Upper Limit of Retinal Nerve Fibre Layer Thickness in Patients with Pseudopapilloedema

ABSTRACT

An initial misdiagnosis of papilloedema in a patient with optic nerve head swelling can be anxiety-provoking and may result in unnecessary, invasive, and costly tests. Cirrus high definition, spectral domain-optical coherence tomography (Cirrus HD-OCT) may provide a rapid and non-invasive test. We sought to determine an upper limit of average retinal nerve fibre layer (RNFL) thickness in patients with pseudopapilloedema without visible drusen using Cirrus HD-OCT that could be utilised in conjunction with the clinical presentation and physical examination when managing patients with optic nerve head swelling. Inclusion criteria consisted of at least two neuro-ophthalmological visits and repeated imaging of the optic nerve head with Cirrus HD-OCT at least 6 months apart. Exclusion criteria included clinically visible drusen along with previous or concomitant diagnosis of retinal or other optic nerve pathology. Thirty-eight eyes from 19 patients with pseudopapilloedema were included in this study. The upper limit of average RNFL thickness was defined as two standard deviations above the mean of the average RNFL thickness and was calculated to be 158.65 µm for scans obtained with Cirrus HD-OCT devices. A patient with suspected optic nerve head swelling, an average RNFL thickness less than 158.65 µm, and no other evidence of papilloedema or neurological signs or symptoms can be managed with serial follow-ups with OCT imaging for at least 6 months. If the patient continues to have no clinical symptoms suggesting increased intracranial pressure and the average RNFL thickness is stable, the likelihood of papilloedema is minimal.

Introduction

Differentiating pseudopapilloedema from true papilloedema in patients with suspected pathological optic nerve head swelling is challenging. Papilloedema, occurring secondarily to increased intracranial pressure (ICP), can have serious medical consequences. On the other hand, although patients with pseudopapilloedema can also develop increased ICP, pseudopapilloedema is a benign condition characterised by congenital optic nerve crowding that may be associated with optic nerve head drusen (ONHD). Pseudopapilloedema without visible optic nerve head drusen is even more commonly misdiagnosed as papilloedema.^1,2

Many patients, especially paediatric patients, with pseudopapilloedema have been commonly mistaken to have papilloedema.³ An initial mischaracterisation of pseudopapilloedema as true papilloedema can induce anxiety in patients and their parent. It can also lead to many unnecessary, invasive and expensive tests such as magnetic resonance imaging (MRI), magnetic resonance venography (MRV) and lumbar punctures (LP), sometimes with falsely elevated opening pressures.⁴ For example, in one study, 28 patients with pseudopapilloedema were mistaken to have had papilloedema over a 3-year period and underwent excessive testing/procedures and many consultations before being confirmed to have pseudopapilloedema.⁵ Therefore, an accurate and quick method to diagnose pseudopapilloedema is needed. Optical coherence tomography (OCT), a fast and non-invasive imaging technique used to obtain cross-sectional images of the retina and the optic nerve, may be useful in distinguishing between possible causes of optic nerve head swelling.^6,7 Using OCT, previous investigators have shown increased swelling of the peripapillary retinal nerve fibre layer over time in patients with papilloedema, as opposed to pseudopapilloedema. Increased subretinal fluid near the optic nerve head has also been seen in patients with papilloedema.^8–11 However, an upper limit of normal or ‘cut-off’ RNFL thickness has not been determined.

The purpose of this study was to evaluate the RNFL thickness in patients with pseudopapilloedema without visible optic nerve head drusen using Cirrus high definition (HD), spectral domain OCT (Cirrus HD-OCT). With these data, we aimed to calculate an upper limit for average RNFL thickness in these patients in order to propose recommendations on how to better approach patients with suspected papilloedema.

Materials and methods

This study was approved by the Institutional Review Board of the Tufts University School of Medicine. It adheres to the tenets of the Declaration of Helsinki and is in accord with the Health Insurance Portability and Accountability Act regulations. Informed consent for this study was considered exempt by the Institutional Review Board due to its retrospective design.

This is a retrospective, longitudinal, observational case series. Data were collected from the database of patients of the Neuro-ophthalmology Service at New England Eye Center at Tufts Medical Center in Boston, Massachusetts, during a 10-year period from 2010 to 2020. Keyword searches of patients in the database include pseudopapilloedema, papilloedema, ONHD and congenitally crowded nerves. All patients included in this study were referred to the Neuro-ophthalmology service due to suspicion of papilloedema. All patients provided detailed medical and ocular history and underwent complete neuro-ophthalmological evaluation including slit-lamp examination, automated visual field testing, OCTs and colour fundus photography.

