HEMORRHAGIC COMPLICATIONS OF OPTIC NERVE HEAD DRUSEN ON SPECTRAL DOMAIN OPTICAL COHERENCE TOMOGRAPHY KYOUNG MIN LEE, MD,*† JEONG-MIN HWANG, MD,*‡ SE JOON WOO, MD*‡ Purpose: To describe the clinical features of hemorrhagic complications secondary to optic nerve head drusen (ONHD) using spectral domain optical coherence tomography (SD-OCT). Methods: Sixty-three consecutive patients with SD-OCT–documented ONHD who presented at Seoul National University Bundang Hospital from December 2009 to July 2012 were included. Full ophthalmologic examinations, including fundus photographs, SD-OCT, fundus angiography, and visual field tests were analyzed in a total of 101 ONHD-positive eyes from 63 patients. Results: Hemorrhagic ONHD complications were found in 7 eyes (7%) from a total of 101 eyes with ONHD. All of them had buried ONHD (visualized with SD-OCT) and myopia (mean spherical equivalent = −4.00 ± 2.35 diopters). Patients with ONHD hemorrhagic complications had smaller disk diameters than patients without hemorrhagic complications (1,308 ± 166 vs. 1,555 ± 217 mm, P = 0.004). Peripapillary hemorrhages were classified into the following 3 types based on SD-OCT findings: subretinal (6 eyes, 86%), retinal (5 eyes, 71%), and vitreous hemorrhage (4 eyes, 57%). Six patients (86%) complained of the recent onset of visual symptoms, but visual acuities at presentation were 20/20 in all patients. In the three patients who were followed up, most hemorrhages were absorbed without complications. Conclusion: Peripapillary hemorrhage can often occur in patients with buried ONHD and small disk diameters. As SD-OCT can be used to visualize ONHD beneath hemorrhage, it is helpful with the differential diagnosis and follow-up evaluation. RETINA 34:1142–1148, 2014
O
ptic nerve head drusen (ONHD) are thought to arise from abnormal axonal metabolism in congenitally predisposed optic disks.1 Peripapillary hemorrhage is a complication of ONHD,2–5 and occurs in 2% to 14% of patients.6–9 With the advent of optical coherence tomography (OCT), the diagnosis of ONHD has reached a new level.10–18 By providing detailed cross-sectional images, OCT can be used to determine
anatomical information about the hemorrhage, including its depth. The earliest OCT reports of ONHD with hemorrhage only showed subretinal hemorrhage because of lower image resolution.3 The introduction of spectral domain OCT (SD-OCT) has allowed for direct visualization of ONHD and each retinal layer.12–18 To the best of our knowledge, there have been no previous reports of SD-OCT findings of ONHD combined with peripapillary hemorrhage. Here, we document ONHD hemorrhagic complications using SD-OCT and describe the associated clinical features.
From the *Department of Ophthalmology, Seoul National University College of Medicine, Seoul, Korea; †Department of Ophthalmology, 15th Special Missions Wing, Republic of Korea Air Force, Seongnam, Korea; and ‡Department of Ophthalmology, Seoul National University Bundang Hospital, Seongnam, Korea. None of the authors have any financial/conflicting interests to disclose. K. M. Lee and J.-M. Hwang equally contributed to the work and therefore should be considered equivalent authors. Reprint requests: Se Joon Woo, MD, Department of Ophthalmology, Seoul National University Bundang Hospital, 300 Gumi-dong, Bundang-gu, Seongnam, Gyeonggi-do 463-707, Korea; e-mail: [email protected]
Patients and Methods This study was approved by the Institutional Review Board of Seoul National University Bundang Hospital and adhered to the tenets of the Declaration of Helsinki. 1142
HEMORRHAGIC COMPLICATIONS OF ONHD ON OCT LEE ET AL
We identified 63 patients with ONHD among the patients referred to our clinic for neuro-ophthalmic examinations from December 2009 to July 2012. Patients with the other diseases that could lead to peripapillary hemorrhage were excluded from our analyses. These included: optic disk edema, ischemic optic neuropathy, optic neuritis, glaucoma, retinal vascular occlusions, and diabetic retinopathy. All patients underwent an initial full ophthalmologic examination including the measurement of best-corrected visual acuity, a dilated funduscopic examination, fundus photography, and SD-OCT (Spectralis OCT, Heidelberg Engineering, Heidelberg, Germany). A diagnosis of ONHD required the direct visualization of ONHD by funduscopy18 or SD-OCT.12–18 To be classified as visible, ONHD must be seen to protrude from the disk on fundus photography. To be classified as buried ONHD, SD-OCT should reveal ONHD to be localized as a mass between the outer retina and the retinal pigment epithelium–choroid boundary.12–18 Multiple horizontal and vertical SD-OCT scans, centered on the optic disk, were used for classification. One hundred OCT frames were averaged to obtain high-resolution horizontal scans dissecting the center of the optic nerve head, and 10 frames were averaged to obtain multiple horizontal or vertical scans. Cases with hemorrhagic complications, as determined by fundus photography, were reviewed retrospectively. The following data were collected for analyses: age, gender, presenting symptoms, visual acuity, refractive error, hemorrhage location, and hemorrhage depth. Using the SD-OCT caliper tool, we measured the height of ONHD about the retinal pigment epithelial layer. Disk diameter was measured in two different planes, the retinal surface plane through confocal infrared images provided by SD-OCT and the choroidal plane by measuring the size of the opening in Bruch membrane bordering the optic nerve. In cases with suspicious disk
1143
edema or the other causes of hemorrhage, the optic disk was evaluated further either at the initial visit or during follow-up examinations. These assessments used fundus fluorescein angiography, indocyanine green angiography, Hardy–Rand–Rittler color vision test, and Goldmann or Humphrey visual field test. Cases with leakage as detected by fluorescein angiography or abrupt changes in retinal nerve fiber layer (RNFL) thickness were excluded to rule out active disk edema as a cause for the hemorrhagic complication. Demographic data were compared between patients with and without hemorrhagic complications secondary to ONHD using Mann–Whitney U test and Fisher exact test. Intraindividual comparisons, in cases of bilateral ONHD, were performed using the Wilcoxon signed-rank test. All statistical analyses were performed using the SPSS software (SPSS version 18.0, SSPS, Inc, Chicago, IL).
