临床指标对急性肾损伤伴心肾综合征患者行肾脏替代治疗时机的预测价值

目的 探讨多项指标在预测急性肾损伤伴心肾综合征(cardiorenal syndrome,CRS)患者行肾脏替代治疗(RRT)时机的价值.方法 选取75例心肾综合征住院患者为对象,所有患者给予保守治疗3d,心功能好转者为对照组(n=39),心功能无好转者进入RRT组(n=36).记录患者入院第1天的一般情况,血白蛋白、血红蛋白、血肌酐-Ⅰ、尿素氮-Ⅰ、B型钠尿肽-Ⅰ (BNP-Ⅰ)、24 h尿量-Ⅰ、呋塞米用量(呋塞米-Ⅰ),治疗第4天的呋塞米用量(呋塞米-Ⅱ)、24 h尿量-Ⅱ、肌酐-Ⅱ、尿素氮-Ⅱ、BNP-Ⅱ等指标,计算治疗前后部分指标的比值:24h尿量Ⅱ/Ⅰ、肌酐Ⅱ/Ⅰ、尿素氮Ⅱ/Ⅰ、BNPⅡ/Ⅰ(第1天指标以“Ⅰ”标记,治疗第4天指标以“Ⅱ”标记,指标的动态变化以第4天指标与第1天指标的比值表示,以“Ⅱ/Ⅰ”标记).运用受试者工作特征曲线(ROC)下面积评价上述各指标预测患者行RRT治疗的敏感性与特异性.结果 对照组与RRT组在24h尿量-Ⅰ、24 h尿量-Ⅱ、肌酐Ⅱ/Ⅰ、BNP-Ⅱ、BNPⅡ/Ⅰ方面的差异有统计学意义(均P＜ 0.01).计算各指标ROC曲线下面积,其中24h尿量-Ⅰ(AUC=0.736)、24h尿量-Ⅱ(AUC=0.875)、肌酐Ⅱ/Ⅰ(AUC=0.747)、BNP-Ⅱ(AUC=0.779)、BNPⅡ/Ⅰ(AUC=0.894)在预测患者行RRT治疗方面均有较高价值.对上述阳性指标,当分别选取截点值为:24h尿量-Ⅰ =905ml(敏感度75.0％,特异度94.9％)、24h尿量-Ⅱ=1450 ml(敏感度75.0％,特异度100％)、BNP-Ⅱ=3360 ng/L(敏感度72.2％,特异度100％)、BNPⅡ/Ⅰ=1.37(敏感度75.0％,特异度100％)、肌酐Ⅱ/Ⅰ=1.25(敏感度72.2％,特异度94.4％)时,对于是否行RRT治疗有较高的预测价值.结论 24 h尿量、治疗后的BNP数值及BNP与肌酐的动态变化,可以较好地预测伴CRS的急性肾损伤患者是否要行RRT治疗.

Value of clinical parameters in predicting the initiation of renal replacement therapy in acute kidney injury patients with cardiorenal syndrome

Objective To investigate the value of clinical parameters in predicting the initiation of renal replacement therapy(RRT) in acute kidney injury (AKI) patients with cardiorenal syndrome (CRS). Methods A total of 75 AKI patients hospitalized with CRS were enrolled. All patients received pharmacologic therapy on the beginning 3 days. The patients whose heart function improved were divided into control group (n=39), and the patients whose heart function worsened were divided into RRT group (n=36). Clinical and laboratory data on the first day and the fourth day were collected and analyzed. The factors on the first day were labeled as “Ⅰ”, and those on the fourth day were labeled as “Ⅱ.” The ratio of some parameters calculated were labeled as “Ⅱ/Ⅰ”. Area under curve (AUC) of receiver operating characteristic curve (ROC) of these factors was used to evaluate the sensitivity and specificity in predicting the initiation of RRT. Results The patients in RRT group had significantly higher levels of BNP-Ⅱ, BNPⅡ/Ⅰand creatinineⅡ/Ⅰ(P＜0.01), and lower levels of 24 hours urine volume-Ⅰ and 24 hours urine volume-Ⅱ(P＜0.01). From ROC curve analysis, the AUC of 24 hours urine volume-Ⅰ, 24 hours urine volume-Ⅱ, creatinineⅡ/Ⅰ, BNP-Ⅱlevels and BNPⅡ/Ⅰ to predict RRT were 0.736, 0.875, 0.747, 0.779 and 0.894 respectively. When the cutoff values of 24 hours urine volume-Ⅰ, 24 hours urine volume-Ⅱ, BNP-Ⅱlevels, BNPⅡ/Ⅰ and creatinineⅡ/Ⅰ were 905 ml (sensitivity 75%, specificity 94.9%), 1450 ml (sensitivity 75%, specificity 100%), 3360 ng/L (sensitivity 72.2%, specificity 100%), 1.37 (sensitivity 75%, specificity 100%) and 1.25 (sensitivity 72.2%, specificity 94.4%) respectively, the value of the parameters to predict RRT was high. Conclusions The 24 hours urine volume, BNP levels after treatment and the dynamic changes of BNP levels and creatinine levels can be used as predictors of the initiation of RRT in the AKI patients with CRS.