Background: Monitoring blood glucose levels is essential for managing hyperglycemia during pregnancy, and inaccuracies in glucose measurement can have harmful consequences for both the mother and the baby. This study evaluates the performance of four commercially available glucose meters in pregnant women, taking into account the impact of hematocrit variations during pregnancy.

Methods: Pregnant women attending the diabetes in pregnancy clinic at Campbelltown Hospital underwent concurrent collection of capillary and venous blood samples. Duplicate measurements of capillary glucose levels were taken using four glucose meters (Ascensia Contour Next, Roche Accu-chek Guide, Abbott Freestyle Neo, and LifeSmart meter) and compared against two reference methods: i-STAT point of care capillary glucose and the laboratory venous plasma glucose concentration measured by hexokinase method. Accuracy evaluation involved analyzing mean absolute relative difference, mean bias, Bland-Altman tool, International Organization for Standardization (ISO) 15197:2013 criteria, and surveillance error grid (SEG) analyses. The influence of haematocrit changes was examined using an independent t-test.

Results: A total of 320 paired samples from 40 women (mean (SD) age 32.4±4.6 years, non-European ethnicity 64%, mean haematocrit 0.35 ±0.02 L/L, mean (SD) glucose 6.6±2.7 (range 2.5–13.9) mmol/l (iSTAT), 6.5±2.9 (range 2.5–14.5) mmol/l (hexokinase)) were analyzed. There were significant differences between meters in terms of the frequency of glucose measurements exceeding the reference method (p <0.001). In comparison to iSTAT, the results indicated the following mean biases and 95% limits of agreement for each meter: Ascensia +1.7% (-10.5 to 14%), Roche -2.5% (-10.8% to 5.8%), Abbott -9.1% (-21.3% to 3.2%), and LifeSmart +8.2% (-6.6% to 22.9%). When compared to the hexokinase method, the biases and limits of agreement were as follows: Ascensia +4% (-18.4 to 26.5%), Roche -0.3% (-19.6% to 18.9%), Abbott -7.1% (-26.9% to 12.7%), and LifeSmart +10.6% (-13.5% to 34.7%). Among the meters evaluated, three (Ascensia, Roche, and Abbott) met the ISO accuracy criteria when compared to iSTAT, while only Roche met the criteria when compared to the hexokinase method. When using iSTAT as the reference method for SEG analysis, only Roche demonstrated a ≥ 97% rate of pairs falling within the SEG no-risk 'green' zone. However, none of the meters achieved >97% within the no-risk zone when compared to the hexokinase method. Meter values did not change with variations in haematocrit levels.

Conclusion: The meters demonstrated better performance when using iSTAT as the reference method. Preliminary results indicate differences between meters without a major impact by differences in haematocrit.