Accepted Manuscript Title: Effect of continuous positive airway pressure on cerebral blood flow in patients with obstructive sleep apnea Author: Jeong Sik Kim, JiHye Seo, Eun Yeon Joo, Seung Bong Hong PII: DOI: Reference:
S1389-9457(16)30006-5 http://dx.doi.org/doi: 10.1016/j.sleep.2016.03.010 SLEEP 3036
To appear in:
Received date: Revised date: Accepted date:
11-9-2015 3-3-2016 4-3-2016
Please cite this article as: Jeong Sik Kim, JiHye Seo, Eun Yeon Joo, Seung Bong Hong, Effect of continuous positive airway pressure on cerebral blood flow in patients with obstructive sleep apnea, Sleep Medicine (2016), http://dx.doi.org/doi: 10.1016/j.sleep.2016.03.010. This is a PDF file of an unedited manuscript that has been accepted for publication. As a service to our customers we are providing this early version of the manuscript. The manuscript will undergo copyediting, typesetting, and review of the resulting proof before it is published in its final form. Please note that during the production process errors may be discovered which could affect the content, and all legal disclaimers that apply to the journal pertain.
Effect of continuous positive airway pressure on cerebral blood flow in patients with obstructive sleep apnea Jeong Sik Kima, JiHye Seoa, Eun Yeon Jooa,b, Seung Bong Honga,b,*
Department of Neurology, Samsung Medical Center, Sungkyunkwan University School of
Medicine, Seoul, Republic of Korea b
Samsung Biomedical Research Institute (SBRI), Seoul, Republic of Korea
*Corresponding author. Department of Neurology, Samsung Medical Center, Sungkyunkwan University School of Medicine, 81 Irwon-ro, Gangnam-gu, Seoul 135-710, Korea. Tel.: +822-3410-3592; fax: +82-2-3410-0052. E-mail address: [email protected]
(Seung Bong Hong)
Page 1 of 29
Highlights There was significantly lower rCBF in multiple brain areas of untreated OSA patients. Long-term CPAP treatment resulted in partial or complete improvement of rCBF abnormalities. Improvements in clinical and PSG variables paralleled rCBF changes after CPAP therapy.
ABSTRACT Objective: Obstructive sleep apnea (OSA) is commonly associated with neural and cognitive deficits induced by recurrent hypoxemia and sleep fragment. The aims of this study were to use statistical parametric mapping (SPM) to analyze changes of regional cerebral blood flow (rCBF) in untreated patients with severe OSA before and after nasal continuous positive airway pressure (CPAP) treatment; examine the impact of OSA-related variables on rCBF; and assess the therapeutic effect of nasal CPAP treatment. Methods: Thirty male patients with severe OSA underwent brain single photon emission computed tomography (SPECT) scans twice before and after CPAP treatment for ≥6 months, whereas 26 healthy controls underwent a single SPECT scan. The rCBF differences were compared between two OSA sub-groups (untreated and treated) and the control group, and correlations between rCBF differences and clinical parameters were analyzed. Results: Compared with the controls, the untreated OSA patients showed a significantly lower rCBF in multiple brain areas. After the treatment, partial reversal of the rCBF decreases was observed in the limbic and prefrontal areas. Moreover, complete reversal of the
Page 2 of 29
rCBF decreases was observed in the medial orbitofrontal, angular and cerebellar areas. Significant improvements in some clinical and polysomnographic variables (Epworth Sleepiness Scale, apnea-hypopnea index, CPAP duration, and arousal index) paralleled the rCBF changes after the treatment. Conclusions: Decreased rCBF in severe OSA was significantly reversible by CPAP treatment and correlated with the improvements in the apnea-hypopnea index, arousal index, CPAP duration and Epworth Sleepiness Scale. These results suggest that long-term CPAP treatment improves rCBF in areas responsible for executive, affective, and memory function.
