Recurrent episodes of apnea or hypopnea due to partial or complete upper airway obstruction during sleep are associated with hypercapnia and hypoxia, leading to impaired sympathetic activity, irregulation of cerebral and ocular blood flow, and disruption of the balance between vasomeditors such as nitric oxide, vascular endothelial growth factor, and endothelium [17-20]. In addition, it has been suggested that the disturbance of the balance between these vasomediators causes an increase in intracranial pressure, which contributes to a decrease in cerebral perfusion and impaired ocular blood flow [21-24].
It has been shown that the neurosensory retina, especially the retinal ganglion cells forming the optic nerve head, is hypersensitive to hypoxia and reduced perfusion [25]. Hypoxia and hypoperfusion have been important triggering factors for inflammation in the nuclear cell membrane and an increase in oxidative stress, leading to impairment of the cell membrane and cellular edema and subsequently resulting in apoptosis seen in the latter stages of hypoxic diseases [26,27]. These alterations in the brain and the orbita lead to papilledema that reflects the increase in RNFL thickness at early stages of the disease which proceeds to neuronal degeneration and apoptosis that later presents as thinning of RNFL in OCT measurements [27,28]. Therefore RNFL thinning has been suggested as the resulting neurodegenerative effect of OSAS in ocular structures [9, 29, 30].
Increasing evidence has shown that the longer the duration of oxygen desaturation, the more severe the disease on the optic nerve, thus the thinner the RNFL become [13,22]. Based on this fact, the current study aimed to establish the effects of PAP treatment on RNFL and macula of patients with severe OSAS by using OCT to understand if the effects of OSAS related hypercapnia and hypoxia on these parameters were reversible by PAP treatment or not. Following a 3-months PAP treatment, thickening of macula as well as thinning of RNFL in the superior nasal sector was found in comparison with OCT measurements taken before treatment. While it was thought that hypoxia-induced inflammation may regress with PAP treatment and a significant improvement in macular thickness can be observed, changes in the upper nasal sectors of the RNFL appear to be irreversible even with an intense PAP given for 3 months.
On the other hand, the effects of both local inflammatory reactions and increased intracranial pressure may have affected RNFL, which initially appeared as thickening in some sectors in the early stages. (9) In connection with this fact, increased intracranial pressure and papilledema in patients with OSAS have been reported to improve after PAP treatment, which may imply that oxygenation therapy through PAP treatment may be effective in the early stages of the disease to reverse the edematous effects of OSAS [31,32]. Accordingly, thinning in the superior nasal sector of RNFL may have occured due to reversible edematous changes at early stages of the disease that recovered after 3-months of PAP treatment. However, the thinning in the superior nasal sector of the RNFL may have also occured due to neurodegeneration developed after irreversible retinal ganglion cells’ axonal loss caused by the acute onset of inflammation, intracranial pressure and hypoxia as retinal ganglion cells are sensitive to even mild hypoxemia [25-28]. Consistent with these findings, O'Donoghue et al showed that after 6 months of PAP treatment, 23 patients with severe OSAS showed a significant reduction in whole brain volume of approximately 4% in high-resolution magnetic resonance scanning brain imaging [32].
It was noteworthy that in this study, thinning seen only in the upper nasal sector of RNFL was not seen in other sectors. This can be attributed to the difference between RNFL sectors in terms of vascularization; This suggests that the duration of the resulting neurodegenerative effect of OSAS on RNFL sectors may also differ. It has been shown that the temporal RNFL has been shown to be more vascularized than the nasal RNFL in young and healthy individuals [33,34]. Therefore, temporal RNFL is thought to be less susceptible to hypoxemic stress as well as to intracranial hypertension caused by hypoxic and hypercapnic episodes of OSAS. Indeed, in a meta-analysis including 10 studies that analyzes a large sample of OSAS patients with healthy controls, it was found that RNFL in OSAS patients had thinning mostly in the mean, upper, nasal and lower sectors [13].
Glaucomatous optic neuropathy is characterized by axonal loss of ganglion cells and visual field defects over the years, which can first be noticed by the thinning of RNFL measured by OCT. (35) Thinning of RNFL in OSAS patients has been associated with a high prevalence of primary open angle glaucoma among OSAS patients in some studies [29,36]. Similar to the pathogenesis of OSAS, the impaired balance between vasodilators such as nitric oxide and vasoconstrictors such as endothelin is considered one of the causes of glaucoma development [37]. Therefore, it is suggested that this mechanism is the cause of glaucoma-like damage resulting in damage to retinal ganglion cells in patients with OSAS. Based on these facts, the present study also investigated the effect of PAP treatment on IOP in patients with severe OSAS, where it was shown that PAP treatment caused a statistically significant IOP reduction in both eyes. Similar to the current study, Casas et al included patients with OSAS without any glaucomatous evidence, such as IOP values >21 mmHg, abnormal gonisoscopy, and abnormal perimetric findings, in order to evaluate the thinning of RNFL seen in OSAS patients [9]. They showed significantly higher IOP values and changes in the visual field index of OSAS patients compared to controls. They also found an increase in some optic nerve head parameters in OSAS patients without glaucoma compared to healthy controls, which were considered signs of neuronal degeneration. As a matter of fact, a positive correlation between IOP and AHI has been shown among OSAS patients in previous studies [38, 39].
