Use of Continuous Glucose Monitor in Critically Ill COVID-19 Patients Requiring Insulin Infusion

An Observational Study

Eileen R. Faulds; Andrew Boutsicaris; Lyndsey Sumner; Laureen Jones; Molly McNett; Keaton S. Smetana; Casey C. May; Elizabeth Buschur; Matthew C Exline; Matthew D. Ringel; Kathleen Dungan


J Clin Endocrinol Metab. 2021;106(10):e4007-e4016. 

In This Article

Abstract and Introduction


Context: The coronavirus disease 2019 (COVID-19) pandemic has created a need for remote blood glucose (BG) monitoring in the intensive care unit (ICU).

Objective: To evaluate feasibility and patient safety of a hybrid monitoring strategy of point-of-care (POC) BG plus continuous glucose monitor (CGM) in the ICU.

Design: Retrospective analysis.

Setting: ICU of an academic medical center.

Patients: Patients with COVID-19 on IV insulin.

Intervention: After meeting initial validation criteria, CGM was used for IV insulin titration and POC BG was performed every 6 hours or as needed.

Main Outcome Measures: Outcomes included frequency of POC BG, workflow, safety, and accuracy measures.

Results: The study included 19 patients, 18 with CGM data, mean age 58 years, 89% on mechanical ventilation, 37% on vasopressors, and 42% on dialysis. The median time to CGM validation was 137 minutes (interquartile range [IQR] 114–206). During IV insulin, the median number of POC values was 7 (IQR 6–16) on day 1, and declined slightly thereafter (71% reduction compared with standard of 24/day). The median number of CGM values used nonadjunctively to titrate IV insulin was 11.5 (IQR 0, 15) on day 1 and increased thereafter. Time in range 70 to 180 mg/dL was 64 ± 23% on day 1 and 72 ± 16% on days 2 through 7, whereas time <70 mg/dL was 1.5 ± 4.1% on day 1 and <1% on days 2 through 7.

Conclusions: This study provides data to support that CGM using a hybrid protocol is feasible, accurate, safe, and has potential to reduce nursing and staff workload.


Patients with diabetes are at increased risk of coronavirus disease 2019 (COVID-19) morbidity and mortality.[1,2] Although the role of interventions to improve glucose control in patients with COVID-19 is not well understood, IV insulin is generally considered standard of care in the intensive care unit (ICU).[3–5] Safe and effective management of IV insulin requires accurate and timely point-of care (POC) blood glucose (BG) testing, typically on an hourly basis, and may cause discomfort, blood loss, and sleep interruption for patients. In addition, IV insulin necessitates substantial nursing and staff resources,[6,7] requires use of personal protective equipment, and increases health care worker exposure to infection. Using a continuous glucose monitor (CGM) may therefore be a useful tool for minimizing the need for POC testing.

Use of a CGM, although US Food and Drug Administration (FDA)-approved in the outpatient setting, is not approved for use in the inpatient setting. Prior studies using older technologies support accuracy and noninferiority to standard care using POC testing, but data using modern technologies are limited.[8–13] There are particular concerns using CGM technologies in the context of critically ill patients.[14] Two prospective studies in this population using these older CGM technologies suggest that their accuracy is acceptable.[13,15] In addition, the intervention reduced nursing workload, blood loss, and hospital costs because of the need for less POC BG monitoring.[15] Taken together, despite using older technologies, these studies suggest safety of using a CGM with reduced requirement for nursing intervention, though there are some limitations in detection of hypoglycaemia.[15,16]

Improvement in CGM technologies has resulted in improved accuracy and eliminated measurement interferences from substances such as acetaminophen.[17] At present, 2 CGMs (Dexcom G and Freestyle Libre) are FDA-approved for nonadjunctive use in the outpatient setting. However, studies in the hospital setting are largely limited to noncritically ill patients.[14,18] The COVID-19 pandemic placed unprecedented strains on the inpatient health systems across the world, particularly for affected patients with diabetes. Thus, on April 1, 2020, the FDA announced that it "will not object to the provision of CGM systems for the treatment of patients in hospital settings to support COVID-19 healthcare-related efforts during the pandemic."[19] Although safe use can be more readily established in noncritically ill patients, there are specific concerns that several factors may affect CGM performance in the ICU, including changes in tissue perfusion or edema, hydration, acid-base balance, and medication interference. One approach to address these factors is to establish a hybrid protocol that incorporates both a CGM and POC BG measurements. In addition, the relatively reduced sensitivity of CGM technologies to hypoglycemia may be offset by setting a higher low glucose alert threshold, as well as using predictive alerts based on trends in glucose values.[20,21] Finally, there is evidence that glucose measurement in frequent intervals may compensate for total error allowance due to multiplicity of readings.[22] Here, we report a retrospective analysis of a hybrid POC plus CGM protocol in COVID-19 patients requiring IV insulin in the ICU using modern CGM technologies and settings to minimize safety concerns. We aimed to assess changes in frequency of POC glucose measurement and patient's glycemic control with the utilization of CGM plus CGM device accuracy.