Bahr et al. Respiratory Research          (2020) 21:258 
https://doi.org/10.1186/s12931-020-01533-7
RESEARCH Open Access
Obstructive sleep apnea as a risk factor for
primary open angle glaucoma and ocular
hypertension in a monocentric pilot study
Katharina Bahr1* , Michael Bopp2, Waeel Kewader1, Henri Dootz1, Julia Döge1, Tilman Huppertz1, Perikles Simon3,
Verena Prokosch-Willing2, Christoph Matthias1 and Haralampos Gouveris1
Abstract
Background: Both glaucoma and obstructive sleep apnea (OSA) are widespread diseases. OSA may presumably
partly cause or worsen glaucoma, although the etiopathogenesis is unclear. Here we analyze for the first time the
possible association between different glaucoma phenotypes and OSA.
Methods: 110 patients (47 females, 63 males; median age 64.3 years, median BMI 26.62 kg/m2) with suspected
glaucoma and without any prior diagnosis of OSA were prospectively studied by one-night home sleep apnea
testing (HSAT), 101 of the patients were analyzed. HSAT parameters, like apnea hypopnea index (AHI) and oxygen
desaturation index as well as opthalmological parameters like intraocular pressure (IOP) and mean defect depth
(MD) were collected. Moreover, HSAT results were compared across four phenotypic groups: primary open angle
glaucoma (POAG), low-tension-glaucoma (LTG), ocular hypertension (OH), and controls.
Results: There was no strong correlation between IOP or MD and AHI. BMI, age and gender did not differ between
groups. Significant differences between POAG and LTG were found for all HSAT parameters. The AHI showed the
most prominent group difference (Wilcoxon-Kruskal-Wallis rank sum test was highly significant with chi2 = 22, df = 3
p < 0.0001) with severely lower event rates in the LTG (9.45/h) compared to POAG (22.7/h) and controls (21.9/h;
p < 0.0001 and 0.02, respectively). Highly significant differences were found between the four groups regarding
AHI (Chi2 = 22, df = 3, p < 0.0001) with significantly lower events per hour in the LTG compared to POAG (Hodges-
Lehmann = − 13.8, 95% CI (− 18.6 – − 8.8; p < 0.0001) and to controls (Hodges-Lehmann = 12.1, 95% CI -19.9 – − 2.4; p <
0.02). Severe and moderate OSA was more prevalent in POAG (69.8%) and OH (33.3%) than in LTG (9%). The effect of the
glaucoma phenotype on the AHI was more prominent in females (p = 0.0006) than in males (p = 0.011).
Conclusion: Although physical endpoints, such as MD and IOP, do not correlate with AHI, there was a strong correlation
between the POAG and OH clinical glaucoma phenotypes and the AHI. Further studies should investigate the necessity
to test routine screening for OSA by HSAT in patients with diagnosed POAG and OH. Besides, some characteristics of LTG
differed widely from other glaucoma types and controls. LTG patients had a significantly lower rate of OSA compared to
other glaucoma types and even controls. This might be due to a different pathogenesis of LTG.
(Continued on next page)
* Correspondence: katharina.bahr@unimedizin-mainz.de
1Sleep Medicine Center, Department of Otorhinolaryngology, Head and Neck
Surgery, Medical Center of the University of Mainz, 55131 Mainz, Germany
Full list of author information is available at the end of the article
© The Author(s). 2020 Open Access This article is licensed under a Creative Commons Attribution 4.0 International License,
which permits use, sharing, adaptation, distribution and reproduction in any medium or format, as long as you give
appropriate credit to the original author(s) and the source, provide a link to the Creative Commons licence, and indicate if
changes were made. The images or other third party material in this article are included in the article's Creative Commons
licence, unless indicated otherwise in a credit line to the material. If material is not included in the article's Creative Commons
licence and your intended use is not permitted by statutory regulation or exceeds the permitted use, you will need to obtain
permission directly from the copyright holder. To view a copy of this licence, visit http://creativecommons.org/licenses/by/4.0/.
The Creative Commons Public Domain Dedication waiver (http://creativecommons.org/publicdomain/zero/1.0/) applies to the
data made available in this article, unless otherwise stated in a credit line to the data.
Bahr et al. Respiratory Research          (2020) 21:258 Page 2 of 8
(Continued from previous page)
Trial registration: Retrospectively registered at DRKS (nr. S00021201) on April 9th 2020.
