Study on the circadian rhythm and sleep state characteristics of patients with acute cerebral infarction accompanied by sleep-breathing disorders

Abstract

There is limited research on the circadian rhythm and sleep state in patients with acute cerebral infarction (ACI) accompanied by sleep-breathing disorders (SDB). This study aims to provide a scientific basis for individualized diagnosis and treatment for stroke-related SDB patients. The SC-500 sleep monitor was used to continuously monitor 1367 ACI patients over 5 days. Based on the apnea–hypopnea index (AHI), patients were divided into non-SDB group (normal) and SDB group (mild, moderate, severe, fluctuating). Interdaily stability (IS) and intradaily variability (IV) were calculated through heart rate monitoring, and sleep states and their correlations were analyzed. Compared to the non-SDB group, patients with moderate-to-severe ACI accompanied by SDB showed decreased IS, increased IV, and sleep fragmentation. Significant statistical differences were observed in total sleep time (TST), rapid eye movement latency (REML), sleep efficiency (SE), non-rapid eye movement stages 1–2 (NREM stages1–2), non-rapid eye movement stages 3–4 (NREM stages 3–4), proportion of non-rapid eye movement (NREM%), wake after sleep onset (WASO), and number of awakenings (NOA) between the SDB group and the non-SDB group (P < 0.05). AHI showed a strong negative correlation with IS and a strong positive correlation with IV. AHI was positively correlated with sleep latency (SL), REML, NREM stages1–2, NREM%, proportion of rapid eye movement (REM%), WASO, time out of bed (TOB), and NOA, and negatively correlated with TST, SE, NREM stages 3–4, and rapid eye movement (REM), all with statistical significance (P < 0.05). There were significant statistical differences in the Mini-Mental State Examination (MMSE) between patients with and without SDB, and among mild, moderate, severe, and fluctuating groups (P < 0.05). Patients with moderate-to-severe ACI accompanied by SDB are more likely to experience changes in circadian rhythm and sleep states, which in turn affect cognitive functions.

Supplementary Information

The online version contains supplementary material available at 10.1007/s41105-024-00516-1.

Introduction

Acute cerebral infarction (ACI) refers to the pathological changes of ischemia and necrosis in neurons within a specific perfusion area due to acute occlusion of cerebral vessels, which triggers the respective neurological and neuroconductive dysfunctions in the brain region [1]. Sleep-disordered breathing (SDB) has been recognized as a common comorbidity in stroke patients [2]. The diagnostic criteria for SDB include a frequency of respiratory cessation and hypoventilation events ≥ 5 times/h, with symptomatic manifestations such as snoring, frequent nocturnal awakenings, and excessive daytime sleepiness [3]. Recent research indicates that approximately 40–70% of stroke patients may suffer from SDB [4]. The investigation into ACI accompanied by SDB is gaining increased attention. However, research concerning the circadian rhythm and sleep status of these patients remains significantly lacking. When the AHI reaches ≥ 15 times/h, it also manifests in declined attention, impaired executive function, memory loss, and emotional dysregulation [5]. This study employs continuous sleep monitoring technology to explore the dynamic changes in circadian rhythm and sleep status among ACI patients with SDB. The aim is to provide a scientific basis for optimizing the clinical management of ACI patients with SDB, and improving their neurological function recovery.

Subjects and methods

Research subjects

Patients diagnosed with ACI, who were admitted to the Department of Neurology at Kailuan General Hospital and underwent cranial magnetic resonance (MR)-diffusion weighted imaging (DWI) between November 2020 and December 2022, were selected for this study. Inclusion criteria: (1) age over 18 years; (2) meeting the diagnostic criteria of the '2021 Guidelines for the Prevention of Stroke and Transient Ischemic Attack: A Guideline from the American Heart Association/American Stroke Association' [6]; (3) continuous monitoring of heart rate and sleep state for ≥ 5 days with valid data ≥ 8 h/day; and patients monitored within 7 days of onset. Exclusion criteria: (1) consciousness disorders; (2) recent use of central nervous system stimulants, anti-anxiety/depression medications, antiepileptic drugs, or sedative-hypnotic drugs; (3) history of psychiatric disorders; (4) severe cognitive impairment and uncooperative with examination; (5) history of other neurological diseases: neuromuscular junction and muscular diseases, neurodegenerative diseases; (6) also having malignant tumors; (7) also having other serious physical diseases and unable to cooperate with the examination.

