生物学标志物在急性肾损伤病因鉴别诊断中的作用

Role of biomarkers in differential diagnosis of causes in acute kidney injury patients

目的 探讨生物学标志物在鉴别急性肾损伤（AKI）病因和明确肾损伤部位中的价值,为临床早期干预治疗和判断预后提供依据。方法 选取诊断明确的AKI住院患者103例,根据AKI病因,将患者分为肾前性AKI（PRE组,34例）、肾后性AKI（POST组,9例）和肾实质性AKI（60例）,肾实质性AKI根据肾损伤部位,分为急性肾小管坏死（ATN组,31例）、急性肾小球和肾小血管性损伤（AGV组,15例）、急性肾间质性损伤（AIN组,14例）。收集所有患者的血液和尿液标本。采用ELISA法检测尿中性粒细胞明胶酶相关脂质运载蛋白（NGAL）、尿视黄醇结合蛋白（RBP）、尿IL-6和IL-18水平,采用比色法检测尿N-乙酰-β-D-氨基葡萄糖苷酶（NAG）水平,采用微粒子增强比浊法检测血清半胱氨酸蛋白酶抑制剂C（sCysC）水平,采用酶法检测血清肌酐（sCr）水平,采用酸碱度指示剂蛋白质误差法检测随机尿蛋白水平。比较各组间生物学标志物的水平,并应用ROC曲线的AUC评价各生物学标志物对诊断AKI病因和明确肾损伤部位的敏感度和特异度。结果ATN组、AGV组、AIN组和POST组患者的sCr水平均显著高于PRE组（P值均〈0.05）。PRE组的sCysC水平显著低于肾实质性AKI患者（P〈0.05）,ATN组的sCysC水平显著高于PRE组（P〈0.05）,AIN组和POST组的sCysC水平均显著低于ATN组和AGV组（P值均〈0.05）。PRE组的尿蛋白水平显著低于ATN组和AGV组（P值均〈0.05）,ATN组和AIN组的尿蛋白水平均显著低于AGV组（P值均〈0.05）。PRE组和AIN组患者的尿IL-6水平均显著低于ATN组（P值均〈0.05）。AGV组、AIN组和POST组的尿NGAL水平均显著低于ATN组（P值均〈0.05）。在接受肾脏活组织检查的肾质性AKI患者中,AIN组的sCysC水平显著低于ATN组和AGV组（P值均〈0.05）,AIN组的尿蛋白水平显著低于AGV组（P〈0.05）,ATN组和AIN组的尿NAG水平均显著低于AGV组（P值均〈0.05）,AGV组和AIN组的尿NGAL水平均显著低于ATN组（P值均〈0.05）。sCysC诊断肾实质性AKI的ROC曲线的AUC为0.626（95%CI为0.515-0.737,P〈0.05）,当sCysC水平〉3.26 mg/L时,其诊断肾实质性AKI的敏感度为61.7%,特异度为62.2%。尿NGAL诊断急性肾小管坏死的ROC曲线的AUC为0.684（95%CI为0.563-0.806,P〈0.01）,当尿NGAL〉36.41μg/mL时,其诊断急性肾小管坏死的敏感度为70.0%,特异度为62.0%。尿蛋白诊断急性肾小球和肾小血管性损伤的ROC曲线的AUC为0.843（95%CI为0.688-0.997,P〈0.01）,当尿蛋白水平〉15mg/L时,其诊断急性肾小球和肾小血管性损伤的敏感度为80.0%,特异度为83.9%。结论 生物学标志物可能可应用于AKI病因和肾损伤部位的鉴别诊断,但尚需要进行大样本的临床研究来证实。

Objective To investigate the value of biomarkers in diagnosing the cause and injured location of acute kidney injury （AKI）. Methods A total of 103 hospitalized patients with definite diagnosis of AKI were enrolled in this study. There were 34 cases of prerenal AKI, 9 of postrenal AKI and 60 of renal parenchymal AKI. Of the 60 renal parenchymal AKI patients, there were 31 of acute tubular necrosis （ATN group）, 15 of acute glomerular vascular injury （AGV group）, and 14 of acute interstitial necrosis （AIN group）. Blood and urine samples were collected when patients were diagnosed as AKI. Enzyme-linked immunosorbent assay （ELISA） was used to detect the concentration of urinary biomarkers including neutrophil gelatinase-associated lipocalin （NGAL）, retinol binding protein （RBP）, interleukin （IL）-6 and IL-18. Colorimetric method was used to measure the level of urinary N-acetyl-β-D-glucosaminidase （NAG）. Turbidimetry was applied to examine the concentration of serum cystatin C （Cys C）. Enzymatic analysis was used to measure the level of serum creatinine （sCr）. PH indicator protein error method was used to randomly detect the urine protein. The levels of biomarkers between different causes of AKI patients were compared. Area under the receiver operating characteristic （ROC） curve （AUC） in these biomarkers was used to evaluate the sensitivity and specificity in diagnosing the cause and renal injured location of AKI. Results The levels of sOr in ATN, AGV, AIN and postrenal AKI patients were significantly higher than that in prerenal AKI patients （all P〈0.05）. The concentration of serum CysC in prerenal AKI patients was significantly lower than that in renal parenchymal AKI patients and ATN patients （both P〈0.05）. Meanwhile, the CysC in AIN and postrenal AKI patients was significantly lower than that in ATN and AGV patients （all P〈0.05）. The urinary protein level in prerenal AKI patients was significantly lower than that in ATN and AGV patients （both P〈0.05）. Moreover, the urinary protein in ATN and AIN patients was significantly lower than that in AGV patients （both P〈 0.05）. The levels of urinary IL-6 in prerenal AKI and AIN patients were significantly lower than that in ATN patients （both P〈0.05）. The concentrations of urinary NGAL in AGV, AIN and postrenal AKI patients were significantly lower than that in ATN patients （all P〈0.05）. In renal parenchymal AKI patients who were diagnosed by renal biopsy, the level of serum OysC in AIN patients was significantly lower than that in ATN and AGV patients （both P 〈0.05）. Meanwhile, the urinary protein in AIN patients was significantly lower than that in AGV patients （P〈 0.05）. The concentrations of urinary NAG in ATN and AIN patients was significantly lower than that in AGV patients （both P〈0.05）. The levels of urinary NGAL in AGV and AIN patients were significantly lower than that in ATN patients （,both P〈0.05）. The AUO was 0. 626 when serum CysC were used to diagnose renal parenchymal AKI （95%CI： 0. 515-0. 737, P〈0.05）. Once the concentration of serum CysC was more than 3.26 mg/L, its sensitivity and specificity for the diagnosis of renal parenchymal AKI was 61.7% and 62.2%, respectively. The AUC was 0. 684 when urinary NGAL was used to diagnose ATN （95% Cl=0. 563- 0. 806, P〈0. 01）. If the concentration of urinary NGAL was more than 36.41 μg/mL, its sensitivity and specificity for diagnosis of ATN was 70.0% and 62.0%, respectively. The AUC was 0. 843 when urinary protein was used to diagnose AGV （95% CI= 0. 688-0. 997, P〈0.01 ）. If the concentration of urinary protein was more than 150 mg/dL, the sensitivity and specificity for diagnosis of AGV was 80.0% and 83.9 %, respectively. Conclusion Biomarkers might be used to diagnose the cause and injured location of AKI, but more studies with large samples should be done to confirm the result.