You are viewing the site in preview mode

Skip to main content
Download PDF

Obstructive sleep apnea and structural and functional brain alterations: a brain-wide investigation from clinical association to genetic causality

BMC Medicine volume 23, Article number: 42 (2025) Cite this article

Abstract

Background

Obstructive sleep apnea (OSA) is linked to brain alterations, but the specific regions affected and the causal associations between these changes remain unclear.

Methods

We studied 20 pairs of age-, sex-, BMI-, and education- matched OSA patients and healthy controls using multimodal magnetic resonance imaging (MRI) from August 2019 to February 2020. Additionally, large-scale Mendelian randomization analyses were performed using genome-wide association study (GWAS) data on OSA and 3935 brain imaging-derived phenotypes (IDPs), assessed in up to 33,224 individuals between December 2023 and March 2024, to explore potential genetic causality between OSA and alterations in whole brain structure and function.

Results

In the cohort study, OSA patients exhibited significantly lower fractional amplitude of low-frequency fluctuation and regional homogeneity in the right posterior cerebellar lobe and bilateral superior and middle frontal gyrus, while showing higher levels in the left occipital lobe and left posterior central gyrus. Decreased fractional anisotropy (FA) but increased apparent diffusion coefficient (ADC) was shown in the bilateral superior longitudinal fasciculus. According to the results of Affiliation file 2: table s6, it is the ADC value of right superior longitudinal fasciculus was shown a positive correlation with the lowest oxygen saturation. In the Mendelian randomization analyses, the area of left inferior temporal sulcus (OR: 0.89; 95% CI: 0.82–0.96), rfMRI connectivity ICA100 edge 893 (OR: 0.88; 95% CI: 0.82–0.96), ICA100 edge 951 (OR: 0.89; 95% CI: 0.82–0.97), and ICA100 edge 1213 (OR: 0.89; 95% CI: 0.82–0.96) were significantly decreased in OSA. Conversely, mean thickness of G-front-inf-Triangul in right hemisphere (OR: 1.14; 95% CI: 1.05–1.23), mean orientation dispersion index in right tapetum (OR: 1.13; 95% CI: 1.04–1.23), and rfMRI connectivity ICA100 edge 258 (OR: 1.13; 95% CI: 1.04–1.22) showed opposite results.

Conclusions

Nerve fiber damage and imbalances in neuronal activity across multiple brain regions caused by hypoxia, particularly the frontal lobe, underlie the structural and the functional connectivity impairments in OSA.

Peer Review reports

Background

Obstructive sleep apnea (OSA) is a common chronic sleep-related breathing disorder characterized by recurrent upper airway obstruction during sleep [1]. This disorder is associated with a range of neurocognitive deficits, including memory loss, cognitive dysfunction, and decline in executive functions [2, 3]. These impairments are thought to arise from hemodynamic changes during apneic episodes, repeated nocturnal hypoxemia and hypercapnia, and frequent awakenings [1]. Recent research utilizing various magnetic resonance imaging (MRI) techniques has aimed to explore the brain structural and functional changes associated with OSA. These studies have revealed disruptions in white matter (WM) integrity [4] and morphological differences in gray matter (GM) regions [5], suggesting that microstructural brain damage may disrupt neural activity within the central autonomic function network. Additionally, functional MRI studies have shown alterations in brain connectivity, which may contribute to the neurocognitive impairments seen in OSA [6]. However, there is limited comprehensive research combining both structural and functional MRI findings, and discrepancies in results have led to debate regarding the link between OSA and brain changes.

Advanced neuroimaging techniques offer valuable and non-invasive tools to investigate cerebral structural and functional integrity. Voxel-based morphometry (VBM) is commonly used to quantify GM and WM volume changes using high-resolution 3D T1-weighted Mendelian randomization images [7]. Resting-state functional MRI (rfMRI) measures the blood oxygenation level-dependent (BOLD) signal, reflecting intrinsic brain activity [8]. rfMRI parameters such as functional connectivity (FC), fractional amplitude of low-frequency fluctuation (fALFF), and regional homogeneity (ReHo) provide a comprehensive view of brain activity [9]. Furthermore, diffusion tensor imaging (DTI) assesses WM microstructural abnormalities through metrics like fractional anisotropy (FA), mean diffusivity (MD), and apparent diffusion coefficient (ADC), offering insights into WM fiber integrity and myelination [10]. These techniques collectively enhance our understanding of brain alterations in various disorders, including OSA.

OSA is a multifactorial disorder with complex development and progression. Clinical observational studies often face confounding factors, making it challenging to establish a clear causal relationship between OSA and brain changes. Mendelian randomization provides a robust method to investigate causality using genetic variants as instrumental variables [11]. Mendelian randomization allows for more accurate estimations of causal relationships [12]. A recent large genome-wide association study (GWAS) dataset containing 3935 brain imaging-derived phenotypes (IDPs) was used to explore the causal links between brain characteristics and specific diseases [13, 14].

In this study, we conducted a two-stage analysis. First, we examined brain changes in OSA patients using multiple MRI techniques and assessed the relationships between MRI parameters and polysomnography (PSG) results. Second, we performed a two-sample Mendelian randomization analysis utilizing GWAS data on 3935 IDPs [15], which represents the largest correlation dataset thus far, and OSA. Our goal was to comprehensively assess the associations between structural and functional brain alterations in OSA, from clinical correlations to genetic causality, to better understand the underlying pathophysiological mechanisms and genetic links and their potential impact on neurological disease development.

Methods

Study design

This two-stage study design is summarized in Fig. 1. In stage 1, an observational study approach was utilized to explore the potential relationships between OSA and brain changes. In stage 2, Mendelian randomization analyses were conducted to assess the causal influence of OSA on brain changes.