Research methods

Baseline data collection

Detailed records were kept of the selected participants' age, gender, body mass index (BMI), educational level, smoking history, and drinking history, as well as histories of hypertension, diabetes, and hyperhomocysteinemia (HHcy).

Localization diagnosis of cerebral infarction and National Institutes of Health Stroke Scale grade assessment

The location of the cerebral infarction lesions was determined by neuroimaging specialists and neurologists based on the patient's MR results and clinical characteristics. The lesions were categorized into cerebral hemisphere, thalamus, brainstem, cerebellum, and multiple locations of infarction. Additionally, the severity of the stroke was assessed using the National Institutes of Health Stroke Scale (NIHSS) grading system.

Diagnosis criteria and grouping of SDB

The diagnosis is based on the criteria set forth in the 'International Classification of Sleep Disorders, Third Edition,' published by the American Academy of Sleep Medicine [7]. The SC-500-type sleep monitor (produced by Nanjing Bochuang Haiyun Electronic Technology Co., Ltd.), a Level III sleep monitoring technology according to the classification of the American Academy of Sleep Medicine, was used. The apnea–hypopnea index (AHI) was measured to assess the condition of sleep apnea [8, 9]. An AHI of ≥ 5 times/h was used as the criterion: patients with AHI consistently < 5 times/h over 5 days were classified as the non-SDB group (normal); those with AHI ≥ 5 times/h were classified as the SDB group. Within the SDB group, patients with AHI ≥ 5 and < 15 times/h were categorized as mild SDB; those with AHI ≥ 15 and < 30 times/h as moderate SDB; and those with AHI ≥ 30 times/h as severe SDB. If the AHI showed different levels over the 5 days of continuous monitoring, the patient was defined as belonging to the SDB fluctuating group.

Circadian rhythm and sleep state data collection

After admission, patients were monitored for 24 h using the SC-500-type sleep monitor starting from their bedtime. Data collection was carried out by five graduate students from the research team under the supervision of their advisor. Heart rate monitoring was conducted continuously for 5 days (average hourly heart rate from 0:00 to 23:59 each day), and the '24-h Chi-square formula' was used in Excel 16.0 to calculate the circadian rhythm—interdaily stability (IS) and intradaily variability (IV) [10, 11]. The sleep monitor, placed under the mattress, used highly sensitive pressure sensors to monitor heart rate, respiratory and body movement signals, identifying different sleep stages [12, 13], and collecting data on total sleep time (TST), sleep latency (SL), rapid eye movement latency (REML), sleep efficiency (SE), non-rapid eye movement stages 1–2 (NREM stages 1–2) (light sleep), non-rapid eye movement stages 3–4 (NREM stages 3–4) (deep sleep), rapid eye movement (REM), wake after sleep onset (WASO), time out of bed (TOB), number of awakenings (NOA), and AHI [14]. Patients were excluded from the group if heart rate data were missing for ≥ 8 h/day due to reasons like examinations or leaving the ward; for missing data < 8 h/day, the 24-h average heart rate was used as a substitute. Sleep states were determined based on the longest duration within 24 h. The criteria for IS/IV assessment: (1) IS measures the coordination of the 24-h biological rhythm pattern, representing the average variation rate between activities every 24 h. Calculated based on overall variability and mean values, it reflects the synchrony between resting and active rhythms and zeitgebers. Scores range from 0 to 1, with higher scores indicating closer alignment with the 24-h biological rhythm pattern. (2) IV measures the stability of the circadian rhythm, representing the frequency of transitions between rest and activity, based on hourly activity data. Its value ranges from 0 to 1, with higher scores indicating an increased number of transitions from rest to activity, thus showing increased circadian rhythm instability [15]. The 24-h Chi-square formula is as follows:

$$\text{IS}=\frac{n{\sum }_{h=1}^n{\left({\overline{\chi }}_h-\overline{\chi }\right)}^2}{p{\sum }_{i=1}^n{\left({\chi }_i-\overline{\chi }\right)}^2}\text{IV}=\frac{n{\sum }_{i=2}^n{\left({\chi }_i-{\chi }_{i-1}\right)}^2}{\left(n-1\right){\sum }_{i=1}^n{\left({\chi }_i-\overline{\chi }\right)}^2}$$

n indicates the total count of heart rates per hour; p indicates the number of hours in a day (24 h); h indicates the average heart rate of the nth hour in the total count of heart rates per hour; $\overline{x}$ indicates the average heart rate per hour; i indicates the heart rate of the ith hour.

