Challenges with point of care glucose measurements for management of hypoglycemia in neonates

Abstract

Guidelines from the Canadian Paediatric Society recommend investigating hypoglycemia at a patient blood glucose concentration of 2.6 mmol/L for patients less than 72 hours of age and 3.3 mmol/L for patients 72 hours of age or older (1). Patients with blood glucose <2.8 mmol/L after 72 hours of age require additional investigation with samples sent to the laboratory for glucose (to confirm), beta-hydroxybutyrate, bicarbonate, lactate, free fatty acids, insulin, growth hormone, cortisol, carnitine, and acylcarnitines (1). Neonatal blood glucose concentrations are frequently monitored using point of care testing (POCT) glucose metres. In Canada, there are currently two glucose metres approved by Health Canada for use in hospitals, the Accu-Chek Inform II (Roche Diagnostics) and the StatStrip (Nova Biomedical). POCT glucose metres are accurate for monitoring glycemic control in patients with diabetes (2). However, the accuracy and precision limitations of these metres struggle to match clinical need in the context of neonatal hypoglycemia. POCT glucose results have been shown to differ by as much as 10–20% from central laboratory methods for measurements in the hypoglycemic range; importantly, this difference, or bias, is not consistent between vendors (3,4). The extent of this bias can also shift over time due to variable performance of different lot numbers of test strips, highlighting the need for ongoing evaluation of metre performance.