Issue Navigator

Volume 14 No. 06
Earn CME
Accepted Papers

Scientific Investigations

Continuous Positive Airway Pressure Treatment May Improve Optic Nerve Function in Obstructive Sleep Apnea: An Electrophysiological Study

Claudio Liguori, MD1; Fabio Placidi, MD, PhD1; Maria Giuseppina Palmieri, MD, PhD1; Francesca Izzi, MD, PhD1; Raffaella Ludovisi, BSc1; Nicola Biagio Mercuri, MD1,2,3; Mariangela Pierantozzi, MD, PhD2
1Sleep Medicine Centre, Neurophysiopathology Unit, Department of Systems Medicine, University of Rome “Tor Vergata,” Rome, Italy; 2Neurology Unit, Department of Systems Medicine, University of Rome “Tor Vergata,” Rome, Italy; 3Fondazione Santa Lucia IRCCS, Rome, Italy


Study Objectives:

Obstructive sleep apnea (OSA) is a sleep disorder frequently associated with optic nerve diseases. Moreover, untreated patients with severe OSA may show optic nerve dysfunction as documented by electrophysiological studies using visual evoked potentials (VEP). Because continuous positive airway pressure (CPAP) treatment has proved to restore the physiologic nocturnal breathing, thus preventing nocturnal hypoxemia and reducing inflammation, in this study we tested whether 1-year CPAP treatment may modify VEP responses in patients with severe OSA.


VEP were recorded at baseline and after 1 year of CPAP treatment in 20 patients with severe OSA, divided in two groups on the basis of CPAP adherence, and compared to a healthy control group.


Patients with good adherence to CPAP therapy (CPAP+; n = 10) showed VEP P100 amplitude significantly higher than patients with poor adherence to CPAP therapy (CPAP−; n = 10). Moreover, the CPAP+ group showed VEP responses similar to those in the control group (n = 26). Considering the mean difference of VEP responses between baseline and follow-up, the CPAP+ group showed a significant increase in VEP P100 amplitude and a significant decrease in VEP P100 latency compared to the CPAP− group.


This study documented that CPAP therapy significantly improves VEP in patients with OSA who are adherent to the treatment. We hypothesize that CPAP treatment, minimizing the metabolic, inflammatory and ischemic consequences of OSA, may normalize the altered VEP responses in patients with OSA by restoring and preserving optic nerve function.


Liguori C, Placidi F, Palmieri MG, Izzi F, Ludovisi R, Mercuri NB, Pierantozzi M. Continuous positive airway pressure treatment may improve optic nerve function in obstructive sleep apnea: an electrophysiological study. J Clin Sleep Med. 2018;14(6):953–958.


Current Knowledge/Study Rationale: Obstructive sleep apnea (OSA) has been associated with optic nerve disorders. Because visual evoked potentials (VEP) are a sensitive instrument to measure optic nerve function, we evaluated VEP in patients with OSA before and after 1 year of continuous positive airway pressure (CPAP) treatment.

Study Impact: Because we documented a marked improvement of VEP responses in patients with OSA who were adherent to CPAP treatment compared to those who were not adherent, we hypothesize that CPAP may restore optic nerve function in patients with OSA. Thus, it is possible that CPAP treatment for patients with OSA could prevent optic nerve diseases.


Obstructive sleep apnea (OSA) is a sleep disorder with widespread prevalence that has been receiving considerable attention because it represents a risk factor for several health consequences, such as cardiovascular and cerebrovascular diseases, diabetes, vitamin D deficiency, osteoporosis, and cognitive decline.18 Continuous positive airway pressure (CPAP) represents the first-line treatment for moderate to severe OSA and plays a key role in improving patients' symptoms, as well as in reducing vascular and metabolic consequences of OSA.912

Recently, OSA has been associated with the occurrence of ocular diseases. In several clinical and electrophysiological controlled studies, a significant prevalence of glaucoma and nonarteritic anterior ischemic optic neuropathy (NAION) was found in patients with OSA.13,14

Visual evoked potentials (VEP) represent a reliable electrophysiological instrument currently used to investigate the functional integrity of the optic nerve.15 In several neurological disorders, VEP have been widely used to detect possible alterations of the optic pathway and to assess changes of optic nerve function due to pharmacological treatment or disease progression.1618

In a recent VEP study from our group, the occurrence of optic nerve dysfunction was documented in patients with untreated severe OSA and was interpreted as the effect of both intermittent nocturnal hypoxemia and increased systemic inflammation.19

Because CPAP treatment can restore physiologic nocturnal breathing—thus preventing nocturnal hypoxemia and reducing inflammation—we sought to test whether 1 year of CPAP treatment could reverse VEP pathological responses in patients with severe OSA.