Fig. 1
figure 1

Flowchart of the analyses conducted in the present study. Stage 1, the observational study. Stage 2, the Mendelian randomization study design. There are three principal assumptions in Mendelian randomization design: (A) instrumental variables (IVs) must be strongly associated with exposure, (B) IVs are not related to confounding factors, and (C) IVs affect outcome risk only through the risk factor rather than any other ways; OSA, obstructive sleep apnea; HCs, healthy controls; T2- FLAIR, T2-fluid attenuated inversion recovery; 3D-T1WI, three-dimensional T1-weighted images; rfMRI, resting-state fMRI; DTI, diffusion tensor imaging; BMI, body mass index; IVs, instrumental variables; SNP, single-nucleotide polymorphisms; LD, linkage disequilibrium

Full size image

The participants of observational study

The potentially eligible participants (the age range from 18 to 60 years old) were recruited from the Sleep Medicine Center of the First Affiliated Hospital of Guangzhou Medical University between August 2019 and February 2020. We defined patients with OSA as those with apnea–hypopnea index (AHI) ≥ 5 events/h [16]. The detail of the inclusion and exclusion criteria is shown in Additional file 1, especially excluding the patients with other sleep disorders, such as narcolepsy, isolated REM sleep behavior disorder (IRBD), and periodic limb movements (PLM), which has already shown some functional or structural changes [9, 17,18,19]. Prior to performing MRI, healthy controls were matched with OSA patients based on sex, age, BMI, and education, which had been demonstrated to influence the brain structure and function [20,21,22]. All participants voluntarily participated in the study and provided informed consent. This study was approved by the Medical Ethics Committee of the First Affiliated Hospital of Guangzhou Medical University (Ethics number: 201705).

Overnight polysomnography (PSG)

PSG was performed using standard scoring techniques as our previous study [23]. Prior to PSG, all participants underwent a complete medical history interview and standard physical examination and completed the Epworth Sleepiness Scale (ESS). Detailed operational procedures of the PSG monitor are provided in Additional file 1 [16, 23, 24].

MRI image acquisition

Within 1 week of the PSG examination, all participants underwent 3.0 T brain MRI using an 8-channel parallel head coil (Philips Achieva, the Netherlands). Additional file 1 provides specific details on MRI image acquisition and the methodologies [25,26,27,28,29]. The regions with significant differences in structure and function basing on the VBM analysis and fALFF and ReHo results were selected as regions of interest (ROIs) to explore the FC change. In DTI analysis, the bilateral superior longitudinal fasciculus, the inferior longitudinal fasciculus, the uncinate fasciculus, pyramidal tract, and the cingulum bundles were chosen as ROIs which were considered particularly vulnerable to hypoxia and sleep fragmentation.

Observational statistical analysis

Demographic data analyses were conducted using the Statistical Package for Social Sciences (SPSS) software, version 16.0. Continuous variables were compared using either a two-sample t-test or Mann–Whitney test, while categorical variables were analyzed using a chi-squared test between the patient group and healthy controls. Pearson or Spearman correlation analyses were used to assess correlations between MRI and PSG parameters in the OSA group, with significance level set at p < 0.05.

For VBM and rfMRI analyses, two-sample t-tests were employed to compare voxel-based maps between groups. Multiple comparison corrections were applied using Gaussian random field (GRF) correction, with a voxel-level threshold of p < 0.01 and a cluster-level threshold of p < 0.05. DTI parameters (FA and ADC) within ROIs were compared using independent-samples t-tests, with statistical significance level set at Bonferroni corrected p < 0.05, accounting for multiple ROI comparisons.

The two-sample Mendelian randomization analysis

Our study adheres to the Strengthening the Reporting of Observational Studies in Epidemiology using Mendelian Randomization (STROBE-MR) checklist [30, 31]. The Mendelian randomization analysis employs single-nucleotide polymorphisms (SNPs) as IVs to estimate causal associations between exposures and outcomes. This method relies on three key assumptions. First, IVs must be strongly associated with exposure. Second, IVs are not related to confounding factors. Third, IVs affect outcome risk only through the risk factor rather than any other ways.

Data sources and IVs selected

All datasets used in this study are publicly available, with comprehensive descriptions provided in Additional file 2: Table S1. The GWAS summary-level data for OSA were acquired from the FinnGen biobank. OSA diagnoses were based on the International Classification of Diseases codes (ICD-10: G47.3, ICD-9: 3472A). The FinnGen study is a large-scale genomics initiative to understand disease mechanisms and predispositions [32]. We utilized statistics on 3935 imaging-derived phenotypes (IDPs) processed from a sample of 33,224 individuals of European ancestry from the UK Biobank. These IDPs encompass six MRI modalities related to brain structure, function, and connectivity [15]. Detailed information on imaging processing, genetic pre-processing, and quality control information is shown in https://biobank.ctsu.ox.ac.uk/crystal/crystal/docs/brain_mri.pdf [15, 33]. The complete description of each brain IDP is shown in Additional file 2: Table S2.

SNPs associated with OSA were selected based on genome-wide significance level (p < 5 × 10−8), clumped within 10,000 kb and linkage disequilibrium (LD) (r2 < 0.001) to ensure the independence. R2 and F-statistic were used for each SNP to evaluate strength of the instruments using the formula R2 = β2/(SE2 × N + β2), F = R2/(1 − R2) * (N − 2), where N represents the sample size. Weak instrumental variables (F-statistics < 10) were be removed to avoid bias in Mendelian randomization analysis [34]. We further employed the LDtrait Tool (https://ldlink.nih.gov/?tab=ldtrait) to remove the IVs related to outcome traits [35].

Main statistical analyses of Mendelian randomization

The inverse variance weighting (IVW) approach was selected as the primary method for causal inference [36]. Then, we used the Cochrane’s Q value to assess the heterogeneity [37]. If significant heterogeneity is observed, the random-effects model is chosen [38]. MR–Egger regression and Mendelian randomization pleiotropy residual sum and outlier (MR-PRESSO) were used to assess horizontal pleiotropy [39]. If significant SNP outliers (p < 0.05) were detected by MR-PRESSO, they were removed [40]. All analyses are conducted using the TwoSampleMR and MR-PRESSO packages in R Software 4.3.1, with forest plot generated using the ggplot2 package.

