Understanding pseudopapilledema on spectral domain optical coherence tomography

Abstract

Purpose:

Optic nerve head drusen (ONHD), peripapillary hyperreflective ovoid mass-like structures (PHOMS), and horizontal hyperreflective lines (HHL) are commonly seen in eyes with pseudopapilledema on enhanced depth imaging (EDI) spectral domain optical coherence tomography (SDOCT). The objective of this study is to assess the frequency of ONHD, PHOMS, and HHL on spectral domain OCT in the eyes diagnosed to have pseudopapilledema.

Methods:

A retrospective case–control study included patients diagnosed as pseudopapilledema and had EDI SD OCT imaging of the optic nerve head (n = 48 eyes) and controls (n = 20 eyes). OCT scans through the optic nerve head were studied to diagnose ONHD, HHL, and PHOMS. One proportion z test was used to find the difference in proportions.

Results:

Forty eight eyes of 27 subjects were studied. ONHD as described by the optic disc drusen Studies consortium was noted in 19 eyes (39.48%), P value-0.032, PHOMS in 31 eyes (64.6%), P value 0.043, HL in 19 eyes (39.48%), P value 0.032, and none of the normals had ONHD, PHOMS, and HHL.

Conclusions:

PHOMS are more frequently seen than ONHD and HHL in eyes with pseudopapilledema.

Pseudopapilledema has been classically described as a noncongested, irregularly elevated optic disc, which lacks a physiologic cup and which has no pathologic alteration in the retinal blood vessels by Hoyt et al.[1] Misdiagnosing pseudopapilledema as true disc edema may lead to an extensive, invasive, and unnecessary work-up for elevated intracranial pressure, including neuroimaging and lumbar puncture.[2] Optic nerve head drusen can be one of the causes of pseudopapilledema. Although superficial drusen can be detected easily during fundus examination, detecting buried deep drusen requires the use of additional imaging methods such as ultrasonography, fundus fluorescein angiography (FFA), computed tomography, and spectral domain optical coherence tomography (SDOCT). SDOCT can display two-dimensional images comparable to histology as well as three-dimensional videos of ONHD.[3] Recently, the optic disc drusen Studies consortium has described the OCT features of the ONHD.[4] The group has also defined horizontal hyperreflective lines (HHL) on EDI SD OCT as an associated finding in eyes with ONHD. Pseudopapilledema in children has also been studied, and the OCT features like peripapillary hyperreflective ovoid mass-like structures (PHOMS) have been reported to be seen more frequently in these eyes.[5] Despite these studies, there continues to be a little ambiguity regarding the frequency of ONHD, PHOMS, and HHL in patients with pseudopapilledema on SDOCT. The aim of this study is to assess the frequency of ONHD, PHOMS, and HHL in eyes diagnosed to have pseudopapilledema on neuro ophthalmological clinical evaluation. The secondary aim was to study the SDOCT features of ONHD with respect to the disc area, retinal nerve fiber (RNFL) thickness, and ganglion cell complex (GCC) thickness.

Methods

This study was conducted at the outpatient department of Vision and Medical Research Foundation, Sankara Nethralaya, Tamil Nadu, a tertiary referral center in southern India. It is a retrospective case control study from January 2011 to March 2022. Vision research foundation's ethics committee's approval was obtained and the research adhered to the tenets of the Declaration of Helsinki. Inclusion criteria—Patients diagnosed to have ONHD on ultrasound B scan and or seen by the neuro ophthalmologists, and diagnosed to have pseudopapilledema on clinical evaluation were included in the study. Exclusion criteria—Patients with optic neuritis, inflammatory optic neuropathy, ischemic optic neuropathy, and true papilledema were excluded from the study. The clinical profile and the clinical evaluation records of all the included patients were retrospectively analyzed. EDI SD OCT was performed in all the eyes using a CIRRUS 5000 OCT machine. All the OCT images were revaluated by the authors to diagnose the presence of ONHD, PHOMS, and HHL. The eyes diagnosed to have ONHD on OCT were further divided into three groups on the basis of the disc area, Group 1 with the area between 1 and 2 mm², Group 2 with disc area between 2 and 3 mm², and Group 3 with disc area 3-4 mm² and also on the basis of the depth of drusen location as superficial drusen, which were above the Bruch's membrane (BM); deep drusen, which was below the BM and eyes with combined drusen that is both deep drusen and superficial drusen. We studied the location of the drusen in nine sectors (center, superior, superonasal, nasal, inferonasal, inferior, inferotemporal, temporal, and superotemporal sectors). The descriptive statistics were computed for continuous variables, and for qualitative variables, frequency distribution were computed to determine the distribution of those variables. Frequency and percentages were used to assess the distribution of qualitative variables. One proportion z test was used to find the difference in proportions.