Patients with prior ocular surgeries, co-existing retinal or other optic nerve pathologies, obvious optic nerve head drusen on fundus examination, age >50 years and follow-up of less than 6 months were excluded from the study. Inclusion criteria consisted of at least two neuro-ophthalmological visits with repeated images of the optic nerve head using Cirrus HD-OCT (version 7.0.319, Carl Zeiss Meditec, Inc., Dublin, CA, USA) more than 6 months apart. Patients with elevated optic discs, blurred optic disc margins, anomalous vascular configuration without presence of cupping, visible drusen or symptoms/signs of papilloedema, such as headaches or LP confirmed increased opening pressures, in addition to minimal variability in average RNFL thickness during serial follow-up visits were determined to have pseudopapilloedema.

Chart review and analysis were done by two observers, and findings consistent with pseudopapilloedema were recorded.

OCT measurement and analysis

Data from longitudinal Cirrus HD-OCT imaging with cube scans consisting of 512 A-scans × 128 B-scans centred on the optic disc and an HD 5-line raster scan centred on the optic disc of both eyes on Cirrus HD-OCT were collected. The 5-line raster protocol is a 6 mm line scan that acquires 4096 A-scans in each of the five lines. The enhanced depth imaging (EDI)-OCT on the Cirrus device was not part of the imaging protocol as it was not available during the time when most of these scans were obtained. OCT was done at each visit to qualitatively and quantitatively identify features of pseudopapilloedema.

Two independent reviewers analysed and reviewed the morphology of RNFL thickness on OCT. RNFL thickness analysis was performed on each eye of all patients. Average RNFL thicknesses at baseline along with average RNFL thickness at subsequent follow up visits were recorded and compared longitudinally and across all patients.

In order to evaluate the variation in average RNFL thickness present on serial OCT imaging, one patient was analysed in detail. Twelve scans were obtained from each eye during one visit. Scans 1–3 were acquired sequentially without the patient repositioning. Scans 4–6 were repeated with the patient repositioning after each scan. Scans 7–9 were repeated with the patient fixating at the temporal aspect of the fixation target and scans 10–12 were repeated with the patient fixating at the nasal aspect of the fixation target.

Statistical analysis

The average RNFL thicknesses, baseline age and follow-up duration of all patients were recorded and analysed. The minimum, maximum, mean, and standard deviation of these quantitative variables were calculated. Using these values, the upper limit of RNFL thickness, defined as two standard deviations above the mean average RNFL thickness, was determined.

Results

Twenty-five patients diagnosed with pseudopapilloedema were identified in the data base between 2010 and 2020. Six of these patients were found to eventually develop papilloedema superimposed on pseudopapilloedema and were excluded. The final results and analysis presented are from 38 eyes from 19 patients with pseudopapilloedema. Eight of these patients had MRIs that were normal. One patient underwent an LP and was found to have a normal opening pressure. The mean age of the patients was 29.47 years (range 9–43 years), and 75% of the patients were females. The average duration of follow-up was 17.35 months. Patient demographics are presented in Table 1. Slit-lamp examination, automated visual field testing, and colour fundus photography showed minimal differences between visits.

Table 1.

Patient demographics.

	Pseudopapilloedema
Number of subjects	38 eyes (from 19 patients)
Mean age at initial visit in years (range)	29.47 (9–43)
Gender (females:males)	15:4
Mean number of clinic visits (range)	3.21 (2–5)
Mean duration of follow-up in months (range)	17.35 (6–68)

The upper limit of average RNFL thickness of these patients was calculated to be 158.65 µm, which is two standard deviations above the mean. Serial OCTs of all patients showed a mean difference of −4.68 µm in average RNFL thickness over longitudinal follow-up between 6 and 68 months (Figures 1 and 2).

Figure 1.

Trend of average retinal nerve fibre layer (RNFL) thickness of 18 eyes with pseudopapilloedema who had longitudinal follow-up of 6–11 months.

Figure 2.

Trend of average retinal nerve fibre layer (RNFL) thickness of 20 eyes with pseudopapilloedema who had longitudinal follow-up of 12–68 months.

Based on the standard deviation of the difference in average RNFL thickness over longitudinal follow-up seen in this cohort, we determined the average RNFL to be ‘stable’ if this difference was within two standard deviations (±23.44 µm). Table 2 describes the statistical analyses of this group.