Results

Demographic and clinical characteristic of observational study

We initially recruited 38 OSA patients and 23 healthy controls who provided informed consent and underwent overnight PSG. After screening, 31 OSA patients and 22 healthy controls met the inclusion and exclusion criteria. Following matching and MRI assessment, the final study population comprised 20 OSA patients and 20 matched healthy controls (Fig. 1A). The demographic and clinical characteristics of the participants are summarized in Table 1. The OSA patients exhibited higher AHI, percentage sleep time with < 90% saturation (T90%), arousal index (AI), and ESS compared to healthy controls. The lowest oxygen saturation (LSaO2) in the OSA patients was lower than that in the healthy controls.

Table 1 Clinical characteristics for OSA patients and healthy controls
Full size table

Multimodal neuroimaging study

Voxel-based morphometry (VBM) analysis

VBM analysis showed no significant differences in GM volume between OSA patients and healthy controls (with GRF correction, cluster-level p > 0.05, voxel-level p > 0.01).

Fractional amplitude of low frequency fluctuation (fALFF)

In comparison to healthy controls, OSA patients showed significantly lower fALFF in the right posterior cerebellar lobe, the bilateral superior and middle frontal gyrus, and the left anterior, middle, and posterior cingulate gyrus. However, higher fALFF was shown in the left occipital lobe. Detailed results are presented in Additional file 2: Table S3 and Fig. 2A.

Fig. 2
figure 2

Difference of fALFF and ReHo values in the OSA patients. A Compared with healthy controls (HCs), OSA patients decreased fALFF in the right posterior cerebellar lobe, the bilateral superior and middle frontal gyrus, and the left anterior, middle, and posterior cingulate gyrus but increased fALFF in the left occipital lobe. B Compared with HCs, OSA patients decreased ReHo in the right posterior cerebellar lobe, left superior and middle frontal gyrus, and right superior and middle frontal gyrus but increased ReHo in the left posterior central gyrus. Red colors show brain regions with increased values; blue colors indicate brain regions with decreased values

Full size image

Regional homogeneity (ReHo)

Compared with healthy controls, OSA patients demonstrated significant lower ReHo in the right posterior cerebellar lobe and bilateral superior and middle frontal gyrus but higher ReHo in the left posterior central gyrus. Detailed findings are shown in Additional file 2: Table S4 and Fig. 2B.

Functional connectivity (FC)

Based on the brain regions identified with different fALFF and ReHo values between OSA patients and health controls, the bilateral superior frontal gyrus, the bilateral middle frontal gyrus, and the right posterior cerebellar cortex were selected as ROIs. FC analysis did not reveal any significant differences between OSA patients and healthy controls in these ROIs (with GRF correction, cluster-level p > 0.05, voxel-level p > 0.01).

Diffusion tensor imaging (DTI) analysis

In ROIs analyses, OSA patients showed a significant decrease in the FA value in both the left and right superior longitudinal fasciculus (both p < 0.001). Conversely, they demonstrated a significant increase in the ACD value in the same areas (both p < 0.001). No significant differences were observed in the DTI parameters of other ROIs between the two groups (Additional file 2: Table S5).

Relationship between multimodal neuroimaging and PSG parameters in OSA patients

Pearson correlation analysis revealed a significant positive correlation between the ACD value in the right superior longitudinal fasciculus and LSaO2 (r = 0.449, p = 0.047) in OSA patients (Fig. 3). However, no significant correlation was observed between the PSG parameters and other distinct fALFF, ReHo, FA, and ACD values in the brain regions of OSA patients and healthy controls (Additional file 2: Table S6).

Fig. 3
figure 3

Correlation analysis between significantly different diffusion tensor imaging (DTI) and LSaO2 in OSA patients. SLF, superior longitudinal fasciculus; LSaO2, lowest oxygen saturation

Full size image

Mendelian randomization analyses

Following LD pruning and removing outliers, 23 SNPs significantly related to OSA were selected as exposures. The F statistics for these SNPs ranged from 30 to 125. Detailed information for each selected SNP is provided in Additional file 2: Table S7. According to the IVW method (Fig. 4 and Additional file 2: Table S8), genetically predicted OSA was found to be causally associated with 7 IDPs, including 4 rfMRI connectivity. Specifically, the area of the left inferior temporal sulcus (OR: 0.89; 95% CI: 0.82–0.96; p = 0.004), rfMRI connectivity ICA100 edge 893 (OR: 0.88; 95% CI: 0.82–0.96; p = 0.003), rfMRI connectivity ICA100 edge 951 (OR: 0.89; 95% CI: 0.82–0.97; p = 0.005), and rfMRI connectivity ICA100 edge 1213 (OR: 0.89; 95% CI: 0.82–0.96; p = 0.004) were significantly decreased in OSA patients; the picture of the detailed area about each edge linked is shown in Additional file 3: Fig. S1-S3. Conversely, the mean thickness of G-front-inf-Triangul in the right hemisphere (OR: 1.14; 95% CI: 1.05–1.23; p = 0.002), mean orientation dispersion index (OD) in the right tapetum (OR: 1.13; 95% CI: 1.04–1.23; p = 0.004), and rfMRI connectivity ICA100 edge 258 (OR: 1.13; 95% CI: 1.04–1.22; p = 0.004) (Additional file 3: Fig. S4) showed opposite results. Results from Cochran’s Q statistical test and MR-Egger intercept analysis (Additional file 2: Table S9) did not indicate evidence of heterogeneity and pleiotropy, except for rfMRI connectivity ICA100 edge 893. In summary, these sensitivity analyses validate the reliability of inferring causal effects.

Fig. 4
figure 4

The forest plot of the Mendelian randomization analyses results about causal associations of OSA on 3935 brain imaging derived phenotypes (IDPs). OR, odds ratio; CI, confidence interval; Significant threshold was set at p-value < 0.05 for the inverse variance weighted method (IVW)

Full size image

Discussion

Our study combined rigorous, paired, and representative observational study and Mendelian randomization analyses to examine the potential link between OSA and brain structural and functional integrity. In a cohort of 20 pairs of sex-, age-, BMI-, and education-matched OSA cases and controls, we observed no significant GM volume changes among OSA patients via VBM analysis. However, rfMRI results indicated simultaneous decreases in neuronal activity in the right posterior cerebellar lobe and bilateral superior and middle frontal gyrus, along with reduced local consistency. Meanwhile, ReHo values increased in the left posterior central gyrus, and fALFF values increased in the left occipital lobe. DTI analysis further revealed decreases in FA but increases in ADC within the bilateral superior longitudinal fasciculus among OSA patients, complemented by a positive correlation between ADC values in the right superior longitudinal fasciculus and LSaO2. Moreover, our Mendelian randomization analyses revealed that OSA was associated with a reduction in the area of the left inferior temporal sulcus and a decline in most brain connectivity. Conversely, it also led to increased mean thickness of G-front-inf-Triangul (indicated triangular part of the inferior frontal gyrus) in right hemisphere and mean OD in right tapetum.

