Multilayer network analysis of dynamic network reconfiguration in patients with moderate-to-severe obstructive sleep apnea and its association with neurocognitive function

Investigating the link between genetic predictive factors of brain functional networks and two specific sleep disorders: Sleep apnoea and snoring

BACKGROUND

Sleep disorders are a widespread public health issue globally. Investigating the causal relationship between resting-state brain functional abnormalities and sleep disorders can provide scientific evidence for precision medicine interventions.

METHODS

We screened single nucleotide polymorphisms (SNPs) associated with rs-fMRI phenotype as instrumental variables Using bidirectional two-sample Mendelian randomization (MR), mediation MR, and multivariate MR based on Bayesian methods, the study tested the causal relationship between genetically predicted rs-fMRI and nine common sleep disorders.

RESULTS

The main inverse variance weighted (IVW) analysis identified four resting state functional magnetic resonance imaging (rs-fMRI) phenotypes that are causally associated with the risk of sleep disorders. For example, increased amplitude in nodes of the parietal, precuneus, occipital, temporal, and cerebellum regions, as well as the default mode network (DMN), central executive network (CEN) and attention network (AN) was associated with an increased risk of sleep apnoea. Enhanced neural activity in the calcarine or lingual and cerebellum regions and increased functional connectivity with the visual and subcortical-cerebellum networks was associated with a reduced risk of snoring. The mediation MR analysis shows that, body mass index (BMI) plays a significant mediating role in the risk of sleep apnoea by modulating the amplitude of nodes in the parietal, temporal, and cerebellum regions, as well as the connectivity changes in the DMN, CEN, and AN.

CONCLUSIONS

This study identified three rs-fMRI phenotypes linked to increased sleep apnoea risk and one associated with decreased snoring risk, providing an important target for the treatment of sleep disorders at the level of brain functional networks.

Machine Learning Classification Based on Individual Whole-Brain Functional Connectivity in Male OSA Patients

Purpose

Previous studies have shown altered paired brain functional connectivity (FC) in obstructive sleep apnea (OSA) patients, linked to cognitive impairment. This study utilized individual FC analysis to investigate the distinctive FC characteristics in OSA and evaluate their classification efficiency.

Methods

We included 82 moderate to severe OSA patients [41 OSA with normal cognition (OSA-NC), 41 OSA with mild cognitive impairments (OSA-MCI)] and 84 healthy control (HC). Resting-state fMRI data and clinical scale data were collected. Individual FC was derived using multi-task learning-based sparse convex alternating structure optimization, with feature selection via the least absolute shrinkage and selection operator. Support vector machine classifiers were used for OSA vs HC and OSA-NC vs OSA-MCI classification. The top 10 FC features contributing to classification were analyzed for group differences. A significance level of p < 0.05 was considered statistically significant.

Results

The study results showed that individual FC achieved higher classification accuracy than traditional Pearson-based FC (OSA vs HC: 91.8% vs 79.5%; OSA-NC vs OSA-MCI: 81.3% vs 63.8%). The top 10 individual-specific FC networks contributing to classification were mainly located in the default mode network, attention network, showing significant inter-group differences in connectivity strength between the two groups.

Conclusion

This study identified static individualized FC characteristics in OSA patients with varying cognitive impairments. Based on individual FC, the classification accuracy of OSA-NC and OSA-MCI was significantly improved, the individual FC may serve as a potential neuroimaging marker for predicting OSA-MCI, providing an individual clinical diagnosis and treatment evaluation.

Distinct Convergent Brain Alterations in Sleep Disorders and Sleep Deprivation: A Meta-Analysis

Importance

Sleep disorders have different etiologies yet share some nocturnal and daytime symptoms, suggesting common neurobiological substrates; healthy individuals undergoing experimental sleep deprivation also report analogous daytime symptoms. However, brain similarities and differences between long-term sleep disorders and short-term sleep deprivation are unclear.

Objective

To investigate the shared and specific neural correlates across sleep disorders and sleep deprivation.

Data Sources

PubMed, Web of Science, Embase, Scopus, and BrainMap were searched up to January 2024 to identify relevant structural and functional neuroimaging articles.

Study Selection

Whole-brain neuroimaging articles reporting voxel-based group differences between patients with different sleep disorders and healthy control participants or between total or partial sleep-deprived and well-rested individuals were included.

Data Extraction and Synthesis

Significant coordinates of group comparisons, their contrast direction (eg, patients < controls), and imaging modality were extracted. For each article, 2 raters independently evaluated eligibility and extracted data. Subsequently, several meta-analyses were performed with the revised activation likelihood estimation algorithm using P < .05 cluster-level familywise error correction.

Main Outcomes and Measures

Transdiagnostic regional brain alterations were identified across sleep disorders and among articles reporting sleep deprivation. Their associated behavioral functions and task-based or task-free connectivity patterns were explored using 2 independent datasets (BrainMap and the enhanced Nathan Kline Institute-Rockland Sample).

Results

A total of 231 articles (140 unique experiments, 3380 unique participants) were retrieved. The analysis across sleep disorders (n = 95 experiments) identified the subgenual anterior cingulate cortex (176 voxels, z score = 4.86), associated with reward, reasoning, and gustation, and the amygdala and hippocampus (130 voxels, z score = 4.00), associated with negative emotion processing, memory, and olfaction. Both clusters had positive functional connectivity with the default mode network. The right thalamus (153 voxels, z score = 5.21) emerged as a consistent regional alteration following sleep deprivation (n = 45 experiments). This cluster was associated with thermoregulation, action, and pain perception and showed positive functional connectivity with subcortical and (pre)motor regions. Subanalyses regarding the direction of alterations demonstrated that the subgenual anterior cingulate cortex exhibited decreased activation, connectivity, and/or volume, while the amygdala and hippocampus cluster and the thalamus cluster demonstrated increased activation, connectivity, and/or volume.