Keywords: Obstructive sleep apnea Single photon emission computed tomography (SPECT) Cerebrovascular circulation Continuous positive airway pressure (CPAP) Introduction Obstructive sleep apnea (OSA) syndrome is the most common type of sleep disorder. It is clinically characterized by repeated partial or complete obstruction of the upper airway during sleep, leading to an intermittent oxygen deficiency in the blood. Obstructive sleep apnea occurs in all age groups from infants to the elderly, but it is more common in men than women . Obstructive sleep apnea is associated with cerebrovascular and cardiovascular diseases, excessive daytime sleepiness (EDS), and cognitive dysfunction [2-4]. The correlations between these deficits and functional and structural brain abnormalities have been controversial. 3
Page 3 of 29
Nasal continuous positive airway pressure (CPAP), which keeps the airway open during the night, is the most common and effective nonsurgical treatment for patients with OSA. Several studies have revealed that CPAP treatment eliminates the respiratory disturbance and sleep fragmentation caused by OSA , resulting in improvements in subjective daytime functioning and cognitive function, and decreased risk of cardiovascular diseases [6-14]. So far, several neuroimaging studies have reported evidence supporting the therapeutic effects of CPAP for OSA. Some functional studies have indicated a persistent lack of prefrontal activation, with partial recovery in the posterior parietal lobes , and decreases in the activation of prefrontal and hippocampal structures after CPAP treatment . A structural magnetic resonance imaging (MRI) study using voxel-based morphometry reported a gray matter (GM) volume increase in hippocampal and frontal structures after 3 months of CPAP , whereas no GM changes were reported after CPAP therapy in another study . A diffusion tensor imaging study reported almost complete reversal of white matter abnormalities after patients with OSA underwent 12 months of CPAP treatment . Another study  investigated regional cerebral blood flow (rCBF) changes by statistical parametric mapping (SPM) analysis of brain single photon emission computed tomography (SPECT) scans to evaluate the therapeutic effects of CPAP in patients with OSA. This study demonstrated deficits in rCBF in the frontal lobes of such patients, which were reversed by good CPAP compliance. However, this study had some limitations: the small number of subjects (patients with OSA, n=15; healthy controls, n=9) and the relatively short duration (3 months) of CPAP treatment. The present study aimed to use SPM to analyze changes of regional cerebral blood flow (rCBF) in untreated patients with severe OSA before and after long-term nasal CPAP treatment (≥6 months); examine the impact of OSA-related variables on rCBF; and assess the 4
Page 4 of 29
therapeutic effect of nasal CPAP treatment.
2.1. Subjects The present study consecutively recruited 34 previously untreated patients with severe OSA (all male, aged 31-60 years) and 30 age-matched and gender-matched, healthy volunteers (aged 31-60) at the sleep disorder clinic of a university hospital located in Seoul, Korea. All patients underwent routine diagnostic work-ups, including: blood tests, electrocardiograms, and chest radiographs. Inclusion criteria were as follows: (1) being male; (2) aged >18 years but ≤60 years; and (3) having an apnea-hypopnea index (AHI) >30. Thirty age-matched, healthy male volunteers were recruited using a local community advertisement. Each candidate was subjected to a detailed clinical interview, sleep questionnaire, and overnight polysomnography (PSG). Control candidates were excluded if they had any evidence of OSA (AHI >5) or other sleep disorders such as periodic limb movement disorder and insomnia. Exclusion criteria for patients and healthy volunteers were as follows: (1) mean daily sleep time <7 hours; (2) abnormal sleep-wake rhythm; (3) other sleep disorders; (4) hypertension, diabetes, heart, and respiratory diseases; (5) history of cerebrovascular or cardiovascular disease; (6) other neurological (epilepsy, neurodegenerative diseases, head injury) or psychiatric diseases (psychosis, current depression); (7) alcohol or illicit drug abuse or current intake of psychoactive medications; and (8) a structural lesion on brain MRI. Three patients and four controls with poor-quality preprocessed SPECT scans were excluded from 5
Page 5 of 29
the patient and control groups, respectively. Finally, 30 patients with OSA and 26 healthy volunteers were included in the present study. Informed consent was obtained from all subjects and the hospital Institutional Review Board approved the study protocol.
2.2. Overnight PSG and CPAP All patients were asked to abstain from alcohol or caffeinated beverages on the day before the sleep studies were conducted. Sleep studies were recorded using a Somnologica (Embla; Denver, CO, USA). Polysomnography was recorded using a six-channel electroencephalogram (EEG) (F3/A2; F4/A1; C3/A2; C4/A1; O1/A2; O2/A1), a four-channel electrooculogram, an electromyogram (EMG) (on submental, intercostal, and anterior tibialis muscles), and an electrocardiogram (ECG) with surface electrodes. Patients were also attached with a thermistor (for monitoring oronasal airflow), nasal air pressure monitor, an oximeter (for measuring oxygen saturation), piezoelectric bands (for determining thoracic and abdominal wall motion), and a body-position sensor. Patients went to bed at 23:00 and were awakened at 07:00 the next day. Sleep architecture was scored in 30-second epochs, and sleep staging was interpreted in accordance with the criteria of Rechtschaffen and Kales . Apneas and hypopneas were defined by standard scoring scales . The American Academy of Sleep Medicine (AASM) rules (AASM manual version 2.0) as published  were used for scoring hypopnea as follows: ≥30% reduction in nasal pressure signal excursions from baseline that lasted ≥10 seconds with a ≥3% desaturation from the pre-event baseline or arousal. To determine the effective pressure required to maintain airway patency, patients underwent CPAP-titration PSG. After the PSG evaluation, the optimal pressure was identified as the minimum pressure capable of eliminating apneas and hypopneas in each 6
Page 6 of 29
patient. All patients were sent home with fixed PAP devices with C-FLEX (Respironics, Mseries). In-home adherence was objectively reported by Encore software, and the patients
Comment [A1]: Please also state place of
were excluded from analysis if their compliance rate was <5 hours per night and <80% of the
days in the study period.
Comment [A2]: Please state name of developer and place
In addition, subjective sleepiness was evaluated using the Epworth Sleepiness Scale (ESS) to assess excessive daytime sleepiness (EDS) .