Keywords: Obstructive sleep apnea, Sleep-disordered breathing, Glaucoma, Primary open angle glaucoma
Introduction/background [12]. Intermittent hypoxia caused by OSA leads to optic
Obstructive sleep apnea (OSA) is a sleep disorder char- nerve damage due to oxygen undersupply. According to
acterized by pauses in breathing or instances of shallow the authors [12], vascular pathological changes that are
breathing during sleep (international classification of dis- demonstrably caused by OSA also lead to optic nerve
eases –ICD- 10 code: G47.31). Glaucoma is a group of neurodegeneration due to nerve hypoperfusion. A fur-
diseases characterized by damage to the optic nerve and ther mechanistic connection is caused by an increased
retinal nerve fibers (ICD-10 code: H40-H42). intraocular pressure due to the mechanical breathing in-
Both glaucoma and OSA are widespread diseases. Due stability associated with OSA. Another hypothesis link-
to its current prevalence of 23% in women and 50% in ing OSA and glaucoma involves increased intermittent
men, public awareness of OSA steadily increases [1]. sympathetic tone during sleep which is due to the re-
The disease is characterized by breathing interruptions, spiratory events and associated arousals, that leads to an
due to a narrowing of the upper respiratory tract causing increased release of vasoconstrictive substances (cate-
reduced peripheral oxygen supply during sleep. Termin- cholamines, angiotensin II, vasopressin) and atrial natri-
ation of these events requires arousal from sleep in order uretic peptide (ANP) [13]. However, the extent of the
to re-establish the upper airway tone and resume ventila- role played by OSA in the development of glaucoma is
tion, resulting in sleep fragmentation and thus poor sleep still rather unclear.
quality [2]. So far, a number of risk factors for OSA have Although there are several studies correlating the
been identified, such as obesity, increasing age and the as- severity of glaucoma with the severity of OSA, there are
sociated reduced muscle tone, as well as anatomical com- only few published investigations of the possible correla-
promise promoting upper airway narrowing [3]. tions between different glaucoma phenotypes and OSA se-
Snoring, sleep disruption, impaired daytime perform- verity [14]. The aim of the present investigation is to study
ance, attention deficits and daytime sleepiness are charac- the correlation of the severities of both entities, e.g. estab-
teristic symptoms of OSA. However, since in principle all lishing a correlation of ophthalmological parameters,
organs can be affected by the nightly intermittent oxygen namely IOP as one of the main risk factors and the mean
deficiency, further effects of the disease are also possible. defect (MD) in visual fields, as a marker for severity of the
Glaucoma, on the other hand, is a group of eye dis- disease, with quantifiable parameters of OSA like the
eases, in which damage to the optic nerve leads to visual apnea/hypopnea index (AHI) and the oxygen desaturation
field disturbances. This damage is caused by a multifac- index (ODI). Besides that, distinct glaucoma phenotypes
torial progressive optic neuropathy that can result in the were analyzed regarding the respective OSA prevalence.
degeneration of retinal ganglion cells, leading to a charac- Moreover, we discuss the potential interactions and the
teristic cupping of the optic nerve and thus in irreversible pathogenesis of OSA and glaucoma and make suggestions
vision loss and blindness [4]. A distinction is made be- for their clinical management.
tween different glaucoma phenotypes, whereby a major
factor is intraocular pressure. Increased intraocular pres-
sure is considered the only modifiable risk factor, as op- Methods
posed to other non-modifiable risk factors, like age and In order to correlate ophthalmological parameters (in
positive family history [5]. However, not every glaucoma particular MD and IOP) with OSA-related respiratory
phenotype is associated with high intraocular pressure; home sleep apnea testing (HSAT) parameters, glaucoma
e.g. in the low-tension glaucoma (LTG) phenotype dam- patients and patients with ocular hypertension (OH)
age to the optic nerve occurs without increased intraocu- with no previous diagnosis of OSA who were hospital-
lar pressure (IOP). ized for glaucoma diagnosis were prospectively recruited
A link between OSA and glaucoma has been proposed to undergo an attended one-night-HSAT screening for
[5–10]. Specifically, it is assumed that OSA might partly OSA. Glaucoma-specific diagnostic procedures included
cause or worsen glaucoma [5]. To date, the exact patho- a full ophthalmological examination, repeated IOP mea-
genesis is unclear. One suggested mechanism is episodic surements during different specified times of the day in
hypoxic damage of the optic nerve [11]. Chaitanya et al. a period of 36 h and MD measurements via perimetry.
