Correlation Between Visual Symptoms and Optic Disc Drusen Phenotype in a Cohort of Patients

ABSTRACT

Optic disc drusen (ODD) is a slowly progressive, neurodegenerative optic neuropathy with incompletely described symptomatology. We aimed to investigate correlations between visual symptoms, optic nerve head anatomy, and visual function in a cohort of patients with ODD. Patients with optical coherence tomography (OCT)-verified ODD enrolled in a prior study (2017–2024) and consenting to recontact were invited to participate. Participants completed a National Eye Institute Visual Function Questionnaire-25 adapted for ODD (VFQ-25+), systematic OCT scan of the macula and optic nerve head, and a 30-degree automated perimetry. We included 118 patients (234 eyes). Comparing patients with bilateral superficial ODD (n = 46) to those with bilateral deep ODD (n = 26), there were no significant differences in symptoms such as transient visual obscurations (73.9% vs. 73.1%, p = .94) or nyctalopia (78.3% vs. 61.5%, p = .13), nor in VFQ-25+ scores (80.7 vs. 85.8, p = .06). Analyzing individual eyes, eyes with superficial ODD (n = 137) showed significantly lower macular ganglion cell layer and inner nuclear layer volumes and peripapillary retinal nerve fiber layer thickness (p < .0001) and a stronger correlation between perimetric mean deviation (MD) and ODD size (R = 0.75, p = 7.4e-26) compared to eyes with deep ODD (n = 97, R = 0.33, p = .0011). Although superficial ODD was associated with more pronounced declines in visual function, symptom reporting was consistent across ODD locations. Similar VFQ-25+ scores among different patient phenotypes suggest perimetric and retinal changes may not be readily perceived by patients.

Introduction

Optic disc drusen (ODD) are small, acellular, calcified deposits within the optic nerve head, consisting of mucopolysaccharides and mucoproteins.¹ These deposits are typically identified during the second to third decades of life, often incidentally during routine ophthalmological examinations.² ODD exist in about 2% in the general population, with an equal sex distribution and supposedly a higher prevalence among white individuals.³ Although more common in ODD family members,⁴ genetic determinants for ODD formation have not been fully elucidated.⁵ The pathogenesis of ODD unfolds by the abnormal calcification and deposition of mitochondrial remnants within the prelaminar optic nerve head, possibly due to axoplasmic transport disruptions in a congenitally narrow scleral canal.⁶ Using optical coherence tomography (OCT), ODD can be classified according to their anatomical location in relation to Bruch’s membrane opening (BMO), with superficial ODD being above and deep ODD being below BMO.^7,8

Though visual field defects are seen upon examination in up to 87%,^4,9 they are progressing at a slow rate,¹⁰ and the central visual field is rarely affected.² Therefore, the condition is largely asymptomatic.¹¹ A symptom classically associated with ODD is transient visual obscurations (TVOs), which are brief episodes of visual loss lasting seconds. Although commonly reported in textbook materials to occur in about 9% of ODD patients,¹² no universally agreed-upon primary study has specifically investigated the prevalence. Additionally, patients may experience nyctalopia, which, although rare, has been reported in cases where ODD is the only finding despite extensive investigation.^4,13

The prevalence of visual symptoms in ODD patients has not previously been systematically examined and no studies have examined the correlation between visual symptoms and ODD phenotype, that is, mainly deep or mainly superficial ODD. In this study we aimed to examine the presence of visual symptoms in a large cohort of patients with ODD, that previously had been thoroughly phenotyped using state-of-the-art enhanced depth imaging (EDI) OCT of the optic nerve head, and we correlated the visual symptoms with the specific anatomical ODD location.

Methods

In this cross-sectional study, ODD patients of all ages who from 30 November 2017, to 29 February 2024, as part of being enrolled in another ODD study at the Department of Ophthalmology, Rigshospitalet, had consented to being contacted for potential inclusion in a future research project on ODD, were contacted and invited to participate. All included patients consented to participate. The handling and storage of data was approved by the Knowledge Center for Data Reviews of the Capital Region of Denmark (RH-2017–316). All procedures adhered to the tenets of the Declaration of Helsinki.

