Intracerebral hemorrhage (ICH) accounts for over 10% of the approximately 17 million stroke cases worldwide each year. Although its incidence is lower than that of ischemic stroke, ICH is associated with a mortality rate exceeding 40%, and only about 20% of survivors regain functional independence within 6 months. This makes ICH one of the most lethal and difficult-to-treat subtypes of stroke (Broderick et al., 2021; Ohashi et al., 2023).

The circadian rhythm system plays a pivotal role in maintaining physiological homeostasis by synchronizing tissue-level functions with behavioral and environmental cycles such as sleep, activity, feeding, and light–dark rhythms. Disruption of circadian rhythms can lead to both acute and cumulative adverse health effects (Patke et al., 2020). The suprachiasmatic nucleus (SCN), located in the hypothalamus, acts as the central pacemaker of the circadian system. It transmits temporal cues to peripheral tissues via autonomic pathways, rhythmic secretion of melatonin from the pineal gland, and other efferent signals, thereby ensuring phase coherence across the body’s biological clocks (Zhou et al., 2022).

Among various circadian-regulated processes, the sleep–wake cycle is the most prominent in humans. This daily rhythm is governed by two opposing regulatory forces: the homeostatic sleep drive and the circadian wake drive (Dopp et al., 2024). The homeostatic drive increases with prolonged wakefulness and exposure to light, counteracting the circadian promotion of alertness. A well-regulated sleep–wake cycle is essential for brain function and overall health, offering neuroprotective benefits through several mechanisms, including reductions in metabolic rate and core temperature, modulation of inflammation and apoptosis, and enhancement of neuroplasticity—all of which contribute to neurological recovery (Pace et al., 2018).

These findings suggest that sleep may influence both the onset and clinical outcomes of stroke. However, existing research has predominantly focused on circadian rhythm disturbances and sleep architecture alterations in patients with ischemic stroke, exploring their potential prognostic implications (Miano et al., 2022; Wang et al., 2024).

Studies have shown that the 72-h period following the onset of intracerebral hemorrhage represents a high-risk time window for hematoma expansion and neurological deterioration, as well as a critical intervention period that determines patient prognosis (Lord et al., 2015; Zang et al., 2019; Anderson et al., 2010). Therefore, investigating alterations in the sleep–wake circadian rhythm during this crucial window is particularly important for understanding the mechanisms underlying central–peripheral rhythm coupling imbalance.

Studies investigating circadian and sleep–wake characteristics in patients with ICH, particularly during the acute phase, remain limited. Specifically, little is known about how the diurnal sleep–wake rhythm is affected in this population. Moreover, it remains unclear to what extent peripheral circadian rhythms can be autonomously maintained during the acute stage of cerebral hemorrhage.

The primary aim of this study is to comprehensively examine alterations in circadian sleep–wake patterns and sleep architecture in patients with ICH during the acute phase, and to characterize the specific manifestations of circadian rhythm disruption in this context.

Our study found that patients in the acute phase of hemorrhagic stroke exhibited a state of profound hypersomnia, with a total sleep duration of 17.81 h per day. This finding aligns with alterations in circadian rhythms observed in this population. Compared to the control group, the stroke group demonstrated significantly lower inter-daily stability (IS), intra-daily variability (IV), and relative amplitude (RA), despite being in the same ward environment. This suggests that environmental or nursing-related physical factors alone cannot fully explain the observed disruption in the sleep–wake circadian rhythm (Table 7).

The regulation of sleep–wake behavior is governed by both circadian and homeostatic mechanisms. Recent research has revealed that the neural circuitry involved in sleep–wake regulation includes multiple interconnected nuclei responsible for promoting either sleep or wakefulness, forming a complex and integrated regulatory network. This network is essential for maintaining the highly conserved and vital behavior of alternating sleep and wake states.