In the current study we included patients with severe OSA (respiratory event index [REI] > 30 events/h) that was diagnosed at our Sleep Medicine Center, who performed VEP recording before starting CPAP treatment (baseline) and after 1 year of CPAP treatment. Patients participating in the study met the following inclusion criteria: diagnosis of severe OSA (REI > 30 events/h), as measured by a home sleep apnea test (Embletta; Embla, Amsterdam, The Netherlands) performed according to criteria published by the American Academy of Sleep Medicine20; no comorbid medical disorders; and no concomitant psychiatric or neurological disorders. The exclusion criteria were the following: heavy smoking; bronchial asthma, chronic obstructive pulmonary disorders, and interstitial lung diseases; autoimmune disorders; malignancies; diabetes; and hypertension and/or history of hypertensive crisis. Notably, all patients underwent a dedicated ophthalmological examination in order to exclude clinical or subclinical visual impairment or conditions affecting visual field such as intracranial or ocular mass lesions, uveitis, optic neuropathy, NAION, and optic disk disorders. Moreover, participants were required to have current systolic blood pressure < 140 mmHg, diastolic blood pressure < 85 mmHg, and fasting blood glucose < 100 mg/dL. Finally, patients with systemic disorders with known influence on the function of the retina and optic nerve as well as poor cooperation with the electrophysiological tests were excluded.

After 1 year, CPAP treatment was documented by the ventilator software report. Exclusively patients with OSA with a REI < 5 events/h, confirmed by a home sleep apnea test performed during CPAP treatment, were admitted at the follow-up and VEP recording performed after 1 year of CPAP treatment. Patients were distributed in two groups on the basis of CPAP adherence: patients showing good adherence to CPAP therapy (CPAP+) and patients with poor adherence to CPAP therapy (CPAP−). Good adherence to CPAP treatment was defined according to previously published criteria: CPAP use for ≥ 4 h/ night and > 5 nights/wk.12

We also enrolled a control group of healthy volunteers who were similar in age, sex, and body mass index (BMI) to the patients with OSA. The inclusion criteria for the control group were the absence of sleep disorders, evaluated by means of both a structured interview and polygraphic cardiorespiratory monitoring (REI < 5 events/h), and the absence of any medical disorder.

We also evaluated daytime somnolence using the Epworth Sleepiness Scale (ESS) at baseline and at 1-year follow-up in patients with OSA and the control group.21,22

All participants gave informed consent to the study, which was approved by the Independent Ethical Committee of the University Hospital of Rome “Tor Vergata.”

VEP Recordings

In each VEP recording session, all participants were seated in a semidark, acoustically isolated room. Each subject was adapted to ambient room lights for 10 minutes. Stimulation was monocular. Visual stimuli were checkerboard patterns (contrast 100%) generated on a television monitor and reversed in contrast at the rate of two reversals/s to obtain transient responses. VEP recordings were performed at 120 cm distance with an optoelectronic stimulator Galileo Mizar Sirius (EBNeuro SpA, Florence, Italy), with high (15' checks) and low (55') spatial frequencies. For each eye and each recording session, responses to two blocks of 100 stimuli were recorded, under full refractive correction. VEP were recorded using gold-cup electrodes located at Oz (active electrode), Fpz (reference electrode), and Cz (ground) (10–20 International System). The interelectrode resistance was kept lower than 3 kΩ. Biological signal was amplified (gain 20,000), filtered (high-pass 1 Hz, low-pass 100 Hz), and 100 artifact-free signals were averaged separately for each eye. For VEP responses, the following standard parameters were recorded: N75, P100, and N145. VEP P100 latency, determined from the visual stimulus onset to the maximal component peak, and VEP P100 amplitude, calculated peak-to-peak (P100-to-N145), were considered for the analysis.23,24

To control vigilance fluctuations during VEP sessions, participants' behavior was continuously checked by a co-registered electroencephalogram and by recording continuous reaction times in order to avoid dozing. Finally, to guarantee the reliability of results, the physicians involved in VEP recordings (RL) as well as in data analysis (MGP; MP) were completely blinded to the patients' condition.

