Which continuous glucose monitoring sites provide the most accurate, sensitive, and specific real-time values of blood glucose?
Continuous or traditional monitoring of blood glucose has played a key role in therapy plans for enhanced diabetes control. Whether it be for type 1 or type 2 diabetes, continuous monitoring allows patients to have a better understanding of their day-to-day glycemic control by providing real-time blood glucose values continuously throughout the day without all the lancet pricks that traditional monitoring of blood glucose requires. Maintaining blood glucose values within normal limits has proven numerous times throughout multiple studies to reduce long term secondary problems that develop with worsening diabetes, such as renal and neuropathic complications.
As we know, devices that continuously monitor blood glucose can do so through the abdomen or upper arm. Researchers from the department of endocrinology at Copenhagen University Hospital in Denmark wanted to determine if the accuracy of a continuous blood glucose monitor was different depending on the application site of the monitor. The study enrolled a total of 14 relatively healthy patients who had suffered from type 1 diabetes for at least 2 years, had an HbA1c of less than or equal to 8.5%, and have been treated with insulin via pump for at least 1 year. Patients were excluded if they were aerobically fit and exercised vigorously for more than 3 total hours per week.
Study participants had continuous blood glucose monitors placed on the abdomen at least two inches away from the site of their insulin infusion pump and on their non dominant upper arm between the shoulder and the elbow. To have data against which to compare the two sets of continuous blood glucose reading sets, the researchers of this study also prompted study subjects to self-monitor blood glucose traditionally by pricking their finger with a lancet and drawing up blood in a test strip to insert into a monitoring device seven times per day for four days total in best attempts to emulate the values reports from the continuous reader. .
The accuracy of blood glucose readings between the device placed on the abdomen and device placed on the upper arm were compared using Skillings-Mack two factor analysis of variance. This test provided the best method to compare the two data sets due to non-normal distribution of absolute relative differences between the two data sets, which was the ultimate primary endpoint. Secondary endpoints observed in this study included incidence, sensitivity, and specificity of hypo and hyperglycemia.
Primary analysis resulted in no statistically significant difference between continuous blood glucose readings from devices placed on the abdomen and devices placed on the upper arm, though the arm sensor outperformed the abdomen sensor in most analyses of secondary endpoint measures, accepting the original null hypothesis. These results are expected as there are differences in interstitial fluid compositions between differing body tissues, i.e. abdomen and upper arm. Though some continuous blood glucose monitoring devices only carry FDA approvals for specific insertion sites, usually the abdomen, results from this study show no significant differences between readings obtained from the abdomen or upper arm.
- Results from 14 patients determined that there were no statistically significant differences in either sensitivity, specificity, or absolute relative difference between continuous blood glucose readings from devices inserted into the abdomen or upper arm.
- Prescribers should understand exactly how continuous blood glucose monitors work and counsel patients appropriately on proper calibration, administration, and reading techniques.
- It is important for patients to understand their specific blood glucose goals, signs and symptoms for hypo and hyperglycemia, and what measures to take when experiencing one of these events.
Steineck IK, Mahmoudi Z, Ranjan A, et al. Comparison of continuous glucose monitoring accuracy between abdominal and upper arm insertion sites. Diabetes Technology & Therapeutics 2019; 21(5). DOI: 10.1089/dia.2019.0014.