Results Of 101 eyes from 63 patients with ONHD, 7 eyes (7%) from 7 patients (3 male and 4 female patients; mean age, 20.4 ± 12.5 years [range, 10–42 years]) experienced hemorrhagic complications. All 7 patients had buried rather than visible ONHD and myopia (spherical equivalent = −4.00 ± 2.35 diopters [D]; range, −2.25 to −7.75 D) except one (Case 7) of near emmetropia because of previous corneal refractive surgery performed 6 years ago (Table 2). Patients with hemorrhagic complications had smaller disk diameters as measured in both retinal surface and choroidal planes (retinal surface plane: 1,148 ± 86 vs. 1,306 ± 216 mm, P = 0.026; choroidal plane: 1,308 ± 166 vs. 1,555 ± 217 mm, P = 0.004; Mann–Whitney U test). Four patients had unilateral ONHD and 3 had bilateral ONHD (Table 1).
Table 1. Clinical Features of Patients With and Without Hemorrhagic Complications Secondary to Optic Nerve Head Drusen
Age, years Male:female Bilaterality BCVA, logMAR Refraction, diopters Buried ONHD, % Disk size R, mm Disk size C, mm ONHD height, mm
ONHD With Hemorrhage (N = 7)
ONHD Without Hemorrhage (N = 94)
P
20.4 ± 12.5 3:4 3 (43%) 0.00 ± 0.00 −4.00 ± 2.35‡ 7 (100%) 1,148 ± 86 1,308 ± 166 356 ± 52
15.2 ± 11.0 15:41 15 (27%) −0.07 ± 0.18 −3.46 ± 2.75 90 (96%) 1,306 ± 216 1,555 ± 217 360 ± 91
0.095* 0.397† 1.000† 0.056* 0.828* 1.000† 0.026* 0.004* 0.883*
*P were calculated using the Mann–Whitney U test. †P were calculated using the Fisher exact test. ‡One case (Case 5) was excluded because of previous refractive surgery. BCVA, best-corrected visual acuity; Disk size C, disk diameter at the choroidal plane on SD-OCT; Disk size R, disk diameter at the retinal surface plane on SD-OCT; ONHD, optic nerve head drusen.
1144 RETINA, THE JOURNAL OF RETINAL AND VITREOUS DISEASES
2014 VOLUME 34 NUMBER 6
Table 2. Clinical Features of Patients With Hemorrhagic Complications Secondary to Buried Optic Nerve Head Drusen Case
Age/ Sex
Refraction, diopters*
Symptoms
Subretinal Hemorrhage
1 2 3 4
18/M 12/F 13/M 10/F
−7.75 −2.50 −6.125 −3.125
Floater (4 days) Floater (4 days) Blurred vision (2 months) Blurred vision (1 day)
Superonasal Superonasal Superonasal Entire disk
5
42/M
−2.25
Superonasal
6
15/F
−2.25
None (on routine examination) Floater (2 weeks)
7
31/F
−0.125†
Floater (4 days)
Retinal Hemorrhage
Vitreous Hemorrhage
Superotemporal — Superonasal Floater — — Entire disk Elongated vitreous opacity Nasal —
—
—
Nasal
Nasal
Floater, Elongated vitreous opacity Elongated vitreous opacity
*Spherical equivalent. †This patient had undergone binocular corneal refractive surgery (Laser Assisted Sub-Epithelial Keratomileusis) 6 years before visiting our clinic.
Hemorrhagic complications were in only one eye despite the fact that the ONHD were bilateral in some cases. In bilateral ONHD, the eye with hemorrhagic complication did not differ from the fellow eye in terms of optic disk size (hemorrhagic eye vs. fellow eye; retinal surface plane: 1,172 ± 138 vs. 1,045 ± 210 mm, P = 0.109; choroidal plane: 1,450 ± 95 vs. 1,283 ± 144 mm, P = 0.109; Wilcoxon signed-rank test) or ONHD height (hemorrhagic eye vs. fellow eye; 360 ± 91 vs. 484 ± 156 mm, P = 0.180; Wilcoxon signed-rank test). Peripapillary hemorrhage was classified as subretinal, retinal (intraretinal/RNFL), or vitreous hemorrhage (Table 2). Subretinal hemorrhage was found in 6 eyes (86%), retinal hemorrhage in 5 (71%), and vitreous hemorrhage in 4 (57%). Except for one patient with an old vitreous hemorrhage (Case 6), all patients had subretinal hemorrhages in the following areas: superonasal area (4 patients), nasal area (1 patient), and throughout the entire disk (1 patient) (Table 2). Additionally, all but 2 patients (Cases 3 and 6) had intraretinal or RNFL hemorrhages in the nasal (2 patients), superonasal (1 patient), and superotemporal (1 patient) areas, or spreading throughout the entire disk (1 patient) (Table 2). In cases with retinal and subretinal hemorrhage, these areas of hemorrhage did not colocalize (Table 2) (Figure 1, A and C). Fundus fluorescein angiography was performed in 3 cases (Cases 1, 3, and 4; Figure 1E). There was no evidence of neovascularization or fluorescein leakage adjacent to the optic nerve head in any of the patients examined. Fundus indocyanine green angiography was performed in 1 patient (Case 3; Figure 1F). The images revealed neither polypoidal structures nor choroidal neovascularization. Color vision and visual field tests were performed in two patients (Cases 4 and 5). The results showed no evidence of a color vision defect in either case. A mas-
sive subretinal hemorrhage led to an enlarged blind spot in 1 patient (Case 4; Figure 2). The small subretinal hemorrhage observed in Case 5 did not result in a visual field defect (Case 5; Figure 3C). Vitreous hemorrhages were divided into 2 types: floating vitreous opacities and elongated opacities attached to the disk (Table 2). Six patients (86%) complained about the sudden onset of visual symptoms (range, 1 day–2 months). Only 1 patient (Case 5) was completely symptom free, with the hemorrhage detected incidentally during a routine annual examination (Table 2). However, the initial best-corrected visual acuity was 20/20 in all cases. Follow-up examinations were performed in 3 patients. The annual follow-up examinations in 2 patients (Cases 5 and 6) showed complete absorption of retinal and subretinal areas of hemorrhage (Figure 3B). All floating vitreous hemorrhages ultimately disappeared, while the elongated opacity that had been observed to be attached to the disk persisted (Case 6; Figure 3E). Follow-up examination after 1 month (Case 1; Figure 1B) revealed that most hemorrhage had been absorbed, but the satellite subretinal hemorrhages remained. Subretinal fluid was detected between the optic disk and the satellite subretinal hemorrhage. Similarly, subretinal fluid was also found in initial SD-OCT images obtained for Cases 2 and 3 (Figure 1, C and D).