2.3. Technetium-99m ethyl cysteinate dimer brain SPECT All participants refrained from consuming caffeinated beverages but were allowed to drink water from 07:00 until the end of the SPECT study. Technetium-99m ethyl cysteinate dimer (ECD) (99mTc-ECD; 925.9 MBq) was administered at 10:00, and SPECT scans were performed 30 minutes after injecting the radioisotope using a three-headed Triad XLT system equipped with low-energy, high-resolution collimators (Trionix Research Laboratory, Twinsburg, OH, USA). This camera has a transaxial system resolution of 6.9 mm full width at half maximum (FWHM). Images were reconstructed using filtered back projection and a Butterworth filter. The attenuation correction was performed using Chang’s method (attenuation coefficient=0.12 cm–1). The reconstructed voxel size for SPECT was 3.56 × 3.56 × 3.56 mm (x, y, and z, respectively). Patients and controls were instructed not to fall asleep after ECD injection. Wakefulness after an ECD injection was monitored using a four-channel EEG, a two-channel ECG, and single-channel EMG. The Institutional Review Board at Samsung Medical Center authorized the form used for informed consent and the study protocol, which included the administration of a radioactive substance and SPECT scanning. All patients underwent brain SPECT scanning twice (before and after CPAP treatment), whereas healthy controls underwent a single 7
Page 7 of 29
2.4. Brain SPECT image preprocessing and statistical analyses All SPECT images from patients and controls were converted from the digital imaging and communications in medicine (DICOM) format to the Analyze format. Preprocessing and statistical analyses were performed using SPM8 (Wellcome Department of Cognitive Neurology, Institute of Neurology, University College London, UK; http://www.fil.ion.ucl.ac.uk/spm) implemented in MATLAB 7.6 (The MathWorks, Natick, MA, USA). For comparing perfusion changes between the two subgroups of patients (untreated and CPAP treated) and healthy controls, a co-registration process was performed to normalize the motion difference between a set of SPECT scans. A customized SPECT template image was then built using SPECT and MRI scans from the 26 healthy controls. The co-registered SPECT images were spatially normalized to the customized SPECT template, and smoothed using a 16-mm FWHM filter. Brain perfusion at each voxel was proportionally scaled according to the mean value of the whole brain to remove any differences in global CBF between subjects. Prior to the statistical analyses, differences were determined between groups for clinical variables such as age, body mass index (BMI), and PSG at the p<0.05 level of statistical significance using a paired t-test or an independent t-test. Furthermore, to identify rCBF differences between untreated patients with OSA and healthy controls, and between post-treatment patients with OSA and healthy controls, two-sample t-tests were applied, with age and BMI as covariates. Moreover, a paired t-test with age and BMI as covariates was performed to directly compare rCBF changes in OSA patients before and after treatment. In 8
Page 8 of 29
SPM analyses, contrasts were computed to represent brain regions showing both hyperperfusion and hypo-perfusion differences of untreated and treated patients with OSA, and of untreated or treated patients with OSA compared with healthy controls. The resulting regions of difference were then examined at the peak level. The SPM results were uncorrected (p<0.001) and corrected for multiple comparisons using a topological false discovery rate statistical threshold of p<0.05, with an extent threshold of at least 50 contiguous voxels (the size each voxel being 2 2 2 mm). A set of Montreal Neurologic Institute (MNI) stereotaxic coordinates (x, y, and z) provided in the SPM results were confirmed by visual analyses and further converted to anatomical names using a free toolbox (xjView8.4) that is
Comment [A3]: Please provide details like
compatible with SPM8.
you have done for the one below
Moreover, to evaluate the degree of rCBF normalization (recovery) after the CPAP treatment, compared with the controls, the regions-of-interest (ROI) analysis for quantifying rCBF differences between healthy controls and OSA patients with CPAP treatment was performed. To do so, each region showing a significant rCBF increase after CPAP treatment was first defined as an ROI (right medial frontal, left cingulate gyrus, right anterior cingulum, left parahippocampal gyrus, right postcentral gyrus) by the MarsBaR toolbox for SPM (http://marsbar.sourceforge.net). Then the mean intensities in the ROI images were calculated from the controls and post-treatment. Finally, the rCBF recovery ratio of the post-treatment [recovery ratio (%) = (mean intensity of post-treatment patients/mean intensity of controls)*100%] was calculated. Correlations between rCBF differences (pre-CPAP and post-CPAP treatment) and the average time and frequency of CPAP usage (per week), ESS, AHI, and BMI were tested using Pearson’s correlation analysis in patients with OSA, with age as a nuisance variable using an uncorrected p-value (p<0.001). The extent threshold for the whole brain was set to 9
Page 9 of 29
KE>50. The results were superimposed on the two-dimensional planes of a single-subject MRI template using SPM8.