Mini-Mental State Examination assessment

Upon admission, patients immediately underwent screening using the Mini-Mental State Examination (MMSE) for cognitive assessment. This includes evaluation of temporal orientation, spatial orientation, memory, attention, calculation, recall, naming, repetition, comprehension, reading, writing, and constructional abilities. Patients were then categorized based on the presence or absence of SDB and the degree of severity—mild, moderate, severe, or fluctuating.

Statistical methodology

The statistical analysis was carried out using SPSS 26.0 software. Quantitative data conforming to normal distribution are represented by ($\overline{x}\pm s$). Comparison between groups was conducted using independent samples t test and one-way analysis of variance (ANOVA), where homogeneity of variances was assessed by the ANOVA method, and multiple comparisons were conducted using the Bonferroni method. In cases of heterogeneity of variance, the Welch method was employed, and multiple comparisons were performed using the Tanhane method. Non-normally distributed quantitative data are represented by median M (Q1–Q3), and intergroup comparisons were conducted using the Kruskal–Wallis H test. Categorical data were represented by the frequency and proportion, and intergroup comparison was performed using the χ² test. Correlation analysis was conducted using Spearman's correlation analysis, with P < 0.05 considered statistically significant.

Results

General information

The study included a total of 1367 cases of ACI, including 704 cases of recurrent stroke. There were 963 male patients (69.9%) and 404 female patients (30.1%), with ages ranging from 33 to 92 years (66.12 ± 10.62 years). The non-SDB group consisted of 147 cases, and the SDB group had 1220 cases; within the SDB group, there were 248 cases in the mild group, 395 in the moderate group, 295 in the severe group, and 282 in the fluctuating group.

Comparison of baseline and clinical data between ACI patients with and without SDB

In the baseline data, there were no statistically significant differences in gender, age, BMI, educational level, smoking history, drinking history, history of hypertension, diabetes, and HHcy between the non-SDB and SDB groups (P > 0.05). In clinical data, compared to the non-SDB group (normal), the SDB group showed a significantly higher incidence of thalamic and brainstem infarctions (P < 0.05). Additionally, the proportion of patients with NIHSS grade-severe group was higher in the SDB group compared to the non-SDB group (normal), with this difference being statistically significant (P < 0.05), as shown in Table 1.

Table 1.

Comparison of baseline and clinical data between ACI patient groups with no SDB and SDB groups

Variables	No SDB group (n = 147)	SDB group (n = 1220)	χ2 (t) value	P value
Gender, n (%)			1.574	0.210
Male	97 (66.0%)	866 (71.0%)
Female	50 (34.0%)	354 (29.0%)
Age ($\overline{x}\pm s$), years	65.63 ± 11.18	66.18 ± 10.56	− 0.602a	0.547
BMI (kg/m2)	27.05 ± 4.50	27.74 ± 4.09	− 1.928a	0.054
Education level, n (%)	–	–	0.512	0.474
Junior high and below	129 (87.8%)	1094 (89.7%)
High school and above	18 (12.2%)	126 (10.3%)
History of smoking, n (%)	–	–	0.039	0.843
No	82 (55.8%)	691 (56.6%)
Yes	65 (44.2%)	529 (43.4%)
History of alcohol consumption, n (%)	–	–	0.747	0.387
No	90 (61.2%)	791 (64.8%)
Yes	57 (38.8%)	429 (35.2%)
History of hypertension, n (%)	–	–	1.833	0.176
No	41 (27.9%)	408 (33.4%)
Yes	106 (72.1%)	812 (66.6%)
History of diabetes, n (%)	–	–	0.056	0.813
No	97 (66.0%)	793 (65.0%)
Yes	50 (34.0%)	427 (35.0%)
History of HHcy, n (%)	–	–	1.018	0.313
No	88 (59.9%)	677 (55.5%)
Yes	59 (40.1%)	543 (44.5%)
Location of cerebral infarction lesions, n (%)	–	–	–	–
Cerebral hemisphere	22 (15.0%)	119 (9.8%)	3.852	0.051
Thalamus	23 (15.6%)	279 (22.9%)	3.976	0.046
Brainstem	54 (36.7%)	555 (45.5%)	5.759	0.044
Cerebellum	20 (13.6%)	106 (8.7%)	3.790	0.052
Multiple locations	28 (19.0%)	161 (13.2%)	3.769	0.052
NIHSS severity, n (%)	–	–	–	–
Normal	21 (14.3%)	118 (9.7%)	3.057	0.080
Mild	43 (29.3%)	293 (24.0%)	3.833	0.164
Moderate	44 (29.9%)	378 (31.0%)	0.068	0.794
Severe	39 (26.5%)	431 (35.3%)	4.500	0.034