Discussion Seven percent of ONHD cases presented with hemorrhagic complications. All the patients had buried ONHD, more myopia, and smaller disk diameters than the respective mean values for patients without hemorrhagic complication of ONHD (Table 1). Although the features of peripapillary
HEMORRHAGIC COMPLICATIONS OF ONHD ON OCT LEE ET AL
1145
Fig. 1. Peripapillary hemorrhage in Cases 1 (A and B), 2 (C), and 3 (D–F). A. Subretinal hemorrhage in the superonasal area (arrowheads), retinal hemorrhage in the superotemporal area (arrows). B. Follow-up examination after 1 month. Satellite subretinal hemorrhage (arrowheads) and subretinal fluid (arrows) can be seen in the intervening area. C. Hemorrhage behind the retinal vasculature (superior arrow), and minimal subretinal hemorrhage as detected by SD-OCT (inferior arrow). D. Satellite subretinal hemorrhage (arrowheads) and subretinal fluid in the intervening area (arrows). E. Fluorescein angiography showed masked background fluorescence by subretinal hemorrhage and retinal pigment epithelial detachment. F. Indocyanine angiography showed defects in Bruch membrane (arrowheads) and the boundary of a past hemorrhage (arrows).
hemorrhage differed among the 7 patients presented here, the SD-OCT findings obtained at 1-month follow-up for Case 1 (Figure 1B) were similar to the initial SD-OCT findings for Case 3 (Figure 1D). Except for two patients (Cases 3 and 6) who pre-
sented at our clinic at least 2 weeks after visual symptoms began, the SD-OCT findings were quite similar among patients. This similarity suggests that the initial peripapillary hemorrhages may have shared common pathogenic mechanism or origin.
1146 RETINA, THE JOURNAL OF RETINAL AND VITREOUS DISEASES
2014 VOLUME 34 NUMBER 6
Fig. 2. Peripapillary hemorrhage in Case 4. A. Fellow eye with optic nerve head drusen (arrowheads). B. Massive subretinal and retinal hemorrhages encircle the entire disk. An elongated vitreous hemorrhage can be seen (arrow). Adjacent to the subretinal fluid and hemorrhage, highly reflective optic nerve head drusen can be seen against the background low reflectivity of the outer nuclear layer (arrowheads). C and D. Goldmann perimetry reveals blind spot enlargement in comparison to the fellow eye. Magnetic resonance imaging of this patient did not show any brain or optic nerve lesion (not shown).
We noticed that the area of subretinal hemorrhage was superonasal or nasal in each case, regardless of where the retinal or vitreous hemorrhage was located. Since ONHD appeared as C-shaped lumps on the nasal side of the disk, the colocalization of ONHD and subretinal hemorrhage suggests that nasal subretinal hemorrhages are associated with ONHD. A subretinal hemorrhage that has developed could subsequently migrate into the retina, RNFL, or vitreous cavity. With time, the hemorrhage was absorbed and a variety of conditions were found to coexist. Sometimes only a vitreous hemorrhage existed (Figure 3D). However, this case also shared similarity with Cases 4 and 7 (Figures 2 and 3F), as both buried ONHD and remnants of an elongated vitreous opacity attached to a protruding disk. Most patients experienced a sudden decrease in visual quality (Table 1). However, best-corrected visual acuity was 20/20 in all cases. Visual field test revealed only enlargement of a blind spot even in 1 patient who happened to have a massive peripapillary hemorrhage (Figure 2B). Our findings showed that mild subretinal hemorrhage had no impact on the patient’s visual field (Figure 3C). Most hemorrhages were absorbed 1 month after the initial presentation (Figure 1B). Two patients underwent follow-up examinations 1 year later. At that time, one patient exhibited no evidence of past subretinal or intraretinal hemorrhage (Figure 3B), but the rem-
nants of an elongated vitreous opacity attached to the disk were observed in another (Figure 3E). The fast recoveries and excellent visual prognoses achieved in many cases suggest that the incidence of ONHD hemorrhagic complications may have been underreported. Notably, hemorrhage stresses local tissue and is toxic to the retinal cells.19,20 Recurrent hemorrhage can induce RNFL thinning and subsequent peripapillary atrophy, as often seen in cases of longstanding ONHD.4 Sanders et al21 classified ONHD hemorrhage into three categories: small, transient asymptomatic splinter hemorrhages on the optic disk head; optic nerve head hemorrhages extending into the vitreous and producing transient symptomatic visual field defects; and deep, peripapillary subretinal hemorrhages associated with severe visual disturbances and permanent field defects. Our cases comprised examples of the latter two groups. Potential explanations for the hemorrhage observed in our patient population include the following. First, the presence of ONHD exerts pressure that can erode the choriocapillaris. This hypothesis is supported by the fact that patients with hemorrhagic complications had smaller disks, which exaggerates the structural stress between ONHD and the choriocapillaris. Second, the rupture of Bruch membrane (lacquer crack) might exist as in Case 3 (Figure 1F). All patients had myopia, which also stretches Bruch membrane, sometimes leading to lacquer cracks and subretinal hemorrhage.22 The
HEMORRHAGIC COMPLICATIONS OF ONHD ON OCT LEE ET AL
1147
Fig. 3. Peripapillary hemorrhage in Cases 5 (A–C), 6 (D and E), and 7 (F). A. Subretinal hemorrhage in the superonasal area (arrows), retinal hemorrhage in the nasal area. B. Follow-up examination after 1 year. All of the hemorrhages have been absorbed. C. Humphrey visual field test results showing a normal visual field. D. A free-floating vitreous hemorrhage (between the dotted lines) and elongated vitreous opacity (arrow) can be seen. E. A 1-year follow-up examination. Although the floating vitreous opacity has disappeared, a portion of the elongated vitreous opacity remains attached to the disk (arrow). F. Subretinal and retinal hemorrhage in the nasal area. An elongated vitreous opacity can also be seen (right dotted line).
