Reviewed by Dr. Chad Guenther
Carey M, et al. Diabetes. 2017 Nov; 66(11):2764-73

Hypoglycemia is a major risk factor for morbidity and mortality in patients on insulin therapy. Two risk factors for hypoglycemia include hypoglycemia-associated autonomic failure (HAAF), associated with a recent episode of hypoglycemia, and exercise-associated autonomic failure (EAAF), associated with vigorous exercise. In both of these conditions, counterregulatory responses to hypoglycemia are blunted. Endogenous opioid pathways have been speculated to play a role in these responses. Carey et al. tested this hypothesis using morphine administration.

12 healthy subjects without diabetes, recent hypoglycemia, family history of diabetes, or present medication usage were randomized to saline or morphine infusion in a crossover design with 5-week washout. Each study period took place across two days. On day one, either normal saline or morphine (0.1 µg/kg/min) were infused over 120 minutes, followed by a 120 minute break with snack, followed by the same infusion over 120 minutes. Day two consisted of insulin and glucose infusions titrating to blood glucose levels of 90, 80, 70, and 60 mg/dL with monitoring of laboratory and symptomatic responses.

The researchers found significantly lower epinephrine concentrations in the morphine versus control group at 70 and 60 mg/dL of blood glucose on day two. At a blood glucose of 60 mg/dL, the epinephrine level in the morphine group was 292.5 +/- 15.7 pg/mL compared to 419.4 +/- 20.4 pg/mL in the control group, which is a 30% reduction in the morphine group (p = 0.02). The morphine group also required higher glucose infusion rates to maintain blood glucose at 70 and 60 mg/dL (p < 0.01) and experienced significantly fewer hypoglycemic symptoms at 60 mg/dL (p = 0.030) compared to the control group. In a subsequent test of 6 additional participants subjected to episodes of hypoglycemia in the laboratory across two days, there was a 45% reduction in epinephrine concentrations during the hypoglycemic episode on day two compared to controls.

Many patients with diabetes mellitus (DM) are treated with insulin, and patients with type I DM have been shown to have fewer diabetic complications with tighter glycemic control. However, hypoglycemia is a cause of death in 6-10% of patients with type I DM and is the cause of 100,000 ED visits and 30,000 hospital admissions annually in the United States. Better understanding of the blunting of counterregulatory response pathways to hypoglycemia in HAAF and EAAF may help to lower morbidity and mortality.

Recent animal models have shown a relationship between endogenous opioids and a blunted hypoglycemic response. Additionally, acute naloxone administration has been shown to prevent the development of HAAF in humans in experimental hypoglycemia, although short courses of naltrexone have not shown similar effects. This study was undertaken to better understand the physiologic role of opioid receptor activation in HAAF.

The authors found a decrease in epinephrine release during hypoglycemia after morphine administration. However, the hypoglycemic epinephrine response was less blunted after one day of morphine administration than after one day of hypoglycemia (30% decrease versus 45% decrease compared to controls). Additionally, a number of other counterregulatory hormones were unaffected by morphine administration. This implies that endogenous opioids play a significant role in the HAAF/EAAF pathways, but other factors are also likely at work.

Based on these findings, it is also important to consider the role prescription opioids may have in hypoglycemia-induced morbidity and mortality. Many diabetic patients managed with insulin also suffer from chronic pain managed with opioid therapy. Assuming an average-sized man, the infusions used in this trial would represent around a 2 mg total dose of morphine, lower than the daily dose of many chronic pain regimens. Given growing national concern about opioid use, the blunted hypoglycemia response represents another pathway by which opioids may harm patients.