Get access

Comparison of fingerstick hemoglobin A1c levels assayed by DCA 2000 with the DCCT/EDIC central laboratory assay: results of a Diabetes Research in Children Network (DirecNet) Study

Authors


  • *

    A listing of the DirecNet Study Group appears in the Appendix.

†William V. Tamborlane, MD
Jaeb Center for Health Research
15310 Amberly Drive, Suite 350
Tampa, FL 33647, USA.
Tel: +1 813 975 8690;
fax: +1 813 903 8227;
e-mail: direcnet@jaeb.org

Abstract

Abstract:  Background:  The Diabetes Control and Complications Trial/Epidemiology of Diabetes Interventions and Complications (DCCT/EDIC) high-performance liquid chromatography (HPLC) method for measuring hemoglobin A1c (HbA1c) serves as a reference standard against which other assays are compared. The DCA 2000® + Analyzer (Bayer Inc., Tarrytown, NY, USA), which uses an immunoassay, is a very popular device for measuring HbA1c levels in pediatric diabetes practices.

Objective:  To determine how HbA1c values measured with the DCA 2000 in a multisite, pediatric diabetes clinic setting compare with corresponding HbA1c values measured in the DCCT/EDIC laboratory.

Design/Methods:  To examine this question, the Diabetes Research in Children Network (DirecNet) used the DCA 2000 in five clinical centers to measure baseline HbA1c levels in 200 youth with type 1 diabetes mellitus (T1DM) (aged 12.5 ± 2.8 yr) who were participating in an outpatient clinical trial. At the same visit, an additional blood sample was obtained, refrigerated, and shipped to the DCCT/EDIC central laboratory for determination of HbA1c values.

Results:  The central laboratory HbA1c value averaged 8.0 ± 0.9% (mean ± SD), with a median (25th and 75th quartiles) of 7.8% (7.3 and 8.5%, respectively). The DCA 2000 HbA1c values were strongly correlated (r = 0.94, p < 0.001), but significantly higher than DCCT/EDIC central laboratory values with a mean difference of +0.2% (95% confidence interval +0.14 to 0.23%, p < 0.001). There was some variation in the differences between DCA 2000 and central laboratory values at the five clinical centers (p < 0.001) with mean differences ranging between 0.0 and 0.3%, but differences between the two methods did not vary significantly by age or gender.

Conclusion:  Measurements of HbA1c by the DCA 2000 compare favorably with the DCCT/EDIC central laboratory method, albeit with slightly higher values.

Ancillary