Factors associated with nocturnal and diurnal glycemic variability in patients with type 2 diabetes: a cross-sectional study.

Purpose: There is little information on factors that influence the glycemic variability (GV) during the nocturnal and diurnal periods. We aimed to examine the relationship between clinical factors and GV during these two periods.

Methods: This cross-sectional study included 134 patients with type 2 diabetes. 24-h changes in blood glucose were recorded by a continuous glucose monitoring system. Nocturnal and diurnal GV were assessed by standard deviation of blood glucose (SDBG), coefficient of variation (CV), and mean amplitude of glycemic excursions (MAGE), respectively. Robust regression analyses were performed to identify the factors associated with GV. Restricted cubic splines were used to determine dose-response relationship.

Results: During the nocturnal period, age and glycemic level at 12:00 A.M. were positively associated with GV, whereas alanine aminotransferase was negatively associated with GV. During the diurnal period, homeostatic model assessment 2-insulin sensitivity (HOMA2-S) was positively associated with GV, whereas insulin secretion-sensitivity index-2 (ISSI2) was negatively associated with GV. Additionally, we found a J-shape association between the glycemic level at 12:00 A.M. and MAGE, with 9.0 mmol/L blood glucose level as a cutoff point. Similar nonlinear associations were found between ISSI2 and SDBG, and between ISSI2 and MAGE, with ISSI2 value of 175 as a cutoff point.

Conclusion: Factors associated with GV were different between nocturnal and diurnal periods. The cutoff points we found in this study may provide the therapeutic targets for beta-cell function and pre-sleep glycemic level in clinical practice.