The MITRE (Minimally Invasive Technology Role and Evaluation) Study, a randomized controlled trial of continuous glucose monitoring device use, showed them to be of no greater benefit than standard care. Stanton Newman, PhD, University College, London, UK, reported that all groups, including controls, had a sustained reduction of HbA1c.

Prof. Newman said that prior to MITRE, clinical trial evidence regarding the use of continuous glucose monitoring devices was limited by small sample size, the inclusion primarily of type 1 diabetes patients, and the increased attention given to those receiving the continuous glucose monitoring device. The objective of MITRE was to compare the benefits of using the GlucoWatch® G2™ Biographer (Animas) and the Continuous Glucose Monitoring System (CGMS®, MiniMed) on HbA1c versus attention control and standard treatment in a randomized controlled trial. Percentage change in HbA1c from baseline to 6, 12, and 18 months was the primary endpoint.

Patients were randomly assigned to one of MITRE's four study arms (∼100 patients each). The groups differed as follows:

• Standard Control (baseline visit only asked to test capillary blood glucose at normal frequency with Lifescan Onetouch Ultra Meter®), continued with standard care.

• Attention Control (feedback based on self-monitoring of blood glucose)

• GlucoWatch (used at times of patient choice; recommended minimum of 4x/month and maximum of 4x/week)

• CGMS (fitted at 3, 6, and 12 weeks and worn for 72 hours each time)

The participants in the two treatment groups and the attention placebo group attended three research visits. The attention control group was included to control for the impact of increased levels of contact with health care professionals in the two device groups. Planned visits were conducted with nurses trained specifically on use of the MITRE devices, interpretation of blood glucose results and delivery of appropriate feedback to patients.

The study population included individuals ≥18 years of age with insulin-requiring diabetes (≥2 injections/day), diabetes duration >6 months, and two consecutive HbA1c measurements ≥7.5%.

Mean age was ∼52 years (55% men) with 57% having type 1 diabetes. Mean HbA1c, which was 9.1% at baseline, declined in all groups, although the effect waned over time. At 18 months, the reduction was 1% for the GlucoWatch group and between 4% and 5% for the other groups, with no significant differences at any time point. In the GlucoWatch group 15% of patients achieved a reduction of 12.5% from baseline HbA1c versus 29% in the CGMS group. None of the differences was significant at any time point.

Monitor use declined over time, with 20% continuing to use the GlucoWatch and 67% continuing to use CGMS. Hypoglycemic events were similar between groups. CGMS-derived information tended to alter clinical feedback more than that from GlucoWatch.

Prof. Newman concluded, “There was no group effect on HbA1c of minimally invasive monitors relative to attention control or standard control.” He commented also that with all groups showing a sustained HbA1c reduction, trial participation may have led to improved metabolic control, obscuring any effects of devices.