Peripapillary Hyperreflective Ovoid Mass-Like Structures: Prevalence and Associations in the Adult Population of the Beijing Eye Study

Abstract

Purpose

To assess prevalence and associations of peripapillary hyperreflective ovoid mass-like structures (PHOMS) on optical coherence tomographic images in a general adult population.

Methods

Participants in the population-based Beijing Eye Study underwent ocular and systemic examinations. Using optical coherence tomographic optic nerve head images, we assessed presence and location of PHOMS.

Results

The study included 963 eyes (age, 64.1 ± 9.5 years; axial length, 23.05 ± 1.02 mm). PHOMS detected in 15 eyes (1.6%; 95% confidence interval [CI], 0.0–2.1) extended over 30°, 60°, and 90° in 4 (27%), 6 (40%), and 5 (33%) eyes, respectively. The PHOMS were located at 2 o’clock (referring to right eyes) in two (13%), at 3:00 h (n = 1), 5:30 h (n = 1), 6:00 h (n = 2), 6:30 h (n = 1), 7:00 h (n = 2), 8:00 h (n = 1), 8:30 h (n = 2), 9:30 h (n = 2), and at 10:30 h (n = 1). On multivariable analysis, higher PHOMS prevalence was associated with smaller optic disc size (odds ratio [OR], 0.05; 95% CI, 0.01–0.28; P < 0.001), thicker peripapillary retinal nerve fiber layer (OR, 1.13; 95% CI, 1.04–1.22; P = 0.004), thicker retinal pigment epithelium–Bruch’s membrane complex thickness (OR, 1.70; 95% CI, 1.19–2.43; P = 0.04), and longer axial length (OR, 2.14; 95% CI, 1.12–4.08; P = 0.02). It was not associated with best-corrected visual acuity, perimetric indices, or any other ocular or systemic parameter or disease examined.

Conclusions

PHOMS are relatively rare in the general adult and elderly population and mainly associated with a small optic disc. They are not related to a decrease in visual function or peripapillary retinal nerve fiber layer thickness, and they do not indicate an optic nerve damage. They are unrelated to other major ocular and systemic diseases, although they may be due to a localized crowding of peripapillary nerve fibers in eyes with small (crowded) optic discs.

Peripapillary hyperreflective ovoid mass-like structures (PHOMS) are toroidal structures with a round mostly oval appearance on cross-sectional optical coherence tomographic (OCT) images of the optic nerve head.^1–19 They have been reported to correlate with several other neuro-ophthalmological and neurological abnormalities and disorders, including optic disc drusen, abnormal optic disc shapes, and optic disc edema. In the latter case, an increased retrolaminar cerebrospinal fluid pressure and secondary impediment of the orthograde axoplasmic flow has been discussed to lead to a thickening of the optic nerve fibers and to a herniation and displacement of the axons away from the optic disc. PHOMS have been reported to be the most common reason for pseudopapilledema in children.⁶ The first observations of PHOMS described their occurrence in eyes of patients with neuro-ophthalmological and neurological diseases, such as intracranial hypertension, multiple sclerosis, neuromyelitis optica spectrum disorders, and myelin oligodendrocyte glycoprotein-immunoglobulin G–associated disease.^1–19

Few studies have so far examined the prevalence and associations of PHOMS in normal individuals, and, if so, the normal individuals were often recruited in a hospital-based manner. The only population-based study so far on the prevalence and associations of PHOMS was conducted by Hamann et al. in children.¹¹ The previous studies had some limitations, such as that the investigations included only relatively small study populations, or were primarily focused on patients with neuro-ophthalmological or neurological abnormalities and diseases, or did not include cohorts of normal healthy individuals; that the study populations were not recruited in a population-based manner or did not include adults; and that the number of parameters additionally examined and tested for their association with PHOMS was relatively small. In addition, information about PHOMS in Asian individuals has remained scarce so far. We, therefore, conducted the present study to examine the prevalence and associations of PHOMS in a population of Chinese adults, recruited in a population-based manner and undergoing a series of ophthalmological and systemic examinations so that PHOMS-related findings could also be examined for their associations with other parameters.