Statistical Analysis

A commercial software program (Statistica 10.0; Stat-soft Inc, Tulsa, Oklahoma, United States) was used for the statistical analysis.

The Kolmogorov-Smirnov test was used to control the normal distribution of the obtained data (Table S1 in the supplemental material). We used analysis of variance to compare demographic, clinical and VEP data among CPAP+, CPAP− and controls. The post hoc analysis was performed using the Tukey test for honestly significant difference. A value of P < .05 was considered to be statistically significant. The t test was used to compare clinical data between baseline and after 1 year of CPAP therapy. Moreover, the mean difference (Δ) of REI and VEP responses between baseline and 1 year of CPAP treatment was assessed for CPAP+ and CPAP− groups. The t test was used to compare Δ of VEP responses between CPAP+ and CPAP− groups. Finally, a multiple regression analysis was also performed including VEP data and BMI.


Demographic, Clinical, and Polygraphic Data of Patients and Controls

Twenty patients with severe OSA met the eligible criteria and thus were included in the study. According to our definition of adherence to CPAP therapy, 10 patients were included in the CPAP+ group, and the remaining 10 patients constituted the CPAP− group. The healthy control population consisted of 26 subjects who were similar in age, sex, and BMI to the patients with OSA.

We documented a significant reduction of ESS scores between baseline and 1-year follow-up in both CPAP+ (12.5 ± 3.88 versus 8.4 ± 3.77, P < .01) and CPAP− groups (11.3 ± 7.49 versus 9.7 ± 5.25, P < .01). We did not find changes in BMI at 1-year follow-up in either the CPAP+ (27.8 ± 2.39 versus 27.7 ± 1.97 kg/m2) or CPAP− groups (29.39 ± 3.91 versus 28.74 ± 3.37 kg/m2).

Demographic, clinical, and polygraphic features of the CPAP+, CPAP−, and control groups are summarized in Table 1.

Clinical and polygraphic data.


table icon
Table 1

Clinical and polygraphic data.

(more ...)

VEP Data

VEP data were similar in the CPAP+ and CPAP− groups at baseline (Table 2). The statistical analysis comparing VEP data after 1 year of CPAP treatment and recorded at low and high spatial frequencies in both eyes among the three studied groups (CPAP+ versus CPAP− versus control) documented no significant difference in VEP P100 amplitude between CPAP+ and control groups. Conversely, VEP P100 amplitude in the CPAP− group was significantly lower in comparison with both CPAP+ and control groups (see Table 3). The same statistical analysis showed that VEP P100 latency measured after 1 year of CPAP treatment was significantly longer in both CPAP+ and CPAP− groups compared to the control group. Moreover, VEP P100 latency was not different between CPAP+ and CPAP− groups after 1 year of CPAP treatment (see Table 3). However, the comparison of the Δ of VEP responses between CPAP+ and CPAP− groups documented the significant decrease of Δ of VEP P100 latency in CPAP+ versus CPAP− groups (see Table 4). Moreover, the CPAP+ group showed a significant increase of Δ of VEP P100 amplitude compared to the CPAP− group.

VEP data recorded at baseline.


table icon
Table 2

VEP data recorded at baseline.

(more ...)

VEP data after 1 year of CPAP treatment.


table icon
Table 3

VEP data after 1 year of CPAP treatment.

(more ...)

Δ of VEP responses.


table icon
Table 4

Δ of VEP responses.

(more ...)

We did not find an association between BMI and VEP data using multiple regression analysis.


It has been documented recently that patients with severe OSA show optic nerve dysfunction as documented by the significant pathological changes of VEP responses, namely the amplitude reduction and latency delay of VEP P100 component.19 These VEP abnormalities have been interpreted as the electrophysiological expression of metabolic and neurological consequences of OSA, possibly causing both axonal and demyelinating damage of the optic nerve.

OSA is currently managed with CPAP, which is largely considered the gold-standard therapy for this sleep disorder.9 The efficacy of CPAP is widely known in patients with OSA, because CPAP treatment is able to restore sleep and to resolve nighttime intermittent hypoxemia, thus ensuring the physiological sleep processes and the optimal nocturnal oxygen saturation.

