CHICAGO -- Hypoglycemic events cause changes in heart rate variability, but were detected with a wearable device, researchers said here.
In 28 of 39 hypoglycemic events, the heart rate variability was detected by the wearable patch (VitalConnect HealthPatch), reported Marleen Bekkink, MD, PhD, of Radboud University Medical Center in Nijmegen, the Netherlands, and colleagues.
"Measuring heart rate variability using a wearable device is feasible and well tolerated," Bekkink said at a press conference at ENDO 2018. "Specific heart rate variability could be found at the initiation of hypoglycemia."
Action Points
- Note that this study was published as an abstract and presented at a conference. These data and conclusions should be considered to be preliminary until published in a peer-reviewed journal.
But she cautioned that the device is not ready for prime time in this setting. "Our algorithm will be further refined to improve accuracy, but continuous measurement of heart rate variability via wearables seems promising for alerting patients to upcoming hypoglycemia."
There has been a push to develop such a warning system after episodes involving, Bekkink added.
In the proof-of-principle study, Bekkink's group enrolled 10 type 1 diabetes patients who were using continuous subcutaneous insulin infusion devices, or who used multiple daily insulin injection. Patients were excluded from the study if they had been diagnosed with cardiac arrhythmia or were taking beta-blockers that interfere with heart rate.
During the study, patients wore the patch for 5 days while also using a continuous glucose infusion device and keeping a glucose diary. The researchers defined a hypoglycemic event as a blood sugar reading of ≤70 mg/dL by finger stick measurement.
Three women and seven men, average age 45, participated in the study. They had received a type 1 diabetes diagnosis for an average of 24 years before entering the trial. Their average HbA1C was 7.9%.
During the course of the study, 10 patients experienced 39 hypoglycemic events, and during those events their average glucose reading was 54 mg/dL. Most hypoglycemic events showed a typical heart rate variability pattern that included various spikes and troughs that were unlike other heart rate readings, Bekkink reported.
The authors reported that the median patient-normalized low and high frequency ratio (LF:HF) was 0.31 (interquartile range 0.17-0.63) 30 minutes before hypoglycemia and increased to 0.37 (IQR 0.18-0.63) during hypoglycemia.
The median root mean square of the successive differences (RMSSD) decreased from 40 (IQR 24-59) to 35 (IQR 25-45).
"Twenty events showed a clear detectable increase in LF:HF leading up to the hypoglycemia and eighteen a decrease in RMSSD. There were 11 events that did not display a change in either LF:HF or RMSSD," they wrote.
Michael Wood, MD, of the University of Michigan in Ann Arbor told Ƶ that the development of the algorithm was necessary in this patient population.
"The problem that insulin pumps have in monitoring blood glucose levels is that at the low end of the scale, they are not that accurate and may pump insulin when it may not be needed and can trigger these events," he explained.
Wood, who was not involved in the study, said that episodes of hyperglycemia are not good for the patients but are unlikely to cause any immediate catastrophe, but hypoglycemic events can be fatal. He said that young children on insulin are particularly vulnerable to hypoglycemic events that occur overnight, so an alert device would be helpful.
Bekkink told Ƶ that similar monitoring for people with type II diabetes might also be possible in the future.
Disclosures
Bekkink disclosed no relevant relationships with industry.
Wood disclosed relevant relationships with Medtronics.
Primary Source
The Endocrine Society
Bekkink M, et al "Early detection of hypoglycemia In type 1 diabetes using a wearable device measuring heart rate variability" ENDO 2018; Abstract PO5-1.