Methods

The population-based Beijing Eye Study was conducted in 2011 in four communities in the urban district of Haidian in the north of central Beijing and in three communities in the village area of Yufa of the Daxing District south of Beijing.^20,21 It was approved by the Medical Ethics Committee of the Beijing Tongren Hospital, and all study participants gave their written informed consent. Inclusion criteria for the Beijing Eye Study were living in the study regions of Haidian or Daxing, and an age of at least 50 years.^20,21

Study participants underwent a series of ophthalmological examinations, including refractometry (for measurement of refractive error) and assessment of best-corrected visual acuity, tonometry, slit lamp–based examination of the anterior and posterior segments, biometry, photography of the cornea and lens, and imaging of the macula and optic nerve head (fundus camera CR6-45NM, Canon Inc, Tokyo, Japan). We additionally performed spectral-domain OCT (Spectralis; wavelength, 870 nm; Heidelberg Engineering Co, Heidelberg, Germany) of the macula and optic nerve head. The OCT-based macular imaging included a macular volume scan (30° × 30° field, 31 B-scan lines) centered on the fovea, a macular star with six radial scans, and an optic nerve head star with six radial scans. Each scan line of the macular volume scan and of the macular and optic nerve head stars was based on 100 averaged scans. We additionally performed a peripapillary scan of the retina for measurement of the peripapillary retinal nerve fiber layer (RNFL) thickness.

All study participants underwent a structured interview by trained research technicians. The interview included more than 200 standardized questions on demographic parameters, socioeconomic background, diet and alcohol consumption, smoking habits, known major systemic diseases, and current systemic medical therapies. Using the Mini-Mental State examination, we assessed the cognitive function. Fasting blood samples were collected for measurement of blood lipids, glucose, glycosylated hemoglobin, and serum creatinine. The blood pressure was measured with the participant sitting for at least 5 minutes. We measured the anthropometric parameters of body height and weight and circumference of waist and hip. Arterial hypertension was defined by a systolic blood pressure 160 mm Hg or higher and/or a diastolic blood pressure of 95 mm Hg or higher, and/or self-reported current treatment for arterial hypertension with antihypertensive medication. Diabetes mellitus was characterized by a blood glucose concentration of 7.0 mmol/L or greater, an glycosylated hemoglobin value of 6% or greater, by a self-reported history of physician diagnosis of diabetes mellitus, or by a history of drug treatment for diabetes (insulin or oral hypoglycemic agents). Depressive symptoms were evaluated using a Chinese depression scale adapted from the Zung self-rated depression scale.^20,21

We measured the degree of nuclear lens opacities in six grades applying the grading system of the Age-Related Eye Disease Study. Using retroilluminated photographs of the lens, we determined the degree of cortical cataract as the percentage of areas with cortical and posterior subcapsular lens opacities. The standard to diagnose a nuclear cataract was a nuclear cataract grade of 4 or higher, the standard to diagnose a posterior subcapsular cataract was a posterior subcapsular opacity amount of 0.01 or higher, and the standard to diagnose a cortical cataract was a cortical opacity amount of 0.05 or greater. The degree of fundus tessellation, defined as the visibility of the large choroidal vessels, was assessed using the fundus photographs of the macula and optic disc as described in detail previously.^20,21 Minimum criterion for the diagnosis of diabetic retinopathy was the presence of at least one microaneurysm. The definition of glaucomatous optic neuropathy was based on the criteria of the International Society of Geographic and Epidemiological Ophthalmology. Pseudoexfoliation was assessed in phakic eyes by an experienced ophthalmologist during slit-lamp–assisted biomicroscopy of the anterior segment after pupillary dilation. Subfoveal choroidal thickness was measured on an OCT scan obtained with the enhanced depth imaging modality after pupil dilation and running through the center of the fovea. Subfoveal choroidal thickness was defined as the vertical distance from the hyperreflective line of the Bruch's membrane to the hyperreflective line of the inner surface of the sclera. We measured the optic disc size on the optic disc photographs and corrected the optic media-related magnification according to Littmann’s and Bennet’s method.^22,23 Frequency doubling threshold perimetry had been performed by the participants during the second survey of the Beijing Eye Study in 2006.

Based on the recommendations made by the Optic Disc Drusen Studies Consortium, we defined PHOMs by the criteria of being located in the peripapillary region on top of Bruch's membrane, with a gap frequently present in the OCT scans of PHOMS aligned through the optic disc center; often an upward deflection of two of the other retinal layers; and an OCT signal appearance similar to the reflectivity of the retinal nerve fiber and ganglion cell layers (Fig. 1).³ For the assessment of PHOMS, we examined all OCT images of the optic nerve head. All images were explored by a trained ophthalmologist (JBJ).