Table 2.

Statistics on average retinal nerve fibre layer thickness in pseudopapilloedema patients included in this study (n = 38 eyes, 19 patients).

	Initial mean RNFL thickness (µm)	Difference in mean RNFL thickness from initial to final visit (µm)
Minimum	63	−53**
Median	106	1.00
Mean	107.05	−4.68
Maximum	165	8
Standard deviation	25.8	11.72

**This minimum was a decrease in mean RNFL thickness over 4 years. Maximum increase in mean RNFL thickness was 8 μm.

RNFL = retinal nerve fibre layer.

For the pseudopapilledema patient who was analysed in detail, 12 scans were obtained from each eye during one visit as described above. Colour fundus photographs taken 1 year apart showed no changes in optic nerve head elevation (Figure 3). However, the variation in average RNFL thickness was found to be 7 µm in the right eye and 9 µm in the left eye (Figures 4 and 5). This case demonstrates how optic nerve head swelling itself, along with variability in the OCT technique, can cause measurement changes from scan to scan.

Figure 3.

Colour fundus photographs of a psuedopapilloedema patient’s right (OD) and left (OS) optic nerve heads at the initial visit and at a follow-up visit 1 year later. No visible drusen are seen and no change in optic nerve head swelling is seen during follow-up.

Figure 4.

The retinal nerve fibre layer (RNFL) thickness presented as RNFL thickness maps showing the variability in measurement over 1 year of follow-up for the right (OD) and left (OS) eyes.

Figure 5.

Variability in average retinal nerve fibre layer (RNFL) thickness from 12 scans from one patient with pseudopapilloedema taken at a single visit. Scans 1–3 were acquired sequentially without the patient repositioning. Scans 4–6 were repeated with the patient leaning back and then being reoriented before each scan. Scans 7–9 were repeated with the patient fixating temporally and scans 10–12 were repeated with the patient fixating nasally.

Discussion

In order to avoid unnecessary stress for patients and their families, in addition to expensive, invasive testing and referrals, a standardised approach to evaluate patients suspected to have papilloedema is needed. We suggest OCT as a method to initially differentiate between patients who require further testing to rule out papilloedema and patients with pseudopapilloedema. Using Cirrus HD-OCT, we determined the upper-limit of average RNFL thickness of patients with pseudopapilloedema in our study to be 158.65 µm with a mean difference of −4.68 µm and a maximum increase in RNFL thickness of 8 µm over longitudinal follow-up. The normal parameters of Cirrus SD-OCT for average RNFL thickness are 75–107.2 µm.¹² In our study, the mean of RNFL thickness in patients with pseudopapilloedema were above these normal parameters of Cirrus SD-OCT. Although an average RNFL thickness of 158 µm can also be due to papilloedema, this upper limit can be used as a starting point to determine whether a patient who presents to clinic due to suspected papilloedema requires further testing or can have routine follow-up with OCT.

Recently, various studies have attempted to determine the most useful imaging modality in differentiation of pseudopapilloedema and papilloedema. For example, B-scan ultrasonography has been suggested as one of the best methods for diagnosing patients with pseudopapilloedema and specifically, patients with pseudopapilloedema and buried drusen.^13,14 The cost-effectiveness of using B-scan ultrasonography to make a diagnosis before neuroimaging or additional invasive tests in patients suspected to have papilloedema has also been documented.¹⁵ Although B-scan ultrasonography has been shown to be useful, it is not available in all clinics and requires a trained technician.^13,16 Unlike B-scan ultrasonography, OCT devices are widely available and easy to use. Because it is non-invasive and can be used as an initial test like B-scan ultrasonography, we can also assume that it would be as cost-effective. Additionally, EDI-OCT has been shown to provide unparalleled optic nerve head drusen visualisation and is able to differentiate pseudopapilloedema from papilloedema with high specificity and sensitivity.^16–18

Other imaging modalities have also been studied to determine their accuracy in differentiating pseudopapilloedema from papilloedema. One study, after comparing the utility of B-scan ultrasonography, fundus photography, autofluorescence, fluorescein angiography (FA) and OCT, found that FA was the most accurate in classification of pseudopapilloedema or papilloedema in children.¹⁹ However, FA is an invasive procedure that involves the injection of dye into a patient’s blood vessel by a trained professional.²⁰ This dye can result in nausea and allergic reactions in addition to anaphylaxis in very rare cases.^21,22 Furthermore, sometimes it is difficult to clearly differentiate staining of a crowded disc from leakage from an oedematous disc.