Results

The study included 48 eyes of 27 subjects fulfilling the inclusion criteria in the time period between January 2011 and March 2022. Twenty eyes of 10 patients were studied as normal controls. The median age was 31 years, whereas the average age was 35.81 years (13–63 years).

The highest number (n = 10) of subjects were aged between 21 and 30 years. Out of 27 subjects in the study, nine (33%) were males and 18 (67%) were females. The mean best corrected visual acuity (BCVA) for the entire study group was 0.065 log Mar. On fundus examination, the optic disc drusen could be visualized on the optic disc surface in four eyes [Fig. 1a], and in the remaining eyes, an elevated disc was noted which underwent an ultrasound B scan, which revealed a high reflective clump echo noted at optic nerve head persisting in low gain suggestive of ONHD. [Fig. 1b] On SD OCT using Cirrus machine, optic nerve head drusen as described by the optic disc drusen Studies consortium was noted in 19 eyes (39.48%) (P value 0.032) out of the 48 eyes. [Fig. 1c], PHOMS [Fig. 2] were seen in 31 eyes (64.6%) (P value 0.043). High reflectivity horizontal lines with and without ONHD were noted in 19 eyes (39.48%) (P value 0.032) [Fig. 3]. Nine eyes had superficial drusen, five deep drusen, and five combined drusen. None of the eyes in the control group had ONHD, PHOMS, or HHL on OCT imaging. Drusen was found to be located most commonly in the center (5 eyes) followed by the inferotemporal and nasal sector (3 eyes each). Drusen were not seen in the temporal sector in any eye in our series. Among the group with drusen in the center, two eyes had a large conglomerate spreading from center to involve all the quadrants partially and another two eyes had a cluster of grapes configuration. [Fig. 1c] Single drusen was identifiable in 12 eyes; in three eyes, two drusen were noted; and in four eyes, multiple drusen were seen. Five eyes had combined drusen, nine eyes had superficial drusen, and five eyes had deep drusen. Superficial drusen and deep drusen were seen in both males and females with a slightly higher frequency of superficial drusen in females and deep drusen in males. Superficial drusen were more common than deep drusen in all the age groups except in 20–29 years.

Figure 1.

(a) Color fundus photograph: Showing the superficial drusen on the disc surface (b) Ultrasound B-scan: Showing a localized very high reflective echo over the optic nerve head persisting even in the low gain suggestive of optic nerve head drusen (c) Spectral domain optical coherence tomography: Showing a cluster of optic nerve head drusen as structures with a hyporeflective core surrounded by the high reflective margins

Figure 2.

Peripapillary hyperreflective ovoid mass-like structures: Spectral domain optical coherence tomography of the optic nerve head showing the peripapillary hyperreflective ovoid mass-like structure in the nasal quadrant

Figure 3.

Horizontal hyperreflective lines: Spectral domain optical coherence tomography of the optic nerve head showing the horizontal hyperreflective lines below the Bruch's membrane in the absence of peripapillary hyperreflective ovoid mass-like structures and drusen

[Fig. 4a] deep drusen were seen most commonly in the 20–29 years age group, and in 50–59 years of age group, the combined drusen were seen more frequently. The mean age for the superficial drusen was 36.44, deep drusen was 29 and for combined drusen was 48.4 years. Pseudopapilledema was seen bilaterally in 87.5% of eyes and ONHD documented on SDOCT were bilateral in 78.94% eyes. [Fig. 4b] More than three drusen were always a bilateral occurrence. Unilateral drusen were found to be superficial. The mean disc area for the SDOCT detected ONHD of 19 eyes was 2.13 mm². In 9 eyes with superficial drusen, the mean disc area was 1.92 mm² ± 0.297 mm², whereas that for the 5 eyes with deep drusen 2.17 mm2 ± 0.81 mm2 and combined drusen was 2.45 mm 2 ± 0.71 There was no significant difference in the disc areas of superficial and deep drusen (P value. 203) and also between deep drusen with and without superficial drusen. (P value. 292). [Fig. 4c] Both superficial drusen and deep drusen were more common in the Group 1. The difference in the disc area of eyes with superficial drusen (1.92 mm²) and eyes with combined drusen (2.45 mm²) was statistically significant. (P value. 035) Maximum number of eyes were in the group 1 (n = 12) followed by group 2 (n = 4) and 3 (n = 3). Larger disc areas of 3–4 mm² did not have single drusen. [Fig. 4d] The mean disc area in the control group was 1.96 mm². The mean RNFL thickness of the entire study population of cases diagnosed to have pseudopapilledema (n = 48) was divided into 3 groups based on the disc area as defined above. The mean RNFL was highest in the Group 3 in the inferior quadrant and least in the temporal quadrant in all three groups. The mean RNFL thickness was the least in the Group 1 in all quadrants except temporally. [Fig. 5a] The mean RNFL thickness for superficial drusen in all quadrants was 160.56 µm (superior), 102.89 µm (nasal), 164.11 µm (inferior), and 70.05 µm (temporal), Whereas the mean thickness for deep drusen in all quadrants was 168.4 µm (superior), 112.2 µm (nasal), 177.4 µm (inferior), and 84 µm (temporal). There were five eyes with combined drusen. The mean RNFL thickness in this group in all four quadrants was 106.2 µm (superior), 76.2 µm (nasal), 64.6 µm (inferior), and 37.4 µm (temporal).