3.1. Clinical characteristics All patients with OSA and the controls were middle-aged men. T1, T2, and FLAIR images were inspected to exclude patients with OSA or controls who had gross structural abnormalities on brain MRI. The clinical characteristics and detailed PSG findings of the patients are summarized and compared in Table 1. The mean duration of CPAP was 7.9±1.5 months (6.03±0.57 hours/night), and thus the mean frequency of CPAP use was 86.4±14.3%. The mean CPAP pressure was 9.5±1.9 mm H2O. After long-term CPAP therapy, daytime sleepiness improved significantly (mean ESS 12.0±5.0 → 7.5±4.0, and mean Stanford Sleepiness Scale (SSS) 3.0±1.2 → 1.8±0.9, both p<0.001). The BMI was not different during the study period (pre-treatment 26.22±3.16 kg/m2, post-treatment 26.38±2.73 kg/m2, p=0.84). Compared with the pre-CPAP PSG, some parameters showed a significant improvement in the post-CPAP PSG data. The mean AHI (63.4±17.5 → 4.0±1.5/hours) and arousal index (57.1±20.8 → 7.1±2.7) were significantly decreased after CPAP (both p<0.001). Sleep efficiency and mean sleep latency (SL) were significantly improved to within normal range. Percentages of non-rapid eye movement (N-REM) sleep stage 1, 2, 3, and REM sleep were significantly improved to within normal range (p<0.001) (see Table 1).
Page 10 of 29
3.2. Wakefulness monitoring during SPECT studies The EEGs, ECGs, and EMGs were monitored during SPECT studies to measure the duration of wakefulness after an ECD injection in all subjects. Mean sleep latency after an ECD injection was 7.6±4.2 minutes (range 4.5-20). Thus, it is confirmed that all subjects underwent brain SPECT during the waking state because brain uptake of the radiotracer is completed within 1 to 2 minutes after injection.
3.3. SPM analyses of 99mTc-ECD brain SPECT Compared with the control group, the rCBF in patients with untreated OSA was significantly decreased in the left parahippocampal gyrus, bilateral medial frontal gyri, right medial orbitofrontal gyrus, right anterior cingulum and cingulate gyrus, right postcentral gyrus, left angular gyrus, and right cerebellum at the uncorrected level of p<0.001. After long-term CPAP treatment, these patients showed a significant rCBF increase completely in the right medial orbitofrontal gyrus, left angular gyrus, and right cerebellum, and partially in the right anterior cingulum and cingulate gyrus, bilateral medial frontal gyri, left parahippocampal gyrus, and right postcentral gyrus (Table 2 and Fig. 1). In the direct comparison between pretreatment and post-treatment, a significant rCBF improvement after CPAP treatment was observed in the same brain regions, with similar or wider extent overlapped with rCBF decrease regions in the pre-treatment patients compared with the controls (p<0.001). Compared with the control group, the rCBF in post-treatment OSA patients was
Comment [A4]: Please check this as im not
sure it makes sense. Do you mean: with similar or greater overlapping with the decreased
recovered to 95.6% in the right medial frontal, left cingulate gyrus and right anterior
cingulum, 94.3% in the left parahippocampal gyrus, and 97% in the right postcentral gyrus. Comparing the rCBF in OSA patients before CPAP treatment with healthy controls, there were negative correlations between the rCBF in the right frontal lobe sub-gyral region 11
Page 11 of 29
and the AHI, between the rCBF in the right middle frontal gyrus and the ESS, and between the rCBF in the right medial frontal gyrus and the SL. With regard to the changes in rCBF between pre-treatment and post-treatment, there were significant positive correlations between the post-CPAP rCBF increase in the right superior temporal gyrus and the difference in AHI, between the post-CPAP increase in rCBF in the left middle frontal gyrus and the difference in ESS, and between the post-CPAP increase in rCBF in the right superior temporal gyrus and the difference in CPAP duration (Table 3 and Fig. 2).
This study investigated the presence of rCBF changes, and the effect of long-term CPAP treatment in a large sample of middle-aged men. Two prior SPECT studies have demonstrated complete recovery of rCBF changes (frontal hyperperfusion and parietal hypoperfusion before treatment) during sleep  and partial recovery of rCBF decrease in the superior and middle frontal gyri of OSA patients with good CPAP compliance during wakefulness . Compared with these studies with relatively shorter treatment duration and smaller sample sizes [20, 24], the present long-term CPAP treatment improved rCBF in brain regions (parahippocampal, prefrontal, cingulate, and postcentral gyri) that were largely overlapped with the regions of initial rCBF decreases in the pre-treatment patients compared with the healthy controls, paralleling with significant improvements in some clinical and PSG variables. Partial recovery of rCBF in the parahippocampal gyrus and prefrontal cortex of the present study is concordant to previous structural and functional neuroimaging [16, 17, 25], 12
Page 12 of 29
but not to prior SPECT studies. One task-based fMRI study  reported evidence for the therapeutic effects of CPAP, suggesting that decreases of activation in prefrontal and hippocampal structures may reflect reduced requirements for additional resources to support normal performance in the working memory task. A recent study  reported significant improvements in memory, attention, and executive function in parallel with GM volume increases in hippocampal and frontal structures after the CPAP treatment. The authors suggested that hippocampal damage, which is most strongly and quickly affected by hypoxic and hypercapnic events , could indirectly alter functional connectivity with the prefrontal cortex. In a previous SPECT study , decreased rCBF was found in the bilateral parahippocampal gyri, but not in other brain regions. It is believed that the results of the two studies are different because they have not only entirely different subjects, but also different severity of sleep apnea and other clinical features. Other CBF studies in OSA patients have
Comment [A5]: Do you mean this or rCBF?