Figure 1.

OCT images of PHOMSs (yellow arrows); red arrows, direction of the OCT scans.

For the present study, we included a subgroup of participants out of the whole cohort of the Beijing Eye Study and reassessed their optic nerve head OCT scans with a special focus of PHOMS. Using a commercially available statistical software package (SPSS for Windows, version 27.0, IBM-SPSS, Chicago, IL, USA), we determined the prevalence of PHOMS and other morphologic characteristics as means and 95% confidence intervals (CIs). Associations between PHOMS prevalence and the occurrence of other fundus lesions and ocular parameters were examined in binary regression analyses, first in a univariable manner and followed by multivariable analyses. The latter included as independent variables, all parameters that were associated significantly with PHOMS occurrence in the univariable analyses. In a step-by-step manner, we then dropped out of the list of independent variables those parameters that were no longer associated with PHOMS prevalence significantly. In a final step, we readded parameters to the model to retest the significance of their potential association with the PHOMS prevalence. A two-sided P value of less than 0.05 was considered statistically significant.

Results

The Beijing Eye Study consisted of 3468 of 4403 eligible persons (78.8%). Of the 3468 individuals, the present study included 963 eyes with a mean age of 64.1 ± 9.5 years (median, 62.0 years; range, 50–91 years) and a mean axial length of 23.05 ± 1.02 mm (median, 23.02 mm; range, 19.39–29.00 mm). The group of 963 individuals included in the present study as compared with the group of 2505 Beijing Eye Study participants not included in the present study had a significantly shorter axial length (23.05 ± 1.02 mm vs. 23.31 ± 1.17 mm; P < 0.001), and a greater relative proportion of men (47.2% vs. 42.8%; P < 0.001). The groups did not differ significantly in age (64.1 ± 9.5 years vs. 64.7 ± 9.9 years; P = 0.06) or IOP (14.5 ± 5.8 mm Hg vs. 14.7 ± 2.8 mm Hg; P = 0.06).

PHOMS were detected in 15 of the 963 eyes (1.6%; 95% CI, 0.0–2.1) with a mean refractive error of 0.13 ± 1.20 diopters (D) (median, +0.25 D; range, −2.88 to +2.00 D) (Table 1). The PHOMS extended over 30° in 4 (27%), 60° in 6 (40%), and 90° in 5 (33%) eyes. They were located at 2 o’clock (referring to right eyes) in two (13%), at 3 o’clock in one (7%), and at 5:30 h (one eye), 6:00 h (two eyes), 6:30 h (one eye), 7:00 h (two eyes), 8:00 h (one eye), 8:30 h (two eyes), 9:30 h (two eyes), and at 10:30 h (one eye). In all eyes with PHOMS, only one PHOMS was detected. None of the eyes with PHOMS showed optic disc drusen.

Table 1.

Baseline Characteristics of the Subgroup of Participants With PHOMS and the Subgroup of Individuals Without PHOMS in the Beijing Eye Study

	PHOMS	Without PHOMS	P Value
No.	15	948	–
Age (years)	65.7 ± 10.8	63.8 ± 9.6	0.44
Sex (men/women)	5/10 (33/67)	450/498 (47.5/52.5)	0.28
Region of habitation (rural/urban)	14/1 (93/7)	676/272 (71.3/28.7)	0.06
Body height (cm)	161.9 ± 9.2	161.2 ± 7.9	0.75
Body weight (kg)	71.3 ± 12.7	67.0 ± 11.4	0.15
Body mass index (kg/m2)	27.1 ± 2.9	25.7 ± 3.9	0.20
Body waist/hip circumference ratio	0.92 ± 0.07	0.90 ± 0.06	0.10
Best-corrected visual acuity (logMAR)	0.06 ± 0.15	0.06 ± 0.19	0.99
Refractive error (D)	0.13 ± 1.20	0.09 ± 1.68	0.95
Axial length (mm)	23.4 ± 1.2	23.0 ± 1.0	0.22
Optic disc size area (mm2)	1.96 ± 0.27	2.34 ± 0.50	0.01
IOP (mm Hg)	13.3. ± 1.7	14.6 ± 2.8	0.09
RNFL thickness (µm)	106 ± 8	101 ± 12	0.07

Values are mean ± standard deviation or number (%).