Figure 4.

(a) Depicting the age and the depth of drusen correlation, (b)Depicting the laterality and number of Drusen correlation (c) Showing the disc area in relation to the depth of drusen, and (d) showing the disc area and number of drusen correlation

Figure 5.

(a) Showing the mean retinal nerve fiber layer (RNFL) thickness in groups on the basis of disc area, (b) Depicting the mean RNFL thickness and depth of drusen correlation, (c) Showing the mean ganglion cell complex (GCC) thickness in groups on the basis of disc area, and (d) Showing the mean GCC thickness and the depth of drusen correlation

The mean RNFL between the superficial and deep drusen was comparable in all four quadrants. (P values superior. 0.844, nasal 0.331. inferior 0.352, and temporal 0.352) Mean RNFL thickness for superficial drusen was low when compared to that for deep drusen in all four quadrants and it was the least for the eyes with combined drusen. Comparison of the mean RNFL thickness in all four quadrants in eyes with deep drusen alone and the eyes with deep and superficial drusen both showed statistical significance (P value. 008) in inferior RNFL and temporal RNFL (P value. 0008) only. (P value superior RNFL.113, nasal RNFL.139) [Fig. 5b]. The mean RNFL in all four quadrants in the control group was 136.9 (superior), 77.75 (nasal), 125.95 (inferior), and 62.3 (temporal). The mean GCC thickness of the entire study population of cases diagnosed to have pseudopapilledema was divided into three groups bases on the disc area as defined above. The mean GCC was the highest in the group 1 in the superonasal sector and least in the superotemporal sector in the Group 3. The mean GCC thickness was the greatest in the group 1 in all sectors except inferiorly.[Fig. 5c] The mean GCC thickness for the superficial drusen in six sectors was 76.22 µm (superior), 85.44 µm (superonasal), 88.11 µm (inferonasal) and 74.89 µm (inferior), 73.11 µm (inferotemporal), and 74 µm (superotemporal), whereas the mean thickness for deep drusen with or without superficial drusen was 91 µm (superior), 92.8 µm (superonasal), 88.2 µm (inferonasal) and 81.6 µm (inferior), 88.4 µm (inferotemporal), and 91.2 µm (superotemporal). And the mean thickness for combined drusen was 59.33 µm (superior), 53 µm (superonasal), 46.67 µm (inferonasal) and 60.33 µm (inferior), 58.33 µm (inferotemporal), and 56 µm (superotemporal). [Fig. 5d] The GCC thickness was found to be low for superficial drusen in all the sectors when compared to that for deep drusen. The GCC thickness was found to be significantly lower in all the sectors in eyes with superficial drusen along with deep drusen than in eyes with only deep drusen. (P value - 0.0143,0.006,0.007,0.049,0.0002,0.002). The mean GCC thickness in all 6 sectors in controls was 84.3 (superior), 85.2 (superonasal), 83.4 (inferonasal), 83.4 (inferior), 83.1 (inferotemporal), and 81.2 (superotemporal).