have reviewed current theories and have distinguished Inclusion criteria for this prospective study were a diag-
between hypoxia and vascular and/or mechanical factors nosis of glaucoma or ocular hypertension made by a
dedicated glaucoma consultant, positive snoring history,
Bahr et al. Respiratory Research          (2020) 21:258 Page 3 of 8
age > 18 years, patient’s compliance and insight and body mentioned shortcomings when using this type of HSAT
mass index (BMI) ≤34.9 kg/m2. Exclusion criteria were devices, we decided to use HSAT technology –and not
active malignant tumors (end of treatment < 5 years), polysomnography- to diagnose OSA. This was because
age < 18 years, COPD (Stadium Gold 2–4) Raynaud’s of its lower cost, its greater convenience for the patients
phenomenon (causing problems of pulse oxymetry mea- and due to the fact, that it is an established and ap-
surements) and BMI > 34.9 kg/m2. Participants who were proved method for the diagnosis of OSA and conse-
initially suspected to suffer from glaucoma or ocular quential treatment decisions.
hypertension but eventually did not meet the criteria ei- Participants entered the study after giving consent and
ther for glaucoma or for ocular hypertension formed the being informed about the purpose of the study. The
control group. The process of recruitment and analysis study protocol and the procedures used were in accord-
is shown as a STROBE flowchart in Fig. 1. ance with the principles of the Declaration of Helsinki.
Relevant parameters in HSAT (polygraphy) were The study was approved by the local ethics committee
apnea-hypopnea-index (AHI), oxygen desaturation index of the respective state medical association (Number
(ODI) and average and minimal arterial oxygen desatur- 837.068.17 [10902]).
ation as measured with pulse oxymetry. Ophthalmo-
logical parameters were intraocular pressure (IOP) in
mmHg for each eye and mean defect depth (MD) as well Statistical analysis
as the glaucoma phenotype, namely primary open angle Group comparisons were done by Kruskall-Wallis
glaucoma (POAG), low-tension-glaucoma (LTG) or pre- testing. Non-parametric comparisons between groups
liminary stages of high-tension glaucoma - so-called as post-hoc testing were done using the Wilcoxon’s
ocular hypertension (OH). General parameters included test and Hodges-Lehmann values including 95%-confi-
in the analysis were age, gender and BMI. dence intervals. Bonferroni corrected p-values were
Miniscreen plus (Loewenstein Medical, EC-certificate also calculated. Wilcoxon’s/Kruskal- Wallis rank sum
acc.93/42/ECC Annex II) HSAT devices were used, test was performed to compare the different glaucoma
equipped with a pulse oxymeter, nasal cannula and phenotypes in terms of OSA-parameters, like AHI,
piezoelectric straps for recording of chest and abdomen apnea-index (AI), hypopnea-index (HI) and oxygen
excursion. Before falling asleep, study participants put desaturation index (ODI) as shown on Table 4. Scal-
on the device themselves after being instructed a few ing of OSA severity between mild (AHI 5–15/h),
hours earlier. This device technology provides a 70–85% moderate (AHI ≥15–30/h), severe (AHI ≥ 30/h) and
validity and efficiency in detecting OSA [15], and has a non-OSA (AHI ≤ 5 /h) was done. We compared their
tendency to overestimate the sleep period by 20% and prevalence in the different glaucoma phenotypes with
underestimate OSA severity by 11% [16]. Albeit the contingency analyses.
Fig. 1 The study flow chart in line with the STROBE (Strengthening the Reporting of Observational Studies in Epidemiology) statement
Bahr et al. Respiratory Research          (2020) 21:258 Page 4 of 8
Results Discussion
A total of 110 glaucoma patients (47 females, 63 males; In this report evidence is provided that only glaucoma
median age 64.3 years, median BMI 26.62 kg/m2) have phenotypes that are associated with increased IOP show
been screened via HSAT. After screening, 101 individuals higher AHI. Obstructive sleep apnea was significantly
were found to meet the inclusion criteria for analysis. more prevalent in patients with high IOP, namely partic-
Table 1 shows the demographic features of the analyzed ipants with POAG or OH. Other glaucoma phenotypes,
cohort. Eventually, 14 participants who were initially sus- like LTG, that by definition also show a damage of the
pected to suffer from glaucoma or ocular hypertension optic nerve but are not associated with increased IOP,
but did not meet the criteria either for glaucoma or for are not associated with OSA. To our knowledge, this is
ocular hypertension comprised the control group. the first report dedicated to the investigation of the asso-
There was no significant correlation between both in- ciation of OSA with different glaucoma phenotypes.
traocular pressure (IOP) or median defect depth (MD) Nonetheless, AHI did not correlate directly with physical
and the AHI. BMI and age did not influence the results. measures associated with glaucoma, namely IOP or MD.
