Cortisol as a Predictor of Nocturnal Hypoglycemia in Insulin-Treated Diabetes: A Cross-Sectional Study.

INTRODUCTION

Nocturnal hypoglycemia (NH) is a major clinical concern in insulin-treated diabetes due to blunted autonomic responses and reduced awareness of hypoglycemia during sleep. We investigated the association between NH and early morning fasting cortisol levels in this population.

METHODS

This case-control study included 30 insulin-treated adults with type 1 diabetes (n = 22) or advanced type 2 diabetes (n = 8) and depleted endogenous insulin secretion. Glucose profiles were assessed using intermittently scanned continuous glucose monitoring. NH was defined as glucose levels <70 mg/dL between 00:00 and 06:00. Fasting-morning serum cortisol, plasma glucagon, and serum C-peptide levels were also measured. The clinical and biochemical parameters were compared between patients with and without NH.

RESULTS

NH occurred in 15 of 30 patients (50.0%), including three (10.0%) with level 2 hypoglycemia (<54 mg/dL). There were no significant differences between the groups in terms of age, diabetes type, disease duration, HbA1c level, body mass index (BMI), insulin dose, or C-peptide level. However, fasting cortisol levels were significantly lower in patients with NH. Logistic regression analysis showed that lower cortisol levels were independently associated with NH (odds ratio: 0.708; 95% confidence interval: 0.52-0.97; p < 0.05). Receiver operating characteristic curve analysis identified a cortisol threshold of 10.7 μg/dL for predicting NH (area under the curve = 0.79, sensitivity = 0.73, specificity = 0.80).

CONCLUSIONS

Lower early morning cortisol levels may serve as an independent risk factor for NH in patients with insulin-treated diabetes. Cortisol measurements may help identify individuals at risk of asymptomatic NH.