At the time of inclusion, all patients had already had a detailed spectral domain OCT (Spectralis OCT; Heidelberg Engineering, Heidelberg, Germany) scan of the optic nerve head and macula performed, according to the guidelines provided by the Optic Disc Drusen Studies Consortium.¹⁴ All included patients had ODD in at least one eye. In addition, all included patients had already performed a reliable (reliability factor under 25%) 30-degree automated perimetric (Octopus; Haag-Streit, Switzerland) examination in our department. In case a patient was 12 or below 12 years old at time of last OCT and/or perimetry, a new examination (OCT and/or perimetry) was performed. If the patient was 13 years or older at the time of last OCT and/or perimetry, these examinations were used if the quality of the examinations was sufficiently good. In case of suboptimal quality examinations, patients were invited in for novel examinations to be performed. Patients with ODD-associated complications, and patients diagnosed with ocular comorbidities such as glaucoma, ocular hypertension, vitreous detachment or other retinal/optic neuropathies were also excluded.

Optical coherence tomography analysis

Using the manufacturer software (Heidelberg Eye Explorer, version 1.10.4.0), a single grader (M.B.M) graded all OCT scans. ODD size was measured as the mean length of 3 differently oriented lines drawn from end to end of the largest ODD observed, using the 97-line scans from the dense optic nerve head OCT. The BMO diameter was measured as the mean length of 3 equally spaced radial measurements connecting the 2 endpoints of the Bruch’s membrane on opposing sides, using the 12- or 24-line radial OCT. It was noted for each eye, using the dense optic nerve head EDI-OCT scans, whether the majority of ODD was above or below BMO, classified as “superficial” or “deep” respectively.

Visual symptoms

Included patients were asked to complete a symptomatology questionnaire based on the National Eye Institute Visual Function Questionnaire-25 (VFQ-25) and customized for use in ODD patients with 7 additional questions (VFQ-25+, Supplementary Figure S1). Each answer was numerically validated from 0 to 100 with either 5 or 6 scales method. Hereby, the mean of the responses was calculated for all patients as prescribed in the manual of VFQ-25 calculations score guideline.¹⁵ For the substudy of visual symptoms, patients with superficial ODD in both eyes (termed superficial group) were compared to patients with deep ODD in both eyes (termed deep group). These subgroups were also analyzed with Pearson correlation regarding the possible correlation between VFQ-25+ score per patient and their OCT measurements per eye. Patients with one eye with superficial ODD and one eye with deep ODD (termed mixed group) as well as patients with unilateral ODD were excluded from this substudy.

Statistical analysis

All variables were tested for their individual normal distribution upon analysis using the Shapiro-Wilks test, which showed a skewed population in all variables (p < .0001) but not for the distribution of the BMO diameter variable and variables with percentage values. Therefore, we used the Mann-Whitney U-test to compare all variables except for the BMO diameter, which was analyzed using a t-test, and the percentage-based variables, which were assessed using a z-test. Additionally, different variables underwent Pearson correlation under subset of ODD-location resulting in Pearson correlation coefficient (R) and a p-value regarding the correlations validity.

Results

A total of 118 patients (234 eyes) were included in the study. Of the 118 patients, according to ODD location, 46 were in the superficial group, 26 were in the deep group, and 46 were in the mixed group. No significant differences were observed in age between the superficial and the deep group (Table 1).

Table 1.

Visual symptoms in relation to optic disc drusen anatomical localization.

