By Alan Z. Segal, MD
Associate Professor of Clinical Neurology, Weill Cornell Medical College
In a detailed clinical and electrophysiological study of sleep patterns in 12 patients with thalamic stroke, comparing them with 11 patients who had extrathalamic stroke, the investigators identified a marked decrease in slow wave sleep activity in the group with thalamic stroke. The clinical significance of this finding is uncertain but may have an effect on daytime cognitive performance.
Jaramillo V, Jendoubi J, Maric A, et al. Thalamic influence on slow wave slope renormalization during sleep. Ann Neurol 2021;90:821-833.
The thalamus is known to play a key role in arousal and sleep-wake states. With modulation of thalamic activity, direct changes in cortical neurons occur — a process elucidated recently in rodents using a technique known as “optogenetics.” By transgenic incorporation of light-sensitive proteins into cell membranes, in vivo nerve circuits can be directly manipulated into states of stimulation or inhibition. Seminal work involving optogenetic manipulation of the thalamus (specifically the centromedial nucleus [CMT]) has shown that, with tonic stimulation of the CMT, there is depolarization of cortical pyramidal neurons and production of spike activity (the so-called UP state).1 This has been demonstrated most prominently at the transition between non-rapid eye movement (NREM) sleep and the waking state. Conversely, burst stimulation of the CMT produces cortical quiescence (the DOWN state), which promotes sleep. When this burst activity is optimally synchronized, there is production of high amplitude, slow-wave activity (SWA), which is the hallmark of deep NREM sleep.
According to the authors, sleep-wake fluctuations of cortical activity, generated by the thalamus, contribute to plastic changes in the structure of synapses in the brain — a concept known as “synaptic homeostasis.” During wakefulness, there is a potentiation of synaptic activity, producing an increase in synaptic strength. With sleep, particularly with pronounced SWA, there is a reduction in synaptic strength — a process known as “synaptic renormalization.” Optimal renormalization theoretically would allow the brain to reboot in preparation for the next day. The study provides various lines of evidence for this process, such as animal studies using two-photon imaging and electron microscopy to show an increase in dendritic spines (“a key marker for synaptic strength”) during wakefulness and a decrease during sleep.
In their investigation, Jaramillo and colleagues studied patients who had thalamic strokes and compared them to stroke patients with lesions outside of the thalamus (“extrathalamic”). Recapitulating thalamic lesional studies done in animals, this research takes advantage of stroke as a “natural experiment” to better define these processes in living humans. The study included 23 patients, 12 with thalamic strokes and 11 with extrathalamic strokes. Thalamic lesions were located as follows: five paramedian, five inferolateral, and two tuberothalamic.
The vast majority of these patients (10/12) also had extrathalamic lesions. The 11 extrathalamic patients had lesions (which could be multiple) in the following regions: cortex (n = 8), basal ganglia (n = 5), midbrain cerebral peduncle, and medulla. Representative diffusion-weighted imaging (DWI) of all patients was included in the manuscript.
The extrathalamic subjects underwent 15 days of polysomnography (PSG) and the thalamic subjects underwent 17 days of PSG. Subjects had enhanced electroencephalogram (EEG) analysis during their PSGs, which included high-density lead placement and spectral analysis to compute the power of SWA. Sleep parameters using standard visual scoring did not differ between the two groups. By contrast, power analysis showed pronounced differences between thalamic and extrathalamic groups, with a marked decrease in SWA (defined as EEG power between 1 Hz and 4.5 Hz) among the thalamic stroke patients. This effect was seen in all cortical leads (frontal, central, temporal, parietal, and occipital) but was most striking in the right frontocentral region. Interestingly, this area has been shown to be the most sensitive brain region to the onset of drowsiness.
Subjects were tested daily using subjective sleepiness ratings (such as the Epworth Sleepiness Scale) and 11 cognitive tasks, such as the Corsi block-tapping task, which assesses visuospatial function and short-term working memory. Other tests captured motor/sensory examinations, verbal abilities, and executive function. The only statistically significant difference between the thalamic and extrathalamic patients was in Epworth score (as measured in the morning but not at other times, such as before bed). However, the authors observed that out of 13 total outcome measures, all but one pointed in the negative direction in thalamic subjects, a result well beyond that which could occur by chance (P = 0.05 by chi-squared test).
The authors outlined two mechanisms to explain their results, referred to as “wake-centered” and “sleep-centered” models, which, stated differently, can be considered a standard chicken-or-egg quandary. Both models focus on the concepts of synaptic potentiation during the day and synaptic renormalization at night. In the “wake-centered” scenario, lesions of the thalamus result in impairments in attention and “experience-dependent learning,” which blunts synaptic growth. This reduction then decreases the need for sleep-associated synaptic renormalization, manifested as decreases in SWA. In the “sleep-centered” version, the authors suggested that decreased SWA among the thalamic subjects creates impairments in renormalization, leading to reductions in daytime alertness and cognitive performance.
COMMENTARY
At a minimum, these data show that subjects with thalamic strokes sleep poorly and experience significant daytime cognitive impairments — a relationship that almost certainly is bidirectional. Whether the former contributes more to the latter or vice versa is not known. However, these data fall short of taking the next step of defining SWA generated by the thalamus as evidence of a restorative physiological role of sleep. The authors attempted to use the concept of synaptic renormalization to take limited data from animal experiments and translate them into humans. Synaptic renormalization is an elegant hypothesis, but it involves multiple assumptions and conjecture, and cannot be accepted as the mechanistic underpinning of the homeostatic need for sleep — a phenomenon that largely has eluded modern neuroscience.
REFERENCE
- Gent TC, Bandarabadi M, Gutierrez Herrera C, Adamantidis AR. Thalamic dual control of sleep and wakefulness. Nat Neurosci 2018;21:974-984.