In univariable analysis, a greater PHOMS prevalence was associated with the systemic parameters of higher body waist and hip circumference, and with the ocular parameters of smaller optic disc, higher cortical cataract prevalence, thicker retinal pigment epithelium–Bruch's membrane complex thickness, thinner retinal outer nuclear layer thickness, and thicker outer and inner plexiform layer thickness (Tables 2 and 3).

Table 2.

Associations (Univariate Analysis) Between the Prevalence of PHOMS and Systemic Parameters in Participants of the Beijing Eye Study

	Odds Ratio	95% CI	P Value
Age (years)	1.03	0.98–1.08	0.27
Sex (men/women)	1.81	0.61–5.33	0.28
Region of habitation (rural/urban)	0.18	0.02–1.36	0.10
Body height (cm)	1.01	0.95–1.08	0.75
Body weight (kg)	1.03	0.99–1.08	0.15
Body mass index (kg/m2)	1.08	0.96–1.22	0.19
Body waist circumference (cm)	1.06	1.01–1.11	0.01
Body hip circumference (cm)	1.07	1.01–1.014	0.04
Body waist/hip circumference ratio	785	0.28–2215220	0.10
Level of education (levels 1–5)	0.76	0.50–1.15	0.19
Cognitive score	0.98	0.86–1.12	0.81
Alcohol consumption frequency (none/<1× per month/1× per month/2–3× per month/1× per week/2–3× per week/daily)	1.00	0.78–1.29	0.98
Smoking (never/former/current)	1.11	0.63–1.98	0.72
Smoking (never/ever)	1.36	0.48–3.72	0.58
Smoking package years	1.01	0.995–1.03	0.17
Serum concentrations of:
Glucose (mmol/L)	0.68	0.35–1.33	0.26
Glycosylated hemoglobin (%)	0.32	0.08–1.28	0.11
High-density lipoproteins (mmol/L)	1.74	0.51–5.93	0.38
Low-density lipoproteins (mmol/L)	0.66	0.34–1.29	0.23
Triglycerides (mmol/L)	0.68	0.29–1.62	0.39
Cholesterol (mmol/L)	0.72	0.40–1.32	0.29
Diabetes mellitus, prevalence	0.00	0.00	1.00
Systolic blood pressure (mm Hg)	1.02	0.997–1.04	0.09
Diastolic blood pressure (mm Hg)	1.00	0.96–1.03	0.82
Mean blood pressure (mm Hg)	1.01	0.98–1.04	0.49
Arterial hypertension, prevalence	0.92	0.33–2.61	0.88
Estimated glomerular filtration rate (mL/min/1.73 m²) (CKDE Formula; China adapted)	1.02	0.99–1.06	0.21
Chronic kidney disease, frequency (GFR<60 mL/min/1.73 m²) (CKDE Formula; China adapted), frequency	0.00	0.00	1.00
Use of aspirin, frequency	0.78	0.32–1.89	0.59

CKDE, Chronic Kidney Disease Epidemiology Collaboration; GFR, glomerular filtration rate.

Table 3.

Associations (Univariate Analysis) Between the Prevalence of PHOMS and Ocular Parameters in Participants of the Beijing Eye Study