Measurements	Total	Superficial group (Mean±SD)		Deep group (Mean±SD)		p values
Overview
ODD location (patients)	72.00	46.00		26.00
Age (years)		45.6 ± 18.1		37.8 ± 18.3		p = .30 κ
Sex-ratio (M/F,1/2)		1.89		1.85
Visual Symptoms
VFQ-25+ score (0–100)		80.7 ± 11.7		85.8 ± 8.8		p = .06 ψ
AQ1: TVO (%)		73.9		73.1		p = .94 γ
AQ2: Peripheral vision loss (%)		58.7		46.2		p = .30 γ
AQ3: Photopsia (%)		60.9		50.0		p = .37 γ
AQ4: Nyctalopia (%)		78.3		61.5		p = .13 γ
AQ5: Perceived visual field defect (%)		37.0		7.7		p < .01 γ
AQ6: Reduced contrast perception (%)		47.8		42.3		p = .65 γ
AQ7: Floaters/shadowing (%)		52.2		38.5		p = .26 γ

Abbreviations: ODD = optic disc drusen; SD = standard deviation; AQ = additional question; TVO = transient visual obscurations; ψ = Mann-Whitney U Test; κ = Kolmogorov-Smirnov test; γ = Z-test.

Visual symptoms

Patients reported a wide variety of symptoms (Table 1), the most frequently reported being transient visual obscurations (73.91 and 73.08% in the superficial and the deep group, respectively) and nyctalopia (78.26 and 61.54% in the superficial and the deep group, respectively). The overall VFQ-25+ scores (80.71 and 85.77), reflecting perceived visual function, did not significantly differ between the superficial and the deep group (p = .06). Patients in the superficial group reported a higher frequency of perceived visual defects (36.96%) than patients in the deep group (7.69%), p < .01. For the remaining symptoms, there was no statistical difference between the two groups.

Correlations between ODD location and retinal layer thicknesses

On OCT, of the 234 eyes with ODD, 137 had superficial ODD while 97 had deep ODD. Eyes with superficial ODD had greater alterations in retinal layers compared to eyes with deep ODD (Table 2). The peripapillary retinal nerve fiber layer (pRNFL) thickness, macular ganglion cell layer (mGCL) volume, and macular inner plexiform layer (mIPL) volume, were all significantly lower in eyes with superficial ODD than in eyes with deep ODD (all p-values < 0.0001), whereas the macular inner nuclear layer (mINL) was significantly thinner in eyes with deep ODD compared to eyes with superficial ODD (p < .01, Table 2).

Table 2.

Comparison of eyes with superficial and eyes with deep optic disc drusen.

Measurements	Total	Superficial.(Mean±SD)	Deep (Mean±SD)	p values
Overview
ODD location (eyes)	234	137	97
Patients	118	69	49
Age (years)	.36	46.1 ± 18.7	41.9 ± 19.4	p = .34 κ
Sex-ratio (M/F,1/2)		1.79	1.73
OCT and Perimetry
ODD size (µm)		481.0 ± 271.4	359.3 ± 215.8	p < .001 ψ
Retinal volume (mm3)		8.45 ± 0.71	8.67 ± 0.56	p < .001 ψ
mGCL volume (mm3)		0.94 ± 0.18	1.09 ± 0.15	p < .0001ψ
mIPL volume (mm3)		0.82 ± 0.12	0.90 ± 0.11	p < .0001ψ
mINL volume (mm3)		1.02 ± 0.10	0.99 ± 0.07	p < .01 ψ
mOPL volume (mm3)		0.79 ± 0.07	0.79 ± 0.06	p = .52 ψ
mONL volume (mm3)		1.81 ± 0.25	1.77 ± 0.23	p = .16 ψ
mRPE volume (mm3)		0.38 ± 0.03	0.38 ± 0.03	p = .37 ψ
pRNFL thickness (µm)		73.23 ± 24.64	94.96 ± 24.47	p < .0001ψ
BMO diameter (µm)		1,534 ± 184	1,426 ± 130	p < .0001 φ
Perimetric Mean Deviation (dB)		7.91 ± 6.23	3.79 ± 3.99	p < .0001ψ

Abbreviations: SD = standard deviation; mGCL = macular ganglion cell layer; mIPL;macular inner plexiform layer; mINL = macular inner nuclear layer; mOPL = macular outer plexiform layer; mONL = macular outer nuclear layer; mRPE = macular retinal pigment epthelium; pRNFL = peripapillary retinal nerve fiber layer;BMO = Bruch’s membrane opening; sq. = square root; κ = Kolmogorov-Smirnov test; ψ = Mann-Whitney U Test; φ = t-test.

