Hypoglycemia Begets More Hypoglycemia, New ASPIRE Data Find
Hypoglycemia Begets More Hypoglycemia, New ASPIRE Data Find
The concept that "hypoglycemia begets hypoglycemia" has been reinforced in a new analysis from a major study investigating the effect of "threshold-suspend" technology incorporated into insulin-pump/glucose-sensor systems.
The findings were published in the March issue of Diabetes Technology & Therapeutics by Satish K. Garg, MD, an endocrinologist at the Barbara Davis Center for Diabetes, Aurora, Colorado, and colleagues.
Initial results of Medtronic's in-clinic Automation to Simulate Pancreatic Insulin Response (ASPIRE) study, published in 2012, showed that automated insulin-pump suspension reduced the severity and duration of hypoglycemia compared with continued basal insulin delivery in adults with type 1 diabetes. More recently, results of the in-home ASPIRE study demonstrated a reduction in nocturnal hypoglycemia with the technology.
This new analysis of data from the in-clinic study suggests that antecedent hypoglycemia may impede recovery from subsequent hypoglycemia, even with the help of the threshold-suspend feature. The finding has important implications for future research in this technology, Dr. Garg and colleagues say.
"The effect of antecedent hypoglycemia should be taken into consideration in the design of future experiments assessing strategies to reduce hypoglycemia," the authors write.
But at the same time, "by mitigating the duration of hypoglycemic episodes, automatic-pump suspension may help to preserve the normal autonomic response to hypoglycemia in patients with type 1 diabetes," they point out.
Crossover Design
The study enrolled 50 adults and teenagers aged 17 to 58 years with type 1 diabetes who wore Medtronic Veo insulin-pump systems, which incorporate threshold-suspend technology.
They exercised until plasma glucose levels dipped below 70 mg/dL.
In the experiments where the threshold-suspend feature was in use, insulin delivery was suspended at that level. Without the threshold-suspend (control period), basal insulin delivery continued.
Subsequent glucose levels were monitored every 5 to 15 minutes for 4 hours using a standard glucose and lactate analyzer.
The subjects were divided into 2 groups and studied in a crossover design, wherein 1 group the threshold-suspend experiments were done first ("period 1") and the control second ("period 2"), and in the other vice versa. The experiments were separated by 3 to 10 days.
In a total of 98 successful experiments, hypoglycemia duration was 63.7 minutes shorter for the 50 first "period-1" experiments (with no preceding experiments) compared with the 48 "period-2" experiments (107.8 vs 171.5 minutes, P < .01).
The duration of antecedent hypoglycemia was also significantly lower when threshold-suspend was studied in period 1 compared with period 2 (16.6 minutes vs 204.6 minutes, P < .001).
Glucose values remained higher and returned to normal faster when the threshold-suspend experiment was done first vs second. In contrast, in the control experiments, the mean glucose values remained in the hypoglycemic range for the entire 4-hour observation period and were similar for periods 1 and 2.
When the threshold-suspend experiment was done in period 1, few patients stayed hypoglycemic beyond 1 hour. When threshold-suspend was the period-2 experiment, most patients experienced more than 2 hours of hypoglycemia and about half stayed hypoglycemic until the end.
In both periods, with the control experiments, most subjects remained in the hypoglycemic range for the entire 4 hours. No serious adverse events occurred, the authors note.
Avoiding Hypoglycemia
The findings suggest that 1 or more glucose homeostasis mechanisms remained impaired following recent severe and prolonged hypoglycemia.
It's possible that in threshold-suspend period 2, the longer hypoglycemic duration may be the result of glycogen depletion from the prior hypoglycemia during the period-1 control period. And, although not evaluated in this study, impaired autonomic responses to prolonged hypoglycemia may also have played a role, Dr. Garg and colleagues suggest.
"In future experiments or clinical trials of hypoglycemia, our observations suggest that the crossover design should be used with caution, if at all, and that avoidance of hypoglycemia for more than 10 days may be necessary to allow for near-normal responses to hypoglycemia," they conclude.
The ASPIRE study was funded by Medtronic. Dr. Garg has received research support from Medtronic. Disclosures for the coauthors are listed in the article.