	Odds Ratio	95% CI	P Value
Best-corrected visual acuity (logMAR)	1.01	0.07–13.8	0.99
Frequency doubling technology perimetry (measured in 2006 for 754 eyes)	1.21	0.15–9.53	0.86
Refractive error (D)	1.01	0.72–1.41	0.95
Axial length (mm)	1.34	0.84–2.14	0.21
Anterior corneal curvature radius (mm)	0.92	0.10–8.76	0.94
Central corneal thickness (µm)	0.99	0.98–1.01	0.46
Anterior chamber depth (mm)	1.46	0.59–3.65	0.42
Lens thickness (mm)	0.42	0.07–2.40	0.33
Optic disc size area (mm2)	0.06	0.01–0.34	0.001
Optic disc shape (vertical/horizontal disc diameter ratio)	13.3	0.41–435	0.15
Parapapillary beta zone width (µm)	1.46	0.80–2.66	0.22
Parapapillary gamma zone width (µm)	1.00	0.998–1.01	0.35
IOP (mm Hg)	0.84	0.69–1.03	0.09
Optic disc fovea distance (mm)	0.93	0.13–6.61	0.95
Angle Kappa (between temporal arterial arcades with the optic disc as angle vertex (°)	0.94	0.86–1.02	0.14
RNFL thickness (µm)	1.05	0.997–1.10	0.07
Subfoveal choroidal thickness (µm), total	0.996	0.991–1.002	0.20
Subfoveal choroidal thickness (µm), large-vessel layer	0.99	0.98–1.002	0.13
Subfoveal choroidal thickness (µm), medium-sized vessel layer	0.99	0.97–1.003	0.12
Subfoveal choroidal thickness (µm), small-vessel layer	0.98	0.92–1.04	0.50
Fundus tessellation (grades 0–3)	1.21	0.63–2.32	0.58
Foveal retinal thickness (µm)	1.01	0.98–1.04	0.66
Nuclear cataract, prevalence	1.04	0.36–2.99	0.95
Nuclear cataract, degree (0–7)	1.04	0.64–1.67	0.88
Cortical cataract, prevalence	3.11	1.03–9.43	0.045
Cortical cataract, degree (percentage of lens cortical opacification)	6.82	0.34–138	0.21
Subcapsular posterior cataract, prevalence	0.00	0.00	1.00
Subcapsular posterior cataract, degree (percentage of lens cortical opacification)	0.00	0.00	1.00
Pseudoexfoliation, frequency	0.00	0.00	1.00
AMD, stage (0–3)	1.32	0.70–2.48	0.40
Diabetic retinopathy	0.00	0.00–0.00	1.00
Open-angle glaucoma	0.00	0.00–0.00	1.00
Angle-closure glaucoma	0.00	0.00–0.00	1.00
Retinal vein occlusion	0.00	0.00–0.00	1.00
Hyperreflective spots in the preretinal vitreous body	0.82	0.45–1.48	0.50
Vitreous body detachment stage	0.97	0.72–1.31	0.83
Hyperreflective spots on retinal surface	2.41	0.78–7.49	0.13
Epiretinal membrane, prevalence	2.46	0.54–11.2	0.24
RPE-Bruch’s membrane band, thickness (µm)	1.44	1.06–1.96	0.02
Photoreceptor outer segment thickness (µm)	0.88	0.77–1.01	0.07
Ellipsoid zone thickness (µm)	0.93	0.34–2.54	0.88
Retinal outer nuclear layer thickness (µm) (including Henle’s fiber layer, outer nuclear layer, external limiting membrane, and myoid zone)	0.93	0.88–0.99	0.02
Outer plexiform layer thickness (µm)	1.25	1.11–1.40	<0.001
Inner nuclear layer thickness (µm)	1.00	0.82–1.23	0.98
Inner plexiform layer thickness (µm)	1.17	1.04–1.33	0.01

In the multivariable analysis, we dropped the following parameters owing to missing statistical significance: body hip circumference (P = 0.73), body waist circumference (P = 0.08), outer plexiform layer thickness (P = 0.69), outer nuclear layer thickness (P = 0.18), inner plexiform layer (P = 0.09), and cortical cataract prevalence (P = 0.21). We then added the following parameters to the model: RNFL thickness (P = 0.004) and axial length (P = 0.02). In the final model, greater PHOMS prevalence was associated with a smaller optic disc size (P < 0.001), thicker peripapillary RNFL (P = 0.004), thicker retinal pigment epithelium–Bruch’s membrane complex thickness (P = 0.04), and longer axial length (P = 0.02) (Table 4). If the parameter of axial length was replaced by the parameter of refractive error, a lower (i.e., more myopic) more myopic refractive showed a tendency of a correlation with a higher PHOMS prevalence in the model (OR, 0.72; 95% CI, 0.45–1.13; P = 0.15). If we then added to the model separately the parameters of best-corrected visual acuity (P = 0.32), perimetric defect (P = 0.77), optic disc fovea distance (P = 0.83), optic disc fovea angle (P = 0.18), and estimated cerebrospinal fluid pressure (P = 0.54), the associations with the PHOMS prevalence were not statistically significant.

Table 4.