For these reasons, the main purpose of the current study was to assess whether CPAP treatment, counteracting the pathological processes because of the OSA condition, may influence optic nerve function.

The findings of the current study were that VEP responses may be significantly influenced by 1 year of optimal CPAP treatment. In particular, good adherence to CPAP treatment may improve VEP P100 latency and normalize VEP P100 amplitude in patients with OSA. We found that VEP P100 amplitude in the CPAP+ group was significantly higher than the CPAP− group after 1 year of CPAP treatment. Notably, in the CPAP+ group, VEP P100 amplitude reached the values of the healthy controls. We also documented the significant reduction of VEP P100 latency between baseline and follow-up (performed after 1 year of CPAP therapy) in the CPAP+ group compared to the CPAP− group. The positive effect of CPAP treatment on VEP was further confirmed by the fact that patients in the CPAP− group continued to show pathological VEP P100 amplitude and latency compared to controls after 1 year of CPAP treatment. Therefore, we can assume that the efficacy of CPAP (lowering REI to < 5 events/h) and also the good adherence to the treatment improve VEP parameters in patients with OSA. However, VEP P100 latency in the CPAP+ group remained longer than in the control group, even after 1 year of optimal CPAP treatment.

VEP represent a highly sensitive and reliable electrophysiological tool providing a powerful indication of abnormal signal abnormalities within the visual pathway.25,26 Hence, VEP recordings have been mostly employed to identify and quantify optic nerve dysfunction in several neurological and ocular disorders, considering that delayed VEP P100 latency suggests optic nerve demyelinization, whereas reduced VEP P100 amplitude is expression of optic nerve axonal damage.1519,27,28 In this context, VEP are useful to evaluate the severity of optic neuritis in multiple sclerosis. In patients with multiple sclerosis, VEP P100 latency is delayed as the effect of the underlying demyelinating process, but pharmacological treatments may reduce VEP P100 latency because of the amelioration of optic nerve demyelination.1618 In NAION, as well as in other optic nerve disorders, VEP may show latency delay and amplitude reduction as the result of both optic nerve demyelinization and axonal damage due to hypoxic-ischemic insults. In this view, it has been shown that VEP P100 amplitude gradually improves in patients with NAION or glaucoma as the effect of treatments.27,28

Current hypotheses highlight that OSA has multisystemic consequences, such as the metabolic, neurocognitive, inflammatory-ischemic insults, which are due to the detrimental effects of nocturnal intermittent hypoxemia. Accordingly, intermittent hypoxemia may also affect the optic nerve by altering its axonal and myelinic components, because of the recurrent microischemic insults. In agreement with this evidence, patients with OSA may show the impairment of VEP P100 amplitude and latency.29,30 Moreover, OSA is associated with chronic inflammation, possibly producing the delay of VEP P100 latency by means of optic nerve demyelination.31

Therefore, we suppose that CPAP therapy may improve VEP responses by reversing the ischemic optic axonal damage and reducing the inflammatory processes. However, our data seem to suggest that the improvement of VEP P100 latency is less impressive if compared to VEP P100 amplitude increase. On one hand, it is possible to explain the improvement of VEP P100 amplitude considering that CPAP therapy permits the recovery of nocturnal oxygen saturation preventing the occurrence of intermittent hypoxemia. On the other hand, the partial improvement of VEP P100 latency may be related to the lower efficacy of CPAP treatment on the inflammatory mediators.32 It has been reported that CPAP therapy alone may not completely counteract all the negative consequences induced by OSA. In particular, inflammation is not completely reversed by CPAP therapy and needs to be addressed by other treatments, including weight loss or lifestyle changes.32 In agreement with this observation, we did not see significant BMI changes in our patient population. Mean BMI was lower, but not statistically significant, in the CPAP+ compared to CPAP− group. Therefore, the lesser effect of CPAP treatment on VEP P100 latency delay may be explained due to the persistence of a mild chronic inflammation related to obesity. However, we did not find an association between BMI and VEP data in the CPAP+ and CPAP− groups.

This study presents some limitations. First, the diagnosis of OSA was performed by using a home sleep apnea test, which is more prone to artifact. The population of patients included was also limited by the inclusion and exclusion criteria. We did not include a group of patients with OSA who were adherent to CPAP treatment and showed weight loss. This would have allowed us to test our hypothesis that both OSA and obesity may alter VEP responses. Finally, patients were not randomized and unmeasured confounders were not excluded by our analysis.

