Academy of Diagnostics & Laboratory Medicine - Scientific Short
Scientific Shorts are brought to you by the
Continuous glucose monitors (CGMs) are minimally invasive devices that measure glucose concentrations in the interstitial fluid using a subcutaneous sensor. CGMs have transformed diabetes care by providing near-real-time glucose measurements at 1-15 minute intervals. CGM use improves glycemic control, reduces severe hypoglycemia, reduces the burden of frequent fingerstick testing, and is associated with lower HbA1c.
It is important to note that CGMs are approved by the U.S. Food and Drug Administration (FDA) for outpatient use in the management of type 1 diabetes. However, their inpatient use expanded under FDA enforcement discretion during the COVID-19 pandemic to reduce staff exposure and conserve personal protective equipment. CGMs have remained integrated into hospital workflows, supported by growing real-world evidence and emerging consensus guidance1,2. These include guidance on appropriate clinical applications and education for healthcare staff on interpreting and applying CGM data1. Pediatric patients represent a unique population in whom the benefits of CGM use may be greatest, given the high monitoring burden, pronounced glycemic variability, and more severe and less readily recognized hypoglycemia.
A key question is whether CGM performance is analytically and clinically acceptable in hospitalized children. To address this, we evaluated CGM accuracy by comparing 2,228 paired CGM and point-of-care (POC) glucose measurements from 72 pediatric patients. The overall mean absolute relative difference (MARD) was 14.8%, indicating moderate agreement, with 99.2% of values falling within clinically acceptable Parkes Error Grid (PEG) zones A and B3. However, accuracy declined in the hypoglycemic range, where MARD increased to 20.2%. Similarly, Garg et al. reported similar findings with an overall MARD of 15.9%, which increased to 26% in the hypoglycemic range, with only 64.9% of values meeting clinically acceptable agreement4. Together, these data highlight reduced CGM reliability at low glucose concentrations. Current consensus recommendations require clinicians to verify CGM-measured hypoglycemia using POC or laboratory methods1. Despite this, our study identified a significant implementation gap, with approximately 1 in 5 hypoglycemic CGM readings unconfirmed. Given the known limitations of interstitial glucose sensing in acutely ill patients, unconfirmed CGM readings carry a meaningful risk of both over-treatment and missed true hypoglycemia.
In summary, CGMs are valuable devices in the pediatric inpatient setting, offering continuous insight into glycemic trends in a high-risk population. However, their limitations are most evident in hypoglycemia, where reduced accuracy may limit clinical reliability. Our findings, together with prior studies, underscore a critical need for consistent confirmatory testing, particularly given that a substantial proportion of hypoglycemic readings are not verified in practice. Hospitals need clearer protocols requiring confirmatory testing before any response to a CGM low alert. Until accuracy improves, CGMs should be used only as adjuncts to POC or laboratory glucose measurements.
