Closed-loop insulin control for T2DM is feasible in hospital setting



New research suggests it’s feasible for hospitals to use an automated closed-loop insulin delivery system – free of prandial boluses and carbohydrate counts – to improve glucose control in noncritical patients with type 2 diabetes mellitus (T2DM).

The findings, released at the annual scientific sessions of the American Diabetes Association and via simultaneous publication in The New England Journal of Medicine, don’t examine cost or clinical outcomes. However, “our results suggest this new technology might be another approach to manage in-patient hypoglycemia in a safe and effective way, lead author Lia Bally, MD, PhD, of the division of endocrinology, diabetes, and clinical nutrition, Bern (Switzerland ) University Hospital, said in an interview.

The video associated with this article is no longer available on this site. Please view all of our videos on the MDedge YouTube channel.

For the open-label trial, the researchers recruited 136 adults with T2DM under noncritical care at two hospitals (one in the England and the other in Switzerland). Some patients had undergone surgery, Dr. Bally said, and some others were being treated for systemic infections. Comorbidities were significantly more severe in the closed-loop group, and 43% had sepsis.

All of the subjects required subcutaneous insulin therapy.

From 2016 to 2017, patients were randomly assigned to receive normal subcutaneous insulin therapy (n = 70) or closed-loop insulin delivery (n = 66).

It took about 15 minutes to perform the procedure to implement the closed-loop insulin delivery system, Dr. Bally said. It featured a subcutaneous cannula inserted into the abdomen, a continuous glucose monitor (a device also used in the control group), and a trial insulin pump.

This was not a hybrid system, and it did not include prandial insulin boluses or input of the timing and carbohydrate content of meals. One reason behind the choice to adopt a fully automated system was to relieve the burden on both health care professionals and patients, coauthor Hood Thabit, PhD, of Wellcome Trust–MRC Institute of Metabolic Science, the Manchester Academic Health Science Center, and University of Manchester, said in an interview.

For up to 15 days or until discharge, researchers tracked how much of the time sensor glucose measurements were in a target range of 100 mg/dL to 180 mg/dL.

In the closed-loop group, glucose measurements were in the target range 66 mg/dL ± 17% of the time compared to 42 mg/dL ± 17% in the control group, a difference of 24 mg/dL ± 3% (95% confidence interval, 19-30; P less than .001).

For the closed-loop group, the average glucose level was 154 mg/dL, and it was 188 mg/dL in the control group (P less than .001).

The researchers didn’t find a statistically significant difference between the groups in duration of hypoglycemia or amount of insulin delivered.

None of the patients suffered from severe hypoglycemia or clinically significant hyperglycemia with ketonemia.

There were 18 incidents of clinically significant hyperglycemia events (capillary glucose levels of more than 360 mg/dL) in the closed-loop group, compared with 41 such events in the control group. (P = .03)

Three patients in each group had adverse trial-related device effects.

Of 62 patients in the closed-loop group who completed the trial, 87% reported being pleased by their glucose levels, and all but one reported being happy to have their levels monitored automatically. All 62 patients said they’d recommend the system to others.

Going forward, the researchers hope to launch a multicenter trial that will examine clinical outcomes such as postoperative complications, infections, mortality, and glucose control after hospital discharge, according to Dr. Bally.

The study was supported by Diabetes UK, the Swiss National Science Foundation, the European Foundation for the Study of Diabetes, the JDRF, the National Institute for Health Research Cambridge Biomedical Research Center, and a Wellcome Strategic Award. Abbott Diabetes Care supplied equipment and guidance regarding connectivity, and representatives reviewed the manuscript before submission.

Dr. Bally reported funding from the University Hospital Bern, University of Bern and the Swiss Diabetes Foundation. Dr. Thabit reported no disclosures. Other authors report no disclosures or various disclosures.

SOURCE: Bally L et al. ADA 2018 Abstract 350-OR. Published simultaneously in The New England Journal of Medicine. June 25, 2018

Recommended Reading

Improve your glycemic control efforts with SHM’s GC eQUIPS program
The Hospitalist
Bedside CGM boosts glucose control in hospital
The Hospitalist
VIDEO: How to discharge new pediatric diabetes cases in 2 days
The Hospitalist
Award for best hospital goes to … the Mayo Clinic
The Hospitalist
How to reduce NICU transfers for asymptomatic hypoglycemia
The Hospitalist
Bariatric surgery comes with some risk of complications
The Hospitalist
Major message: Most heart failure is preventable
The Hospitalist
Lower glucose target linked to improved mortality in critically ill
The Hospitalist
High risk of low glucose? Hospital alerts promise a crucial heads-up
The Hospitalist
Fluoroquinolones can cause fatal hypoglycemia, FDA warns
The Hospitalist
   Comments ()