Key brain regions involved in sleep–wake regulation include the brainstem, hypothalamus, dorsal thalamus, and basal forebrain. These regions contain neurons that release various neurotransmitters, such as noradrenaline (NA), gamma-aminobutyric acid (GABA), acetylcholine, and histamine. These functionally distinct neuronal populations are primarily divided into two systems: the wake-promoting and sleep-promoting systems. These systems interact dynamically to regulate transitions between sleep and wake states (Scammell et al., 2017). The wake-promoting system comprises NA-releasing neurons in the locus coeruleus, serotonin (5-HT) neurons in the dorsal raphe nucleus (DRN), and neurons in the paraventricular nucleus of the dorsal thalamus (Ren et al., 2018). The sleep-promoting system includes GABAergic and galaninergic neurons in the ventrolateral preoptic area (VLPO) of the anterior hypothalamus (Saper and Fuller, 2017), as well as GABAergic neurons in the dorsal and ventral striatum and substantia nigra pars reticulata (Yuan et al., 2017; Liu et al., 2020; Oishi et al., 2017).

In our cohort, the predominant hemorrhage sites were the basal ganglia and thalamus—regions intimately involved in sleep–wake regulation (Figure 6). Dopamine D1 receptor-expressing neurons in the dorsal striatum initiate and maintain wakefulness by integrating upstream inputs from the cortex, thalamus, and substantia nigra, and inhibiting sleep-promoting neurons in the globus pallidus internus and substantia nigra pars reticulata via GABA release (Dong et al., 2022). The dorsal thalamus also plays a pivotal role in modulating cortical oscillations during wakefulness and sleep. Optogenetic activation of the central medial nucleus of the dorsal thalamus has been shown to induce rapid transitions from non-REM (NREM) sleep to wakefulness (Gent et al., 2018). Moreover, calretinin-positive neurons in the mediodorsal thalamic nucleus regulate arousal levels by integrating subcortical inputs and synchronizing signals to the cortex (Mátyás et al., 2018).

We observed that the sleep architecture of hemorrhagic stroke patients reflected a hypersomnolent state, with prolonged durations across all sleep stages and shortened latencies to sleep onset, consistent in both daytime and nighttime recordings. These findings are supported by animal studies; one such mouse model (Giordano et al., 2020) reported post-hemorrhagic increases in sleep due to microglial activation and elevated pro-inflammatory cytokines, which can modulate sleep quantity and quality during recovery (Rowe et al., 2019).

When the sleep–wake regulatory network is disrupted, significant diurnal and nocturnal alterations in sleep–wake behavior are evident. In our study, patients exhibited increased durations and proportions of light sleep, deep sleep, NREM and REM stages, as well as elevated sleep efficiency. Conversely, sleep latency, REM latency, and the proportion of NREM sleep were reduced during both day and night (Figures 3, 4). These results are consistent with a recent study (Maas et al., 2020) using actigraphy, which showed that patients with intracerebral hemorrhage rapidly entered a state of behavioral quiescence, suggesting an adaptive brain response to severe acute stress that favors immobility rather than wakefulness—potentially beneficial for recovery. This hypothesis is supported by animal models exploring sleep under stress, which indicate that sleep plays a conserved role in mitigating cellular homeostatic disturbances (Hill et al., 2014). Moreover, a retrospective cohort study (Alabdali et al., 2025)reported that such hypersomnolence during the acute phase could persist into the rehabilitation period after discharge in a subset of patients, thereby adversely affecting quality of life and long-term outcomes.

This hypersomnolent state may also involve GABAergic neurons in the dorsal raphe nucleus (DRN), a key component of the sleep–wake regulatory network. The DRN contains serotonergic (5-HT), dopaminergic (DA), GABAergic, and glutamatergic neurons (Monti, 2010; Liu et al., 2020). GABAergic neurons are the second most abundant population and have been shown to play diverse roles in sleep–wake modulation (Sulaman et al., 2023). Specifically, glutamic acid decarboxylase 2 (GAD2)-positive neurons in the DRN (DRN^GAD2) are activated under acute stress in mouse models and suppress arousal by inhibiting wake-promoting neurons in the paraventricular thalamus (PVT) (Ren et al., 2024).