Diabetes Technol Ther. 2014;16:125-130. Abstract
The concept that "hypoglycemia begets hypoglycemia" has been reinforced in a new analysis from a major study investigating the effect of "threshold-suspend" technology incorporated into insulin-pump/glucose-sensor systems.
The findings were published in the March issue of Diabetes Technology & Therapeutics by Satish K. Garg, MD, an endocrinologist at the Barbara Davis Center for Diabetes, Aurora, Colorado, and colleagues.
Initial results of Medtronic's in-clinic Automation to Simulate Pancreatic Insulin Response (ASPIRE) study, published in 2012, showed that automated insulin-pump suspension reduced the severity and duration of hypoglycemia compared with continued basal insulin delivery in adults with type 1 diabetes. More recently, results of the in-home ASPIRE study demonstrated a reduction in nocturnal hypoglycemia with the technology.
This new analysis of data from the in-clinic study suggests that antecedent hypoglycemia may impede recovery from subsequent hypoglycemia, even with the help of the threshold-suspend feature. The finding has important implications for future research in this technology, Dr. Garg and colleagues say.
"The effect of antecedent hypoglycemia should be taken into consideration in the design of future experiments assessing strategies to reduce hypoglycemia," the authors write.
But at the same time, "by mitigating the duration of hypoglycemic episodes, automatic-pump suspension may help to preserve the normal autonomic response to hypoglycemia in patients with type 1 diabetes," they point out.
Crossover Design
The study enrolled 50 adults and teenagers aged 17 to 58 years with type 1 diabetes who wore Medtronic Veo insulin-pump systems, which incorporate threshold-suspend technology.
They exercised until plasma glucose levels dipped below 70 mg/dL.
In the experiments where the threshold-suspend feature was in use, insulin delivery was suspended at that level. Without the threshold-suspend (control period), basal insulin delivery continued.
Subsequent glucose levels were monitored every 5 to 15 minutes for 4 hours using a standard glucose and lactate analyzer.
The subjects were divided into 2 groups and studied in a crossover design, wherein 1 group the threshold-suspend experiments were done first ("period 1") and the control second ("period 2"), and in the other vice versa. The experiments were separated by 3 to 10 days.
In a total of 98 successful experiments, hypoglycemia duration was 63.7 minutes shorter for the 50 first "period-1" experiments (with no preceding experiments) compared with the 48 "period-2" experiments (107.8 vs 171.5 minutes, P < .01).
The duration of antecedent hypoglycemia was also significantly lower when threshold-suspend was studied in period 1 compared with period 2 (16.6 minutes vs 204.6 minutes, P < .001).
Glucose values remained higher and returned to normal faster when the threshold-suspend experiment was done first vs second. In contrast, in the control experiments, the mean glucose values remained in the hypoglycemic range for the entire 4-hour observation period and were similar for periods 1 and 2.
When the threshold-suspend experiment was done in period 1, few patients stayed hypoglycemic beyond 1 hour. When threshold-suspend was the period-2 experiment, most patients experienced more than 2 hours of hypoglycemia and about half stayed hypoglycemic until the end.
In both periods, with the control experiments, most subjects remained in the hypoglycemic range for the entire 4 hours. No serious adverse events occurred, the authors note.
Avoiding Hypoglycemia
The findings suggest that 1 or more glucose homeostasis mechanisms remained impaired following recent severe and prolonged hypoglycemia.
It's possible that in threshold-suspend period 2, the longer hypoglycemic duration may be the result of glycogen depletion from the prior hypoglycemia during the period-1 control period. And, although not evaluated in this study, impaired autonomic responses to prolonged hypoglycemia may also have played a role, Dr. Garg and colleagues suggest.
"In future experiments or clinical trials of hypoglycemia, our observations suggest that the crossover design should be used with caution, if at all, and that avoidance of hypoglycemia for more than 10 days may be necessary to allow for near-normal responses to hypoglycemia," they conclude.
The ASPIRE study was funded by Medtronic. Dr. Garg has received research support from Medtronic. Disclosures for the coauthors are listed in the article.
Diabetes Technol Ther. 2014;16:125-130. Abstract
Source...