Different glaucoma phenotypes, namely primary open It may be suggested, that OSA, in particular moderate
angle glaucoma (POAG), low-tension-glaucoma (LTG) and severe OSA forms, leads to higher IOP and thus to
and ocular hypertension (OH), were identified and ana- the two aforementioned phenotypes. However, higher
lyzed separately in terms of AHI and OSA prevalence AHI did not correlate with higher IOP. The reason why
(shown on Tables 2 and 3). Since the data related sleep glaucoma phenotypes correlate with OSA severity, in-
apnea showed signs of non-normal distribution and un- stead of physical measures associated with glaucoma,
equal variance across groups, we used Wilcoxon-Testing such as IOP or MD, remains unclear. Notably, this find-
providing respective Z-values as well as Hodges- ing is in line with the few studies that showed a lack of
Lehmann estimator as a rank based estimator for the correlation of IOP with the AHI/RDI [17]. For example,
median difference between two groups “Group” sub- Geyer et al. could neither find a higher incidence of
tracted by “-Group” (Tables 3, 4 and 5). OSA in open-angle glaucoma nor a correlation of the
The presented p-value is the post-hoc p-Value for the RDI with IOP. Xin et al. on the other hand showed that
comparison of the groups, while the “ Probability >Chi2” OSA had a significantly higher prevalence in POAG and
provides the p-value for the global group comparison in particular severe OSA was associated with higher IOP
across the four different groups (control, TLG, POAG, and optic nerve damage, although no correlation could
OH) for the respective “analyzed parameter”. Obstruct- be drawn [18].
ive sleep apnea was significantly more prevalent in pa- Women present a higher degree of systemic inflamma-
tients with high IOP, namely participants with POAG or tion than males despite the same degree of OSA gravity
OH, whereas participants with LTG had significantly [19]. In this present study OSA was even more clearly
fewer and less severe OSA cases. The median AHI was associated with hypertensive glaucoma types in women
22.7/h in the POAG group and 21.9/h in the in OH than in men. This should be considered when women
group. In contrast, the median AHI in LTG was 9.45/h are diagnosed with glaucoma or ocular hypertension.
(p = 0.0009) (Fig. 2). Highly significant differences were There are already various studies dealing with a pos-
found between the four groups with regard to the AHI sible correlation between the two diseases. In spite of
(Chi2 = 22, df = 3, p < 0.0001) with significantly fewer some large-scale studies and meta-analyses, the facts
events per hour in the LTG compared to POAG have not yet been fully explored. In their publication Shi
(Hodges-Lehmann = − 13.8, 95% CI (− 18.6 – − 8.8; p < et al. (2015) prepared a meta-analysis from 16 studies
0.0001) and the controls (Hodges-Lehmann = 12.1; 95% with a total of 2,278,832 participants [8]. The pooled
CI -19.9 – − 2.4; p < 0.02). Additionally, this effect was analysis showed a statistically relevant correlation be-
much more obvious in females (p = 0.0006) than in tween glaucoma and sleep apnea with an OR (odds ra-
males (p = 0.011) (Fig. 2). tio) of 1.96 in the case-control studies and an OR of 1.67
Table 1 Epidemiologic features and ranges of AHI, and ODI of the analyzed groups
Group Gender (f/m) Age median (Quartiles) BMI median (Quartiles) AHI median (Quartiles) ODI median (Quartiles)
POAG 21/31 67 (55–77) 26 (24–29) 22.7 (12.4–30.3) 21.8 (12.2–29.0)
LTG 14/8 63.5 (57.8–73.5) 25 (24–27) 9.45 (5.7–11.4) 10.6 (8.1–17.3)
OH 4/10 64.5 (50.5–68) 29 (24–31) 10.9 (5.9–29.1) 14.65 (6.7–19.8)
Control 8/5 70 (56–73) 26 (24–31) 21.9 (6.1–30.5) 16.7 (7.7–29.5)
Due to the quite different variance of the respective values within each one of the groups, median values and quartiles are presented. POAG Primary open angle
glaucoma, LTG Low-tension glaucoma, OH Ocular hypertension, AHI Apnea hypopnea index, ODI Oxygen desaturation index, BMI Body mass index.