presence of ONHD can exaggerate this stress and produce subretinal hemorrhages. Third, the presence of ONHD could induce a retinal vein occlusion. We excluded the patients with retinal vein occlusions from this study because of the inability to determine whether ONHD was the cause of vascular occlusion or just a coincidence was not certain. However, retinal vascular occlusion by large ONHD could be another possible
cause of peripapillary hemorrhage. Finally, despite the absence of choroidal neovascularization in any of the patients presented here, new blood vessels adjacent to the optic nerve head are often observed in patients with ONHD.23,24 Sudden massive peripapillary hemorrhage with near complete absorption within months was a characteristic and unique feature of ONHD hemorrhagic complications.
1148 RETINA, THE JOURNAL OF RETINAL AND VITREOUS DISEASES
Peripapillary hemorrhage, in our study, had similar features with previously reported intrapapillary hemorrhage with adjacent peripapillary subretinal hemorrhage25,26 and asymptomatic peripapillary subretinal hemorrhage.27 Since they did not perform SD-OCT in their studies, we suspected that some of these cases might have had undetected buried ONHD. In this study, SD-OCT was useful in the documentation of buried ONHD, in localizing the observed hemorrhage, and in confirming the absence of choroidal neovascular membranes. In conclusion, ONHD can present as massive peripapillary, subretinal, retinal, or vitreous hemorrhage. Patients with buried ONHD, myopia, and small disk diameters are at increased risk. The hemorrhages are absorbed without visual deterioration, and SD-OCT provides information necessary for the differential diagnosis, follow-up evaluation, and efforts to elucidate the underlying pathophysiology. Key words: complication, hemorrhage, optic nerve head drusen, spectral domain optical coherence tomography. References 1. Tso MO. Pathology and pathogenesis of drusen of the optic nervehead. Ophthalmology 1981;88:1066–1080. 2. Neffendorf JE, Mulholland C, Quinlan M, Lyons CJ. Disc drusen complicated by optic disc hemorrhage in childhood. Can J Ophthalmol 2010;45:537–538. 3. Romero J, Sowka J, Shechtman D. Hemorrhagic complications of optic disc drusen and available treatment options. Optometry 2008;79:496–500. 4. Aumiller MS. Optic disc drusen: complications and management. Optometry 2007;78:10–16. 5. Rubinstein K, Ali M. Retinal complications of optic disc drusen. Br J Ophthalmol 1982;66:83–95. 6. Flores-Rodriguez P, Gili P, Martin-Rios MD. Ophthalmic features of optic disc drusen. Ophthalmologica 2012;228:59–66. 7. Borruat FX, Sanders MD. [Vascular anomalies and complications of optic nerve drusen]. Klin Monbl Augenheilkd 1996; 208:294–296. 8. Mustonen E. Pseudopapilloedema with and without verified optic disc drusen. A clinical analysis I. Acta Ophthalmol (Copenh) 1983;61:1037–1056. 9. Harris MJ, Fine SL, Owens SL. Hemorrhagic complications of optic nerve drusen. Am J Ophthalmol 1981;92:70–76.
2014 VOLUME 34 NUMBER 6
10. Johnson LN, Diehl ML, Hamm CW, et al. Differentiating optic disc edema from optic nerve head drusen on optical coherence tomography. Arch Ophthalmol 2009;127:45–49. 11. Choi SS, Zawadzki RJ, Greiner MA, et al. Fourier-domain optical coherence tomography and adaptive optics reveal nerve fiber layer loss and photoreceptor changes in a patient with optic nerve drusen. J Neuroophthalmol 2008;28:120–125. 12. Lee KM, Woo SJ, Hwang JM. Differentiation of optic nerve head drusen and optic disc edema with spectral-domain optical coherence tomography. Ophthalmology 2011;118:971–977. 13. Slotnick S, Sherman J. Disc drusen. Ophthalmology 2012;119: 652. author reply 652-653 e651. 14. Wester ST, Fantes FE, Lam BL, et al. Characteristics of optic nerve head drusen on optical coherence tomography images. Ophthalmic Surg Lasers Imaging 2010;41:83–90. 15. Patel NN, Shulman JP, Chin KJ, Finger PT. Optical coherence tomography/scanning laser ophthalmoscopy imaging of optic nerve head drusen. Ophthalmic Surg Lasers Imaging 2010;41:614–621. 16. Murthy RK, Storm L, Grover S, et al. In-vivo high resolution imaging of optic nerve head drusen using spectral-domain optical coherence tomography. BMC Med Imaging 2010;10:11. 17. Yi K, Mujat M, Sun W, et al. Imaging of optic nerve head drusen: improvements with spectral domain optical coherence tomography. J Glaucoma 2009;18:373–378. 18. Lee KM, Woo SJ, Hwang JM. Morphologic characteristics of optic nerve head drusen on spectral-domain optical coherence tomography. Am J Ophthalmol 2013;155:1139–1144. 19. Glatt H, Machemer R. Experimental subretinal hemorrhage in rabbits. Am J Ophthalmol 1982;94:762–773. 20. Doly M, Bonhomme B, Vennat JC. Experimental study of the retinal toxicity of hemoglobinic iron. Ophthalmic Res 1986;18: 21–27. 21. Sanders TE, Gay AJ, Newman M. Hemorrhagic complications of drusen of the optic disk. Am J Ophthalmol 1971;71:204–217. 22. Hirata A, Negi A. Lacquer crack lesions in experimental chick myopia. Graefes Arch Clin Exp Ophthalmol 1998;236:138–145. 23. Wilson GA, Lloyd C, Moore AT. Optic disc drusen and peripapillary subretinal neovascular membranes in children. J Pediatr Ophthalmol Strabismus 2002;39:351–354. 24. Diduszyn JM, Quillen DA, Cantore WA, Gardner TW. Optic disk drusen, peripapillary choroidal neovascularization, and POEMS syndrome. Am J Ophthalmol 2002;133:275–276. 25. Kokame GT, Yamamoto I, Kishi S, et al. Intrapapillary hemorrhage with adjacent peripapillary subretinal hemorrhage. Ophthalmology 2004;111:926–930. 26. Katz B, Hoyt WF. Intrapapillary and peripapillary hemorrhage in young patients with incomplete posterior vitreous detachment. Signs of vitreopapillary traction. Ophthalmology 1995; 102:349–354. 27. Sibony P, Fourman S, Honkanen R, El Baba F. Asymptomatic peripapillary subretinal hemorrhage: a study of 10 cases. J Neuroophthalmol 2008;28:114–119.