To the best of our knowledge, this study is the first to comprehensively assess brain structural and functional integrity in OSA patients, integrating clinical and genetic research through a brain-wide neuroimaging investigation combined with Mendelian randomization analysis. Overall, our findings suggest that OSA primarily induces brain functional changes, with significant structural alterations observed only in the pars triangularis region of the right inferior frontal gyrus and the left inferior temporal sulcus.

The conclusions regarding brain morphometric changes in OSA patients are varied. Previous VBM studies have reported inconsistent findings on GM volume abnormalities in OSA. Some studies have noted reductions in GM volume in regions such as the frontal and parietal cortex [41], hippocampus [42], thalami [5], and cerebellum [43]. However, our observational study cannot find any significant GM volume changes among OSA patients. These discrepancies may stem from methodological differences, varying ages of onset, disease severity, and duration of OSA. Therefore, rigorous Mendelian randomization analysis as a dependable research method was performed to mitigate confounding factors and investigate the causal association between exposure and outcomes at genetic level. Our Mendelian randomization analyses revealed that OSA is associated with an increased mean thickness of triangular part of the inferior frontal gyrus in right hemisphere, consistent with findings by Baril et al., who correlated thicker pars triangularis in right inferior frontal gyrus with increased sleep fragmentation [44]. Additionally, our findings of decreased area in the left inferior temporal sulcus align with previous observations [44,45,46]. Increased GM may signify cerebral edema and/or elevated β-amyloid deposition due to intermittent hypoxia and sleep fragmentation during sleep, whereas decreased GM may indicate neuronal damage [44]. These distinct pathophysiological processes can predict varying disease severities and treatment responses. Despite differing from Mendelian randomization analyses, our observational study predominantly included middle-aged patients, potentially limiting observations on brain morphometric change. Our results indirectly suggest that the age of onset, duration, and patient demographics play crucial roles in OSA’s impact on brain structure. Future research should carefully consider these factors.

Regarding functional changes, we employed ReHo and fALFF to comprehensively assess brain activity regulation. Lower ReHo values indicate reduced synchronous neuronal activity [47]. Our observational study results demonstrated parallel declines in ReHo and fALFF values in right posterior cerebellar lobe and bilateral superior and middle frontal gyrus, suggesting decreased neuronal activity and weakened local connections in these areas. However, FC within these ROIs did not significantly differ. In contrast, Mendelian randomization analyses indicated that rfMRI connectivity analysis demonstrated decreased connectivity across most connections, particularly affecting the frontal and parietal lobes. Consistent with previous studies, our findings have observed changes in spontaneous neural activity across multiple brain regions in OSA, such as right superior frontal gyrus [48] and left middle and medial superior frontal gyrus [49, 50]. Specifically, decreased FCs, such as between the medial prefrontal cortex and left medial temporal lobe [51], the right dorsolateral prefrontal cortex and left precentral gyrus [52], and the medial prefrontal cortex and bilateral hippocampi [53], had been demonstrated involved with the prefrontal cortex. Although specific areas of interest may differ, including our study, evidence suggests that the frontal gyrus is particularly susceptible to hypoxia, impacting higher cognitive functions, especially working memory [54]. The cerebellum, crucial for motor control and interconnected with various brain regions, also shows signs of impairment. Its involvement in respiratory regulation [55], sensation, learning, memory, emotion, executive ability [56], and social function [57] suggests that chronic nocturnal hypoxia in OSA may predominantly affect memory, attention, cognitive functions, and sensorimotor abilities.

In our study, we utilized DTI to indirectly investigate the microstructural integrity of the WM pathways, focusing on the bilateral superior longitudinal fasciculus, inferior longitudinal fasciculus, uncinate fasciculus, pyramidal tract, and cingulum bundles as ROIs. These regions are known to be susceptible to hypoxia and sleep fragmentation and are involved in motor regulation, cognitive function, and emotional processing [58,59,60]. Our study indicated damage to the bilateral superior longitudinal fasciculus in OSA, consistent with findings from previous researches [60, 61]. The superior longitudinal fasciculus is a major association fiber bundle connecting the frontal, parietal, temporal lobes, and occipital lobes, which are crucial for functions such as memory integration, emotional regulation, and language abilities, and their abnormalities may lead to cognitive impairment [62, 63]. Previous reports have shown positive and negative correlations between several WM and sub-cortical sites and AHI and SaO2 change [64]. Our findings also provide evidence of a positive correlation between ADC values and LSaO2, suggesting that intermittent hypoxia may play a primary role in WM microstructural impairments in OSA. Combining our observational studies with Mendelian randomization analyses, we hypothesize that intermittent hypoxia may be the primary factor contributing to WM microstructural impairments in OSA. As the disease progresses, ongoing damage may lead to more extensive abnormalities in brain functional connectivity, potentially resulting in cognitive decline and structural changes in the brain.

However, our study also has several limitations. First, the overall sample size was relatively small in our observational study. Second, the observational study primarily included cases of moderate and severe OSA, and we did not perform stratified analysis by disease severity, which could introduce potential statistical bias. Additionally, we selected a limited number of ROIs based on previous studies to focus on meaningful and feasible scopes rather than conducting whole-brain analyses, which might overlook some regions with potential influences. Third, we did not explore the clinical symptoms of cognitive impairments caused by OSA, which requires a large-scale cohort study for further investigation in the future. Fourth is the dataset availability. The population enrolled for GWAS were mainly Europeans, and ethnicity difference may weaken the reliability of Mendelian randomization analyses results. Thus, wider population research is needed. Finally, despite rigorous Mendelian randomization analyses performed, there are potential confounders that could affect Mendelian randomization analyses results.