Bahr et al. Respiratory Research          (2020) 21:258 Page 5 of 8
Table 2 Contingency Table: Distribution of obstructive sleep apnea (OSA) severity levels by glaucoma group; AHI = apnea-
hypopnea-index
Mild Moderate Severe No OSA Total (%)
AHI 5–15/h AHI ≥15–30/h AHI ≥ 30/h AHI ≤ 5 /h
Primary open angle glaucoma 11 23 14 5 53 (52.48)
Low-tension glaucoma 16 2 0 4 22 (21.78)
Ocular hypertension 6 2 2 2 12 (11.88)
No Glaucoma 2 6 3 3 14 13.86)
Total (%) 35 (34.65) 33 (32.67) 19 (18.81) 14 (13.86) 101 (100)
in the retrospective cohort study. There was no distinc- between POAG and OH as well as POAG and LTG
tion between glaucoma types and it is not clear whether (Table 3). Our patients were naïve to OSA therapy and
the patients were naïve to OSA therapy. OSA diagnosis built an appropriate cohort for further follow-up after ini-
was also not uniformly made via in-home polysomno- tiation of OSA-therapy. To the best of our knowledge, this
graphy, but via questionnaires, HSAT (polygraphies) and is the first prospective study that investigated glaucoma
/ or polysomnography or even just with the question “do patients for the absence or presence of OSA. Most cohort
you have sleep apnea”. studies and meta-analyses analyzed OSA cohorts in terms
In a multicenter study from France with 9580 partici- of the presence of glaucoma. Glaucoma criteria were often
pants, Aptel et al. (2014) could not identify any statistically not specified and there was often no distinction between
relevant relationship between the two diseases (OR 1.13) the different glaucoma phenotypes.
[20]. However, not all ophthalmological parameters had Weaknesses of this study are the rather sparse back-
been included in that study. There could have been un- ground information on other concomitant, e.g. cardio-
known glaucoma cases included in their cohort. Besides, vascular, diseases of the glaucoma patients and on
there was no differentiation between glaucoma phenotypes previously performed glaucoma therapy (e.g. previous
and there was no information on possible exclusion of pa- outpatient procedures), which were not documented in
tients with previous OSA treatment. Moreover, only pa- detail in the medical records and which could hence
tients aged over 50 years had been included in that study. have been a confounder of the results. Moreover, a fur-
Additionally, Morsy et al. (2019) found in their study with ther limitation of the study is that full-night polysomno-
80 participants that OSA is associated with a higher risk of graphy was not performed. The sensors included in the
vision-threatening and non-threatening ocular disorders. HSAT device used in our study were also unable to de-
The lowest oxygen desaturation index in OSA patients was tect hypopneas that are only associated with cortical
a significant predictor of vision-threatening disorders [9]. In arousals. Due to these limitations, such a HSAT device
our study, on the other hand, highest ODIs were found in may underestimate the severity of OSA [21]. Notably,
the POAG group, which made it possible to distinguish we did not use automatically scored HSAT data; on the
Table 3 Non-parametric comparisons with Wilcoxon-Method
Analyzed Parameter Group - Group Z p-value Hodges-Lehmann Chi-Square-Approximation: Probability > Chi2
AHI Control LTG 2.4 0.019 12.1 < 0.0001
AHI LTG POAG −4.9 < 0.001 −13.8
AI Control LTG 2.2 0.0253 4.1 0.0054
AI LTG POAG −3.50 0,0004 −6.5
HI Control LTG 2.4 0.0176 6.6 0.0024
HI OH LTG 2.3 0.0203 2.9
HI LTG POAG −3.6 0.0003 −4.9
SI LTG POAG −2.4 0.0183 −3.9 0.0741
ODI OH POAG −2.3 0.0207 −7.4 0.0025
ODI LTG POAG −3.6 0.0004 −9.2
AHI Apnea hypopnea index, AI Apnea index, HI Hypopnea index, SI Snoring index, ODI Oxygen Desaturation index, LTG Low-tension-glaucoma, POAG Primary
open-angle-glaucoma, OH Ocular hypertension. Positive Hodges-Lehmann values represent positive differences between “Group” and “-Group”.