ACI acute cerebral infarction, SDB sleep-disordered breathing, BMI body mass index, HHcy hyperhomocysteinemia, NIHSS National Institutes of Health Stroke Scale

^aUsing t value

Comparison of circadian rhythms between ACI patients with no SDB (normal) and SDB groups (mild, moderate, severe, fluctuating)

Statistically significant differences were observed in IS and IV across the aforementioned five groups (P < 0.05). In intergroup comparisons, there were significant statistical differences between the non-SDB group (normal) and the moderate-to-severe SDB groups (P < 0.05); between the mild SDB group and the moderate-to-severe SDB groups (P < 0.05); and between the SDB fluctuating group and the moderate-to-severe SDB groups (P < 0.05), as shown in Table 2.

Table 2.

Comparison of circadian rhythms between ACI patient groups with no SDB (normal) and SDB groups (mild, moderate, severe, fluctuating)

No SDB		SDB				F value	P value
	Normal group (n = 147)	Mild group (n = 248)	Moderate group (n = 395)	Severe group (n = 295)	Fluctuating group (n = 282)
IS	0.47 ± 0.16	0.46 ± 0.12	0.39 ± 0.18ab	0.32 ± 0.17abc	0.44 ± 0.12 cd	42.869	< 0.001
IV	0.66 ± 0.14	0.69 ± 0.14	0.74 ± 0.16ab	0.82 ± 0.13abc	0.79 ± 0.16ab	47.118	< 0.001

ACI acute cerebral infarction, SDB sleep-disordered breathing, IS interdaily stability, IV intradaily variability

^aIndicates comparison with the normal group, P < 0.05

^bIndicates comparison with the mild group, P < 0.05

^cIndicates comparison with the moderate group, P < 0.05

^dIndicates comparison with the severe group, P < 0.05

Comparison of sleep states between ACI patients with no SDB (normal) and SDB groups (mild, moderate, severe, fluctuating)

Statistically significant differences were observed in TST, REML, SE, NREM stages 1–2, NREM stages 3–4, NREM%, WASO, and NOA among the aforementioned five groups (P < 0.05). In intergroup comparisons, compared to the non-SDB group (normal), significant differences were observed in the moderate group for TST, in the severe group for REML and SE, in the moderate group for NREM%, in the mild-to-severe groups for WASO, and in the moderate group for NOA (P < 0.05). Compared to the mild group, there were significant differences in NREM stages 1–2 in the moderate-to-severe groups, in NREM stages 3–4 in the severe group, and in WASO in the moderate-to-severe groups (P < 0.05). Compared to the moderate group, there was a significant difference in NREM stages 3–4 in the severe group (P < 0.05), as shown in Table 3.

Table 3.

Comparison of sleep status between ACI patient groups with no SDB (normal) and SDB (mild, moderate, severe, fluctuating)

