Share on Facebook
Share on Twitter
Share on LinkedIn
Share by Email
Print
Dusk phenomenon among individuals with type 2 diabetes (T2D) was observed among 27.3% and was not related with dawn phenomenon, according to results of an observational study, published in Medicine.
While the dawn phenomenon in diabetes is well-established and marked by a spontaneous rise in insulin in the early morning hours, little is known if the dusk phenomenon, or pre-dinner hyperglycemia, is true. In this featured study, researchers describe the dusk phenomenon among 348 patients with type 2 diabetes.
“The concept of dusk phenomenon is both familiar and unfamiliar to endocrinologists. Although there has been a lack of definition for a long time, the dusk phenomenon is often encountered in clinical practice,” write researchers who were led by Xiaoming Tao, of Fudan University in Shanghai.
In this study, patients (N=348) with T2D who were stable for ³3 months were recruited at the Huadong Hospital in China. Patients were given nutritional therapy and retrospectively monitored continuously for glucose levels over 2 days. Dawn phenomenon was calculated by pre-breakfast glucose minus nocturnal nadir glucose and dusk phenomenon by pre-dinner glucose minus 2-hour post-lunch glucose.
Patients were majority women (n=190), aged mean 63.1±11.5 years, with a BMI of 25.1±3.2 kg/m2, and 24-hour mean glucose of 7.5±1.2 mmol/L.
A minority of patients exhibited dusk phenomenon (n=66). Compared with an age- and gender-matched cohort of patients without dusk phenomenon, patients with dusk phenomenon had lower homeostasis model assessment-insulin resistance (HOMA)-b (median, 48.6 vs 63.6; P =.009) and higher glycated hemoglobin (mean, 7.1%±0.8% vs 6.7%±0.8%; P =.004) and 24-hour mean glucose (8.1±1.0 vs 7.1±0.9 mmol/L; P <.0001).
These differences corresponded with a median effect of the dusk phenomenon of 0.9 (interquartile range [IQR], -0.3 to 1.7; P <.0001) on overall nocturnal nadir; and specific effects of 0.8 (IQR, 0-1.8; P =.001) on pre-breakfast, 1.2 (IQR, -0.6 to 2.8; P =.003) on 2 hours after breakfast, 0.6 (IQR, -0.8 to 2.2 ; P =.004) on pre-lunch, 2.1 (IQR, 1.2-3.3; P <.0001) on pre-dinner, and 1.2 (IQR, -0.1 to 2.8; P <.0001) on 2 hours after dinner nadir. In a logistic analysis, HOMA-b was the only significant independent variable associated with the dusk phenomenon (odds ratio [OR], 0.989; 95% CI, 0.980-0.998; P =.017).
These data indicated over a quarter of patients with T2D had dusk phenomenon which altered their mean glucose by 0.9 mmol/L and glycated hemoglobin by 0.4%.
“Another important finding for our study was that the dusk phenomenon could not only cause hyperglycemia before dinner, but also affect blood glucose in different degrees throughout the day, including nocturnal nadir, fasting, post-breakfast, pre-lunch, and post-dinner glucose,” the authors wrote.
The dawn phenomenon can be attributed to the loss of β cell function, growth hormone-induced insulin resistance, increasing of circulating insulin-like growth factor binding protein-1 in the morning, poor sleep quality, and the impairment of circadian clock. But this study showed no correlation between the dawn phenomenon and the dusk phenomenon.
“So, we can only speculate on the mechanism of dusk phenomenon. In our study, HOMA- β was significant independent correlates with the dusk phenomenon, which means that loss of β cell function could contribute substantially to the pre-dinner hyperglycemia. In addition, there is a circadian rhythm in human insulin secretion and insulin sensitivity, and the circadian rhythm of insulin secretion may be one of the causes of the dusk phenomenon. Furthermore, we speculate that the dusk phenomenon may be related to the weakening of the ambient light during the dinner time,” the authors wrote.
This study was limited by not assessing growth hormone, which can impact the dawn phenomenon.
Disclosures: n/a
Reference
Huang Y, Xu Y, Qiu J, et al. The impact of dusk phenomenon on total glucose exposure in Chinese people with type 2 diabetes. Medicine.. 2021;100(13):e25298. doi:10.1097/MD.0000000000025298.