Bahr et al. Respiratory Research          (2020) 21:258 Page 6 of 8
Table 4 Non-parametric comparisons with Wilcoxon-Method in females
Analyzed Parameter Group - Group Z p-value Hodges-Lehmann Chi-Square-Approximation: Probability > Chi2
AHI Controls POAG −2.3 0.0233 −11 −0.0006
AHI LTG POAG −4.1 < 0.0001 −14.8
AI LTG POAG −3.0 0.0032 −6.5 0.0195
HI LTG POAG −2.9 0.0032 −6.9 0.0199
ODI Control POAG −2.2 0.0281 −9.0 0.0183
ODI LTG POAG −2.7 0.0078 −8.9
AHI Apnea hypopnea Index, AI Apnea Index, HI Hypopnea Index, ODI Oxygen Desaturation Index, LTG Low-tension-glaucoma, POAG Primary open-angle-glaucoma.
Positive Hodges-Lehmann values represent positive differences between “Group” and “-Group”.
contrary, the raw data from the HSAT device were involving high-tension glaucoma phenotypes. A further
reviewed and interpreted by physicians who are either possibility appears to be that high-IOP glaucoma pheno-
board-certified in sleep medicine or overseen by a types and OSA may share common pathogenetic mecha-
board-certified sleep medicine physician [22]. nisms or that IOP- and AHI- measures may be subject
Further studies should include a detailed medical to the same confounders.
history of the patients and should include a follow-up One such possible common pathogenetic mechan-
of glaucoma- and OSA-specific parameters under ism may involve inflammatory pathways that are more
OSA-therapy, e.g. positive-airway-pressure (PAP). relevant to the development of optic nerve neur-
Besides, different glaucoma therapies, surgical or opathy and loss of vision in POAG than in LTG. A
pharmacological, should be additionally considered in higher level of systemic inflammation in OSA pa-
the analysis and patients could be followed-up in tients, typically associated with higher systemic blood
terms of the evolution of their AHI (with or without pressure, may therefore only explain the pathogenesis
PAP-therapy). of POAG. However, a large population-based Chinese
On the basis of these results, it can be suggested that cohort study could not find any association between
glaucoma phenotypes with increased IOP and even ocu- the different glaucoma phenotypes and systemic in-
lar hypertension are associated with a higher AHI in flammation, as depicted by C-reactive-protein values
snoring patients and thus with a higher risk of OSA. [23]. Other authors have speculated that an elevated
Moreover, it is suggested that OSA, and in particular its neutrophil-to-lymphocyte ratio and the systemic im-
severe forms, may lead to an increase in IOP and thus to mune inflammation index might serve as predictors
high IOP-associated clinical glaucoma phenotypes. for POAG [24]. Kondkar et al. found high systemic
There is also a remarkable negative association between level of the inflammatory cytokine TNF-α in their
low-tension-glaucoma and OSA, since there was not a study of 51 POAG patients, compared to 88 controls
single case of severe OSA in the group of LTG- [25]. Nonetheless, the mechanisms linking increased
participants. This fact may suggest that low-tension- intraocular pressure to increased systemic inflamma-
glaucoma pathogenesis is quite different from the one tion have not been sufficiently elucidated so far.
Table 5 Non-parametric comparisons with Wilcoxon-Method in males
Analyzed Parameter Group - Group Z p-value Hodges-Lehmann Chi-Square-Approximation: Probability > Chi2
AHI Control LTG 2.9 0.0042 22.6 0.0111
AHI LTG POAG −2.5 0.0117 −12.4
AI Control LTG 2.3 0.0233 12.9 0.0958
HI Control POAG 2.2 0.0316 8.6 0.0141
HI Control LTG 2.9 0.0043 11.7
HI Control OH 2.3 0.0235 9.6
ODI Control LTG 2.6 0.0103 9.7 0.015
ODI Control OH 2.0 0.0433 15.7
ODI LTG POAG −2.3 0.0198 −10.0
AHI Apnea hypopnea Index, AI Apnea Index, HI Hypopnea Index, SI Snoring Index, ODI Oxygen Desaturation Index, LTG Low-tension-glaucoma, POAG Primary
open-angle-glaucoma, OH Ocular hypertension. Positive Hodges-Lehmann values represent positive differences between “Group” and “-Group”.
Bahr et al. Respiratory Research          (2020) 21:258 Page 7 of 8
Availability of data and materials
All data generated or analysed during this study are included in this
published article.
Ethics approval and consent to participate
The study was approved by the ethics committee of the State Medical
Association of Rhineland-Palatinate (Number 837.068.17 [10902]). All patients
gave their written and oral consent to participate after detailed information
given by the participating study physicians.
