Evaluation of the relationship between capillary and venous plasma glucose concentrations obtained by the HemoCue Glucose 201+ system during an oral glucose tolerance test

Abstract

In 55 women with previous gestational diabetes mellitus, simultaneous capillary and venous plasma glucose concentrations were measured at 0, 30 and 120 min during a 75 g oral glucose tolerance test (OGTT). The aims of the study were to examine the relationship between capillary and venous glucose measurements, and to establish equations for the conversion of capillary and venous glucose concentrations using the HemoCue Glucose 201+ system. Additionally, the correlation between the capillary and venous glucose concentrations with the diagnostic cut-off limits proposed by the World Health Organization (WHO) in 1999 was evaluated. Capillary glucose concentrations were consistently higher than venous glucose concentrations at all time points of the OGTT (p < 0.001), and the correlations between the measurements were statistically highly significant (p < 0.001). The differences between the samples were greatest in the non-fasting state as revealed by the 95% prediction intervals (mmol/L) in Bland-Altman plots; ± 0.54 at 0 min, ± 2.01 at 30 min, and ± 1.35 at 120 min. Equivalence values for capillary plasma glucose concentrations derived from this study tended to be higher than those proposed by the WHO as diagnostic cut-off limits. Stratifying subjects by glucose tolerance status according to the WHO criteria revealed disagreements related to glucose values close to the diagnostic cut-off points. The study findings highlight the uncertainty associated with derived equivalence values. However, capillary plasma glucose measurements could be suitable for diagnostic purposes in epidemiological studies and when translating results on a group basis.

KeyWords::

• Agreement
• Bland-Altman plot
• comparison method
• conversion algorithm
• correlation

Acknowledgements

This study was supported by grants from Thelma Zoéga's Foundation, the Research Funds of Malmö University Hospital and the Foundation of the County of Skåne. Vera Gunnarsson is acknowledged for skilful technical assistance. We are also indebted to Helene Jacobsson, biostatistician, Competence Centre for Clinical Research, Skåne University Hospital, Lund, Sweden, for good advice when performing the statistical analyses.

Declaration of interest: The authors report no conflict of interest. The authors alone are responsible for the content and writing of the paper.