| Background and Purpose:Previous studies have found that obstructive sleep apnea(OSA)alters spontaneous brain activity and is associated with impaired cognitive function,but its precise neuropathological mechanisms are not clear.Percentage amplitude of fluctuation(Per AF)can directly reflect the fluctuation of the blood oxygenation level dependent signal at rest without the confounding effect of the original signal and frequency band,and this study aims to confirm the existence of localized spontaneous brain activity changes in OSA using Per AF and further investigate the relationship between such spontaneous brain activity changes and cognitive impairment and provide a new perspective on the neuropathological mechanisms of cognitive impairment in OSA.Methods:In this study,OSA patients aged 18~60 years diagnosed by the Department of Otolaryngology or the Sleep Monitoring Unit of the Department of Respiratory Medicine of the First Affiliated Hospital of Nanchang University from November 2020 to January 2022 were retrospectively recruited as the OSA group and healthy volunteers of similar age and education level as the control group.Finally,52 male patients with moderate-to-severe OSA(age 37.6 ± 9.6 years)and 61 healthy controls(HCs)(age 40.9 ± 11.9 years)were included in this study.All these subjects underwent resting-state functional magnetic resonance imaging and T1-scan imaging,and clinical data were collected.Resting functional MRI data were preprocessed in MATLAB using the DPARSF 6.0,SPM12 and RESTplus software packages,and Per AF values were calculated using the RESTplus software package.Differences in Per AF values between OSA patients and HCs were compared using an independent samples t-test corrected for Gaussian random effects,with P < 0.001 at the voxel level and P < 0.05 at the cluster level considered statistically different.Partial correlation analysis was used to assess the association between changes in Per AF and clinical scores in patients with OSA.Finally,a support vector machine was used to classify OSA and HCs according to Per AF values in different brain regions.Results:(1)Compared to HCs,OSA patients had lower Per AF values in the right rectus gyrus(t=-4.744,P<0.001)and left superior frontal gyrus(t=-5.255,P<0.001),but higher Per AF values in the right posterior cerebellum(t=5.001,P<0.001)and left middle frontal gyrus(t=4.002,P<0.001).(2)Per AF values in the right rectal gyrus of OSA patients were positively correlated with sleep efficiency(r=0.299,P=0.035)and Per AF values in the left superior frontal gyrus were positively correlated with sleep efficiency and disorientation scores on the Montreal Cognitive Assessment(r=0.296,P=0.037).(3)Support vector machine analysis showed that the accuracy of classification of Per AF values in the above differentiated brain regions reached 88.70%(sensitivity 84.91%,specificity 91.94%,AUC 96.59%),allowing a more accurate differentiation between OSA patients and HCs.Conclusion:(1)The frontal lobes and cerebellum of OSA patients show abnormal neuronal spontaneous activity,and these abnormal spontaneous activities may be associated with cognitive decline in OSA patients,providing a new perspective for understanding OSArelated cognitive impairment.(2)On the basis of support vector machines,Per AF values in different brain regions were able to discriminate between OSA and HCs,suggesting that Per AF can be a potential neuroimaging marker to distinguish OSA from HCs and provide a better method for the diagnosis of OSA. |
PDF Full Text Request