Associations (Multivariable Analysis) Between the Prevalence of PHOMS and Systemic and Ocular Parameters in Participants of the Beijing Eye Study

Parameter	Odds Ratio	95% CI	P Vale
Optic disc area (mm2)	0.05	0.01–0.28	<0.001
Peripapillary RNFL thickness (µm)	1.13	1.04–1.22	0.004
Retinal pigment epithelium – Bruch’s membrane complex thickness (µm)	1.70	1.19–2.43	0.04
Axial length (mm)	2.14	1.12–4.08	0.02

Discussion

In our population-based study, the prevalence of PHOMS (15/963 [1.6%]) was relatively low. The PHOMS extended usually over 90° or less and did not show a preferential location. A greater PHOMS prevalence was associated with smaller optic disc size, thicker peripapillary RNFL, thicker retinal pigment epithelium–Bruch’s membrane complex thickness, and longer axial length. PHOMS were not associated with best-corrected visual acuity, perimetric indices, or the prevalence of any other major ocular or systemic parameter or disease examined.

The observations made in our study agree with findings reported form previous investigations, in particular with respect to the association between a greater PHOMS prevalence and a small optic disc. In the population-based Copenhagen Child Cohort 2000 Eye Study, PHOMS were found in 8.9% of 1407 children aged 11 to 12 years, with a PHOMS location predominantly in the superior-nasal region.¹¹ Associated factors were myopia, an optic nerve head tilt, and prelaminar hyperreflective lines on the OCT scans.¹¹ The PHOMS prevalences was not associated with any abnormality, including prenatal factors. The discrepancy between the greater PHOMS prevalence in the children study from Denmark and the lower PHOMS prevalence in our older study cohort may be due to several reasons, in particular owing to differences between the study cohorts in age and ethnicity-related optic disc size.²⁴ In the Beijing Eye Study, the RNFL thickness decreased by approximately 0.5 µm per year of life or 0.36% of an original RNFL thickness of 137 µm in individuals aged 50 years and older.²⁵ The age-related RNFL thinning corresponds with a histomorphometrically examined yearly loss of approximately 0.3% of optic nerve fibers, assessed in individuals aged 50 years and older.²⁶ A lower volume of optic nerve fiber tissue at the optic nerve head may lead to a lower prevalence of PHOMS. Correspondingly, in a recent study by Jørgensen et al.,²⁷ the PHOMS prevalence in 411 eyes with optic disc drusen decreased with older age, and the RNFL decreased from approximately 150 µm in children younger the age of 10 to approximately 70 µm in adults aged 60 years. In addition, Chinese as compared with Caucasians have a larger optic disc as a geometrical reason for a lower density of retinal nerve fibers in the parapapillary region and, thus, lower PHOMS prevalence.²⁴ A further reason may be an age-related difference between children and adults in axial length and globe size, including a longer optic disc–fovea distance. In the Beijing Eye Study, the RNFL decreased with longer axial length, by approximately 2.4 µm for each millimeter enlargement of axial length.²⁵ It again leads to a geometrically decreased RNFL density.

As in the Copenhagen Children Study and in our investigation, myopia or a longer axial length was a factor associated with higher PHOMS prevalence. In addition, optic nerve tilting was a factor in the Copenhagen Study. The reason may be that, in moderately myopic eyes (in contrast with highly myopic eyes), the Bruch’s membrane opening does not yet show an axial elongation-related enlargement, but does experience a shift in direction to the macula.²⁸ It leads to an overhanging of Bruch’s membrane into the nasal intrapapillary compartment and to an absence of Bruch’s membrane in the temporal parapapillary region, that is, the temporal parapapillary gamma zone. Simultaneously, the horizontal optic disc diameter shortens, leading to a vertically oval optic disc shape, a decrease in optic disc area, and a crowding of the peripapillary RNFLs. A markedly oval optic disc has also been termed tilted disc. As in our study, the Copenhagen Study did not find associations between the PHOMS prevalence and systemic abnormalities.

