Abstract:
Background: Estimation of plasma glucose using glucose oxidase/peroxidase is one of the
most frequently used assays in clinical biochemistry laboratories. Although, commercial
reagents are widely used, for large scale analysis, use of in-house reagents can be more cost
effective.
Objectives: To compare the performance characteristics of a specific in-house glucose
reagent (glucose oxidase/peroxidase) with commercial glucose reagent (glucose
oxidase/peroxidase) on a fully automated biochemistry analyser and also to assess the
stability of the specified formula of the in-house glucose reagent.
Methodology: In this study, 200 retained blood samples received for estimation of plasma
glucose were analysed using the in-house glucose reagent and the commercial glucose
reagent within 6 hours of collection on fully automated biochemistry analyzer. The analyzer
was simultaneously loaded with both reagents and calibrated with the same commercial
calibrator. Daily two levels of Internal Quality Controls and monthly External Quality
Assurance samples were run to assure the precision and accuracy. Stability of the in-house
reagent was assessed within a period of three months.
Results and conclusions: Mean values of the in-house glucose reagent and the commercial
glucose reagent were 142.28 and 141.44 mg/dL; respectively (P= 0.001) and showed a
positive correlation of 0.9993. There was no significant difference between the two methods
in the range between 0-300 mg/dL (P >0.05). The in-house reagent showed a slightly higher
glucose concentration compared to the commercial reagent in patients with glucose levels
above 300 mg/dL which was statistically significant (P= 0.007). The agreement between two
methods was compared using Bland-Altman plot and it showed a positive bias of 0.835 in
the range from 0-500 mg/dL. Accuracy, sensitivity and specificity of the in-house reagent
were 96.5%, 96.15% and 97.14%; respectively. The reagent was stable for three months
period at 2-8 ˚C. Performance of in-house glucose reagent is well correlated with that of
commercial reagent for the range up to 300 mg/dL.