also shown different patterns of CBF decrease [28, 29]. These different results of rCBF
Comment [A6]: As before
changes in OSA patients may be related to the difference of clinical characteristics of the patients in each study. From these results, it is speculated that the intermittent hypoxia and sleep disruption accompanying oxygen desaturation in OSA may be strongly associated with rCBF decreases in brain regions involved in cognitive dysfunctions, and that these rCBF abnormalities can be reversible by long-term CPAP treatment. The present patients demonstrated partial recovery of rCBF changes in the cingulate and postcentral gyrus after CPAP treatment. The cingulate cortices are also associated with the modulation of respiration and blood pressure . It has been reported that neurons in the cingulate gyrus and cingulum are vulnerable to excitotoxic processes during intermittent hypoxia or impaired perfusion in patients with OSA . A PET study  reported that
Comment [A7]: Please write all new
critical glucose metabolism in the anterior cingulate cortex was modestly increased after 13
abbreviations in full on first use
Page 13 of 29
adequate CPAP treatment. The authors suggested that the region is involved in a form of attention that serves to regulate both cognitive and emotional processing . One recent resting-state functional MRI study  reported a significant abnormal change in brain activity in the bilateral precentral and postcentral gyri in OSA patients, postulating that these changes could be linked to the increased mechanical effort that may occur during pathological arousals in sleep in response to repetitive choking sensations and physical efforts to restore breathing . Thus, the rCBF recovery of the postcentral gyrus in the present study could be partly explained by a compensation process decreasing oxidative stress exposed to arousal and hypoxic events according to the effective CPAP treatment. In the present study, it is less likely that the conditions might have affected the results
Comment [A8]: Please check to make sure that this makes sense
because all subjects had ≥7 hours of sleep per night and no subjects fell asleep or became drowsy at least 4 minutes after the radiotracer injection for brain SPECT. Sleep deprivation of normal subjects may induce a lot of functional deficit in the human brain. Actually, two brain perfusion studies [36, 37] on sleep deprivation or sleep restriction have shown different patterns of brain perfusion decrease compared to the present study. One study investigated the daytime effects of one night of total sleep deprivation on brain function in sleepwalkers and reported that sleep deprivation conditions induced significant hypoperfusion in the bilateral inferior temporal gyri . A previous functional neuroimaging study reported that this region is particularly vulnerable to sleep deprivation and activation of this region was found to be decreased in the sleep deprived state proportionally to performance decrease during a non-verbal recognition task . There were several limitations to the present study. The disease severity and good CPAP compliance in the patients group limited generalizability. Measurements of cognitive function could have provided a more comprehensive understanding of the hypoxic cerebral 14
Page 14 of 29
perfusion impairment and its reversibility by CPAP treatment in OSA patients. Nonetheless, the study examined rCBF changes in untreated OSA patients and the effect of CPAP on rCBF in a larger number of patients with longer periods of CPAP treatment, and revealed new findings.
Conclusions In conclusion, the present study hypothesized that the presence of cerebral perfusion changes in severe OSA patients compared with healthy subjects may be reversed by long-term CPAP treatment. To verify this, ECD SPECT images of the brain were compared before and after long-term nasal CPAP treatment (at least 6 months) in patients with severe OSA, and abnormal cerebral perfusion was identified in untreated patients with OSA. Furthermore, long-term CPAP treatment resulted in partial or complete improvement of these rCBF abnormalities. The differences of the AHI before and after CPAP and the duration of CPAP use had a positive correlation with the post-CPAP rCBF increase in the right superior temporal gyrus. In addition, the pre-CPAP and post-CPAP differences in ESS correlated with the post-CPAP rCBF increase in the left middle frontal gyrus. These findings suggest that long-term CPAP treatment can partially or completely reverse the reduced rCBF in patients with severe OSA in brain regions that are responsible for executive, affective, and memory function.
Acknowledgments This study was supported by a grant of the Korean Health Technology R&D Project, Ministry of Health & Welfare, Republic of Korea (A110097; to S.B.H.) and by a Grant of Basic 15
Page 15 of 29
Science Research Program through the National Research Foundation of Korea funded by the Ministry of Science, ICT & Future Planning, Republic of Korea (No. 2014 R1A1A3049510; to E.Y.J.).
Conflicts of Interest: The authors report no conflicts of interest.