O
ptic nerve head drusen (ONHD) are thought to arise from abnormal axonal metabolism in congenitally predisposed optic disks.1 Peripapillary hemorrhage is a complication of ONHD,2–5 and occurs in 2% to 14% of patients.6–9 With the advent of optical coherence tomography (OCT), the diagnosis of ONHD has reached a new level.10–18 By providing detailed cross-sectional images, OCT can be used to determine
anatomical information about the hemorrhage, including its depth. The earliest OCT reports of ONHD with hemorrhage only showed subretinal hemorrhage because of lower image resolution.3 The introduction of spectral domain OCT (SD-OCT) has allowed for direct visualization of ONHD and each retinal layer.12–18 To the best of our knowledge, there have been no previous reports of SD-OCT findings of ONHD combined with peripapillary hemorrhage. Here, we document ONHD hemorrhagic complications using SD-OCT and describe the associated clinical features.
From the *Department of Ophthalmology, Seoul National University College of Medicine, Seoul, Korea; †Department of Ophthalmology, 15th Special Missions Wing, Republic of Korea Air Force, Seongnam, Korea; and ‡Department of Ophthalmology, Seoul National University Bundang Hospital, Seongnam, Korea. None of the authors have any financial/conflicting interests to disclose. K. M. Lee and J.-M. Hwang equally contributed to the work and therefore should be considered equivalent authors. Reprint requests: Se Joon Woo, MD, Department of Ophthalmology, Seoul National University Bundang Hospital, 300 Gumi-dong, Bundang-gu, Seongnam, Gyeonggi-do 463-707, Korea; e-mail: [email protected]
Patients and Methods This study was approved by the Institutional Review Board of Seoul National University Bundang Hospital and adhered to the tenets of the Declaration of Helsinki. 1142
HEMORRHAGIC COMPLICATIONS OF ONHD ON OCT LEE ET AL
We identified 63 patients with ONHD among the patients referred to our clinic for neuro-ophthalmic examinations from December 2009 to July 2012. Patients with the other diseases that could lead to peripapillary hemorrhage were excluded from our analyses. These included: optic disk edema, ischemic optic neuropathy, optic neuritis, glaucoma, retinal vascular occlusions, and diabetic retinopathy. All patients underwent an initial full ophthalmologic examination including the measurement of best-corrected visual acuity, a dilated funduscopic examination, fundus photography, and SD-OCT (Spectralis OCT, Heidelberg Engineering, Heidelberg, Germany). A diagnosis of ONHD required the direct visualization of ONHD by funduscopy18 or SD-OCT.12–18 To be classified as visible, ONHD must be seen to protrude from the disk on fundus photography. To be classified as buried ONHD, SD-OCT should reveal ONHD to be localized as a mass between the outer retina and the retinal pigment epithelium–choroid boundary.12–18 Multiple horizontal and vertical SD-OCT scans, centered on the optic disk, were used for classification. One hundred OCT frames were averaged to obtain high-resolution horizontal scans dissecting the center of the optic nerve head, and 10 frames were averaged to obtain multiple horizontal or vertical scans. Cases with hemorrhagic complications, as determined by fundus photography, were reviewed retrospectively. The following data were collected for analyses: age, gender, presenting symptoms, visual acuity, refractive error, hemorrhage location, and hemorrhage depth. Using the SD-OCT caliper tool, we measured the height of ONHD about the retinal pigment epithelial layer. Disk diameter was measured in two different planes, the retinal surface plane through confocal infrared images provided by SD-OCT and the choroidal plane by measuring the size of the opening in Bruch membrane bordering the optic nerve. In cases with suspicious disk
1143
edema or the other causes of hemorrhage, the optic disk was evaluated further either at the initial visit or during follow-up examinations. These assessments used fundus fluorescein angiography, indocyanine green angiography, Hardy–Rand–Rittler color vision test, and Goldmann or Humphrey visual field test. Cases with leakage as detected by fluorescein angiography or abrupt changes in retinal nerve fiber layer (RNFL) thickness were excluded to rule out active disk edema as a cause for the hemorrhagic complication. Demographic data were compared between patients with and without hemorrhagic complications secondary to ONHD using Mann–Whitney U test and Fisher exact test. Intraindividual comparisons, in cases of bilateral ONHD, were performed using the Wilcoxon signed-rank test. All statistical analyses were performed using the SPSS software (SPSS version 18.0, SSPS, Inc, Chicago, IL).
Results Of 101 eyes from 63 patients with ONHD, 7 eyes (7%) from 7 patients (3 male and 4 female patients; mean age, 20.4 ± 12.5 years [range, 10–42 years]) experienced hemorrhagic complications. All 7 patients had buried rather than visible ONHD and myopia (spherical equivalent = −4.00 ± 2.35 diopters [D]; range, −2.25 to −7.75 D) except one (Case 7) of near emmetropia because of previous corneal refractive surgery performed 6 years ago (Table 2). Patients with hemorrhagic complications had smaller disk diameters as measured in both retinal surface and choroidal planes (retinal surface plane: 1,148 ± 86 vs. 1,306 ± 216 mm, P = 0.026; choroidal plane: 1,308 ± 166 vs. 1,555 ± 217 mm, P = 0.004; Mann–Whitney U test). Four patients had unilateral ONHD and 3 had bilateral ONHD (Table 1).