Our study employed the most comprehensive MRI techniques to investigate brain structural and functional changes in patients with OSA. Additionally, utilizing the largest IDP database to explore the causal relationship between OSA and brain structural changes further enhanced the reliability of our observational findings. Although results from the two stages of the study showed partial differences, numerous factors influence the onset and progression of OSA and brain function changes. These confounding factors are challenging to control in observational studies, potentially leading to unreliable results. Our rigorous Mendelian randomization studies address these limitations and offer valuable directions for future research.

Conclusions

Our study suggests that nerve fiber damage and imbalances in neuronal activity across multiple brain regions which associated with hypoxia served as underlying neurophysiological mechanisms contributing to the structural and functional impairments in OSA. Furthermore, the most interesting result is that frontal lobe may be the significant damage area, and further studies should pay more attention to the changes of this region. The functional impairment precedes structural impairment; more attention should be paid to changes in brain function in OSA patients.

Data availability

No datasets were generated or analysed during the current study.

Abbreviations

ADC:

Apparent diffusion coefficient

AHI:

Apnea-hypopnea index

AI:

Arousal index

ALFF:

Low-frequency fluctuation

BOLD:

Blood oxygenation level-dependent

DTI:

Diffusion tensor imaging

ESS:

Epworth Sleepiness Scale

FA:

Fractional anisotropy

fALFF:

Fractional ALFF

FC:

Functional connectivity

GM:

Gray matter

GRF:

Gaussian random field

GWAS:

Genome-wide association study

LD:

Linkage disequilibrium

LSaO2 :

Lowest oxygen saturation

IDPs:

Imaging-derived phenotypes

IVs:

Instrumental variables

MD:

Mean diffusivity

MRI:

Magnetic resonance imaging

OD:

Orientation dispersion index

OSA:

Obstructive sleep apnea

PSG:

Polysomnography

ReHo:

Regional homogeneity

rfMRI:

Resting-state functional MRI

SLF:

Superior longitudinal fasciculus

SNPs:

Single-nucleotide polymorphisms

T90:

Percentage sleep time with < 90% saturation

VBM:

Voxel-based morphometry

WM:

White matter

References

  1. McNicholas WT, Pevernagie D. Obstructive sleep apnea: transition from pathophysiology to an integrative disease model. J Sleep Res. 2022;31(4): e13616.

    Article  PubMed  PubMed Central  Google Scholar 

  2. Bubu OM, Andrade AG, Umasabor-Bubu OQ, et al. Obstructive sleep apnea, cognition and Alzheimer’s disease: a systematic review integrating three decades of multidisciplinary research. Sleep Med Rev. 2020;50: 101250.

    Article  PubMed  Google Scholar 

  3. Beaudin AE, Raneri JK, Ayas NT, et al. Cognitive function in a sleep clinic cohort of patients with obstructive sleep apnea. Ann Am Thorac Soc. 2021;18(5):865–75.

    Article  PubMed  Google Scholar 

  4. Lee MH, Lee SK, Kim S, et al. Association of obstructive sleep apnea with white matter integrity and cognitive performance over a 4-year period in middle to late adulthood. JAMA Netw Open. 2022;5(7): e2222999.

    Article  PubMed  PubMed Central  Google Scholar 

  5. Joo EY, Tae WS, Lee MJ, et al. Reduced brain gray matter concentration in patients with obstructive sleep apnea syndrome. Sleep. 2010;33(2):235–41.

    Article  PubMed  PubMed Central  Google Scholar 

  6. Siddiqi SH, Kandala S, Hacker CD, et al. Precision functional MRI mapping reveals distinct connectivity patterns for depression associated with traumatic brain injury. Sci Transl Med. 2023;15:eabn0441.https://doiorg.publicaciones.saludcastillayleon.es/10.1126/scitranslmed.abn0441

  7. Ashburner J, Friston KJ. Voxel-based morphometry–the methods. Neuroimage. 2000;11(6 Pt 1):805–21.

    Article  CAS  PubMed  Google Scholar 

  8. Barkhof F, Haller S, Rombouts SA. Resting-state functional MR imaging: a new window to the brain. Radiology. 2014;272(1):29–49.

    Article  PubMed  Google Scholar 

  9. Xu L, Xue R, Ai Z, et al. Resting-state functional magnetic resonance imaging as an indicator of neuropsychological changes in type 1 narcolepsy. Acad Radiol. 2024;31(1):69–81.

    Article  PubMed  Google Scholar 

  10. Tae WS, Ham BJ, Pyun SB, Kang SH, Kim BJ. Current clinical applications of diffusion-tensor imaging in neurological disorders. J Clin Neurol. 2018;14(2):129–40.

    Article  PubMed  PubMed Central  Google Scholar 

  11. Emdin CA, Khera AV, Kathiresan S. Mendelian randomization. JAMA. 2017;318(19):1925–6.

    Article  PubMed  Google Scholar 

  12. Burgess S, Butterworth A, Malarstig A, Thompson SG. Use of Mendelian randomisation to assess potential benefit of clinical intervention. BMJ. 2012;345: e7325.

    Article  PubMed  Google Scholar 

  13. He J, Cabrera-Mendoza B, De Angelis F, et al. Sex differences in the pleiotropy of hearing difficulty with imaging-derived phenotypes: a brain-wide investigation. Brain. 2024;147(10):3395-408.

  14. Zanoaga MD, Friligkou E, He J, et al. Brainwide Mendelian randomization study of anxiety disorders and symptoms. Biol Psychiatry. 2024;95(8):810–7.

    Article  PubMed  Google Scholar 

  15. Smith SM, Douaud G, Chen W, et al. An expanded set of genome-wide association studies of brain imaging phenotypes in UK Biobank. Nat Neurosci. 2021;24(5):737–45.

    Article  CAS  PubMed  PubMed Central  Google Scholar 

  16. Sunderram J, Weintraub M, Black K, et al. Chronic rhinosinusitis is an independent risk factor for OSA in World Trade Center responders. Chest. 2019;155(2):375–83.

