% (419/862). The mortality was higher in patients of AKI than in non-AKI patients [71.0% (164/231) vs. 40.4% (255/631)]. Moreover, the mortality increased with primary serum creatinine (Scr), the changes of Scr before and after the treatment, the onset time of AKI,the severity of AKI, the number of organs involved, and the APACHEⅡscore. There were significant improvements in AKI patients after CRRT, including APACHEⅡscore, mean artery pressure (MAP), Scr,blood urea nitrogen, urine output, and serum potassium. But the mortality was irrelevant to whether CRRT was used or not. Logistic analysis showed that hyperkalemia (OR = 4.282, 95%CI = 1.519 - 12.070), high APACHEⅡscore (OR = 1.318，95%CI = 1.192 - 1.457) were the independent risk factors for mortality,while higher MAP was associated with lower mortality (OR = 0.972, 95%CI = 0.946 - 0.999). Cox regression analysis indicated that the number of organs involved and APACHEⅡscore were influencing factors for the survival time of patients, and CRRT was associated with long survival time.

Conclusion

ICU patients had higher incidence and mortality of AKI. Hyperkalemia, higher APACHEⅡscores and lower MAP after treatment were the risk factors for mortality of AKI patients. Though CRRT was not beneficial to AKI, it could improve the survival time of patients during the hospitalization.