Although other studies have suggested that average RNFL thickness assessed with OCT is unreliable in differentiating papilloedema from pseudopapilloedema, these studies failed to mention whether the patients with papilloedema presented with clinical signs such as headache, nausea or vomiting.^19,23 For example, one study in which OCT and ultrasonography were analysed independently showed both to be insufficient in diagnosing all cases of papilloedema; all of the misdiagnosed cases were symptomatic.²³ Evaluation of clinical findings plays a critical role in diagnosing these patients accurately.^24,25 In our study, patients with any clinical signs of intracranial pressure were excluded unless they had a normal opening pressure on LP. It is important to avoid discounting a patient’s history when determining the accuracy of a diagnostic method as these methods should always be used alongside the clinical evaluation.

Based on our findings, when ophthalmologists or neurologists encounter a new patient with suspected papilloedema, we recommend incorporating the use of SD-OCT as part of the initial workup. The results of the SD-OCT can be used in conjunction with a thorough history and focused physical examination to determine further workup. If the patient’s RNFL thickness is less than the calculated upper limit of pseudopapilloedema, 158.65 µm, and the patient is asymptomatic, serial follow-up is recommended for at least 6 months with repeated complete ophthalmological examination and OCT imaging. If the difference in average RNFL thickness is less than approximately 23.44 µm at the follow-up visit and the patient has had no symptoms of increased intracranial pressure, the likelihood that the patient has pseudopapilloedema will be increased. However, if the average RNFL thickness of initial visit is above the determined upper limit and/or the patient is symptomatic, additional diagnostic tests may be necessary to rule out true papilloedema.

The limitations of this study include the small sample size due to it being a single centre study and the exclusion of many participants because of the lack of adequate follow-up. We also only obtained imaging using a Cirrus HD-OCT device and therefore, the upper limit of average RNFL thickness calculated in this study can only be utilised when patients are imaged with Cirrus HD-OCT devices. There is poor reproducibility of average RNFL thickness measurements across different OCT devices. Other limitations include lack of standardisation of reported clinical findings and tests due to the nature of retrospective studies, variability in diagnostic modalities, and difference in timing of OCT imaging between both groups. Additionally, OCT imaging itself has limitations. For example, inherent variation in average RNFL thickness can be present between multiple scans of the same patient taken by the same operator.²⁶ Differences in quantitative measurements can also exist in scans taken by different operators and in scans of the same patient obtained from different OCT devices.^26,27 Another limitation of OCT is the presence of image artefacts, which can affect RNFL thickness measurements.²⁸ These can be due to acquisition error, patient motion or improper fixation in addition to segmentation errors as a result of software algorithm failures.²⁹ Due to these limitations, variability in average RNFL thickness measurements at different time points in the same patient may be misleading. Therefore, it is important that OCT scans are carefully interpreted and used only in conjunction with the patient’s history and clinical examination findings. In fact, optic nerve head swelling itself can create artefacts that affect accurate nerve fibre layer analysis.⁸ To demonstrate an example of this limitation, we included one case of a patient with pseudopapilloedema who underwent multiple scans at the same visit. After repeated scans with the patient sitting still between scans, repositioning between scans, and looking at the temporal or nasal aspect of the fixation target, we found that the average RNFL thickness varied between 1 and 9 µm. Therefore, an increase in average RNFL microns over 9 µm may be indicative that the patient is more likely to be diagnosed with papilloedema.

Furthermore, a subset of patients initially diagnosed with pseudopapilloedema were excluded from this study due to these patients eventually developing papilloedema superimposed on their pseudopapilloedema. This finding emphasises the importance of follow-up with OCT in patients diagnosed with pseudopapilloedema.

In this study, the variability and upper limit of average RNFL thickness of patients with pseudopapilloedema was calculated and used to recommend a standardised method in managing the care and determining the diagnosis of patients who present with elevated optic nerve head. Based on these findings, we believe that utilising Cirrus HD-OCT in addition to accurate history and focused physical exam for initial evaluation and follow-up of these patients may promise a rapid and easy alternative to the use of unnecessary, expensive neuroimaging or invasive procedures.

Funding Statement

The work was supported by the Harper-Inglis Memorial for Eye Research, The Peierls Foundation, That Man May See, and Research to Prevent Blindness [Research to Prevent Blindness Challenge Grant].

Disclosure statement

No potential conflict of interest was reported by the authors.

Data availability statement

The data that support the findings of this study are available from the corresponding author, TRH, upon reasonable request.