Page 16 of 29
 Young T, Peppard PE, Gottlieb DJ. Epidemiology of obstructive sleep apnea: a population health perspective. Am J Respr Crit Care Med 2002;165:1217-39.  Bennett LS, Barbour C, Langford B, Stradling JR, Davies RJ. Health status in obstructive sleep apnea: relationship with sleep fragmentation and daytine sleepiness, and effects of continuous positive airway pressure treatment. Am J Respr Crit Care Med 1999;159:1884-90.  Dempsey JA, Veasey SC, Morgan BJ, O'Donnell CP. Pathophysiology of sleep apnea. Physiol Rev 2010;90:47-112.  Yaggi H, Mohsenin V. Obstructive sleep apnoea and stroke. Lancet Neurol 2004;3:33342.  Loredo JS, Ancoli-Israel S, Dimsdale JE. Effect of continuous positive airway pressure vs placebo continuous positive airway pressure on sleep quality in obstructive sleep apnea. Chest 1999;116:1545-9.  Sin DD, Logan AG, Fitzgerald FS, Liu PP, Bradley TD. Effects of continuous positive airway pressure on cardiovascular outcomes in heart failure patients with and without Cheyne-Stokes respiration. Circulation 2000;102:61-6. 17
Page 17 of 29
 Sanchez AI, Martinez P, Miro E, Bardwell WA, Buela-Casal G. CPAP and behavioral therapies in patients with obstructive sleep apnea: effects on daytime sleepiness, mood, and cognitive function. Sleep Med Rev 2009;13:223-33.  Naegele B, Pepin JL, Levy P, Bonnet C, Pellat J, Feuerstein C. Cognitive executive dysfunction in patients with obstructive sleep apnea syndrome (OSAS) after CPAP treatment. Sleep 1998;21:392-7.  Munoz A, Mayoralas LR, Barbe F, Pericas J, Agusti AG. Long-term effects of CPAP on daytime functioning in patients with sleep apnoea syndrome. Eur Respir J 2000;15:676-81.  Montserrat JM, Ferrer M, Hernandez L, Farre R, Vilagut G, Navajas D, et al. Effectiveness of CPAP treatment in daytime function in sleep apnea syndrome: a randomized controlled study with an optimized placebo. Am J Respr Crit Care Med 2001;164:608-13.  Lau EY, Eskes GA, Morrison DL, Rajda M, Spurr KF. Executive function in patients with obstructive sleep apnea treated with continuous positive airway pressure. JINS 2010;16:1077-88.  Ferini-Strambi L, Baietto C, Di Gioia MR, Castaldi P, Castronovo C, Zucconi M, et al. Cognitive dysfunction in patients with obstructive sleep apnea (OSA): partial reversibility after continuous positive airway pressure (CPAP). Brain Res Bull 2003;61:87-92. 18
Page 18 of 29
 Bedard MA, Montplaisir J, Malo J, Richer F, Rouleau I. Persistent neuropsychological deficits and vigilance impairment in sleep apnea syndrome after treatment with continuous positive airways pressure (CPAP). J Clin Exp Neuropsychol 1993;15:330-41.  Bardwell WA, Ancoli-Israel S, Berry CC, Dimsdale JE. Neuropsychological effects of one-week continuous positive airway pressure treatment in patients with obstructive sleep apnea: a placebo-controlled study. Psychosom Med 2001;63:579-84.  Thomas RJ, Rosen BR, Stern CE, Weiss JW, Kwong KK. Functional imaging of working memory in obstructive sleep-disordered breathing. J Appl Physiol 2005;98:222634.  Castronovo V, Canessa N, Strambi LF, Aloia MS, Consonni M, Marelli S, et al. Brain activation changes before and after PAP treatment in obstructive sleep apnea. Sleep 2009;32:1161-72.  Canessa N, Castronovo V, Cappa SF, Aloia MS, Marelli S, Falini A, et al. Obstructive sleep apnea: brain structural changes and neurocognitive function before and after treatment. Am J Respr Crit Care Med 2011;183:1419-26.  O'Donoghue FJ, Briellmann RS, Rochford PD, Abbott DF, Pell GS, Chan CH, et al. Cerebral structural changes in severe obstructive sleep apnea. Am J Respir Crit Care Med 2005;171:1185-90. 19
Page 19 of 29
 Castronovo V, Scifo P, Castellano A, Aloia MS, Iadanza A, Marelli S, et al. White matter integrity in obstructive sleep apnea before and after treatment. Sleep 2014;37:146575.  Shiota S, Inoue Y, Takekawa H, Kotajima M, Nakajyo M, Usui C, et al. Effect of continuous positive airway pressure on regional cerebral blood flow during wakefulness in obstructive sleep apnea. Sleep & breathing = Schlaf & Atmung 2014;18:289-95.  Rechtschaffen A, Dement W. Studies on the relation of narcolepsy, cataplexy, and sleep with low voltage random EEG activity. Research publications - Association for Research in Nervous and Mental Disease 1967;45:488-505.  Berry RB, Brooks R, Gamaldo CE, Harding SM, Marcus CL, Vaughn BV. The AASM manual for the scoring of sleep and associated events: rules, terminology and technical specifications, Version 2.0. Darien, Illinois: American Academy of Sleep Medicine; 2012.  Johns MW. A new method for measuring daytime sleepiness: the Epworth sleepiness scale. Sleep 1991;14:540-5.  Ficker JH, Feistel H, Moller C, Merkl M, Dertinger S, Siegfried W, et al. [Changes in regional CNS perfusion in obstructive sleep apnea syndrome: initial SPECT studies with injected nocturnal 99mTc-HMPAO]. Pneumologie 1997;51:926-30.  O'Donoghue FJ, Wellard RM, Rochford PD, Dawson A, Barnes M, Ruehland WR, et 20