Table 1. Clinical Features of Patients With and Without Hemorrhagic Complications Secondary to Optic Nerve Head Drusen
Age, years Male:female Bilaterality BCVA, logMAR Refraction, diopters Buried ONHD, % Disk size R, mm Disk size C, mm ONHD height, mm
ONHD With Hemorrhage (N = 7)
ONHD Without Hemorrhage (N = 94)
P
20.4 ± 12.5 3:4 3 (43%) 0.00 ± 0.00 −4.00 ± 2.35‡ 7 (100%) 1,148 ± 86 1,308 ± 166 356 ± 52
15.2 ± 11.0 15:41 15 (27%) −0.07 ± 0.18 −3.46 ± 2.75 90 (96%) 1,306 ± 216 1,555 ± 217 360 ± 91
0.095* 0.397† 1.000† 0.056* 0.828* 1.000† 0.026* 0.004* 0.883*
*P were calculated using the Mann–Whitney U test. †P were calculated using the Fisher exact test. ‡One case (Case 5) was excluded because of previous refractive surgery. BCVA, best-corrected visual acuity; Disk size C, disk diameter at the choroidal plane on SD-OCT; Disk size R, disk diameter at the retinal surface plane on SD-OCT; ONHD, optic nerve head drusen.
1144 RETINA, THE JOURNAL OF RETINAL AND VITREOUS DISEASES
2014 VOLUME 34 NUMBER 6
Table 2. Clinical Features of Patients With Hemorrhagic Complications Secondary to Buried Optic Nerve Head Drusen Case
Age/ Sex
Refraction, diopters*
Symptoms
Subretinal Hemorrhage
1 2 3 4
18/M 12/F 13/M 10/F
−7.75 −2.50 −6.125 −3.125
Floater (4 days) Floater (4 days) Blurred vision (2 months) Blurred vision (1 day)
Superonasal Superonasal Superonasal Entire disk
5
42/M
−2.25
Superonasal
6
15/F
−2.25
None (on routine examination) Floater (2 weeks)
7
31/F
−0.125†
Floater (4 days)
Retinal Hemorrhage
Vitreous Hemorrhage
Superotemporal — Superonasal Floater — — Entire disk Elongated vitreous opacity Nasal —
—
—
Nasal
Nasal
Floater, Elongated vitreous opacity Elongated vitreous opacity
*Spherical equivalent. †This patient had undergone binocular corneal refractive surgery (Laser Assisted Sub-Epithelial Keratomileusis) 6 years before visiting our clinic.
Hemorrhagic complications were in only one eye despite the fact that the ONHD were bilateral in some cases. In bilateral ONHD, the eye with hemorrhagic complication did not differ from the fellow eye in terms of optic disk size (hemorrhagic eye vs. fellow eye; retinal surface plane: 1,172 ± 138 vs. 1,045 ± 210 mm, P = 0.109; choroidal plane: 1,450 ± 95 vs. 1,283 ± 144 mm, P = 0.109; Wilcoxon signed-rank test) or ONHD height (hemorrhagic eye vs. fellow eye; 360 ± 91 vs. 484 ± 156 mm, P = 0.180; Wilcoxon signed-rank test). Peripapillary hemorrhage was classified as subretinal, retinal (intraretinal/RNFL), or vitreous hemorrhage (Table 2). Subretinal hemorrhage was found in 6 eyes (86%), retinal hemorrhage in 5 (71%), and vitreous hemorrhage in 4 (57%). Except for one patient with an old vitreous hemorrhage (Case 6), all patients had subretinal hemorrhages in the following areas: superonasal area (4 patients), nasal area (1 patient), and throughout the entire disk (1 patient) (Table 2). Additionally, all but 2 patients (Cases 3 and 6) had intraretinal or RNFL hemorrhages in the nasal (2 patients), superonasal (1 patient), and superotemporal (1 patient) areas, or spreading throughout the entire disk (1 patient) (Table 2). In cases with retinal and subretinal hemorrhage, these areas of hemorrhage did not colocalize (Table 2) (Figure 1, A and C). Fundus fluorescein angiography was performed in 3 cases (Cases 1, 3, and 4; Figure 1E). There was no evidence of neovascularization or fluorescein leakage adjacent to the optic nerve head in any of the patients examined. Fundus indocyanine green angiography was performed in 1 patient (Case 3; Figure 1F). The images revealed neither polypoidal structures nor choroidal neovascularization. Color vision and visual field tests were performed in two patients (Cases 4 and 5). The results showed no evidence of a color vision defect in either case. A mas-
sive subretinal hemorrhage led to an enlarged blind spot in 1 patient (Case 4; Figure 2). The small subretinal hemorrhage observed in Case 5 did not result in a visual field defect (Case 5; Figure 3C). Vitreous hemorrhages were divided into 2 types: floating vitreous opacities and elongated opacities attached to the disk (Table 2). Six patients (86%) complained about the sudden onset of visual symptoms (range, 1 day–2 months). Only 1 patient (Case 5) was completely symptom free, with the hemorrhage detected incidentally during a routine annual examination (Table 2). However, the initial best-corrected visual acuity was 20/20 in all cases. Follow-up examinations were performed in 3 patients. The annual follow-up examinations in 2 patients (Cases 5 and 6) showed complete absorption of retinal and subretinal areas of hemorrhage (Figure 3B). All floating vitreous hemorrhages ultimately disappeared, while the elongated opacity that had been observed to be attached to the disk persisted (Case 6; Figure 3E). Follow-up examination after 1 month (Case 1; Figure 1B) revealed that most hemorrhage had been absorbed, but the satellite subretinal hemorrhages remained. Subretinal fluid was detected between the optic disk and the satellite subretinal hemorrhage. Similarly, subretinal fluid was also found in initial SD-OCT images obtained for Cases 2 and 3 (Figure 1, C and D).