Discussion

Various imaging techniques have been evaluated to differentiate between papilledema and pseudopapilledema.[6,7,8,9,10,11,12,13,14,15,16,17,18,19] Our study included 48 eyes with a pseudopapilledema with evidence of ONHD on ultrasound in 44 eyes (in the remaining four eyes, the ultrasound was not done as the drusen were visible on the optic disc surface). SDOCT using Cirrus machine revealed the ONHD as described by the optic drusen studies consortium only in 39.48% eyes. PHOMS were seen in 64.6% of eyes and the high reflectivity horizontal lines were noted in 39.48% of eyes. Merchant et al.[20] have reported that the enhanced depth imaging OCT detects ONHD more often than conventional tests like ultrasound. The reason for the discrepancy between the results can be the false positives with the diagnosis of ONHD with USG. Though a previous study by Kurz-Levin et al.[10] states that the chances of false positives with USG are small if USG is performed properly, with the ophthalmoscopic appearance and the clinical presentation taken into account. A recent study by Farazdaghi et al.[21] concluded that optic nerve sheath widening >4.5 mm on USG has the best sensitivity and specificity for diagnosing papilledema. The results of our study and Farazdaghi's indicate, at present, a better support from USG than SDOCT in ruling out true papilledema in patients suspected to have ONHD on clinical evaluation. At the same time, authors understand that the OCT as a technology has a better potential to diagnose pseudopapilledema without ONHD. At this time, we have an OCT description of ONHD but we do not have an OCT description of the pseudopapilledema without ONHD. We need studies with larger number of patients to diagnose this group of patients and come to a consensus regarding the diagnosis of OCT description of pseudopapilledema without ONHD. We found PHOMS to be more frequently associated with pseudopapilledema, a recent paper by Mezad-Koursh et al.[5] also states that PHOMS are seen in 98.4% of children with pseudopapilledema. The authors suspect PHOMS to be a precursor of buried optic nerve head drusen, which can lead to visual field defects, hemorrhages, and choroidal neovascular membranes.

Interestingly, the HHL has also been reported to be the precursors of ONHD as they are also reportedly common in children.[22] They are often found in children with drusen and might represent early drusen in evolution. Skougaard M et al.[23] have reported that superficial drusen were found in older age groups and median age for deep drusen was 21 years.

In our study, we found that the mean age for the presence of deep drusen was 29 ± 13.56 years, for superficial drusen, 36.44 ± 17.38 years, and that for combined drusen, 48.4 ± 12.56 years. These results were similar to that obtained in the above studies. We also found that the combined drusen (both superficial and deep) are seen in the oldest age group suggesting that it may be a continuous process with migration of the ONDH from deep to superficial. The speed at which this process occurs may define the rate at which the retinal fiber layer gets compressed and damage.