Page 20 of 29
al. Magnetic resonance spectroscopy and neurocognitive dysfunction in obstructive sleep apnea before and after CPAP treatment. Sleep 2012;35:41-8.  Kloepfer C, Riemann D, Nofzinger EA, Feige B, Unterrainer J, O'Hara R, et al. Memory before and after sleep in patients with moderate obstructive sleep apnea. JCSM 2009;5:540-8.  Joo EY, Tae WS, Han SJ, Cho JW, Hong SB. Reduced cerebral blood flow during wakefulness in obstructive sleep apnea-hypopnea syndrome. Sleep 2007;30:1515-20.  Baril AA, Gagnon K, Arbour C, Soucy JP, Montplaisir J, Gagnon JF, et al. Regional Cerebral Blood Flow during Wakeful Rest in Older Subjects with Mild to Severe Obstructive Sleep Apnea. Sleep 2015;38:1439-49.  Innes CR, Kelly PT, Hlavac M, Melzer TR, Jones RD. Decreased Regional Cerebral Perfusion in Moderate-Severe Obstructive Sleep Apnoea during Wakefulness. Sleep 2015;38:699-706.  Brannan S, Liotti M, Egan G, Shade R, Madden L, Robillard R, et al. Neuroimaging of cerebral activations and deactivations associated with hypercapnia and hunger for air. Proceedings of the National Academy of Sciences of the United States of America 2001;98:2029-34.  Ang CW, Carlson GC, Coulter DA. Massive and specific dysregulation of direct 21
Page 21 of 29
cortical input to the hippocampus in temporal lobe epilepsy. J Neurosci 2006;26:11850-6.  Ju G, Yoon IY, Lee SD, Kim YK, Yoon E, Kim JW. Modest changes in cerebral glucose metabolism in patients with sleep apnea syndrome after continuous positive airway pressure treatment. Respiration 2012;84:212-8.  Bush G, Luu P, Posner MI. Cognitive and emotional influences in anterior cingulate cortex. Trends in Cognitive Sciences 2000;4:215-22.  Santarnecchi E, Sicilia I, Richiardi J, Vatti G, Polizzotto NR, Marino D, et al. Altered cortical and subcortical local coherence in obstructive sleep apnea: a functional magnetic resonance imaging study. J Sleep Res 2013;22:337-47.  Smejkal V, Druga R, Tintera J. Control of breathing and brain activation in human subjects seen by functional magnetic resonance imaging. Academia Scientiarum Bohemoslovaca 1999;48:21-5.  Dang-Vu TT, Zadra A, Labelle MA, Petit D, Soucy JP, Montplaisir J. Sleep Deprivation Reveals Altered Brain Perfusion Patterns in Somnambulism. PloS one. 2015;10:e0133474.  Bell-McGinty S, Habeck C, Hilton HJ, Rakitin B, Scarmeas N, Zarahn E, et al. Identification and differential vulnerability of a neural network in sleep deprivation. Cereb Cortex 2004;14:496-502. 22
Page 22 of 29
FIGURE LEGENDS Fig. 1. A statistical brain map revealing rCBF changes between patients with OSA (before and after CPAP treatment) and healthy controls. (A) The rCBF was decreased in patients before CPAP treatment compared with healthy controls in the left parahippocampal gyrus, right medial orbitofrontal cortex, right anterior cingulum and cingulate gyrus, bilateral medial frontal gyrus, and right postcentral gyrus, left angular gyrus, and right cerebellum (at the level of p<0.001, uncorrected for multiple comparisons, two-sample t-test). (B) The rCBF was restored in patients after CPAP treatment compared with healthy controls (at the level of p<0.001, uncorrected for multiple comparisons, two-sample t-test). After CPAP treatment, the rCBF in the left parahippocampal gyrus, right medial orbitofrontal cortex, left angular gyrus, and right cerebellum recovered completely, whereas the rCBF in the right anterior cingulum and cingulate gyrus, bilateral medial frontal gyrus, and right postcentral gyrus recovered partially. (C) Compared with pre-treatment baseline rCBF, patients showed a significant improvement in rCBF after CPAP treatment in the same brain regions with similar or wider extent overlapped with rCBF decrease regions in the pre-treatment patients compared with controls
Comment [A9]: As my query in main text
(at the level of p<0.001, uncorrected for multiple comparisons, paired t-test). The color bar indicates the T-value. Left side of brain image depicts the left hemisphere.
