This study was aimed at assessing the impact of the dusk phenomenon on the total
glucose exposure in Chinese people with
type 2 diabetes.A total of 380
type 2 diabetes who received a retrospective continuous
glucose monitoring system (CGMs) for 72 hours were enrolled in our study, 32 of them failed in CGMs. The patients were first divided into 2 groups: dusk phenomenon (n = 95) and non dusk phenomenon group (n = 253). The magnitude of the dusk phenomenon (δDusk) was quantified by pre-dinner
glucose minus post-lunch 2 hours
glucose. A persistent δDusk ≥ 0 or a once only δDusk < 0 can be diagnosed with the dusk phenomenon. The participants were secondarily matched for the post-lunch 2 hours
glucose to assess the impact of the dusk phenomenon on the overall
glucose exposure. The impact of the dusk phenomenon was assessed on high-performance liquid chromatography assay (HbA1c) and 24-hour mean
glucose.There were 95 of 348 (27.3%) participants with the dusk phenomenon in the overall population, and the median of δDusk level was -0.8 (-1.8, 0.2) mmol/L. The median of
glucose differences between the 2 paired groups were 0.4 (-0.4, 1.0)% for HbA1c, 0.9 (0.2, 1.4) mmol/L for 24 hours mean
glucose. The correlation analysis showed no relationship between the magnitude of dawn phenomenon and the dusk phenomenon (r = 0.052, P = .472).The incidence of dusk phenomenon is about 27.3% in people with
type 2 diabetes. The impacts of dusk phenomenon on HbA1c and 24-hour mean
glucose were about 0.4% and 0.9 mmol/L and the dusk phenomenon was not related with the dawn phenomenon.