Stanford researchers reveal surprising links between brain activity and glucose dynamics
May 23, 2022
A Stanford study uncovers new links between brain activity and glucose levels, which may pave the way for innovative approaches to metabolic health.
Stanford researchers conducted the first in-human study investigating how glucose levels in the brain are related to specific patterns of brain activity. Mounting evidence in animal studies have shown that the central nervous system plays a critical role in regulation of peripheral metabolism, however studies in human have been sparse. The study, recently published in Nature Communications, combined intracranial monitoring of brain activity and continuous interstitial glucose monitoring to quantify the relationship between brain activity and glucose dynamics in unprecedented detail.
Yuhao (Danny) Huang, MD
Stanford Neurosurgery PGY-4 resident Yuhao (Danny) Huang, MD, is the lead author.
The study involved three patients with epilepsy who were monitored in a hospital over a period of several days. They underwent routine brain activity using stereo-electroencephalography (sEEG), a method of recording brain activity using depth electrodes implanted in the brain. In one subject, the electrodes recorded from the hypothalamus. As the hypothalamus has been long implicated in central control of peripheral glucose, this presented a unique opportunity to study how hypothalamic activity was related to glucose metabolism.
The results showed that glucose levels were significantly correlated with variations in activity over multiple brain regions and was influenced by a multitude of factors including sleep-wake cycle, circadian rhythm and meals. In particular, the regional degree of correlation found was influenced by its connectivity to the hypothalamus. Finally, the research team applied a machine learning algorithm trained on power features of the recorded brain activity to learn their relationship with glucose variations. They found that this algorithm was able to not only decode current glucose levels using intracranial brain activity but also predict future glucose levels.
The researchers hope that their findings could inspire novel strategies for monitoring and treating metabolic conditions by modulating the brain.
In the below Q&A with Dr. Huang, we discussed the study’s significance, clinical relevance, and an unexpecting finding.
What is the significance of your publication?
We provide the first in-human study combining longitudinal intracranial brain recordings with continuous glucose monitoring (CGM), which allowed us to describe the relationship between brain activity and glucose dynamics in unprecedented detail. We found that correlations between brain activity and glucose levels are present in multiple brain regions, occur over multiple time scales and are modulated by sleep-wake cycles. Importantly, we found that it was possible to use brain activity to accurate predict glucose levels not only in the present but also up to hours in advance in the future.
What is the clinical relevance?
Metabolic diseases such as diabetes have traditionally been considered as a peripheral organ issue. There are now emerging evidence demonstrating targeting the central nervous system can be an effective option, especially for refractory disease. Our study provides further evidence that there is a close relationship between the central nervous system and peripheral metabolism. We hope this inspire future strategies that take into account brain activity in managing refractory metabolic diseases.
How does this research fit into previous or ongoing work on this topic?
The human brain is an avid consumer of the total body glucose – in fact, it accounts for 80% of the total body glucose utilization. This means that the brain has a strong incentive to monitor of the body glucose, such that it has consistent access to its preferred energy item. Although the pancreas is well known for its involvement in glucose homeostasis, there is mounting evidence indicate that the brain is also heavily involved. Studies in animals have shown that there are glucose-sensing neurons in brain, which are responsible for monitoring glucose levels in the body, and enacting change in the body when there are deviations from the normal glucose levels. These neurons are scattered throughout the brain, but their role in glucose homeostasis has been best established in the hypothalamus. Despite the ample evidence in animals, there is limited data in humans whether brain activity is related to glucose levels. Our study builds on prior animal studies by finding that there is strong coupling between brain activity and peripheral glucose dynamics, particularly in the hypothalamus.
Were there surprising or unexpected findings in your study?
Yes! Although we hypothesized that brain activity would be correlated to peripheral glucose variations, we did not expect that brain activity was sufficient for decoding of glucose levels, especially for future glucose levels. This was surprising to us, but it underscores that the brain may be engaged in a homeostatic role for metabolic processes hours in advance.