Performance of point-of-care glucose testing for the diagnosis of gestational diabetes in South Africa.

OBJECTIVE

Lack of accessibility to oral glucose tolerance tests (OGTTs) in South Africa means many pregnant women go without testing for gestational diabetes mellitus (GDM). This study evaluated point-of-care (POC) glucometers against the laboratory-based glucose method in pregnant women.

METHODS

This was a cross-sectional study on pregnant women attending the prenatal clinic in Johannesburg who were recommended for the OGTT. OGTTs were conducted as per International Association of Diabetes and Pregnancy Study Groups (IADPSG) guidelines. Women who consented to the study donated both venous and capillary blood for laboratory-based and POC glucose measurements using seven POC glucometers: I-STAT, Xpress, LDX, VivaChek-Ino, Accu-Chek Active, StatStrip, and Codefree. By assessing sensitivity, specificity, and area under the receiver operating characteristic curve (AUC) and comparing Bland-Altman plots, the diagnostic accuracy of each glucose meter was compared with the reference method, the laboratory-based glucose method.

RESULTS

Data were analyzed for 1076 pregnant women. Based on OGTT testing, 83 women had GDM (7.7%). Overall, the POC glucometers performed poorly, with sensitivity ranging from 17.6% to 87.18% and specificity ranging between 62.7% and 99.8%. The AUC ranged from 0.59 to 0.79. All POC glucometers showed moderate to poor reliability. Laboratory-based fasting plasma glucose (FPG) surpassed the POC glucometers in sensitivity, specificity, and AUC, with values of 94.0%, 100%, and 0.98, respectively.

CONCLUSION

We demonstrated that laboratory-based FPG has the potential to be used as a diagnostic test for GDM and that the POC glucometers cannot replace OGTT laboratory-based measurements.