Correlation between visual symptoms and OCT measurements

When examining the relationship between VFQ-25+ scores and OCT measurements of mGCL volume, the analysis revealed differing results based on drusen depth. For superficial ODD, a lower mGCL volume correlated significantly with more reported symptoms (R = 0.38, p = .00017), while this correlation did not reach statistical significance for deep ODD (R = 0.23, p = .11), Figure 1. Correlation analysis between VFQ-25+ scores and pRNFL thickness revealed a modest but statistically significant correlation for the superficial ODD group (R = 0.23, p = .026), and no correlation for the deep ODD group (R = 0.01, p = .95), Figure 2.

Figure 1.

Correlation between VFQ-25+ score and macular ganglion cell layer volume in eyes with superficial optic disc drusen (A) and in eyes with deep optic disc drusen (B). mGCL = macular ganglion cell layer; R = Pearson correlation coefficient; p = p-value.

Figure 2.

Correlation between VFQ-25+ score and peripapillary retinal nerve fiber layer thickness in eyes with superficial optic disc drusen (A) and in eyes with deep optic disc drusen (B). pRNFL = peripapillary retinal nerve fiber layer; R = Pearson correlation coefficient; p = p-value.

Correlation between visual field and OCT measurements

Strong correlations were observed between perimetric mean deviation (MD) and ODD size, more pronounced in the superficial ODD group (R = 0.75, p < 7.4e-26) than in the deep ODD group (R = 0.33, p = .0011), Figure 3. Additionally, Pearson correlations between ODD size and pRNFL thickness revealed a stronger negative correlation in superficial ODD (R = −0.60, p = 1.2e − 14) than in deep ODD (R = −0.45, p = 4.8e − 06), Figure 4. Similarly, correlations between ODD size and mGCL volume showed a stronger negative correlation in superficial ODD (R = −0.58, p = 6.1e − 14) compared to deep ODD (R = −0.45, p = 3.6e − 06) as represented in Figure 5.

Figure 3.

Correlation between perimetric mean deviation and size of optic disc drusen in eyes with superficial optic disc drusen (A) and in eyes with deep optic disc drusen (B). ODD = optic disc drusen; R = Pearson correlation coefficient; p = p-value.

Figure 4.

Correlation between size of optic disc drusen and peripapillary retinal nerve fiber layer thickness in eyes with superficial optic disc drusen (A) and in eyes with deep optic disc drusen (B). pRNFL = peripapillary retinal nerve fiber layer; ODD = optic disc drusen; R = Pearson correlation coefficient; p = p-value.

Figure 5.

Correlation between size of optic disc drusen and macular ganglion cell layer volume in eyes with superficial optic disc drusen (A) and in eyes with deep optic disc drusen (B). mGCL = macular ganglion cell layer; ODD = optic disc drusen; R = Pearson correlation coefficient; p = p-value.

Discussion

The findings from this study offer insights into the differential impacts of superficial and deep ODD on retinal structure and visual symptoms. Despite the lack of significant differences in the prevalence of frequently reported symptoms such as transient visual obscurations and nyctalopia, as well as in VFQ-25+ scores across groups, structural and functional discrepancies were evident in OCT and perimetry analyses. This suggests that while patient-reported symptoms and overall perceived visual quality might not differ significantly between patients with a majority of superficial ODD versus patients with a majority of deep ODD, the only reported symptom to be significantly different between groups being “perceived visual field defect,” the underlying retinal changes and their functional consequences do vary.

