Materials and Methods
The study design was cross-sectional. The period for patient inclusion was September 2008 to December 2011. We included 152 patients who had been admitted for OSA evaluation at the Arnau de Vilanova University Hospital and Santa Maria Hospital, Lleida, Spain and who agreed to undergo an ophthalmologic examination to diagnose glaucoma. Seventy-five patients had OSA without FES; 52 patients had OSA and FES; and 25 patients did not have OSA. Two of the non-OSA patients had FES. All the patients were subjected to a complete ophthalmologic examination to diagnose glaucoma. They were examined by a glaucoma expert (MaJ.M.R.), who was blinded to the results of the sleep study, according to standard diagnostic criteria.
None of the patients included had undergone any form of surgical intervention to the eyelid or had suffered from any type of ocular trauma or abnormality that could have affected eyelid function, nor did any of them have a history of ocular surgery in the previous 12 months. Patients with known neurological or psychiatric disorders were not included in the study. The protocol and informed consent were both approved by the Ethics Committee of our hospital, and informed consent was obtained in all cases.
Sleep Studies
Diagnosis of OSA was made on the basis of either conventional polysomnography or a cardiorespiratory sleep study. All the sleep studies were individually analyzed at each participating center by applying standard criteria. The polysomnographies included the continuous recording of neurological variables: electroencephalogram (C3/A2 and C4/A1), electrooculogram, and electromyogram. Breathing variables were scored according to a flow tracing provided by a nasal cannula and thermistor. Thoracoabdominal motion was measured with thoracic and abdominal bands. Oxygen saturation was recorded with a finger-pulse oximeter. The cardiorespiratory sleep study included (as a minimum): a continuous recording from the nasal cannula, thoracoabdominal motion, oxygen saturation, and body position. Apnea was defined as an absence of airflow for at least 10 seconds, and hypopnea was defined as a clear (50%) airflow reduction for at least 10 seconds, with a drop in oxygen saturation of at least 4% on arousal. OSA was defined as the absence of airflow in the presence of chest or abdominal wall motion. The apnea-hypopnoea index (AHI) was calculated according to the average number of episodes of apnea plus hypopnea per hour of sleep or recording time. Sleep stages were scored as: normal, AHI < 10; mild OSA, 10 ≥ AHI < 20; moderate OSA, 20 ≥ AHI < 30; and severe OSA, AHI ≥ 30.
Ophthalmic Examination
Eyelid examination was specifically conducted to evaluate eyelid laxity and to obtain an FES diagnosis. FES was defined as easily evertible lids and the presence of papillary conjunctivitis in the same upper eyelid, which is the clinical definition of FES. Easy lid eversion was characterized by increased laxity in the upper lids that became easily distorted and everted with only minimal superolateral traction.
The eye examination included: best-corrected visual acuity with a recording of the refractive correction; slit-lamp biomicroscopy of the anterior segment; Goldmann applanation tonometry performed with the same tonometer; ultrasonic corneal pachymetry; gonioscopy; and visual field analysis with the Humphrey Field Analyzer (SITA-standard program, central 24-2 threshold test). Visual field tests were repeated 1 week later if significant fixation losses and/or false positives or negatives were detected to be >15%. Patients with OSA also often tend to exhibit losses of attention during these tests. We also repeated the visual field test 4 months later in the case of patients suspected of suffering from glaucoma. After pupil dilatation, the morphology of the optic disc was assessed by stereoscopic slit-lamp biomicroscopy, using a fundus 3-mirror lens. The thickness of the retinal nerve fiber layer (RNFL) was assessed using optical coherence tomography (Stratus OCTm Carl Zeiss Ophthalmic Systems Inc.).
The following criteria were used to define glaucoma in our study, irrespective of the intraocular pressure (IOP): a normal-appearing anterior chamber angle on gonioscopy; glaucomatous optic disc damage with asymmetric cupping and thinning of the neuroretinal rim and/or optic disc hemorrhage and/or defects in RNFL thickness; glaucomatous visual field defects (Bjerrum and/or paracentral scotoma, nasal step, and altitudinal defects); and the progression of optic nerve damage and visual field defects. For the diagnosis of primary open-angle glaucoma (POAG), untreated IOPs had to remain at above 21 mm Hg and below 22 mm Hg for the diagnosis of NTG. If the patients had been diagnosed with glaucoma before the present study, we considered that they had previously diagnosed glaucoma (PDG).
Statistical Analysis
All the data were expressed as mean±SD. Statistical analyses of the demographic data, polysomnographic recordings, and ophthalmologic examinations were performed using the unpaired Student t test. The prevalence of glaucoma was calculated from the proportion of patients with evidence of glaucoma. We used the Mann-Whitney test to analyze data between groups. Right and left eyes were analyzed separately. The 95% confidence interval was used to compare the prevalence of glaucoma in our OSA patients with the prevalence of glaucoma in OSA reported in previous publications. Spearman Rank correlation was also used whenever appropriate. Each correlation was controlled for age and body mass index (BMI). P < 0.05 was considered statistically significant. The language and environment for the R Foundation for Statistical Computing (Vienna, Austria) was also used.