No SDB		SDB				F(H) value	P value
	Normal group (n = 147)	Mild group (n = 248)	Moderate group (n = 395)	Severe group (n = 295)	Fluctuating group (n = 282)
TST (min)	516.84 ± 92.36	509.48 ± 101.03	488.56 ± 108.06a	501.36 ± 101.51	507.38 ± 97.39	2.996	0.018
SL (min)	16(7–37)	15(6–33)	15(4–31)	15(5–27)	14(4–32)	#6.942	0.139
REML (min)	76(65–106)	80(67–108)	82(63–104)	85(68–108)a	82(63–104)	#9.599	0.048
SE (%)	0.83 ± 0.08	0.79 ± 0.13	0.79 ± 0.12	0.77 ± 0.14a	0.80 ± 0.11	31.996	< 0.001
NREM1–2 (min)	305.37 ± 90.6	300.41 ± 84.04	320.52 ± 79.72b	325.66 ± 88.32b	308.39 ± 81.48	3.892	0.004
NREM3–4 (min)	57(29–78)	60(38–85.5)	60(39–88)	55(36–88)bc	56(33–80)	#14.150	0.007
REM (min)	112.96 ± 32.91	113.36 ± 34.84	107.48 ± 32.01	109.45 ± 32.86	106.88 ± 34.56	2.039	0.089
NREM (%)	0.71 ± 0.16	0.74 ± 0.13	0.75 ± 0.10a	0.74 ± 0.12	0.74 ± 0.11	2.463	0.044
REM (%)	0.23 ± 0.07	0.23 ± 0.06	0.22 ± 0.05	0.22 ± 0.06	0.22 ± 0.03	2.091	0.081
WASO (min)	54(44–64)	60(47–75)a	65(50–85)ab	66(48–128)ab	62(47.75–121.25)	#42..880	< 0.001
TOB (min)	38(29–52)	37(29–45)	39(22–52)	38(23–48)	37(18–60)	#2.803	0.591
NoA (times/h)	14(4–20)	16(9–22)	17(7–24)a	17(8–24)	16(6–26)	#17.783	< 0.001

ACI acute cerebral infarction, SDB sleep-disordered breathing, TST total sleep time, SL sleep latency, REML rapid eye movement latency, SE sleep efficiency, NREM1–2 non-rapid eye movement 1–2 (light sleep), NREM3–4 non-rapid eye movement 3–4 (deep sleep), REM rapid eye movement, NREM% proportion of non-rapid eye movement, REM% proportion of rapid eye movement, WASO wake after sleep onset, TOB time out of bed, NoA number of awakenings

^#Using H value

^aIndicates comparison with the normal group, P < 0.05

^bIndicates comparison with the mild group, P < 0.05

^cIndicates comparison with the moderate group, P < 0.05

^dIndicates comparison with the severe group, P < 0.05

Spearman correlation analysis of AHI with circadian rhythm and sleep state parameters in ACI with SDB

AHI showed a strong negative correlation with IS (r_s = − 0.817) and a strong positive correlation with IV (r_s = 0.835). AHI was positively correlated with SL (r_s = 0.231), REML (r_s = 0.302), NREM stages 1–2 (r_s = 0.742), NREM% (r_s = 0.251), REM% (r_s = 0.139), WASO (r_s = 0.114), TOB (r_s = 0.341), and NOA (r_s = 0.254). AHI was negatively correlated with TST (r_s = − 0.119), SE (r_s = − 0.609), NREM stages 3–4 (r_s = − 0.361), and REM (r_s = − 0.536). All the above comparisons were statistically significant (P < 0.001), as shown in Table 4.

Table 4.

Spearman correlation analysis results of AHI with circadian rhythm and sleep state parameters in ACI patients with SDB

Variables	rs value	P value
IS	− 0.817	< 0.001
IV	0.835	< 0.001
TST	− 0.119	< 0.001
SL	0.231	< 0.001
REML	0.302	< 0.001
SE	− 0.609	< 0.001
NREM1–2	0.742	< 0.001
NREM3–4	− 0.361	< 0.001
REM	− 0.536	< 0.001
NREM%	0.251	< 0.001
REM%	0.139	< 0.001
WASO	0.114	< 0.001
TOB	0.341	< 0.001
NoA	0.254	< 0.001

ACI acute cerebral infarction, SDB sleep-disordered breathing, AHI apnea–hypopnea index, IS interdaily stability, IV intradaily variability, TST total sleep time, SL sleep latency, REML rapid eye movement latency, SE sleep efficiency, NREM1–2 non-rapid eye movement 1–2 (light sleep), NREM3–4 non-rapid eye movement 3–4 (deep sleep), REM rapid eye movement, NREM% proportion of non-rapid eye movement, REM% proportion of rapid eye movement, WASO wake after sleep onset, TOB time out of bed, NoA number of awakenings

MMSE score results for ACI patients with no SDB (normal) and SDB groups (mild, moderate, severe, fluctuating)