Consent for publication
All patients gave their oral and written consent to publish the collected data;
individual figures or photographs were not included in this study.
Competing interests
The authors declare that they have no competing interests.
Fig. 2 Distribution of the apnea hypopnea index in different
glaucoma groups; females in black, males in grey. OSA severity was Author details1
classified as mild (AHI 5 15/h), moderate (AHI 15 30/h), severe Sleep Medicine Center, Department of Otorhinolaryngology, Head and Neck– ≥ –
Surgery, Medical Center of the University of Mainz, 55131 Mainz, Germany.
(AHI≥ 30/h) and non-OSA (AHI≤ 5 /h). POAG = primary open angle 2Department of Ophthalmology, Medical Center of the University of Mainz,
glaucoma, LTG = low-tension glaucoma, OH = ocular hypertension Mainz, Germany. 3Institute for Sports Science, Johannes
Gutenberg-University, Mainz, Germany.
Conclusion
Received: 10 April 2020 Accepted: 4 October 2020
Ophthalmologic parameters like intraocular pressure or
mean defect depth were not strongly associated with the
AHI, but there was a strong correlation between the References
1. Heinzer R, Vat S, Marques-Vidal P, Marti-Soler H, Andries D, Tobback N, et al.
clinical classification of the glaucoma phenotypes and Prevalence of sleep-disordered breathing in the general population: the
the AHI. Especially the group of patients with LTG dif- HypnoLaus study. Lancet Respir Med. 2015;3(4):310–8.
fered significantly from the group with POAG and 2. Wenner JB, Cheema R, Ayas NT. Clinical manifestations and consequences
of obstructive sleep apnea. J Cardiopulm Rehabil Prev. 2009;29(2):76–83.
healthy controls in terms of OSA-defining parameters, 3. Senaratna CV, Perret JL, Lodge CJ, Lowe AJ, Campbell BE, Matheson MC,
such as the AHI. We conclude that OSA leads to an in- et al. Prevalence of obstructive sleep apnea in the general population: a
crease of IOP and to high-IOP associated phenotypes, systematic review. Sleep Med Rev. 2017;34:70–81.
4. Weinreb RN, Khaw PT. Primary open-angle glaucoma. Lancet. 2004;
namely POAG and OH. Moreover, there is no correl- 363(9422):1711–20.
ation between OSA and LTG, which might support a 5. Swaminathan SS, Bhakta AS, Shi W, Feuer WJ, Abreu AR, Chediak AD, et al.
different pathogenesis of LTG. We propose that further Is obstructive sleep apnea associated with progressive glaucomatous optic
neuropathy? J Glaucoma. 2018;27(1):1–6.
studies should be done to investigate the necessity for 6. Lin CC, Hu CC, Ho JD, Chiu HW, Lin HC. Obstructive sleep apnea and
routine screening for OSA by HSAT specifically targeted increased risk of glaucoma: a population-based matched-cohort study.
to patients with definite POAG and definite ocular Ophthalmology. 2013;120(8):1559–64.
7. Huon LK, Liu SY, Camacho M, Guilleminault C. The association between
hypertension. ophthalmologic diseases and obstructive sleep apnea: a systematic review
and meta-analysis. Sleep Breath. 2016;20(4):1145–54.
Abbreviations 8. Shi Y, Liu P, Guan J, Lu Y, Su K. Association between Glaucoma and
AHI: Apnea hypopnea index; AI: Apnea index; BMI: Body mass index; CRP: C- obstructive sleep apnea syndrome: a meta-analysis and systematic review.
reaktive-protein; HI: Hypopnea index; IOP: Intra-ocular pressure; LTG: Low- PLoS One. 2015;10(2):e0115625.
tension glaucoma; MD: Mean defect depth; ODI: Oxygen desaturation index; 9. Morsy NE, Amani BE, Magda AA, Nabil AJ, Pandi-Perumal SR, BaHammam
OH: Ocular hypertension; OSA: Obstructive sleep apnea; PAP: Positive airway AS, et al. Prevalence and predictors of ocular complications in obstructive
pressure; POAG: Primary open angle glaucoma sleep apnea patients: a cross-sectional case-control study. Open Respir Med
J. 2019;13:19–30.
Acknowledgements 10. Teberik K, Eski MT, Balbay EG, Kaya M. Evaluation of intraocular pressure,
The authors acknowledge the English language editing by Constantinos corneal thickness, and retinal nerve fiber layer thickness in patients with
Contocristos. obstructive sleep apnea syndrome. Pakistan J Med Sci. 2018;34(4):817–22.