Nyctalopia, typically a rare symptom in ODD, appeared with notable frequency in our cohort. This is an important observation as nyctalopia is often associated with disorders with more extensive retinal damage, such as retinitis pigmentosa. The presence of this symptom could theoretically indicate a more significant retrograde impact of ODD on retinal architecture and thus function than previously understood. Interestingly, ODD are about 15 times more prevalent in patients with retinitis pigmentosa, where nyctalopia is a common symptom, compared to the background population.¹⁶

Transient visual obscurations, characterized by brief episodes of visual loss lasting seconds, are classically associated with ODD and have previously been thought to occur in about 9% of cases.¹² The very high prevalence of this symptom in our cohort may be due to the fact that we only examined patients with bilateral ODD, whereas previous reports may have included unilateral cases as well.

Superficial ODD were associated with overall worsened perimetric results and more significant structural alterations in the retina, including notable thinning in pRNFL, mGCL and mIPL (Table 2). These findings are consistent with previous studies that suggest that ODD can affect the extent of structural disruption,¹⁷ likely due to variations in mechanical compression and vascular compromise at the optic nerve head, leading to retrograde degeneration.¹⁸ These results also support, using more advanced methods of diagnosing and quantifying ODD, earlier reports about the lower impact of deep ODD compared with superficial ODD on pRNFL, mGCL and mIPL thicknesses in smaller cohorts.^7,19,20

Interestingly, the stronger correlations were observed between the size of superficial ODD and perimetric MD, mGCL volume and pRNFL thickness, compared to the size of deep ODD, emphasizing the impact of ODD location on visual field impairments and retinal thinning. The higher correlation coefficient in the superficial ODD group (R = 0.75, −0.58, −0.60) compared to the deep ODD group (R = 0.33, −0.45, −0.45) underscores the clinical significance of phenotyping ODD characteristics in the management of patients with ODD. This is a particularly intriguing finding considering that previous studies have reported a higher prevalence of non-arteritic anterior ischemic optic neuropathy (NA-AION) in patients with deep ODD.^21,22

The weaker correlation between VFQ-25+ scores and mGCL volume in patients with deep ODD compared to those with superficial ODD further indicates that deep ODD might be less bothersome to patients despite their presence, which could be due to the more complex extent of retinal disruption they cause. The comparable measurements across ODD-location groups could be useful in counseling patients about their condition and expected outcomes, especially in terms of visual quality and potential progression, which needs a follow up study for clarification.

A strength of this study is the use of advanced imaging techniques such as EDI-OCT, which provides precise classification and measurements of ODD. This approach yields detailed insights into the structural changes associated with different anatomical ODD locations. Additionally, the large and well-defined cohort enhances the reliability of the findings.

The cross-sectional design of the study limits the ability to understand the long-term impact and progression of ODD on visual symptoms. Longitudinal data would provide deeper insights into the natural history of the condition and the durability of the observed symptoms. Larger sample sizes could provide a more robust statistical power to detect smaller effects, particularly in subgroup analyses where the number of subjects might be considered limited.

In conclusion, this study highlights the importance of considering ODD phenotyping in the diagnostic process and in predicting visual symptoms for patients with ODD. These results also suggest potential opportunities for future research, including the exploration of underlying mechanism that could cause the specific impacts of superficial versus deep ODD on the retina, visual function and ODD symptoms in general.

Funding Statement

The VELUX Foundation, Lundbeck Foundation, Framework Grant from Rigshospitalet. The sponsor or funding organization had no role in the design or conduct of this research; Synoptik-Fonden, Velux Fonden.

Disclosure statement

No potential conflict of interest was reported by the author(s).

Meeting presentation

The study was presented at the North American Neuro-Ophthalmology Society (NANOS) Annual Meeting, 2025, Tucson, Arizona, United States.

Supplementary material

Supplemental data for this article can be accessed online at https://doi.org/10.1080/01658107.2025.2553584.