| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
|
|
|
|||||||
|
|||||||||
Technical Briefs |
1 Department of Clinical Chemistry, Altnagelvin Area Hospital, Glenshane Rd., Londonderry BT47 6SB, United Kingdom
aauthor for correspondence: fax 44-2871-313036, e-mail mlynch{at}alt.n-i.nhs.uk
During the investigation of a neonate with NADH-methemoglobin (NADH-MetHb) reductase deficiency (baseline MetHb, 28%), we noticed disparate readings between the ward-based, whole-blood HemoCue B-Glucose Analyzer (HemoCue AB) and the main laboratory plasma glucose method (glucose oxidase) performed on a Synchron CX7 Delta analyzer (Beckman Coulter). The HemoCue results suggested profound hypoglycemia, which was out of keeping with the clinical status of the patient, whereas the Synchron CX7 Delta results were consistently within the pediatric reference interval. These findings were contrary to the previously documented overestimation of the HemoCue analyzer at low glucose concentrations (1)(2)(3). Because the HemoCue analyzer (1)(2)(3)(4)(5)(6)(7)(8)(9)(10)(11) uses a mutarose–glucose dehydrogenase enzyme system to produce a colored formazan, which is then quantified photometrically, the possibility of spectral interference by MetHb with the HemoCue analyzer was raised and investigated as follows.
A pilot study was performed by collecting 30 mL of adult blood into lithium heparin-containing tubes. Two pools of red blood cells were formed, one with a nominal MetHb value of 0% and the other with a nominal value of 100% [prepared with potassium nitrite, as outlined previously (12)(13)]. Both pools, adjusted to the same hemoglobin concentration (Sysmex SE-900 Hematology Analyzer; TOA Medical Electronics), were then diluted with an equal volume of a saline solution (150 mmol/L) containing glucose (
4.0 mmol/L) to produce pools with identical hematocrit (0.5) and glucose concentrations. The final hemoglobin concentration of these pools was 142 g/L. The pools were then used to prepare samples with nominal 0%, 5%, 10%, 15%, 25%, 50%, and 100% MetHb values. The whole-blood samples were analyzed on the HemoCue analyzer, and the corresponding "plasma" glucose was measured on the main laboratory analyzer within 30 min of the analysis of the whole-blood samples. MetHb was measured on an IL 682 CO-Oximeter (Instrumentation Laboratory).
HemoCue glucose concentrations decreased as MetHb values increased, whereas the corresponding plasma concentrations remained unchanged (Fig. 1
). The results confirmed the interference of MetHb with the HemoCue analyzer suggested from the findings of the original neonatal blood samples. The difference between the whole-blood HemoCue glucose result and its corresponding plasma result for the nominal 0% MetHb sample of 1.1 mmol/L (20 mg/dL) parallels the overestimation of the HemoCue analyzer mentioned above.
|
The study was then expanded with 0%, 5%, 10%, 15%, 20%, 25%, 30%, 35%, 40%, 50%, 75%, and 100% MetHb samples (prepared from a second lithium-heparin, whole-blood pool as outlined above, aside from a 8.0 mmol/L glucose solution; the final hemoglobin concentration for these pools was 135 g/L). In addition to measuring whole-blood glucose with the HemoCue analyzer and plasma glucose with the main laboratory method, the following point-of-care testing blood glucose meters were used to measure whole-blood glucose: Accu-Chek Advantage (electrochemical methodology; Roche Diagnostics); Accu-Chek Glucotrend 2 (reflectance photometry; Roche Diagnostics); Medisense Precision QID (electrochemical methodology; Abbott Laboratories); and GlucoMen GlycO (electrochemical methodology; A. Menarini Diagnostics). All glucose measurements were performed within 50 min of one another (Table 1
).
|
HemoCue glucose results decreased as MetHb increased [P <0.001; ANOVA, Analyze-It Software for Microsoft Excel]. At a measured MetHb value of 8%, the HemoCue glucose value was 10% lower than the glucose value of the 0% MetHb sample; and at a measured MetHb value of 30% (i.e., approximately the concentration in our patient), the HemoCue glucose value fell by one-third. Minimal change was seen with the other glucose meters or, as described above, with the main laboratory method. These results confirm that MetHb interferes with the HemoCue glucose analyzer. The HemoCue glucose result for the 0% MetHb sample was higher than any of the results found with the other methods, again in keeping with other studies (1)(2)(3). The measured MetHb values agree well with the nominal values.
We have demonstrated that MetHb interferes with the HemoCue B-Glucose Analyzer, and we speculate that this error originates either from direct interference with the color development of the tetrazolium salt or, more likely, from the absorption of the MetHb species at >630 nm (14). Previously, Zijlstra et al. (15) suggested photometric interference for the differences they observed between adult and neonatal blood with the HemoCue analyzer.
We recommend that the HemoCue B-Glucose Analyzer should not be used for patients with MetHb concentrations >10% and that an alternative method should be used for glucose determination in such patients.
References
The following articles in journals at HighWire Press have cited this article:
|
|
 |
|
  K. Dungan, J. Chapman, S. S. Braithwaite, and J. Buse Glucose Measurement: Confounding Issues in Setting Targets for Inpatient Management Diabetes Care, February 1, 2007; 30(2): 403 - 409. [Full Text] [PDF]
|
|
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