11. Golbidi S, Badran M, Ayas N, Laher I. Cardiovascular consequences of sleep
Authors’ contributions apnea. Lung. 2012;190(2):113–32.
HD and WK collected the data, MB helped with the methods and revised 12. Chaitanya A, Pai VH, Mohapatra AK, Ve RS. Glaucoma and its association
the manuscript, TH, CM, JD and VP revised and reviewed the manuscript. PS with obstructive sleep apnea: a narrative review. Oman J Ophthalmol. 2016;
provided data analysis, presentation, and interpretation and co-designed the 9(3):125–34.
Figs. HG is responsible for study conception and design, helped on methods 13. Baumane K, Ranka R, Laganovska G. Association of NT-proANP level in
and results and revised the manuscript. KB collected and extracted the data, plasma and humor aqueous with primary open-angle Glaucoma. Curr Eye
performed analysis, presentation and interpretation, created the tables and Res. 2017;42(2):233–6.
figures, wrote and revised the manuscript. All authors approved the submitted 14. Balbay EG, Balbay O, Annakkaya AN, Suner KO, Yuksel H, Tunç M, et al.
version of the manuscript and take full responsibility for it. Obstructive sleep apnoea syndrome in patients with primary open-angle
glaucoma. Hong Kong Med J. 2014;20(5):379–85.
Funding 15. Portier F, Portmann A, Czernichow P, Vascaut L, Devin E, Benhamou D, et al.
The authors declare declare the absence of any financial support in this Evaluation of home versus laboratory polysomnography in the diagnosis of
project. Open Access funding enabled and organized by Projekt DEAL. sleep apnea syndrome. Am J Respir Crit Care Med. 2000;162(3 Pt 1):814–8.
Bahr et al. Respiratory Research          (2020) 21:258 Page 8 of 8
16. Light MP, Casimire TN, Chua C, Koushyk V, Burschtin OE, Ayappa I, et al.
Addition of frontal EEG to adult home sleep apnea testing: does a more
accurate determination of sleep time make a difference? Sleep Breath. 2018;
22(4):1179–88.
17. Geyer O, Cohen N, Segev E, Rath EZ, Melamud L, Peled R, et al. The
prevalence of glaucoma in patients with sleep apnea syndrome: same as in
the general population. Am J Ophthalmol. 2003;136(6):1093–6.
18. Xin C, Zhang W, Wang L, Yang D, Wang J. Changes of visual field and optic
nerve fiber layer in patients with OSAS. Sleep Breath. 2015;19(1):129–34.
19. Gouveris H, Bahr K, Jahn C, Matthias C, Simon P. The apnea-hypopnea index
underestimates systemic inflammation in women with sleep-disordered
breathing. J Women's Health (Larchmt). 2018;27(7):920–6.
20. Aptel F, Chiquet C, Tamisier R, Sapene M, Martin F, Stach B, et al.
Association between glaucoma and sleep apnea in a large French
multicenter prospective cohort. Sleep Med. 2014;15(5):576–81.
21. Kapur VK, Auckley DH, Chowdhuri S, Kuhlmann DC, Mehra R, Ramar K, et al.
Clinical practice guideline for diagnostic testing for adult obstructive sleep
apnea: an American Academy of sleep medicine clinical practice guideline.
J Clin Sleep Med. 2017;13(3):479–504.
22. Rosen IM, Kirsch DB, Carden KA, Malhotra RK, Ramar K, Aurora RN, et al.
Clinical use of a home sleep apnea test: an updated American Academy of
sleep medicine position statement. J Clin Sleep Med. 2018;14(12):2075–7.
23. Jonas JB, Wei WB, Xu L, Wang YX. Systemic inflammation and eye diseases.
The Beijing Eye Study. PLoS One. 2018;13(10):e0204263.
24. Tang B, Li S, Han J, Cao W, Sun X. Associations between blood cell profiles
and primary open-angle Glaucoma: a retrospective case-control study.
Ophthalmic Res. 2020;63(4):413–22.
25. Kondkar AA, Sultan T, Almobarak FA, Kalantan H, Al-Obeidan SA, Abu-Amero
KK. Association of increased levels of plasma tumor necrosis factor alpha
with primary open-angle glaucoma. Clin Ophthalmol. 2018;12:701–6.
Publisher’s Note
Springer Nature remains neutral with regard to jurisdictional claims in
published maps and institutional affiliations.