In conclusion, OSA has been associated recently with several optic nerve disorders, such as glaucoma, NAION, papilledema, and optic disc derangement.13 Considering the detrimental effects of OSA on the optic nerve, the current electrophysiological VEP study demonstrated that CPAP therapy may restore optic nerve function earlier than the occurrence of visual symptoms. Therefore, we reaffirm the importance of adherence to CPAP therapy in patients with OSA. This treatment, aside from its other benefits, may preserve and recover the optic nerve from hypoxic and inflammatory insults, thus restoring subclinical optic nerve damage.


All authors have seen and approve the manuscript. The authors report no conflicts of interest.



mean difference


continuous positive airway pressure


Epworth Sleepiness Scale


nonarteritic ischemic optic neuropathy


obstructive sleep apnea


respiratory event index


visual evoked potentials


Authors contributions: Claudio Liguori: study concept, acquisition of data, data analysis and interpretation, drafting the manuscript; Raffaella Ludovisi: acquisition of data; Maria Giuseppina Palmieri: acquisition of data, data analysis; Francesca Izzi: critical revision of the manuscript for important intellectual content; Fabio Placidi: study supervision; Nicola Biagio Mercuri: study supervision, critical revision of the manuscript for important intellectual content; Mariangela Pierantozzi: study concept and supervision, data analysis and interpretation, drafting the manuscript.



Peppard PE, Young T, Palta M, Skatrud J. Prospective study of the association between sleep-disordered breathing and hypertension. N Engl J Med. 2000;342(19):1378–1384. [PubMed]


Peker Y, Carlson J, Hedner J. Increased incidence of coronary artery disease in sleep apnoea: a long-term follow-up. Eur Respir J. 2006;28(3):596–602. [PubMed]


Placidi F, Diomedi M, Cupini LM, Bernardi G, Silvestrini M. Impairment of daytime cerebrovascular reactivity in patients with obstructive sleep apnoea syndrome. J Sleep Res. 1998;7(4):288–292. [PubMed]


Silvestrini M, Rizzato B, Placidi F, Baruffaldi R, Bianconi A, Diomedi M. Carotid artery wall thickness in patients with obstructive sleep apnea syndrome. Stroke. 2002;33(7):1782–1785. [PubMed]


Kent BD, McNicholas WT, Ryan S. Insulin resistance, glucose intolerance and diabetes mellitus in obstructive sleep apnoea. J Thorac Dis. 2015;7(8):1343–1357. [PubMed Central][PubMed]


Liguori C, Romigi A, Izzi F, et al. Continuous positive airway pressure treatment increases serum vitamin D levels in male patients with obstructive sleep apnea. J Clin Sleep Med. 2015;11(6):603–607. [PubMed Central][PubMed]


Liguori C, Mercuri NB, Izzi F, et al. Obstructive sleep apnoea as a risk factor for osteopenia and osteoporosis in the male population. Eur Respir J. 2016;47(3):987–990. [PubMed]


Liguori C, Mercuri NB, Izzi F, et al. Obstructive sleep apnea is associated with early but possibly modifiable Alzheimer's disease biomarkers changes. Sleep. 2017;40(5). [PubMed]


Sullivan CE, Issa FG, Berthon-Jones M, Eves L. Reversal of obstructive sleep apnoea by continuous positive airway pressure applied through the nares. Lancet. 1981;1(8225):862–865. [PubMed]


Weaver TE, Maislin G, Dinges DF, et al. Relationship between hours of CPAP use and achieving normal levels of sleepiness and daily functioning. Sleep. 2007;30(6):711–719. [PubMed Central][PubMed]


Barbé F, Durán-Cantolla J, Sánchez-de-la-Torre M, et al. Effect of continuous positive airway pressure on the incidence of hypertension and cardiovascular events in nonsleepy patients with obstructive sleep apnea: a randomized controlled trial. JAMA. 2012;307(20):2161–2168. [PubMed]


Liguori C, Izzi F, Mercuri NB, et al. Vitamin D status of male OSAS patients improved after long-term CPAP treatment mainly in obese subjects. Sleep Med. 2017;29:81–85. [PubMed]


Fraser CL. Obstructive sleep apnea and optic neuropathy: is there a link? Curr Neurol Neurosci Rep. 2014;14(8):465. [PubMed]