Conclusions and Relevance

Distinct convergent brain abnormalities were observed between long-term sleep disorders (probably reflecting shared symptoms) and short-term sleep deprivation.

Disrupted Cross-Scale Network Associated With Cognitive-Emotional Disorders in Sudden Sensorineural Hearing Loss

BACKGROUND

Sudden sensorineural hearing loss (SSNHL) is associated with abnormal changes in the brain's central nervous system. Previous studies on the brain networks of SSNHL have primarily focused on functional connectivity within the brain. However, in addition to functional connectivity, structural connectivity also plays a crucial role in brain networks. Moreover, traditional functional connectivity analyses often overlook the spatial and temporal characteristics of connectivity changes and fail to provide directional information and causal relationships.

AIMS

This study utilized Structural Covariance Network (SCN), multilayer network analysis, and Dynamic Causal Modeling (DCM) to investigate the cross-scale changes in neural network structure and function in SSNHL patients with accompanying cognitive and emotional disorders.

MATERIALS & METHODS

We collected 3D-T1 structural magnetic resonance image data and functional magnetic resonance image data from 70 SSNHL patients and 81 healthy controls (HCs). SCN analysis was performed based on gray matter volume, and multilayer network analysis was used to calculate node switching rates. Based on the results of multilayer network analysis, six nodes exhibiting significant inter-group differences in node switching rates were selected as regions of interest (ROIs). DCM was then conducted to explore the causal relationships of functional connectivity between these nodes.

RESULTS

Based on SCN, there were no significant inter-group differences in global network properties between SSNHL and HCs. At the node level, the left precentral gyrus in SSNHL showed a significant decrease in node efficiency. In the multilayer network analysis, SSNHL showed a significantly increased node switching rate at the level of the Left Superior Frontal Gyrus (L.SFG), Left Supplementary Motor Area (L.SMA), Left Superior Parietal Gyrus (L.SPG), Right Superior Parietal Gyrus (R.SPG), Right Inferior Parietal Lobe(R.IPL), and Left Thalamus (L.THA). Furthermore, the node switching rate of L.SFG showed a significant negative correlation with the Self-Rating Anxiety Scale (SAS) scores. DCM analysis of these six nodes revealed differences in the functional effective connectivity between the left superior parietal gyrus (L.SPG) and the left supplementary motor area (L.SMA), which were positively correlated with the AVLT-delay scores.

DISCUSSION

These findings suggest that SSNHL patients experience structural and functional remodeling of the cerebral cortex, with hearing loss leading to the reallocation of cognitive resources.

CONCLUSION

This provides new insights into understanding the potential mechanisms between cross-scale networks and cognitive-emotional disorders in SSNHL.

Abnormal dynamic functional connectivity and topological properties of cerebellar network in male obstructive sleep apnea

PURPOSE

To investigate dynamic functional connectivity (dFC) within the cerebellar-whole brain network and dynamic topological properties of the cerebellar network in obstructive sleep apnea (OSA) patients.

METHODS

Sixty male patients and 60 male healthy controls were included. The sliding window method examined the fluctuations in cerebellum-whole brain dFC and connection strength in OSA. Furthermore, graph theory metrics evaluated the dynamic topological properties of the cerebellar network. Additionally, hidden Markov modeling validated the robustness of the dFC. The correlations between the abovementioned measures and clinical assessments were assessed.

RESULTS

Two dynamic network states were characterized. State 2 exhibited a heightened frequency, longer fractional occupancy, and greater mean dwell time in OSA. The cerebellar networks and cerebrocerebellar dFC alterations were mainly located in the default mode network, frontoparietal network, somatomotor network, right cerebellar CrusI/II, and other networks. Global properties indicated aberrant cerebellar topology in OSA. Dynamic properties were correlated with clinical indicators primarily on emotion, cognition, and sleep.

CONCLUSION

Abnormal dFC in male OSA may indicate an imbalance between the integration and segregation of brain networks, concurrent with global topological alterations. Abnormal default mode network interactions with high-order and low-level cognitive networks, disrupting their coordination, may impair the regulation of cognitive, emotional, and sleep functions in OSA.

Altered functional connectivity of cerebellar subregions in male patients with obstructive sleep apnea: A resting-state fMRI study

PURPOSE

Previous studies have demonstrated impaired cerebellar function in patients with obstructive sleep apnea (OSA), which is associated with impaired cognition. However, the effects of OSA on resting-state functional connectivity (FC) in the cerebellum has not been determined. The purpose of this study was to investigate resting-state FC of the cerebellar subregions and its relevance to clinical symptoms in patients with OSA.

METHODS

Sixty-eight patients with OSA and seventy-two healthy controls (HCs) were included in the study. Eight subregions of the cerebellum were selected as regions of interest, and the FC values were calculated for each subregion with other voxels. A correlation analysis was performed to examine the relationship between clinical and cognitive data.

RESULTS

Patients with OSA showed higher FC in specific regions, including the right lobule VI with the right posterior middle temporal gyrus and right angular gyrus, the right Crus I with the bilateral precuneus/left superior parietal lobule, and the right Crus II with the precuneus/right posterior cingulate cortex. Furthermore, the oxygen depletion index was negatively correlated with aberrant FC between the right Crus II and the bilateral precuneus / right posterior cingulate cortex in OSA patients (p = 0.004).

CONCLUSION

The cerebellum is functionally lateralized and closely linked to the posterior default mode network. Higher FC is related to cognition, emotion, language, and sleep in OSA. Abnormal FC may offer new neuroimaging evidence and insights for a deeper comprehension of OSA-related alterations.