Vulnerable Hours of the Day for Severe Iatrogenic Hypoglycaemia in Type 2 Diabetes: A Single-Centre Retrospective Study.

Introduction: Iatrogenic hypoglycaemia is an event that should be avoided in the treatment of diabetes, but the pathophysiology thereof has been poorly examined and reported. There is no established method for preventing iatrogenic hypoglycaemia and the current approach is a reactive response following onset of the disease. In this study, we aimed to explore the existence of 'hypoglycaemia-vulnerable hours of the day' in patients with type 2 diabetes, with the ultimate goal of preventing the onset of iatrogenic hypoglycaemia by clarifying the time when severe hypoglycaemia is likely to occur.

Methods: Of the 553,201 patients who visited the Critical Care and Emergency Center of Aizawa Hospital between 2008 and 2019, patients with proven hypoglycaemia (blood glucose level <3.0 mmol/L) and those using insulin or oral hypoglycaemic agents for the treatment of type 2 diabetes were included: 146 insulin users and 148 oral hypoglycaemic agent users. Cosinor analysis was employed to identify hypoglycaemia-vulnerable hours of the day.

Results: Patients with type 2 diabetes and severe hypoglycaemia had two peaks: at 8:00 and 18:00-19:00. Hypoglycaemia was observed as quadra-peaked in insulin users and double-peaked in oral hypoglycaemic agent users. Single-cosinor analysis revealed that the cycle was 5.83 hours (R=0.417) in insulin users, whereas it was 11.0 hours (R=0.717) in oral hypoglycaemic agent users. In insulin users, a significant periodicity of six hours (=0.003) was observed in the cosinor detection analysis, and a significant correlation (<0.05) was present in the cosinor percent rhythmicity analysis. In contrast, in oral hypoglycaemic agent users, a significant periodicity of 11 hours (=0.03) was ascertained in the cosinor detection analysis, and there was a significant correlation (<0.001) in the cosinor percent rhythmicity analysis. There were different hypoglycaemia-vulnerable hours of the day in the patients with type 2 diabetes, suggesting an interaction between disease pathophysiology and pharmacology.

Conclusions: These results can help elucidate the trend of the development of iatrogenic hypoglycaemia and contribute to the prevention of the onset thereof.