Evaluation of cystatin C and β-2 microglobulin as markers of renal function in patients with type 2 diabetes mellitus
Evaluation of cystatin C and β-2 microglobulin as markers of renal function in patients with type 2 diabetes mellitus
复制标题
DOI：
10.1016/s1056-8727(02)00177-0
复制
发表时间：
2003-05-01
影响因子：
3
通讯作者：
Abdella, N
中科院分区：
文献类型：
--
作者：
Mojiminiyi, OA;Abdella, N
背景资料：尽管最近的研究表明，血清胱抑素C（CC）是一个更好的标记GFR比普遍使用的血清肌酐，其临床效用仍在评估中。研究方法：为了评价其在2型糖尿病（DM）患者中的有用性，测定了105例（38例男性，67例女性）科威特2型糖尿病患者的CC、β 2微球蛋白（B2 M）和肌酐的血清浓度。将结果与肌酐清除率（Ccr）进行比较，肌酐清除率（Ccr）采用Cockroft-Gault公式测量（mCcr）和估计（eCcr），并与24 h尿蛋白和清晨尿白蛋白/肌酐排泄比相关。结果如下：在具有eCcr和mCcr结果的患者中（n = 51），eCcr和mCcr显示彼此显著相关（r = .86，P < .0001），两者之间无显著差异。在所有患者（n = 105）中，CC和B2 M显示彼此之间（r = 0.82，P <0.0001）以及与血清肌酐浓度（r = 0.77和0.84，P <0.0001）显著相关。血清CC、B2 M和肌酐与eCcr呈显著负相关（r分别为-0.63、-61和-0.76）（P < .001）。校正年龄和性别后的偏相关改善了血清肌酐与eCcr的相关性（r = -.81，P < .0001），但CC和B2 M与eCcr的相关性无显著变化（分别为r = -.65，P < .0001和r = -.62，P < .0001）。血清CC、B2 M和肌酐检测eCcr变化的受试者工作特征（ROC）曲线显示，ROC曲线下面积±S.E. CC为0.897 ± 0.119，B2 M为0.871 ± 0.091，血清肌酐为0.785 ± 0.087。血清肌酐和CC（P = 0.07）和B2 M（P = 0.12）的曲线下面积（AUC）之间无统计学显著差异。在常规使用的临界值下，CC检测eCcr的灵敏度最高（< 60 mt/min/1.73 m2）。当正常白蛋白尿患者与微量白蛋白尿患者比较时，CC也是最好的治疗方法。结论：虽然CC、B2 M和肌酐检测GFR变化的总体诊断准确性无显著差异，但CC是常规使用的临界值下最敏感的标志物，在糖尿病患者中比B2 M或血清肌酐更有临床意义。(C)2003年爱思唯尔科学公司All rights reserved.
Background: Despite recent studies showing that serum cystatin C (CC) is a better marker for GFR than the ubiquitously used serum creatinine, its clinical utility remains under evaluation. Methods: To evaluate their usefulness in patients with type 2 diabetes mellitus (DM) serum concentrations of CC, beta-2 microglobulin (B2M) and creatinine were measured in 105 (38 males, 67 females) Kuwaiti patients with type 2 DM. The results were compared with creatinine clearance (Ccr), which was measured (mCcr) and estimated (eCcr) with the Cockroft-Gault formula, and correlated with 24-h urine protein and early morning urine albumin/creatinine excretion ratio. Results: In patients with eCcr and mCcr results (n = 51), eCcr and mCcr showed significant correlation with each other (r's = .86, P < .0001) with no significant difference between the two. In all patients (n = 105), CC and B2M showed significant correlation with each other (r's = .82, P < .0001) and with serum creatinine concentration (r's = .77 and .84, respectively, P < .0001). Serum CC, B2M and creatinine showed significant (P < .001) inverse correlation with eCcr (r's = -.63, -61 and -.76, respectively). Partial correlations after correcting for age and sex improved the correlation of serum creatinine with eCcr (r = -.81, P < .0001), but there was no significant change in the correlations of CC and B2M with eCcr (r = -.65, P < .0001 and r = -.62, P < .0001, respectively). Receiver operating characteristic (ROC) plots for serum CC, B2M and creatinine for detection of changes in the eCcr showed that the area under the ROC curve ±S.E. is 0.897 ± 0.119 for CC, 0.871 ± 0.091 for B2M and 0.785 ± 0.087 for serum creatinine. There was no statistically significant difference between the areas under the curve (AUC) for serum creatinine and CC (P = .07) and B2M (P = .12). CC had the highest sensitivity for detection of eCcr (< 60 mt/min/1.73 m(2)) at routinely used cutoff values. CC was also the best discriminator when patients with normoalbuminuria were compared with patients with microalbuminuria. Conclusion: Although there is no significant difference in the overall diagnostic accuracies of CC, B2M and creatinine for the detection of changes in the GFR, CC is the most sensitive marker at routinely used cutoff values and would be more clinically useful than B2M or serum creatinine in diabetic patients. (C) 2003 Elsevier Science Inc. All rights reserved.