Glucose Testing Methods: A Systematic Review and Meta-Analysis of Diagnostic Accuracy of Point-of-Care Devices for Neonatal Hypoglycemia

To evaluate the accuracy of various point-of-care device methodologies for measuring blood glucose concentrations in babies at risk of neonatal hypoglycemia. This systematic review and meta-analysis included studies from Ovid MEDLINE, Ovid Embase, and Web of Science up to May 20, 2024. Studies comparing point-of-care testing methods for neonatal blood glucose to a standard laboratory method were included, excluding those on continuous glucose monitoring or conducted before 1990. Two researchers independently assessed inclusion and evaluated risk of bias using QUADAS-2. Sensitivity and specificity were calculated using contingency tables, and diagnostic accuracy was analyzed using hierarchical random-effects modelling. Studies with insufficient data were summarized by estimation direction. Seventy-one studies were included. The quantitative analysis (n = 31) evaluated glucose oxidase (GO) + photometry (n = 8), glucose-1-dehydrogenase (GDH) + photometry (n = 6), GO + electrochemistry (n = 13), GDH + electrochemistry (n = 12), and hexokinase (HK) + electrochemistry (n = 2). All methods showed high specificity (≥93%), with GO + electrochemistry, GDH + electrochemistry, and HK + electrochemistry showing superior sensitivity. The summary receiver operating characteristic curve confirmed HK + electrochemistry as the most accurate. Certain point-of-care device methodologies demonstrate greater accuracy in measuring neonatal blood glucose concentrations. Of the methods evaluated, HK + electrochemistry proved to be the most reliable. However, the limited number of studies using this method suggests the need for further research to confirm these findings across diverse settings and populations. PROSPERO on December 29, 2023 (CRD42023488539).