This article is more than 5 years old. Information may no longer be current.
New formula may predict overnight hypoglycemia in type 2 diabetes without CGM
Click Here to Manage Email Alerts
Researchers in Japan developed a predictive formula for lowest nocturnal glucose levels that could be used to address nocturnal hypoglycemia for patients with type 2 diabetes using insulin, according to findings published in the Journal of Diabetes and its Complications.
“Hypoglycemia can be associated with cardiovascular events, dementia, coma and death, resulting in decreased quality of life and poor prognosis,” Kazunori Sakurai, of the department of diabetes, endocrinology and metabolism at Shinshu University School of Medicine in Matsumoto, Japan, and colleagues wrote. “It is particularly notable that nocturnal hypoglycemia is experienced by 41.1% of [type 2 diabetes] patients and 57.1% of patients with severe hypoglycemia. Patients and their families often fail to recognize long-lasting [nocturnal hypoglycemia]. Furthermore, [nocturnal hypoglycemia] can also cause hyperglycemia the next morning by the Somogyi effect.”
Sakurai and colleagues used two groups to develop the formula. A formula-making group included 29 patients (12 women; median age, 64 years) with type 2 diabetes treated with insulin and using continuous glucose monitoring. Data from these participants were gathered based on hospital visits from February 2012 to July 2013. A second formula-validating group included 21 patients (13 women; median age, 64 years) with type 2 diabetes using CGM who made at least one hospital visit between August 2013 and March 2016. Age, physical information, biochemical test data, insulin dose and CGM readings were used as the predictors for lowest nocturnal blood glucose.
The researchers found that age and high levels of HDL cholesterol were associated with lowest nocturnal blood glucose levels, with age having a negative correlation (Spearman rank correlation = –0.42; P = .024) and higher HDL cholesterol connected with lower levels of lowest nocturnal blood glucose (Spearman rank correlation = –0.403; P = .033). The researchers said mean glucose during CGM had a positive correlation with lowest nocturnal blood glucose (Spearman rank correlation = 0.432; P = .019).
In multivariate analysis with self-monitored blood glucose used as an additional predictor of lowest nocturnal blood glucose, researchers used three models, including age, fasting blood glucose and basal insulin dose.
The researchers were able to use the formula to predict lowest nocturnal blood glucose levels with a standard error of 31 mg/dL.
“The present study was performed to develop a formula for prediction of [lowest nocturnal blood glucose] using patients’ medical information even when the CGM systems were not available to prevent [nocturnal hypoglycemia] in [type 2 diabetes] patients treated with insulin,” the researchers wrote. “The formula indicated that low [fasting blood glucose] level could reduce [lowest nocturnal blood glucose] level, leading to increased risk of hypoglycemia. ... The prediction formula can be expected to facilitate early and appropriate prevention of [nocturnal hypoglycemia] in [type 2 diabetes] patients treated with insulin.” – by Phil Neuffer
Disclosures: The authors report no relevant financial disclosures.
Click Here to Manage Email Alerts