Division of Endocrinology and Metabolism, Department of Medicine, Taipei Veterans General Hospital, Taipei, Taiwan.
Am J Kidney Dis. 2010 May;55(5):867-74. doi: 10.1053/j.ajkd.2009.10.064. Epub 2010 Mar 3.
Hemoglobin A(1c) (HbA(1c)) and fructosamine can be used to monitor glycemic control in diabetic patients with normal kidney function, but their validity in patients with chronic kidney disease (CKD) has not been evaluated. In this study, we evaluated the correlation and accuracy of these 2 measures of glycemic control in type 2 diabetic patients with CKD stages 3-4.
Diagnostic test study.
SETTING & PARTICIPANTS: Type 2 diabetic patients with normal (n = 30) and abnormal kidney function (n = 30) were recruited in Taipei Veterans General Hospital, Taiwan.
HbA(1c) and fructosamine.
Self-monitoring of blood glucose levels.
Blood glucose measurements consisted of 6 preprandial, 6 postprandial, and 2 bedtime assessments in a week with a cycle of 4-week intervals for 12 weeks.
Correlation coefficients between HbA(1c) level or fructosamine-albumin ratio and mean blood glucose levels were 0.836 and 0.645 in participants with normal kidney function and 0.813 and 0.649 in participants with CKD stages 3-4, respectively. In patients with CKD stages 3-4, mean blood glucose levels in weeks 1-12 were 21.9 mg/dL (95% CI, 11.6-32.5) higher than estimated average glucose (eAG) levels calculated from HbA(1c) levels in participants with normal kidney function. In patients with CKD stages 3-4, mean blood glucose levels in weeks 10-12 were 15.5 mg/dL (95% CI, 5.2-30.5) higher than eAG levels calculated from fructosamine levels in participants with normal kidney function, but without statistical significance when eAG calculated from fructosamine level was corrected for serum albumin level (difference of 5.6 mg/dL; 95% CI, -8.6 to 19.8).
Relatively small number of participants with limited amount of blood glucose measurement data.
Our data show that eAG calculated from HbA(1c) and fructosamine levels might underestimate mean blood glucose levels in patients with CKD stages 3-4. References ranges may need to be modified when interpreting results of measurements of glycemic control in type 2 diabetic patients with CKD.
血红蛋白 A（1c）（HbA（1c））和果糖胺可用于监测肾功能正常的糖尿病患者的血糖控制情况，但尚未评估它们在慢性肾脏病（CKD）患者中的有效性。在这项研究中，我们评估了 HbA（1c）和果糖胺这两种血糖控制指标在 2 型糖尿病合并 CKD 3-4 期患者中的相关性和准确性。
诊断测试研究。
在台湾台北荣民总医院招募了肾功能正常（n=30）和异常（n=30）的 2 型糖尿病患者。
HbA（1c）和果糖胺。
自我监测血糖水平。
在 12 周的时间里，每周进行 6 次餐前、6 次餐后和 2 次睡前血糖测量，间隔 4 周一个周期。
肾功能正常的患者中，HbA（1c）水平或果糖胺-白蛋白比值与平均血糖水平的相关系数分别为 0.836 和 0.645，CKD 3-4 期患者的相关系数分别为 0.813 和 0.649。在 CKD 3-4 期患者中，第 1-12 周的平均血糖水平比肾功能正常患者的 HbA（1c）水平估算的平均血糖水平高 21.9mg/dL（95%CI，11.6-32.5）。在 CKD 3-4 期患者中，第 10-12 周的平均血糖水平比肾功能正常患者的果糖胺水平估算的平均血糖水平高 15.5mg/dL（95%CI，5.2-30.5），但当根据血清白蛋白水平校正果糖胺水平估算的平均血糖水平时，差异无统计学意义（5.6mg/dL；95%CI，-8.6 至 19.8）。
参与者数量相对较少，血糖测量数据有限。
我们的数据表明，HbA（1c）和果糖胺估算的平均血糖水平可能低估了 CKD 3-4 期患者的平均血糖水平。在解释 2 型糖尿病合并 CKD 患者的血糖控制测量结果时，可能需要修改参考范围。