The MMSE score for the non-SDB group (normal) was 26.40 ± 4.17. In the SDB group (mild, moderate, severe, fluctuating), the MMSE scores were 25.80 ± 5.26, 21.70 ± 4.82, 18.84 ± 4.64, and 22.72 ± 5.63, respectively. In comparison with the non-SDB group (normal), statistically significant differences were found in the moderate, severe, and fluctuating SDB groups (P < 0.05). Compared to the mild group, there were statistically significant differences in the moderate, severe, and fluctuating SDB groups (P < 0.05). When compared with the moderate group, there was a statistically significant difference in the severe group (P < 0.05), and when compared with the severe group, there was a statistically significant difference in the fluctuating group (P < 0.05) (Table 5).

Table 5.

MMSE score results for ACI patient groups with no SDB (normal) and SDB (mild, moderate, severe, fluctuating)

Severity of SDB	Cognitive function assessment value	F value	P value
Normal group	26.40 ± 4.17	94.429	< 0.001
Mild group	25.80 ± 5.26
Moderate group	21.70 ± 4.82ab
Severe group	18.84 ± 4.64abc
Fluctuating group	22.72 ± 5.63abd

MMSE Mini-Mental State Examination, ACI acute cerebral infarction, SDB sleep-disordered breathing

^aIndicates comparison with the normal group, P < 0.05

^bIndicates comparison with the mild group, P < 0.05

^cIndicates comparison with the moderate group, P < 0.05

^dIndicates comparison with the severe group, P < 0.05

Discussion

The core regulatory structure of the sleep–wake circadian rhythm comprises the retina, the suprachiasmatic nucleus (SCN) of the hypothalamus, and the pineal gland. These components regulate the rhythmicity of physiology and behavior through neurochemical interactions [16]. Among them, the SCN is considered the dominant center, acting as the 'pacemaker' of the rhythm [17]. It is regulated by both external environmental light cycles and internal neurotransmitters and hormones [18]. The SCN plays a leading role in regulating the autonomic nervous system. Heart rate is regarded as an indicator of the balance between the sympathetic and parasympathetic nervous systems. In recent years, many studies have used heart rate variability as a predictor of circadian rhythm parameters [19–21]. Measurement of IS and IV enables a more precise assessment of sleep–wake circadian rhythm disorders and their impact on sleep state [22]. Disruption or imbalance of circadian rhythms can widely affect disease pathology, physiological functions, mental health, and emotions, including sleep disorders, memory decline, mood instability, cognitive impairment, and hormonal imbalances [23].

In this study, we conducted a 5-day continuous observation of 1367 ACI patients, grouping them based on their AHI into non-SDB (normal) and four SDB severity categories: mild, moderate, severe, and fluctuating. We discovered that the degrees of IS and IV changes, as observed through continuous heart rate monitoring, are closely related to the severity of SDB. All patients with IS scores within the 0–1 range maintained a 24-h circadian rhythm. However, in patients with moderate-to-severe SDB, daytime stability was significantly impaired, showing a deviation in the sleep–wake pattern's synchronization with the external environment. Additionally, among the 1067 patients with IV scores between 0 and 1, the diurnal rhythm showed fragmentation, and 110 patients with IV scores greater than 1, particularly pronounced in severe SDB patients, indicated increased circadian rhythm fragmentation. This is attributed to frequent nocturnal snoring, partial airway obstruction, increased sensitivity of peripheral chemoreceptors, fluctuations in oxygen partial pressure and HCO3⁻ concentration, and adjustments in thoracic negative pressure, all contributing to sleep–wake circadian rhythm disorder [24, 25]. Our study results indicate a significant negative correlation of SDB severity with IS (r_s = − 0.817) and a significant positive correlation with IV (r_s = 0.835). That is, the higher the AHI, the worse is the IS and the more significant is the IV. Previous studies have shown a correlation between increased per-hour apnea–hypopnea events and the secretion patterns of IS/IV neurotransmitters and hormones [26]. Ahmad et al. reported that pinealocytes in the suprachiasmatic nucleus uptake tryptophan, which is then converted to melatonin under the catalysis of tryptophan hydroxylase. Regulated by the SCN, the paraventricular nucleus of the hypothalamus, and the lateral nuclei in the spinal cord brain region, melatonin's influence on the circadian rhythm changes of the sleep–wake cycle is modulated via the STAT3 pathway [27]. Following a stroke-related SDB complication, melatonin secretion rhythm is impaired, with decreased nighttime secretion amplitude or altered peak secretion phase, leading to early awakening, daytime fatigue, fragmented sleep, inattention, decreased judgment, difficulty in multi-tasking situations, and sustained attention to tasks [28]. Furthermore, neurotransmitter (serotonin, dopamine, gamma-aminobutyric acid) secretion changes along with IS/IV abnormalities, resulting in irritability, restlessness, lack of interest, and mood depression, weakening stress response. These physiological changes lead to various psychological disorders, including anxiety, depression, seasonal affective disorder, and bipolar disorder, causing patients to lose confidence and hope in life over time [29].