In contrast with our study and the Copenhagen Study, most other previous studies reported on the concurrence of PHOMS with other optic nerve head anomalies and neurological diseases.^{2,4–8,10,14,16–18,29,30} In a hospital-based study, Maalej et al.¹⁶ examined 154 patients with prominent optic discs and found that PHOMS were associated with idiopathic intracranial hypertension in 38% of the cohort, posterior uveitis (11%), optic disc drusen (10%), or tilted optic discs (5%) or were isolated (36%). PHOMS were over-represented in the nasal region (96%). PHOMS location in the superior or inferior parapapillary region was associated with idiopathic intracranial hypertension or optic disc drusen, whereas PHOMS location in the temporal or nasal region was associated with an isolated occurrence. The authors concluded that PHOMS could occur isolated without other optic nerve head abnormalities, in particular if located in the nasal or temporal regions, with a small volume and a stable aspect in follow-up. In another hospital-based investigation, Ahn et al.⁹ reported on the occurrence of PHOMS with a small scleral canal and a concurrence with buried optic disc drusen in children. In a similar manner, Vienne-Jumeau et al.¹⁸ examined 27 patients with concomitant PHOMS and optic disc drusen. These associations between a greater PHOMS prevalence and concurrence of disc drusen agrees with a small optic disc as the main risk factor for both entities.¹⁹ One may infer that it is mainly the small optic disc that may lead to the concurrence. In addition, one may discuss that the space-occupying effect of disc drusen may be an additional factor leading to a peripheral displacement of optic nerve fibers in the parapapillary region. Interestingly, the PHOMS location in our study cohort did not show a preference for any special location. The reason for the discrepancy between the studies in the predominant location of the PHOMS has remained unclear.

Other studies reported about the concurrence of PHOMS not only with optic disc drusen, but also with nonarteritic anterior ischemic optic neuropathy.^6,31 Again, with a small optic disc size being the most important risk factor for both entities, nonarteritic anterior ischemic optic neuropathy, and PHOMS, it may be primarily the small optic disc that led to the concurrence.³²

In a study by Gernert et al.¹⁷ on a large group of patients with a broad spectrum of neurological disorders, PHOMS were detected in 7% of the eyes of patients with a variety of disorders, including neuroimmunological diseases, epilepsy, movement disorders, intracranial hypertension, and inborn errors of metabolism, with a PHOMS occurrence in all subgroups. The PHOMS location was predominantly in the nasal region. The PHOMS volume correlated positively with the opening of Bruch’s membrane and negatively with the age of the cohort after the exclusion of patients with intracranial hypertension. The authors concluded that, owing to the association between larger PHOMS volumes and intracranial hypertension, elevated intracranial pressure is pathogenetically related to the PHOMS occurrence, in particular in eyes with papilledema grade 3 or larger.¹⁷ It may be postulated that the retinal nerve fibers, secondarily thickened owing to the increased cerebrospinal fluid pressure, need more space and may eventually arrange in a PHOMS-like manner. As in previous studies, we detected at maximum one PHOM per eye. It fits with previous reports, in which PHOMS, in contrast with optic disc drusen with potential multiple occurrences per eye, were described to be present as only a single lesion per eye, often having a torus (doughnut) or toroid (a partial torus) configuration.^33,34

When the results of our study are discussed, its limitations have to be taken into account. First, PHOMS have been defined by a peripapillary location, abutting on the retina, a hyperreflectivity on the OCT images, an ovoid shape on linear OCT scans taken through the optic disc center, and mass-like, space-filling structural characteristic of displacing the adjacent retina away from the disc and often expanding behind the retina, sometimes leading to a ski slope sign.^3,33 Although we applied these diagnostic criteria, the detection and delineation of PHOMS from the neighboring tissue in the peripapillary region, in particular in eyes with small (crowded) optic discs, can be difficult, potentially leading to an inaccuracy in the diagnosis of PHOMS. It may be considered that PHOMS can sometimes more easily be detected using a dense optic nerve head OCT scan first, and that then radial OCT scans are used for measuring. Second, the participants in our study were Han Chinese, so that the observations made in our investigation may not be transferred directly to individuals of other ethnicities. Third, the group of individuals included in the present study had a shorter axial length than the remaining group of participants of the Beijing Eye Study (23.05 ± 1.02 mm vs. 23.31 ± 1.17 mm). Because the PHOMS prevalence was associated (to a minor degree) with axial length, a selection bias might have occurred leading to an underestimation of the PHOMS prevalence. Differences in other parameters between the included and not-included individuals may not have influenced the results of our study because the PHOMS prevalence was not related to these parameters. Fourth, we did not apply a special OCT protocol for examining the optic nerve head, so that relatively small lesions such as single hidden optic disc drusen might have remained undetected. Because PHOMS usually have a size of at least 30°, it is unlikely that some small PHOMS went undetected.

Conclusions

PHOMS were relatively rare in the adult and elderly population of our Chinese cohort and were mainly associated with a small optic disc. Not related with a decrease in visual function and peripapillary RNFL thickness, they did not indicate an optic nerve damage. Unrelated to other major ocular and systemic diseases, they might be due to a localized crowding of peripapillary nerve fibers in eyes with small (crowded) optic discs.