Discussion Seven percent of ONHD cases presented with hemorrhagic complications. All the patients had buried ONHD, more myopia, and smaller disk diameters than the respective mean values for patients without hemorrhagic complication of ONHD (Table 1). Although the features of peripapillary
HEMORRHAGIC COMPLICATIONS OF ONHD ON OCT LEE ET AL
1145
Fig. 1. Peripapillary hemorrhage in Cases 1 (A and B), 2 (C), and 3 (D–F). A. Subretinal hemorrhage in the superonasal area (arrowheads), retinal hemorrhage in the superotemporal area (arrows). B. Follow-up examination after 1 month. Satellite subretinal hemorrhage (arrowheads) and subretinal fluid (arrows) can be seen in the intervening area. C. Hemorrhage behind the retinal vasculature (superior arrow), and minimal subretinal hemorrhage as detected by SD-OCT (inferior arrow). D. Satellite subretinal hemorrhage (arrowheads) and subretinal fluid in the intervening area (arrows). E. Fluorescein angiography showed masked background fluorescence by subretinal hemorrhage and retinal pigment epithelial detachment. F. Indocyanine angiography showed defects in Bruch membrane (arrowheads) and the boundary of a past hemorrhage (arrows).
hemorrhage differed among the 7 patients presented here, the SD-OCT findings obtained at 1-month follow-up for Case 1 (Figure 1B) were similar to the initial SD-OCT findings for Case 3 (Figure 1D). Except for two patients (Cases 3 and 6) who pre-
sented at our clinic at least 2 weeks after visual symptoms began, the SD-OCT findings were quite similar among patients. This similarity suggests that the initial peripapillary hemorrhages may have shared common pathogenic mechanism or origin.
1146 RETINA, THE JOURNAL OF RETINAL AND VITREOUS DISEASES
2014 VOLUME 34 NUMBER 6
Fig. 2. Peripapillary hemorrhage in Case 4. A. Fellow eye with optic nerve head drusen (arrowheads). B. Massive subretinal and retinal hemorrhages encircle the entire disk. An elongated vitreous hemorrhage can be seen (arrow). Adjacent to the subretinal fluid and hemorrhage, highly reflective optic nerve head drusen can be seen against the background low reflectivity of the outer nuclear layer (arrowheads). C and D. Goldmann perimetry reveals blind spot enlargement in comparison to the fellow eye. Magnetic resonance imaging of this patient did not show any brain or optic nerve lesion (not shown).
We noticed that the area of subretinal hemorrhage was superonasal or nasal in each case, regardless of where the retinal or vitreous hemorrhage was located. Since ONHD appeared as C-shaped lumps on the nasal side of the disk, the colocalization of ONHD and subretinal hemorrhage suggests that nasal subretinal hemorrhages are associated with ONHD. A subretinal hemorrhage that has developed could subsequently migrate into the retina, RNFL, or vitreous cavity. With time, the hemorrhage was absorbed and a variety of conditions were found to coexist. Sometimes only a vitreous hemorrhage existed (Figure 3D). However, this case also shared similarity with Cases 4 and 7 (Figures 2 and 3F), as both buried ONHD and remnants of an elongated vitreous opacity attached to a protruding disk. Most patients experienced a sudden decrease in visual quality (Table 1). However, best-corrected visual acuity was 20/20 in all cases. Visual field test revealed only enlargement of a blind spot even in 1 patient who happened to have a massive peripapillary hemorrhage (Figure 2B). Our findings showed that mild subretinal hemorrhage had no impact on the patient’s visual field (Figure 3C). Most hemorrhages were absorbed 1 month after the initial presentation (Figure 1B). Two patients underwent follow-up examinations 1 year later. At that time, one patient exhibited no evidence of past subretinal or intraretinal hemorrhage (Figure 3B), but the rem-
nants of an elongated vitreous opacity attached to the disk were observed in another (Figure 3E). The fast recoveries and excellent visual prognoses achieved in many cases suggest that the incidence of ONHD hemorrhagic complications may have been underreported. Notably, hemorrhage stresses local tissue and is toxic to the retinal cells.19,20 Recurrent hemorrhage can induce RNFL thinning and subsequent peripapillary atrophy, as often seen in cases of longstanding ONHD.4 Sanders et al21 classified ONHD hemorrhage into three categories: small, transient asymptomatic splinter hemorrhages on the optic disk head; optic nerve head hemorrhages extending into the vitreous and producing transient symptomatic visual field defects; and deep, peripapillary subretinal hemorrhages associated with severe visual disturbances and permanent field defects. Our cases comprised examples of the latter two groups. Potential explanations for the hemorrhage observed in our patient population include the following. First, the presence of ONHD exerts pressure that can erode the choriocapillaris. This hypothesis is supported by the fact that patients with hemorrhagic complications had smaller disks, which exaggerates the structural stress between ONHD and the choriocapillaris. Second, the rupture of Bruch membrane (lacquer crack) might exist as in Case 3 (Figure 1F). All patients had myopia, which also stretches Bruch membrane, sometimes leading to lacquer cracks and subretinal hemorrhage.22 The
HEMORRHAGIC COMPLICATIONS OF ONHD ON OCT LEE ET AL
1147
Fig. 3. Peripapillary hemorrhage in Cases 5 (A–C), 6 (D and E), and 7 (F). A. Subretinal hemorrhage in the superonasal area (arrows), retinal hemorrhage in the nasal area. B. Follow-up examination after 1 year. All of the hemorrhages have been absorbed. C. Humphrey visual field test results showing a normal visual field. D. A free-floating vitreous hemorrhage (between the dotted lines) and elongated vitreous opacity (arrow) can be seen. E. A 1-year follow-up examination. Although the floating vitreous opacity has disappeared, a portion of the elongated vitreous opacity remains attached to the disk (arrow). F. Subretinal and retinal hemorrhage in the nasal area. An elongated vitreous opacity can also be seen (right dotted line).
presence of ONHD can exaggerate this stress and produce subretinal hemorrhages. Third, the presence of ONHD could induce a retinal vein occlusion. We excluded the patients with retinal vein occlusions from this study because of the inability to determine whether ONHD was the cause of vascular occlusion or just a coincidence was not certain. However, retinal vascular occlusion by large ONHD could be another possible
cause of peripapillary hemorrhage. Finally, despite the absence of choroidal neovascularization in any of the patients presented here, new blood vessels adjacent to the optic nerve head are often observed in patients with ONHD.23,24 Sudden massive peripapillary hemorrhage with near complete absorption within months was a characteristic and unique feature of ONHD hemorrhagic complications.