Research indicates a close connection between ACI and SDB. Apnea episodes lead to reduced blood oxygen saturation causing cerebral vasodilation, while the resumption of breathing and resultant hyperoxemia cause cerebral vasoconstriction. This repetitive vascular response exacerbates cerebral vascular stress, affects brain cell metabolism, disrupts autonomic nervous system balance, increases systemic inflammatory responses and oxidative stress, impairs endothelial function, and aggravates atherosclerosis, which is one of the main causes of ACI [30–32]. Mild SDB in ACI patients might not show obvious clinical symptoms, whereas severe SDB patients often experience frequent snoring, daytime sleepiness, inattention, and memory decline. In our study, MMSE scoring screening showed that the non-SDB group (normal) had the highest scores (26.40 ± 4.17), while the severe SDB group had the lowest scores (18.84 ± 4.64). Except for the lack of significant differences between the fluctuating and moderate SDB groups, other groups showed significant differences (P < 0.05). These results suggest that SDB, especially in severe cases, is associated with cognitive decline. Decreased blood oxygen saturation and reduced sleep quality negatively impact the brain, leading to weakened memory, attention, judgment, and thinking abilities. Long-term SDB can also exacerbate the severity and recurrence risk of cerebral infarction. This study's inclusion of ACI patients with a history of stroke (704 cases) significantly increases its clinical relevance, as a history of stroke is widely recognized as an important predictor of cognitive impairment risk. Previous clinical data indicate [33] that such patients are more likely to develop cognitive impairments in the future, and existing cognitive deficits may worsen with recurrent strokes. Additionally, this group also faces the risk of new cognitive issues, potentially due to cumulative neuronal damage caused by recurrent stroke events. Therefore, the results of this study provide important insights into understanding the developmental pathway of cognitive impairments in ACI patients and devising more effective prevention and intervention strategies for this specific population.

In this study, ① except for the normal group, the TST in all SDB severity groups from mild to fluctuating was shortened, with the TST in the moderate group significantly lower than that of the normal group (P < 0.05). ② The REML in the severe group was significantly longer than that in the mild group, with a corresponding increase in REM%. ③ There were no significant differences in SL, which may be due to daytime drowsiness in moderate-to-severe SDB patients, leading to a shortened sleep latency and rapid onset of sleep at night. Additionally, some SDB patients may experience difficulty falling asleep due to anxiety or concern about potential apneic events at night, leading to variability in sleep latency [34]. ④ Compared to the normal group, SE significantly decreased in the severe group; NREM stages 1–2 significantly increased in the moderate-to-severe groups; corresponding NREM stages 3–4 decreased; WASO time increased in the moderate and severe groups, and the NOA increased in the moderate group. ⑤ Although there were no significant differences in TOB among the groups, there was an increasing trend in the moderate-to-severe groups. ⑥ Compared to the non-SDB group, the SDB group had an increased proportion of brainstem infarctions, thalamic infarctions, and NIHSS grade—severe group (P < 0.05), indicating an increased risk of SDB with brainstem and thalamic infarctions and higher NIHSS severity upon admission. The pathogenesis involves prolonged expiration and disrupted respiratory centers in the brainstem following infarction, along with damage to the vestibular areas in the medulla and their chemoreceptors, affecting the regulation of acid-sensitive ion channels, vasoactive intestinal peptide, and serotonin, leading to abnormal respiratory rhythms, shallow breathing, or apnea, thus inducing SDB [35, 36]. Thalamic infarction disrupts the connection with the brainstem that maintains normal sleep–wake states and damages the autonomic respiratory regulatory signals, affecting the respiratory rhythm generator in the brainstem and stabilizing respiratory rhythms [37]. In summary, compared to the non-SDB group (normal), moderate-to-severe SDB patients showed significant differences in overall sleep state parameters (P < 0.05). Such patients may experience impaired memory and cognitive functions due to reduced total sleep time, difficulty falling asleep, prolonged light sleep, shortened deep sleep, frequent micro-awakenings, and overall decreased sleep quality, particularly affecting encoding of new information and recall of past events, decision-making, and lack of adaptability in executing strategies [38, 39]. The etiology involves pharyngeal muscle relaxation in moderate-to-severe SDB patients during NREM stages 3–4 and REM, leading to airway narrowing or obstruction, impeded pulmonary gas exchange, periodic decreases in blood oxygen saturation, affecting brain oxygenation levels related to the frontal lobe, amygdala, and hippocampus [40]. This decrease in blood oxygen saturation also interferes with the secretion of cortisol and melatonin, which regulate circadian rhythms, further promoting inflammation cell activation and the release of inflammatory mediators and reactive oxygen species, cumulatively increasing the risk of stroke. It is evident that ACI and SDB are closely related, not only in terms of changes in circadian rhythm and sleep structure but also significantly affecting cognitive functions.