    Article  PubMed  Google Scholar 

  17. Bourgouin PA, Rahayel S, Gaubert M, et al. Neuroimaging of rapid eye movement sleep behavior disorder. Int Rev Neurobiol. 2019;144:185–210.

    Article  PubMed  Google Scholar 

  18. Pyatigorskaya N, Gaurav R, Arnaldi D, et al. Magnetic Resonance Imaging Biomarkers to Assess Substantia Nigra Damage in Idiopathic Rapid Eye Movement Sleep Behavior Disorder. Sleep. 2017;40(11). https://doiorg.publicaciones.saludcastillayleon.es/10.1093/sleep/zsx149.

  19. Zhang H, Xu L, Ai Z, et al. The brain topological alterations in the structural connectome and correlations with clinical characteristics in type 1 narcolepsy. Neuroimage Clin. 2024;44: 103697.

    Article  CAS  PubMed  PubMed Central  Google Scholar 

  20. Binnewies J, Nawijn L, van Tol MJ, et al. Associations between depression, lifestyle and brain structure: a longitudinal MRI study. Neuroimage. 2021;231: 117834.

    Article  PubMed  Google Scholar 

  21. Han YP, Tang X, Han M, et al. Relationship between obesity and structural brain abnormality: accumulated evidence from observational studies. Ageing Res Rev. 2021;71: 101445.

    Article  PubMed  Google Scholar 

  22. Seyedsalehi A, Warrier V, Bethlehem R, et al. Educational attainment, structural brain reserve and Alzheimer’s disease: a Mendelian randomization analysis. Brain. 2023;146(5):2059–74.

    Article  PubMed  Google Scholar 

  23. Wu K, Su X, Li G, Zhang N. Antioxidant carbocysteine treatment in obstructive sleep apnea syndrome: a randomized clinical trial. PLoS ONE. 2016;11(2): e0148519.

    Article  PubMed  PubMed Central  Google Scholar 

  24. Ruehland WR, Rochford PD, O’Donoghue FJ, et al. The new AASM criteria for scoring hypopneas: impact on the apnea hypopnea index. Sleep. 2009;32(2):150–7.

    Article  PubMed  PubMed Central  Google Scholar 

  25. Yan CG, Wang XD, Zuo XN, et al. DPABI: data processing & analysis for (resting-state) brain imaging. Neuroinformatics. 2016;14(3):339–51.

    Article  PubMed  Google Scholar 

  26. Shu Y, Liu X, Yu P, et al. Inherent regional brain activity changes in male obstructive sleep apnea with mild cognitive impairment: a resting-state magnetic resonance study. Front Aging Neurosci. 2022;14:1022628.

    Article  PubMed  PubMed Central  Google Scholar 

  27. Yu H, Chen L, Li H, et al. Abnormal resting-state functional connectivity of amygdala subregions in patients with obstructive sleep apnea. Neuropsychiatr Dis Treat. 2019;15:977–87.

    Article  PubMed  PubMed Central  Google Scholar 

  28. Zou W, Song P, Lu W, et al. Global hippocampus functional connectivity as a predictive neural marker for conversion to future mood disorder in unaffected offspring of bipolar disorder parents. Asian J Psychiatr. 2022;78: 103307.

    Article  PubMed  Google Scholar 

  29. Ashburner J. A fast diffeomorphic image registration algorithm. Neuroimage. 2007;38(1):95–113.

    Article  PubMed  Google Scholar 

  30. Skrivankova VW, Richmond RC, Woolf B, et al. Strengthening the reporting of observational studies in epidemiology using Mendelian randomization: the STROBE-MR statement. JAMA. 2021;326(16):1614–21.

    Article  PubMed  Google Scholar 

  31. Skrivankova VW, Richmond RC, Woolf B, et al. Strengthening the reporting of observational studies in epidemiology using Mendelian randomisation (STROBE-MR): explanation and elaboration. BMJ. 2021;375: n2233.

    Article  PubMed  PubMed Central  Google Scholar 

  32. Kurki MI, Karjalainen J, Palta P, et al. FinnGen provides genetic insights from a well-phenotyped isolated population. Nature. 2023;613(7944):508–18.

    Article  CAS  PubMed  PubMed Central  Google Scholar 

  33. Elliott LT, Sharp K, Alfaro-Almagro F, et al. Genome-wide association studies of brain imaging phenotypes in UK Biobank. Nature. 2018;562(7726):210–6.

    Article  CAS  PubMed  PubMed Central  Google Scholar 

  34. Burgess S, Thompson SG. Avoiding bias from weak instruments in Mendelian randomization studies. Int J Epidemiol. 2011;40(3):755–64.

    Article  PubMed  Google Scholar 

  35. Lin SH, Brown DW, Machiela MJ. LDtrait: an online tool for identifying published phenotype associations in linkage disequilibrium. Cancer Res. 2020;80(16):3443–6.

    Article  CAS  PubMed  PubMed Central  Google Scholar 

  36. Burgess S, Butterworth A, Thompson SG. Mendelian randomization analysis with multiple genetic variants using summarized data. Genet Epidemiol. 2013;37(7):658–65.>

    Article  PubMed  PubMed Central  Google Scholar 

  37. Greco MFD, Minelli C, Sheehan NA, Thompson JR. Detecting pleiotropy in Mendelian randomisation studies with summary data and a continuous outcome. Stat Med. 2015;34(21):2926–40.

    Article  Google Scholar 

  38. Bowden J, Del Greco MF, Minelli C, Davey Smith G, Sheehan N, Thompson J. A framework for the investigation of pleiotropy in two-sample summary data Mendelian randomization. Stat Med. 2017;36(11):1783–802.

    Article  PubMed  PubMed Central  Google Scholar 

  39. Burgess S, Thompson SG. Interpreting findings from Mendelian randomization using the MR-Egger method. Eur J Epidemiol. 2017;32(5):377–89.

