Preferences for Blood Glucose Management in Adult Intensive Care Unit Patients-An International Survey

Hyperglycaemia is common in intensive care unit (ICU) patients and blood glucose management practices likely vary, but there are limited contemporary data on ICU doctors' and nurses' preferences.

We conducted an international online survey of ICU doctors and nurses. The 16-question survey covered respondent characteristics, glucose management practices, perceived challenges with intermittent point of care (iPOC) glucose monitoring and continuous glucose monitoring (CGM), and preferences for a future trial on CGM versus usual care. Data were reported descriptively for all respondents and stratified by profession.

We received 1424 responses from 12 countries, of which 63% were from nurses. The overall response rate was 36% and the highest proportion of missing data for any question was 9%. Most respondents (92%) reported that their ICU had a glucose management protocol. The median reported insulin initiation threshold was blood glucose of 10 mmol/L. Long-acting insulin was reported to be used occasionally by 68% of respondents. As needed pro re nata insulin was reported as most often given subcutaneously (43%) or intravenously (25%). Overall, 61% of ICU nurses reported concerns related to iPOC use versus 53% among ICU doctors (concerns among nurses versus doctors included risk of hypoglycaemia in 41% vs. 28%; risk of hyperglycaemia in 28% vs. 16%; patient discomfort in 26% vs. 27%). Overall, 75% of respondents never used CGM and 18% of ICU nurses reported concerns related to CGM use versus 22% of ICU doctors (accuracy and reliability in 14% vs. 18%; calibration and maintenance in 9% versus 16%; patient discomfort in 5% vs. 6%, respectively). Most respondents (89%) supported a randomised trial on CGM versus usual care in ICU and 68% preferred an intervention arm with a specific CGM-treatment protocol.

Glucose management preferences varied among ICU staff, particularly in the administration of as needed doses and long-acting insulin. ICU nurses appeared more concerned about iPOC use than ICU doctors. The concerns about use of CGM appeared less common than concerns about iPOC. Most nurses and doctors would support a randomised trial on CGM versus usual care for glucose management in ICU and reported a preference for CGM to be used with a specific treatment protocol.

This international survey highlights substantial professional differences and heterogeneity in ICU glucose management practices, particularly regarding as-needed and long-acting insulin use. Nurses expressed greater concern than doctors about intermittentpoint point-of-care glucose monitoring, especially the risks of hypoglycaemia and hyperglycaemia. Although continuous glucose monitoring was rarely used, it was viewed favourably overall, with broad support for a future protocolised randomised CGM trial.