The results of this study indicate that there is a correlation between AHI in ACI with SDB and various sleep parameters (P < 0.05), and this correlation is also evident in the sleep parameters of the moderate-to-severe SDB groups. Titova et al. found that the probability of occurrence in stroke patients with SDB is increased during the REM stage, which is attributed to decreased upper airway muscle activity during REM, resulting in increased critical airway closure pressure and causing airway collapse [41]. However, our results showed no statistically significant differences in REM between the mild, moderate, severe, and fluctuating SDB groups (P > 0.05), which is inconsistent with their findings. Tian et al., through rodent experiments, discovered that damage to the medullary reticular structure can relieve inhibition of spinal neurons, resulting in REM muscle atonia. This phenomenon is regulated by the “REM sleep switch” of the subcoeruleus nucleus in the pontine tegmentum on the ventromedial nucleus of the medulla [42]. Therefore, we infer that in the acute phase of stroke, even if brain damage involves the pontine and medullary infarction areas, significant changes in sleep structure during REM do not necessarily occur if the “core” areas are not damaged. This conclusion requires further exploration in future studies.

Limitations

This study is designed as a single-center trial, which may present regional bias in its data, limiting the extensive applicability of the findings. To enhance the representativeness and statistical power of the research, future studies should employ a multi-center collaborative approach, expand the sample size, and utilize analytical methods such as stratified random sampling and mixed effects models suitable for repeated measures data.

Conclusion

Patients with moderate-to-severe ACI accompanied by SDB exhibit significant changes in circadian rhythm, particularly associated with IS and IV. Symptoms include reduced daytime stability, increased diurnal variability, and fragmented sleep–wake circadian rhythm disorders. Additionally, sleep parameters are characterized by shortened TST, prolonged REML, decreased SE, increased NREM stages 1–2, decreased NREM stages 3–4, and increased WASO. There is a higher proportion of brainstem infarctions, thalamic infarctions, and NIHSS grade—severe group in the SDB groups; cognitive impairments are more likely to occur in the moderate-to-severe SDB groups.

Supplementary Information

Below is the link to the electronic supplementary material.

Author contributions

Lianhui Wang was responsible for conceptualization, design, and drafting of the manuscript. Jing Xue, Qian Ma, and Yongshan Fu were responsible for data collection, analysis, and interpretation. Xiaodong Yuan, Pingshu Zhang, and Ya Ou were responsible for manuscript revision.

Funding

This study was funded by the Hebei Province Medical Science Research Project Grant for the year 2023 (Project No: 20231854), and the Tangshan City Level Science and Technology Project for the year 2022 (Project No: 22130212G).

Declarations

Conflict of interest

The authors declare that they have no known competing financial interests or personal relationships that could have appeared to influence the work reported in this paper.

Ethical approval

This study fully adheres to the ethical principles of the Helsinki Declaration. It has been approved by the Medical Ethics Committee of Kailuan General Hospital (Approval No. 2023005), and informed consent has been obtained from all participants.