    Article  PubMed  PubMed Central  Google Scholar 

  40. Verbanck M, Chen CY, Neale B, Do R. Detection of widespread horizontal pleiotropy in causal relationships inferred from Mendelian randomization between complex traits and diseases. Nat Genet. 2018;50(5):693–8.

    Article  CAS  PubMed Central  Google Scholar 

  41. Philby MF, Macey PM, Ma RA, Kumar R, Gozal D, Kheirandish-Gozal L. Reduced regional grey matter volumes in pediatric obstructive sleep apnea. Sci Rep. 2017;7:44566.

    Article  CAS  PubMed  PubMed Central  Google Scholar 

  42. Torelli F, Moscufo N, Garreffa G, et al. Cognitive profile and brain morphological changes in obstructive sleep apnea. Neuroimage. 2011;54(2):787–93.

    Article  PubMed  Google Scholar 

  43. Xiao P, Hua K, Chen F, et al. Abnormal cerebral blood flow and volumetric brain morphometry in patients with obstructive sleep apnea. Front Neurosci. 2022;16: 934166.

    Article  PubMed  PubMed Central  Google Scholar 

  44. Baril AA, Gagnon K, Brayet P, et al. Gray matter hypertrophy and thickening with obstructive sleep apnea in middle-aged and older adults. Am J Respir Crit Care Med. 2017;195(11):1509–18.

    Article  CAS  PubMed  Google Scholar 

  45. Yu C, Fu Y, Lu Y, et al. Alterations of brain gray matter volume in children with obstructive sleep apnea. Front Neurol. 2023;14:1107086.

    Article  PubMed  PubMed Central  Google Scholar 

  46. Gao J, Cao J, Chen J, et al. Brain morphology and functional connectivity alterations in patients with severe obstructive sleep apnea. Sleep Med. 2023;111:62–9.

    Article  PubMed  Google Scholar 

  47. Sun Y, Lei F, Luo L, Zou K, Tang X. Effects of a single night of continuous positive airway pressure on spontaneous brain activity in severe obstructive sleep apnea. Sci Rep. 2023;13(1):8950.

    Article  CAS  PubMed  PubMed Central  Google Scholar 

  48. Peng DC, Dai XJ, Gong HH, Li HJ, Nie X, Zhang W. Altered intrinsic regional brain activity in male patients with severe obstructive sleep apnea: a resting-state functional magnetic resonance imaging study. Neuropsychiatr Dis Treat. 2014;10:1819–26.

    PubMed  PubMed Central  Google Scholar 

  49. Bai J, Wen H, Tai J, et al. Altered spontaneous brain activity related to neurologic and sleep dysfunction in children with obstructive sleep apnea syndrome. Front Neurosci. 2021;15: 595412.

    Article  PubMed  PubMed Central  Google Scholar 

  50. Ji T, Li X, Chen J, et al. Brain function in children with obstructive sleep apnea: a resting-state fMRI study. Sleep. 2021;44(8):zsab047.

  51. Li HJ, Nie X, Gong HH, Zhang W, Nie S, Peng DC. Abnormal resting-state functional connectivity within the default mode network subregions in male patients with obstructive sleep apnea. Neuropsychiatr Dis Treat. 2016;12:203–12.

    Article  PubMed  PubMed Central  Google Scholar 

  52. Zhang Q, Wang D, Qin W, et al. Altered resting-state brain activity in obstructive sleep apnea. Sleep. 2013;36(5):651-659B.

    Article  CAS  PubMed  PubMed Central  Google Scholar 

  53. Martinez Villar G, Daneault V, Martineau-Dussault MÈ, et al. Altered resting-state functional connectivity patterns in late middle-aged and older adults with obstructive sleep apnea. Front Neurol. 2023;14:1215882.

    Article  PubMed  PubMed Central  Google Scholar 

  54. Schneider S, Strüder HK. Monitoring effects of acute hypoxia on brain cortical activity by using electromagnetic tomography. Behav Brain Res. 2009;197(2):476–80.

    Article  PubMed  Google Scholar 

  55. Xu F, Frazier DT. Role of the cerebellar deep nuclei in respiratory modulation. Cerebellum. 2002;1(1):35–40.

    Article  PubMed  Google Scholar 

  56. Strick PL, Dum RP, Fiez JA. Cerebellum and nonmotor function. Annu Rev Neurosci. 2009;32:413–34.

    Article  CAS  PubMed  Google Scholar 

  57. Jack A, Pelphrey KA. Neural correlates of animacy attribution include neocerebellum in healthy adults. Cereb Cortex. 2015;25(11):4240–7.

    Article  PubMed  Google Scholar 

  58. Chen HL, Lu CH, Lin HC, et al. White matter damage and systemic inflammation in obstructive sleep apnea. Sleep. 2015;38(3):361–70.

    Article  PubMed  PubMed Central  Google Scholar 

  59. Chen HL, Huang CC, Lin HC, et al. White matter alteration and autonomic impairment in obstructive sleep apnea. J Clin Sleep Med. 2020;16(2):293–302.

    Article  PubMed  PubMed Central  Google Scholar 

  60. Koo DL, Cabeen RP, Yook SH, et al. More extensive white matter disruptions present in untreated obstructive sleep apnea than we thought: a large sample diffusion imaging study. Hum Brain Mapp. 2023;44(8):3045–56.

    Article  PubMed  PubMed Central  Google Scholar 

  61. Li Y, Wen H, Li H, et al. Characterisation of brain microstructural alterations in children with obstructive sleep apnea syndrome using diffusion kurtosis imaging. J Sleep Res. 2023;32(2): e13710.

    Article  PubMed  Google Scholar 

  62. Schurr R, Zelman A, Mezer AA. Subdividing the superior longitudinal fasciculus using local quantitative MRI. Neuroimage. 2020;208: 116439.

    Article  PubMed  Google Scholar 

  63. Briggs RG, Khan AB, Chakraborty AR, et al. Anatomy and white matter connections of the superior frontal gyrus. Clin Anat. 2020;33(6):823–32.

    Article  PubMed  Google Scholar 

  64. Sahib A, Roy B, Kang D, Aysola RS, Wen E, Kumar R. Relationships between brain tissue damage, oxygen desaturation, and disease severity in obstructive sleep apnea evaluated by diffusion tensor imaging. J Clin Sleep Med. 2022;18(12):2713–21.