CPAP, continuous positive airway pressure; OSA, obstructive sleep apnea, rCBF, regional cerebral blood flow
Page 23 of 29
Fig. 2. Brain areas showing significant relationships between rCBF changes and patients’ clinical parameters. There were significant positive correlations of rCBF changes with preCPAP and post-CPAP AHI differences (Pearson’s r=0.507), ESS differences (Pearson’s r=0.546), and durations of CPAP use (Pearson’s r=0.589) at the level of p<0.001 after accounting for the effects of age and BMI scores.
The x-axis indicates clinical parameters; the y-axis indicates standardized residuals of rCBF at the peak level. AHI, apnea-hypopnea index; BMI, body mass index; CPAP, continuous positive airway pressure; ESS, Epworth Sleepiness Scale; rCBF, regional cerebral blood flow
Page 24 of 29
Table 1. Comparison of demographics and overnight PSG of patients with OSA before and after CPAP therapy. Before CPAP
Demographic data 46.33±7.96 (31-60)
Mean BMI, kg/m
Mean AHI, per hour
Mean AI, per hour
Total sleep time, min
Age, years 2
Comment [A10]: Please ensure that it is clear what all of these numbers are for eg: mean ± SD (range)
398.9±19.3 (369.5– 433) Sleep efficiency, %
NREM sleep 1, %
NREM sleep 2, %
NREM sleep 3, %
Mean sleep latency, minutes
All values are expressed as mean (SD)
Comment [A11]: As my comment above: is this mean ± SD (range)?
AHI, apnea-hypopnea index; AI, arousal index; BMI, body mass index; CPAP, continuous positive airway pressure; ESS, Epworth Sleepiness Scale; NREM, non-REM; OSA, 25
Page 25 of 29
obstructive sleep apnea; PSG polysomnography; REM, rapid eye movement; SSS, Stanford Sleepiness Scale * Independent t-test, p<0.001.
Page 26 of 29
Table 2. Brain regions showing significant rCBF differences between patients with OSA (pre-treatment and post-treatment) and controls, and between pre-treatment and posttreatment OSA groups.
Comment [A12]: Please ensure that these
Peak level KE
groups are clearly shown in the table
MNI (mm) p
Comment [A13]: Please define in footnotes if necessary
Normal > Pre-treatment Cluster #1 R. Anterior cingulum
R. Cingulate gyrus
R. Medial orbitofrontal gyrus
B. Medial frontal gyri
B. Medial frontal gyrus
R. Anterior cingulum
R. Cingulate gyrus
Cluster #3 L. Angular gyrus Cluster #4 R. Postcentral gyrus Cluster #5 L. Parahippocampal gyrus Cluster #6 R. Cerebellum Normal > Post-treatment Cluster #1
Comment [A14]: If there are no data,
Cluster #2 R. Postcentral gyrus
please do not leave blank, insert an em dash –
Cluster #3 L. Parahippocampal gyrus
Page 27 of 29
Pre-treatment < Post-treatment Cluster #1 R. Anterior cingulum
R. Cingulate gyrus
R. Medial orbitofrontal gyrus
B. Medial frontal gyri
L. Angular gyrus
L. Postcentral gyrus
L. Parahippocampal gyrus
L. Fusiform gyrus
Cluster #4 R. Postcentral gyrus Cluster #5
Cluster #6 R. Cerebellum
Each cluster contains connected voxels, which can have several anatomical labels. The peak MNI coordinate for each cluster is the peak voxel in the cluster.
B, bilateral; L, left; MNI, Montreal Neurologic Institute; OSA, obstructive sleep apnea; R, right; rCBF, regional cerebral blood flow
Extent threshold KE>50
Page 28 of 29
Table 3. Correlation analysis of clinical parameters and rCBF changes after CPAP treatment. MNI (mm) Parameter
Correlation with difference in rCBF (Post CPAP Pre CPAP) ESS difference (+) AHI difference, per hour (+) CPAP duration, hours/night (+)
Middle frontal gyrus
Superior temporal gyrus
Superior temporal gyrus
Correlation with rCBF before CPAP ESS ()
Middle frontal gyrus
SL, minutes ()
Medial frontal gyrus
AHI, per hour ()
Sub-gyral of frontal lobe
Frontal lobe sub-gyral
Inferior frontal gyrus
Correlation with rCBF after CPAP ESS ()
Precentral gyrus Mean CPAP pressure,
Inferior frontal gyrus
Middle frontal gyrus
Note that all reported peak areas are significant at the uncorrected level of p<0.001 (+), positive correlations; (), negative correlations AHI, apnea-hypopnea index; CPAP, continuous positive airway pressure; ESS, Epworth Sleepiness Scale; L, left; MNI, Montreal Neurological Institute; R, right; rCBF, regional cerebral blood flow; SL, sleep latency; SSS, Stanford sleepiness scale
Page 29 of 29