Secondary Logo

Journal Logo

Institutional members access full text with Ovid®

Quality Control of Monitoring Glycemia in Patients with Diabetes

Solnica, Bogdan MD, PhD*; Naskalski, Jerzy W PhD, DSc*; Hebda-Szydlo, Alicja MD; Sieradzki, Jacek MD, DSc

Point of Care: The Journal of Near-Patient Testing & Technology: March 2005 - Volume 4 - Issue 1 - p 45-48
doi: 10.1097/01.poc.0000157181.57778.ba
Symposium Article
Buy

We present results for quality control (QC) on glucometers used in a diabetes outpatient clinic. The QC included inspection of glucometer blood glucose measurement by patients and a comparison of glucose results from the patients' meters with the point-of-care analyzer used in a clinic and with a central laboratory method. In the study, results from 86 patient glucometers were compared with the reference HemoCue glucose analyzer, and results from 72 glucometers were compared with the reference laboratory method. Glucometers studied were Accutrend (11), Glucotrend (32), Precision QiD (22), One Touch (11), and Glucocard II (11). Reference glucose assays were performed using the glucose oxidase method on a Hitachi 912 analyzer. Glucose concentrations measured using patient glucometers ranged from 36 to 425 mg/dL. The analytical bias of glucometers varied from −1.45% to 9.54%. Correlation coefficient between results obtained by the patient glucometers and HemoCue analyzer or laboratory method ranged from 0.964 to 0.983 and from 0.955 to 0.986, respectively. Passing-Bablok agreement test and Deming regression analysis indicated good concordance of results between all the patient glucometers and the HemoCue analyzer, whereas good agreement with the laboratory method was found only for the Accutrend, Glucotrend, and One Touch glucometers. Differences between results did not exceed 20 mg/dL, and the mean difference between results for the patient glucometers and HemoCue analyzer ranged from 0.06 to 5.55 mg/dL and between the glucometers and the laboratory method from 1.82 to 12.9 mg/dL. In conclusion, good analytical performance of the glucometers studied and a bias of <10% from the reference values were found. Results of this study show the possibility for routine QC of glucometers in an outpatient clinic.

From the *Department of Clinical Biochemistry, Jagiellonian University Medical College, Krakow, Poland; and †Clinic of Metabolic Diseases, Jagiellonian University Medical College, Krakow, Poland.

Reprints: Dr. Bogdan Solnica, Department of Clinical Biochemistry, Jagiellonian University Medical College, ul. Kopernika 15b, 31-501 Krakow, Poland (e-mail: mbsolnic@cyf-kr.edu.pl).

© 2005 Lippincott Williams & Wilkins, Inc.