The circadian rhythm system regulates a wide array of physiological functions, including arousal, sympathetic tone, cardiovascular activity, coagulation, immune function, glucose regulation, and metabolism (Zaki et al., 2020; Nishimon et al., 2021; Coiffard et al., 2021). During critical illness, encephalopathy is a major risk factor for rapid central circadian disruption (Maas et al., 2020). By analyzing rest-activity rhythms (RARs), we found that IS, IV, and RA were significantly reduced in patients with hemorrhagic stroke, indicating a loss of rhythmicity in the activity-rest cycle and a blunting of circadian fluctuations. These results are consistent with the observed hypersomnolent state in these patients.

By examining rest–activity rhythms (RARs) under the same environmental conditions, we observed that patients with intracerebral hemorrhage (ICH) exhibited lower IS, IV, and RA values compared with controls. These findings indicate weakened synchronization with external environmental cues, flattened within-day activity–rest fluctuations, and an overall reduction in circadian differentiation and rhythm strength. This phenomenon suggests that peripheral rhythms have only limited autonomous maintenance capacity under acute stress, with their expression dependent on intact central–peripheral coupling.

ICH may disrupt the synchronization between the suprachiasmatic nucleus (SCN) and peripheral clocks through mechanisms such as increased intracranial pressure, systemic inflammatory responses, neuroendocrine activation (e.g., glucocorticoid and catecholamine release), and autonomic imbalance, thereby leading to circadian destabilization at behavioral and physiological output levels. In other words, although peripheral molecular clocks may continue to oscillate at low amplitude, the lack of effective central driving and integration with external zeitgebers prevents sustained behavioral rhythms.

Furthermore, prior studies in patients with anterior circulation ischemic stroke (ACIS) and posterior circulation ischemic stroke (PCIS) (Yuan et al., 2025) demonstrated that ACIS was associated with poorer daytime stability, greater circadian variability, reduced light sensitivity, and sleep–wake instability, whereas PCIS patients retained better daytime coordination and circadian recognition. These differences highlight distinct patterns of impairment in the central circadian pacemaker and its downstream neurovascular circuits. ACIS more frequently affects the anterior circulation and SCN, resulting in phase shifts and sleep–wake imbalance, while PCIS relatively preserves central driving and maintains entrainment to external cues. In contrast, ICH exacerbates SCN and downstream circuit dysfunction through acute intracranial hypertension, hematoma compression, widespread autonomic imbalance, and systemic inflammation, leading to consistent reductions in IS, IV, and RA. Collectively, this suggests that acute-phase ICH exerts a circadian impact at least comparable to, if not more severe than, acute ischemic stroke, driven by both focal neural damage and systemic stress–metabolic perturbations.

Moreover, the stroke group demonstrated a higher prevalence of increased daytime sleep, overall prolonged sleep, and reversed sleep–wake cycles. Circadian misalignment is closely linked to the development of sleep disorders (Yamanaka, 2020). To investigate whether acute disruptions in RARs are related to the emergence of sleep–wake disturbances, we conducted logistic regression analyses using IS and IV as independent variables and increased daytime sleep and reversed sleep–wake cycle as dependent variables. The results revealed that a lower IS was significantly associated with reversed sleep–wake patterns. This finding indicates that IS may serve as a non-invasive and observable indicator of clinical risk for sleep disturbances.

Our study demonstrated that patients with acute ICH exhibited excessive sleepiness, disrupted sleep architecture, and attenuated circadian rhythms, characterized by significant reductions in IS, IV, and RA, as well as inverted sleep cycles. These results indicate that peripheral rhythms lack the capacity for independent maintenance under acute stress and rely on intact central–peripheral coupling. Multiple factors, including hematoma compression, intracranial pressure fluctuations, inflammatory responses, and autonomic imbalance, may jointly impair the circadian regulatory network. Clinically, such disturbances in sleep and circadian rhythms not only hinder neurological recovery in the acute phase but may also adversely affect long-term functional outcomes and quality of life. Future studies integrating circadian biomarkers (e.g., melatonin, orexin), genetic and molecular markers may further elucidate differential effects of ICH on central and peripheral rhythms, while personalized circadian-based interventions—such as light therapy, chronotherapy, and pharmacological modulation—could potentially improve sleep quality and neurological prognosis.