    Article  PubMed  PubMed Central  Google Scholar 

Download references

Acknowledgements

The authors would like to thank all the study participants who dedicated to our research and all the experts who helped to complete this research. We are also grateful to the participants and investigators of the FinnGen and UKB studies.

Funding

This work was supported by the Young Scientist Fund of National Natural Science Foundation of China (NO. 82100062) and school-enterprise cooperation funding project of Guangzhou (NO. SL2023A03J01372).

Author information

Author notes

  1. Kang Wu, Qiming Gan, Yuhong Pi and Yanjuan Wu contributed equally to this work.

Authors and Affiliations

  1. Sleep Medicine Center, State Key Laboratory of Respiratory Disease, National Clinical Research Center for Respiratory Disease, National Center for Respiratory Medicine, Guangzhou Institute of Respiratory Health, The First Affiliated Hospital of Guangzhou Medical University, NO.28 Qiaozhong Mid Road, Guangzhou, Guangdong, 510160, China

    Kang Wu, Qiming Gan, Yuhong Pi, Yanjuan Wu, Xiaofen Su, Sun Zhang, Xinni Wang & Nuofu Zhang

  2. The Affiliated Brain Hospital of Guangzhou Medical University, Guangzhou, Guangdong, China

    Wenjin Zou

  3. Department of Radiology, The First Affiliated Hospital of Guangzhou Medical University, Guangzhou, Guangdong, China

    Xinchun Li

Authors
  1. Kang Wu
    View author publications

    You can also search for this author inPubMed Google Scholar

  2. Qiming Gan
    View author publications

    You can also search for this author inPubMed Google Scholar

  3. Yuhong Pi
    View author publications

    You can also search for this author inPubMed Google Scholar

  4. Yanjuan Wu
    View author publications

    You can also search for this author inPubMed Google Scholar

  5. Wenjin Zou
    View author publications

    You can also search for this author inPubMed Google Scholar

  6. Xiaofen Su
    View author publications

    You can also search for this author inPubMed Google Scholar

  7. Sun Zhang
    View author publications

    You can also search for this author inPubMed Google Scholar

  8. Xinni Wang
    View author publications

    You can also search for this author inPubMed Google Scholar

  9. Xinchun Li
    View author publications

    You can also search for this author inPubMed Google Scholar

  10. Nuofu Zhang
    View author publications

    You can also search for this author inPubMed Google Scholar

Contributions

KW and NFZ conceptualized and supervise the study, and acquired the fundings. KW, QMG, YHP, YJW, WJZ, XCL curated data, designed the methodology and provided project administration. KW, QMG and QMG performed the formal statistical analysis. KW, QMG, YHP, YJW, WJZ, XFS, SZ, XNW, XCL, and NFZ reviewed results and verified the conclusions. KW and NFZ provided overall supervision. KW, QMG, and YJW wrote the original draft. KW, QMG, YHP, YJW, WJZ, XFS, SZ, XNW, XCL, and NFZ edited the original and revised manuscripts. All authors have read and approved the final version of the manuscript.

Corresponding author

Correspondence to Nuofu Zhang.

Ethics declarations

Ethics approval and consent to participate

The ethics committee of the First Affiliated Hospital of Guangzhou Medical University strictly follows the Declaration of Helsinki and International Ethical Guidelines for Health-related Research Involving Humans, etc., to perform independent ethical review duties (Ethics number: 201705). All genome-wide association studies included in this study all originate from publicly published GWAS summary databases, which complies with the conditions for exemption from review as stated in the “Ethical Review Measures for Life Sciences and Medical Research Involving Humans.”

Consent for publication

Not applicable.

Competing interests

The authors declare no competing interests.

Additional information

Publisher’s Note

Springer Nature remains neutral with regard to jurisdictional claims in published maps and institutional affiliations.

Supplementary Information

12916_2025_3876_MOESM1_ESM.docx

Additional file 1: The detail information about methods, including inclusion and exclusion criteria in observational study, overnight polysomnography and MRI image acquisition.

Additional file 2. The added information about GWAS datasets and the detail results Mendelian Randomization analyses.

12916_2025_3876_MOESM3_ESM.docx

Additional file 3. The supplementary figures. Fig. S1. The partial correlation of edge 893 in dimensionality 100 linked node 32 and node 43. Fig. S2. The partial correlation of edge 951 in dimensionality 100 linked node 5 and node 45. Fig. S3. The partial correlation of edge 1213 in dimensionality 100 linked node 37 and node 50. Fig. S4. The partial correlation of edge 258 in dimensionality 100 linked node 5 and node 100.

Rights and permissions

Open Access This article is licensed under a Creative Commons Attribution-NonCommercial-NoDerivatives 4.0 International License, which permits any non-commercial use, sharing, distribution and reproduction in any medium or format, as long as you give appropriate credit to the original author(s) and the source, provide a link to the Creative Commons licence, and indicate if you modified the licensed material. You do not have permission under this licence to share adapted material derived from this article or parts of it. The images or other third party material in this article are included in the article’s Creative Commons licence, unless indicated otherwise in a credit line to the material. If material is not included in the article’s Creative Commons licence and your intended use is not permitted by statutory regulation or exceeds the permitted use, you will need to obtain permission directly from the copyright holder. To view a copy of this licence, visit http://creativecommons.org/licenses/by-nc-nd/4.0/.

Reprints and permissions

About this article

Check for updates. Verify currency and authenticity via CrossMark

Cite this article

Wu, K., Gan, Q., Pi, Y. et al. Obstructive sleep apnea and structural and functional brain alterations: a brain-wide investigation from clinical association to genetic causality. BMC Med 23, 42 (2025). https://doiorg.publicaciones.saludcastillayleon.es/10.1186/s12916-025-03876-8

Download citation

  • Received:

  • Accepted:

  • Published:

  • DOI: https://doiorg.publicaciones.saludcastillayleon.es/10.1186/s12916-025-03876-8

Keywords

BMC Medicine

ISSN: 1741-7015

Contact us