Authors (including presenting author) :
Yeung MW(1), Liu SKB(1), Lo WP(1), Kwan YF(1), Shum HP(1)
Affiliation :
(1)Pamela Youde Nethersole Eastern Hospital, Intensive Care Unit
Introduction :
Critical illness can impair the body's ability to regulate blood glucose, resulting in hyperglycemia or hypoglycemia. Both conditions can adversely affect the outcomes of critically ill patients. Several studies recommend that critically ill patients with hyperglycemia should be monitored and treated within a target range. Therefore, managing glycemic control in these patients requires frequent glucose monitoring, which increases nurse workload.
Continuous glucose monitoring (CGM) may offer benefits for achieving glycemic control in critically ill patients. CGM measures glucose levels in the interstitial fluid just under the skin continuously while the device is worn. It can track glucose levels over time and detect fluctuations and hypoglycemic events, providing insight into glucose trends. Compared to using a blood glucose meter for monitoring, CGM can reduce nurse workload.
Objectives :
To explore the feasibility, accuracy, and reliability of using CGM for glycemic control in critically ill patients with septic shock.
Methodology :
Since September 2023, 15 patients were recruited from PYNEH ICU. Their glycemic control and hemodynamic parameters were unstable due to septic shock. They had their glucose levels measured frequently based on doctor's orders. Arterial blood samples were used with a POCT standard blood glucose meter (BGM) (Model:StatStrip Glucose/Ketone Meters) to obtain glucose readings. These 15 patients were fitted with the 7th generation (G7) Dexcom CGM by well-trained nurses. CGM glucose readings were retrieved for comparison with BGM blood glucose readings. Both monitoring devices' readings were retrieved from the clinical information system (CIS) and the CGM records retrospectively.
Result & Outcome :
Result
We analyzed 202 paired measurements from 15 critically ill patients. Bland–Altman plots and Clarke error grid analysis were employed for data analysis. For the Bland-Altman plots, 82% of data points were within the limits of agreement. For the Clarke error grid analysis, most data points fell in Zone A (46%) and Zone B (51%). This suggests that CGM can provide clinically acceptable readings, and the monitoring system is accurate enough for clinical use but only moderate reliability in patients with septic shock.
Conclusion
CGM can be a valuable tool for monitoring blood glucose trends in critically ill patients, allowing for early detection of abnormalities and timely interventions. However, there are limitations to consider: (1) CGM devices are currently costly, (2) the patient's clinical status may affect CGM reliability, such as hypotension, tissue perfusion, or the use of vasoactive medication, and (3) the device should not be used during radiological procedures or in operating theatres, leading to potential monitoring interruptions. Randomized controlled trials and larger sample size research in different clinical status are needed in the future to establish CGM's reliability before full clinical integration.