There are only a few studies evaluating IOP, macula, and RNFL of patients with severe OSAS before and after PAP treatment [40,41,42]. Batum et al. investigated the effects of 6-month PAP treatment on patients with severe OSAS [40]. Similarly, the authors reported a significantly thicker macula in the foveal area after treatment compared to OCT measurements at diagnosis. They stated that other sectors of the macula were also measured thicker after PAP treatment, but these differences were not found to be statistically significant. Similar to the pathophysiology seen in RNFL, the authors attributed these changes in macular thickness to the effects of PAP treatment, which increased patients' oxygen saturation to counteract the inflammatory effects associated with hypoxia. However, in contrast to the current study, Batum et al found a significant increase in the average, nasal, and inferior sectors of RNFL thickness after 6 months of PAP treatment compared to thickness measurements at the time of diagnosis. In addition, the authors stated that PAP was effective in preventing hypoxia in retinal tissues, although it did not cause any change in intraocular pressure in patients with severe OSAS. Supporting this finding, the authors reported significant changes in visual evoked potential testing in patients with severe OSAS after PAP treatment, indicating that the axonal and myelin component of the optic nerve was damaged by recurrent microischemia due to intermittent hypoxemia [30,43].
In contrast to the current study and the study of Batum et al., Yuvacı et al. reported significantly thinner macula in the superior, superior nasal, inferior nasal and temporal superior sectors measured after 3 months of PAP treatment compared to the measurements at the time of diagnosis [41]. Differently, they evaluated only the right eyes of patients with severe OSAS. The authors also found that after 3 months of PAP treatment, RNFL thickness was significantly reduced in the average, inferior, superior nasal, and superior temporal quadrants, but noted that clinical improvement was observed in patients with severe OSAS. The researchers also noted that the average RNFL thickness decreased after both 4 weeks and 12 weeks of PAP treatment. Also, unlike the current study, Yuvacı et al. showed a statistically significant increase in mean IOP that did not reach a normal range after PAP treatment. Nevertheless, RNFL thinning was attributed not only to glaucoma, but also to apnea-hypopnea induced hypoperfusion and ischemia, vasomediator imbalance such as an imbalance between NO and endothelin, and nocturnal hypotension [41].
Zengin et al. demonstrated changes in RNFL thickness after one year of follow-up in mild, moderate, and severe OSAS patients [42]. Similar to the study of Yuvacı et al., after 12 months of PAP treatment, they showed thinning of the RNFL in the measurements of the average, inferior nasal, superior nasal, and superior temporal sectors. They also observed a progressive reduction in RNFL thickness during follow-up, which was shown to be statistically insignificant compared to healthy controls. The study of Zengin et al., unlike the current study and the studies mentioned above, consisted of 44 OSAS patients, including only 14 severe OSAS patients. In addition, Zengin et al. showed a statistically significant increase in the mean IOP of OSAS patients during the course of the disease, in contrast to the decrease in IOP levels of OSAS patients after 3 months of PAP treatment. The authors concluded that the thinning observed in RNFL could be caused by both glaucoma and OSAS [42].
Indeed, a relatively high incidence of glaucoma in OSAS patients has been documented in the literature [36]. However, studies investigating the relationship between glaucoma and OSAS still reveal conflicting results. While some studies associated normotensive glaucoma or primary open angle glaucoma with OSAS [29,36, 38,39,44,45]; some studies did not find any correlation between OSAS and glaucoma, or some studies found similar prevalence of glaucoma in OSAS patients compared to normal subjects [22,46, 47, 48,49]. In addition, there are reports in the literature documenting improvement of visual defects in OSAS patients following PAP treatment [50,51]. Further, the thinner RNFL in patients with OSAS without optic neuropathy symptoms or visual field defects suggests that OSAS may have an effect on the optic nerve for other reasons, but not glaucoma [9,30]. The diversity of all these results and interpretations in the literature can be attributed to patients who differ in terms of OSAS severity and duration, as well as study design, sample size and demographic characteristics. Therefore, we believe that further studies with large homogeneous groups are needed to establish the relationship between intraocular pressure and OSAS and to distinguish RNFL changes from glaucoma changes.
The greatest strength of this study is that all patients with OSAS are free from other known diseases and have never received sleep apnea treatment before. On the other hand, the most important limiting factor of the study was the relatively small sample size and the lack of long-term follow-up (longitudinal design) of patients with severe OSAS. In addition, since a vascular pathogenesis or change has been mentioned in the pathogenesis of the neurodegenerative effects of OSAS, the fact that brain and ocular vascularity have not been measured by imaging and measurement techniques such as OCTA or MRI can be considered as a deficiency.
Our study is important because it is one of the few studies evaluating the effectiveness of PAP treatment on IOP, macula, and RNFL in patients with severe OSAS. The effects of OSAS on IOP and macula have been shown to be reversible with a 3-month PAP treatment, but the reversibility of the neurodegenerative effects of OSAS on RNFL with this treatment still seems controversial. In summary, the findings of this study suggest that OCT measurements will be candidates for use as a biomarker to evaluate disease progression in patients with severe OSAS under PAP treatment. However, as this issue has not been investigated in depth in the literature, future studies with a longitudinal design and larger sample size are needed to distinguish the effects of OSAS and PAP treatment on IOP, macula, and RNFL.