We also noticed that the superficial drusen were more frequently seen in females and deep drusen were more frequent in males. Lee KM et al.[18] have reported nasal quadrant was usually involved. We noticed that the ONHD was most commonly located in the center of the optic nerve head (two eyes). We also noticed that it had the least predilection for the temporal sector. Lee KM et al.[18] have reported that, of 99 eyes in their study, 95 eyes had deep and only 4 eyes had visible ONHD. In our study also, four eyes out of 48 eyes had drusen on the disc surface. Skougaard M et al.[23] have reported about 77% of ONHD to be deep in their study of enucleated eyes and 52.9% by Friedman et al.[24] also in their study of cadaveric eyes. In our study, 52.63% of eyes with ONHD detected on SDOCT had deep drusen with or without superficial drusen and 47.37% eyes had only superficial drusen. Our results were similar to that of Friedman et al.[24] and suggest nearly equal affliction of eyes with superficial drusen and deep drusen. Lee KM et al.[18] have reported that 62% of their patients had bilateral ONHD and 38% had unilateral ONHD. Wester et al.[25] have reported that 13 patients (99%), out of the total 14 patients, had bilateral ONHD except one patient. Lasse Malmqvist et al.[26] have reported that 95% of their patients had bilateral ONHD. Štrofová H and Jarošová A[27] have reported that 22 (95.6%) out of 23 patients in their study had bilateral ONHD. In our study 78.94% cases with evidence of ONHD on OCT. Our study showed similar results to the previous studies with a majority of bilateral ONHD. Interestingly, Friedman et al.[24] study on cadaveric eyes revealed a presence of bilateral ONHD only in two cases out of 15. We also observed that 63.16% (12 eyes) of cases had single ONHD. In 15.79% (three eyes), two ONHDs were noted and in 21.05% (four eyes) multiple ONHDs were seen. Friedman et al.[24] reported 41.17% of eyes to have a solitary ONHD and more recently Skougaard M et al.[23] reported single drusen in 64.51% of eyes in their series. G C Sekhar et al.[28] have reported the mean disc area in the south Indian population as 3.37 ± 0.68 mm². Shaun Dacosta et al.[29] have reported the mean disc area in the Indian population as 2.63 ± 0.55 mm². The mean disc area for the SDOCT detected ONHD of 19 eyes was 2.13 mm². Lee KM et al.[18] have reported the mean disc diameter in superficial drusen as 1643 ± 265 µm and that in deep drusen as 1287 ± 185 µm. In our study, we observed the mean disc area for superficial drusen to be 1.92 ± 0.297 mm² and the mean disc area for deep drusen with and without superficial drusen to be 2.31 ± 0. 0.731 mm². There was no statistically significant difference in the mean disc area between the superficial and deep drusen. In eyes with deep drusen only it was 2.17 mm² ± 0.81 mm² and combined drusen was 2.45 mm 2 ± 0.71. Eyes with combined drusen had a larger disc area than eyes with superficial drusen. Flores-Rodríguez et al.[30] have depicted optic disc area measurement to be smaller in ONHD compared to healthy subjects with fundus photography and vice versa with SDOCT. Probably, their study included more eyes with combined drusen where the average disc diameter is larger than that of eyes with superficial drusen. In our study, also the mean disc area of the eyes with ONHD was larger than the mean disc area of the control group. Gutenberg Health Study elucidated a correlation between the smaller disc size and optic nerve head drusen.[31] It found a lower prevalence of ONHD in those with Chinese or African heritage while Caucasians tended to have an increased prevalence. The reason was suggested to be due to optic disc diameter, which is known to be smaller in Caucasians compared with other ethnicities. The study was based on the fundus photography. The explanation for this contradictory results on fundus photography and OCT might be that the fundus photography does not capture the surface area of the lumpy bumpy disc as accurately as the OCT. Lee KM et al.[18] have reported that there was no significant difference in the average RNFL thickness of all the quadrants for superficial drusen and deep drusen. Bassi ST and Mohana KP[32] have reported that for the cases with pseudopapilledema, average RNFL thickness in the superior quadrant was 167.2 ± 52.6, in the inferior quadrant was 153, in the temporal quadrant was 75, and that in the nasal quadrant was 100.55 ± 37.49. The overall average RNFL thickness was 122.3 µm. Lenworth N Johnson et al.[33] have reported average RNFL thickness for ONHD in the superior quadrant to be 121.7, in the nasal quadrant to be 78.6, in the inferior quadrant to be 153, and that in the temporal quadrant to be 85 µm. Alfonso Casada et al. have reported that 4% of deep drusen showed abnormally reduced average RNFL thickness, overall 33% of cases showed thinning of RNFL and that the mean RNFL thickness was 88.8 ± 24.6 µm.[34] Lasse Malmqvist et al.[26] have reported significantly thinner peripapillary RNFL thickness in cases of superficial ONHD. Mean RNFL thickness for superficial drusen was low when compared to that for deep drusen in all four quadrants. The RNFL thickness in all four quadrants in combined drusen was significantly lesser when compared to the deep drusen in the respective four quadrants. The reason could be more axonal compression by the more number of drusen in the optic nerve head. Alfonso Casada et al. have reported that overall 33% ONHD cases showed abnormally reduced average GCC thickness values and the mean GCC thickness was 77.9 ± 12.2 µm.[34] Masoud Aghsaei et al. have reported that average GCC thickness was 90.8 ± 6.1 µm.[35] Lasse Malmqvist et al.[26] have reported that visible ONHD showed significantly reduced GCC thickness compared to deep drusen. Our results are also indicating a low mean GCC thickness for superficial drusen in all the sectors when compared to that for deep drusen. In our study, the GCC thickness was found to be significantly lower in all the sectors in eyes with superficial drusen along with deep drusen than in eyes with only deep drusen. Hence, in eyes with pseudopapilledema, thinning of GCC should prompt the examiner to look for the ONHD in the superficial layers if not evident on the clinical evaluation.

Limitations of our study: The number of patients included in the study is less. The study could have given better insights with larger number of patients as we divided the patients further into groups to study the patterns to understand the pathophysiology of pseudopapilledema. Since pseudopapilledema is an uncommon affliction of the eyes, the neuro ophthalmologists from multiple centers should collaborate to consolidate the data on OCT to predict patterns on the images. The second limitation is that the optic disc drusen studies consortium utilized the spectralis OCT machine (Heidelberg Engineering, Heidelberg, Germany), whereas our study was done using Cirrus spectral domain OCT machine (Carl Zeiss Meditec, Inc., Dublin, CA, USA). But till date to the best of our knowledge, there are no studies performed to prove the superiority of one over the other to detect optic nerve head drusen.

Conclusion

Pseudopapilledema on OCT is characterized by ONHD and HHL less frequently than PHOMS. Sixty percent of eyes with pseudopapilledema are without ONHD. Eyes with ONHD have a larger disc area than the normal population. Superficial ONHD have thinner RNFL and GCC compared to the deep ONHD.

Financial support and sponsorship

Nil.

Conflicts of interest

There are no conflicts of interest.