BARCELONA – Real-time continuous glucose monitoring (rtCGM) was better than self-monitoring of blood glucose (SMBG) in reducing hemoglobin A1c (HbA1c) and other glycemic endpoints in people with type 1 diabetes, regardless of the type of insulin delivery method used in a 3-year follow-up of a prospective, real-world clinical trial.
Long-term results from the COMISAIR study showed that the end-of-study HbA1c values were significantly lower, compared with baseline values, in people with type 1 diabetes who used rtCGM with multiple daily injections (MDI) of insulin (7.0% [53 mmol/mol], P = .0002) or an insulin pump (6.9% [52 mmol/mol], P less than .0001). There was no significant difference between the two rtCGM delivery-method groups.
Final HbA1c values for those who used SMBG with multiple daily injections or an insulin pump were 8.0% (64 mmol/mol) and 7.7% (61 mmol/mol), respectively, but were not significantly different from baseline (P = .3574 and P = .1, respectively).
These findings could help guide physicians when discussing treatment and monitoring options with their patients, suggested study investigator Jan Šoupal, MD, PhD, of Charles University in Prague, when he presented the findings at the annual meeting of the European Association for the Study of Diabetes.
Dr. Šoupal and associates have previously reported data from the COMISAIR (Comparison of Different Treatment Modalities for Type 1 Diabetes Including Sensor-Augmented Insulin Regimens) study at 1 year of follow-up for 65 patients (Diabetes Technol Ther. 2016;18:532-8). The findings he presented at the EASD meeting, simultaneously published online Diabetes Care, were for the full cohort of 94 patients and, with 3 years of follow-up, makes it “the longest CGM trial ever,” he said.
At the time the COMISAIR study was initiated, in 2013, “we knew that insulin pump therapy, especially in combination with real-time CGM, can improve several outcomes of patients with type 1 diabetes,” Dr. Šoupal observed. However, the effectiveness of CGM in patients with MDI was not widely described, and comparisons between continuous subcutaneous insulin infusion (CSII) and insulin MDI with rtCGM were lacking. “Moreover, we didn’t have any comparison between insulin pump therapy alone, without CGM, and MDI with CGM, and there were no long-term trials with real-time [continuous glucose monitoring].”
The aim of the COMISAIR study was therefore to compare four different treatment strategies in people with type 1 diabetes who had an HbA1c of 7%-10% (53-86 mmol/mol), despite MDI treatment with insulin analogues and SMBG. The treatment strategies tested were CSII plus rtCGM (n = 26), MDI plus rtCGM (n = 22), CSII plus SMBG (n = 25), and MDI plus SMBG (n = 21). Patients were not randomized to these treatment arms but exposed to all of them during a 4-day DAFNE-like training session and then allowed to choose which they would like to use according to their individual needs and preferences, reflecting real-life practice.
Dr. Šoupal pointed out that two different continuous glucose monitoring devices had been used in the trial, and that 100% of the CGM groups wore a sensor for more than 70% of the time, which was one of the prerequisites for inclusion in the trial. Good adherence was observed, with 93% of patients completing all study visits, and CGM users wearing their sensors on average 88% of the time. “This nice adherence may be connected to the pretty good results,” he observed.
In discussing the HbA1c results, Dr. Šoupal noted that “improvement observed in patients with [continuous glucose monitoring] is stable throughout 3 years, which is not always a reality for different types of treatment for diabetes.” In addition, “it is not so important how insulin is delivered, what is more important is how patients with type 1 diabetes monitor their glucose.”
Another key endpoint of the trial was time in range (70-180 mg/dL [3.9-10 mmol/L]). Results showed significantly more patients achieving this with rtCGM than with SMBG, regardless of whether they were using pump therapy or MDI. Comparing 3-year with baseline values, time in range was 72.3% versus 50.9% for rtCGM with CSII and 69% versus 48.7% for rtCGM with MDI (P less than .0001 for both). Results with SMBG with CSII or MDI were a respective 57.8% versus 50.6% (P = .0114) and 54.7% versus 51.8% (P = 1.0).
Glycemic variability was reduced in patients using insulin pumps with SBMG, and “not surprisingly, there was a reduction in both CGM-augmented groups,” Dr. Šoupal stated.
There was a reduction in the time spent in hypoglycemia from baseline to year 3 in all four groups, but that was significant only for the two rtCGM groups. Overall, there were seven severe hypoglycemia episodes, five in the SMBG groups (two in the CSII group, three in the MDI group) and two in the rtCGM groups (one each in the CSII and MDI groups), with one episode only occurring when the CGM sensor was not being worn.
Three episodes of ketoacidosis were reported – one each in the SMBG-pump, SMBG-MDI, and rtCGM-pump groups.
In summing up, Dr. Šoupal said that “real-time CGM, both with insulin pumps and with [multiple daily injections], provided significant, comparable, and stable improvement of glycemic outcomes.” He added that “treatment with CGM and MDI was more effective than treatment with insulin pump therapy alone, and that CGM and MDI can even be considered as a suitable alternative to treatment with insulin pumps and CMG for some patients.”
With many treatment options available, some will suit patients better than others, he suggested, but although “individualization of our treatment is important”, the COMISAIR data show that “it is CGM that makes the difference”.
The study was supported by the Agency for Healthcare Research and the Ministry of Health of the Czech Republic. Dr. Šoupal reported receiving honoraria from Abbott, AstraZeneca, Boehringer Ingelheim, Dexcom, Eli Lilly, Medtronic, Novo Nordisk, and Roche. Dexcom also paid for the development of the manuscript published in Diabetes Care.
SOURCES: J et al. EASD 2019, Abstract 40; J et al. Diabetes Care. 2019 Sep 17. doi: 10.2337/dc19-0888.