Moghimi S, Ahmadraji A, Sotoodeh H, et al. Retinal nerve fiber thickness is reduced in sleep apnea syndrome. Sleep Med. 2013;14(1):53–57. [PubMed]


Martínez-Lapiscina EH, Sanchez-Dalmau B, Fraga-Pumar E, et al. The visual pathway as a model to understand brain damage in multiple sclerosis. Mult Scler. 2014;20(13):1678–1685. [PubMed]


Niklas A, Sebraoui H, Hess E, Wagner A, Then Bergh F. Outcome measures for trials of remyelinating agents in multiple sclerosis: retrospective longitudinal analysis of visual evoked potential latency. Mult Scler. 2009;15(1):68–74. [PubMed]


Liscić RM, Brecelj J. Visual evoked potentials in multiple sclerosis patients treated with interferon beta-1a. Croat Med J. 2004;45(3):323–327. [PubMed]


Anderson DC, Slater GE, Sherman R, Ettinger MG. Evoked potentials to test a treatment of chronic multiple sclerosis. Arch Neurol. 1987;44(12):1232–1236. [PubMed]


Liguori C, Palmieri MG, Pierantozzi M, et al. Optic nerve dysfunction in obstructive sleep apnea: an electrophysiological study. Sleep. 2016;39(1):19–23. [PubMed Central][PubMed]


Berry RB, Budhiraja R, Gottlieb DJ, et al. Rules for scoring respiratory events in sleep: update of the 2007 AASM Manual for the Scoring of Sleep and Associated Events. Deliberations of the Sleep Apnea Definitions Task Force of the American Academy of Sleep Medicine. J Clin Sleep Med. 2012;8(5):597–619. [PubMed Central][PubMed]


Johns MW. A new method for measuring daytime sleepiness: the Epworth sleepiness scale. Sleep. 1991;14(6):540–545. [PubMed]


Vignatelli L, Plazzi G, Barbato A, et al. Italian version of the Epworth sleepiness scale: external validity. Neurol Sci. 2003;23(6):295–300. [PubMed]


Parisi V, Pierelli F, Malandrini A, et al. Visual electrophysiological responses in subjects with cerebral autosomal arteriopathy with subcortical infarcts and leukoencephalopathy (CADASIL). Clin Neurophysiol. 2000;111(9):1582–1588. [PubMed]


Parisi V, Gallinaro G, Ziccardi L, Coppola G. Electrophysiological assessment of visual function in patients with non-arteritic ischaemic optic neuropathy. Eur J Neurol. 2008;15(8):839–845. [PubMed]


Walsh P, Kane N, Butler S. The clinical role of evoked potentials. J Neurol Neurosurg Psychiatry. 2005;76 Suppl 2:ii16–ii22. [PubMed Central][PubMed]


Holder GE. Electrophysiological assessment of optic nerve disease. Eye. 2004;18(11):1133–1143. [PubMed]


Parisi V, Centofanti M, Ziccardi L, et al. Treatment with citicoline eye drops enhances retinal function and neural conduction along the visual pathways in open angle glaucoma. Graefes Arch Clin Exp Ophthalmol. 2015;253(8):1327–1340. [PubMed]


Di Censo F, Di Censo M, Colasante G, Bordin M, Al Oum M. Bilateral nonarteritic ischemic optic neuropathy treated with HBO2 therapy: a case report of angiographic and electrodiagnostic findings. Undersea Hyperb Med. 2016;43(2):167–173. [PubMed]


Kalesnykas G, Tuulos T, Uusitalo H, Jolkkonen J. Neurodegeneration and cellular stress in the retina and optic nerve in rat cerebral ischemia and hypoperfusion models. Neuroscience. 2008;155(3):937–947. [PubMed]


LoPachin RM Jr, Stys PK. Elemental composition and water content of rat optic nerve myelinated axons and glial cells: effects of in vitro anoxia and reoxygenation. J Neurosci. 1995;15(10):6735–6746. [PubMed]


Chen HL, Lin HC, Lu CH, et al. Systemic inflammation and alterations to cerebral blood flow in obstructive sleep apnea. J Sleep Res. 2017;26(6):789–798. [PubMed]


Chirinos JA, Gurubhagavatula I, Teff K, et al. CPAP, weight loss, or both for obstructive sleep apnea. N Engl J Med. 2014;370(24):2265–2275. [PubMed Central][PubMed]