1148 RETINA, THE JOURNAL OF RETINAL AND VITREOUS DISEASES
Peripapillary hemorrhage, in our study, had similar features with previously reported intrapapillary hemorrhage with adjacent peripapillary subretinal hemorrhage25,26 and asymptomatic peripapillary subretinal hemorrhage.27 Since they did not perform SD-OCT in their studies, we suspected that some of these cases might have had undetected buried ONHD. In this study, SD-OCT was useful in the documentation of buried ONHD, in localizing the observed hemorrhage, and in confirming the absence of choroidal neovascular membranes. In conclusion, ONHD can present as massive peripapillary, subretinal, retinal, or vitreous hemorrhage. Patients with buried ONHD, myopia, and small disk diameters are at increased risk. The hemorrhages are absorbed without visual deterioration, and SD-OCT provides information necessary for the differential diagnosis, follow-up evaluation, and efforts to elucidate the underlying pathophysiology. Key words: complication, hemorrhage, optic nerve head drusen, spectral domain optical coherence tomography. References 1. Tso MO. Pathology and pathogenesis of drusen of the optic nervehead. Ophthalmology 1981;88:1066–1080. 2. Neffendorf JE, Mulholland C, Quinlan M, Lyons CJ. Disc drusen complicated by optic disc hemorrhage in childhood. Can J Ophthalmol 2010;45:537–538. 3. Romero J, Sowka J, Shechtman D. Hemorrhagic complications of optic disc drusen and available treatment options. Optometry 2008;79:496–500. 4. Aumiller MS. Optic disc drusen: complications and management. Optometry 2007;78:10–16. 5. Rubinstein K, Ali M. Retinal complications of optic disc drusen. Br J Ophthalmol 1982;66:83–95. 6. Flores-Rodriguez P, Gili P, Martin-Rios MD. Ophthalmic features of optic disc drusen. Ophthalmologica 2012;228:59–66. 7. Borruat FX, Sanders MD. [Vascular anomalies and complications of optic nerve drusen]. Klin Monbl Augenheilkd 1996; 208:294–296. 8. Mustonen E. Pseudopapilloedema with and without verified optic disc drusen. A clinical analysis I. Acta Ophthalmol (Copenh) 1983;61:1037–1056. 9. Harris MJ, Fine SL, Owens SL. Hemorrhagic complications of optic nerve drusen. Am J Ophthalmol 1981;92:70–76.
2014 VOLUME 34 NUMBER 6
10. Johnson LN, Diehl ML, Hamm CW, et al. Differentiating optic disc edema from optic nerve head drusen on optical coherence tomography. Arch Ophthalmol 2009;127:45–49. 11. Choi SS, Zawadzki RJ, Greiner MA, et al. Fourier-domain optical coherence tomography and adaptive optics reveal nerve fiber layer loss and photoreceptor changes in a patient with optic nerve drusen. J Neuroophthalmol 2008;28:120–125. 12. Lee KM, Woo SJ, Hwang JM. Differentiation of optic nerve head drusen and optic disc edema with spectral-domain optical coherence tomography. Ophthalmology 2011;118:971–977. 13. Slotnick S, Sherman J. Disc drusen. Ophthalmology 2012;119: 652. author reply 652-653 e651. 14. Wester ST, Fantes FE, Lam BL, et al. Characteristics of optic nerve head drusen on optical coherence tomography images. Ophthalmic Surg Lasers Imaging 2010;41:83–90. 15. Patel NN, Shulman JP, Chin KJ, Finger PT. Optical coherence tomography/scanning laser ophthalmoscopy imaging of optic nerve head drusen. Ophthalmic Surg Lasers Imaging 2010;41:614–621. 16. Murthy RK, Storm L, Grover S, et al. In-vivo high resolution imaging of optic nerve head drusen using spectral-domain optical coherence tomography. BMC Med Imaging 2010;10:11. 17. Yi K, Mujat M, Sun W, et al. Imaging of optic nerve head drusen: improvements with spectral domain optical coherence tomography. J Glaucoma 2009;18:373–378. 18. Lee KM, Woo SJ, Hwang JM. Morphologic characteristics of optic nerve head drusen on spectral-domain optical coherence tomography. Am J Ophthalmol 2013;155:1139–1144. 19. Glatt H, Machemer R. Experimental subretinal hemorrhage in rabbits. Am J Ophthalmol 1982;94:762–773. 20. Doly M, Bonhomme B, Vennat JC. Experimental study of the retinal toxicity of hemoglobinic iron. Ophthalmic Res 1986;18: 21–27. 21. Sanders TE, Gay AJ, Newman M. Hemorrhagic complications of drusen of the optic disk. Am J Ophthalmol 1971;71:204–217. 22. Hirata A, Negi A. Lacquer crack lesions in experimental chick myopia. Graefes Arch Clin Exp Ophthalmol 1998;236:138–145. 23. Wilson GA, Lloyd C, Moore AT. Optic disc drusen and peripapillary subretinal neovascular membranes in children. J Pediatr Ophthalmol Strabismus 2002;39:351–354. 24. Diduszyn JM, Quillen DA, Cantore WA, Gardner TW. Optic disk drusen, peripapillary choroidal neovascularization, and POEMS syndrome. Am J Ophthalmol 2002;133:275–276. 25. Kokame GT, Yamamoto I, Kishi S, et al. Intrapapillary hemorrhage with adjacent peripapillary subretinal hemorrhage. Ophthalmology 2004;111:926–930. 26. Katz B, Hoyt WF. Intrapapillary and peripapillary hemorrhage in young patients with incomplete posterior vitreous detachment. Signs of vitreopapillary traction. Ophthalmology 1995; 102:349–354. 27. Sibony P, Fourman S, Honkanen R, El Baba F. Asymptomatic peripapillary subretinal hemorrhage: a study of 10 cases. J Neuroophthalmol 2008;28:114–119.
