Acta Ophthalmologica 2015
Autofluorescence imaging in the differential diagnosis of optic disc melanocytoma Panagiotis Salvanos,1,2 Tor P. Utheim,3,4 Morten C. Moe,1,2 Nils Eide1* and Ragnheiður Bragadόttir1,2* 1
Department of Ophthalmology, Oslo University Hospital, Oslo, Norway University of Oslo, Oslo, Norway 3 Department of Medical Biochemistry, Oslo University Hospital, Oslo, Norway 4 Institute of Oral Biology, Faculty of Dentistry, University of Oslo, Oslo, Norway 2
ABSTRACT. Purpose: Optic disc melanocytoma (ODM) is a benign tumour that usually occurs on or adjacent to the optic nerve head. The aim of the study was to evaluate fundus autofluorescence (FAF) imaging as a diagnostic tool in ODM. Methods: Retrospective comparative case series study of six patients with ODM and a comparing group of four patients with juxtapapillary choroidal nevus (JCN) and four with juxtapapillary uveal melanoma (JUM). Clinical examination was supplemented with ultrasound B-scan examination and spectral-domain optical coherence tomography. FAF images were obtained with the 532-nm laser (Optomap P200Tx) from all patients. Results: Clinical examination in the ODM group revealed a dome-shaped, darkly pigmented tumour on or adjacent to the optic disc in all patients, with a mean tumour basal dimension 1.4 mm and mean tumour thickness by ultrasonography of 1.0 mm. FAF revealed a totally hypofluorescent mass with sharply demarcated, feathery edges. No hyperfluorescent changes due to orange pigment or subretinal fluid were seen. In contrast, patients with JCN and JUM manifested focal hyperfluorescence as well as larger hyperfluorescent areas at the tumour and its borders. Conclusion: Fundus autofluorescence imaging is a non-invasive adjuvant tool in the differential diagnosis of ODM characterized by lack of hyperfluorescence compared to JCN and JUM. Key words: autofluorescence – imaging – melanocytoma – nevus – optic disc – uveal melanoma
Acta Ophthalmol. 2015: 93: 476–480 ª 2015 Acta Ophthalmologica Scandinavica Foundation. Published by John Wiley & Sons Ltd
doi: 10.1111/aos.12725
*These two authors are equally contributing senior authors.
Introduction Optic disc melanocytoma (ODM) is a rare, benign, dark brown- to blackcoloured tumour that usually occurs on or adjacent to the optic nerve head (Zimmerman & Garron 1962; Shields 1978) and is unilateral in almost all cases
476
(Shields et al. 2004, 2012). Although most tumours are asymptomatic, some patients may experience visual symptoms related to neural or vascular compression or tumour necrosis (Shields et al. 2001). In addition, it has been estimated that 1–2% of ODM can undergo malignant transformation
(Sharma et al. 2002; Shields et al. 2004, 2006). This intraocular tumour must be clinically differentiated from juxtapapillary uveal melanoma (JUM), juxtapapillary choroidal nevus (JCN), hyperplasia of the retina pigment epithelium (RPE), hypertrophy of the RPE, combined hamartoma of the retina and RPE and adenoma of the RPE (Shields et al. 2006; Krohn & Kjersem 2011). Standard fundus photography has been used for follow-up to document any changes of the tumour. Optical coherence tomography (OCT) has been utilized to characterize ODM, showing a gradual transition from normal retina to nodular tumour, and the mass displays a bright anterior border layer with optically empty internal details (Shields et al. 2008; Finger et al. 2010). Fundus autofluorescence (FAF) imaging is a non-invasive modality that is highly correlated to the lipofuscin content at the RPE level (Arroyo et al. 2005). Formation of lipofuscin is an indirect marker of metabolic activity between the photoreceptor outer segment turnover and RPE phagocytic ability. The FAF signal can thus indicate the balance between the formation and disposal mechanisms, and can be used as a hallmark of metabolic stress on the RPE (Schmitz-Valckenberg et al. 2008; Sparrow et al. 2012). However, dysfunction in the outer retina and subretinal space may also lead to FAF abnormalities (Spaide 2008). The distribution and intensity of FAF may thus help elucidate pathophysiological mechanisms
Acta Ophthalmologica 2015
(Meyerle et al. 2007). The fluorescence properties are often altered in intraocular malignancy, and the detection of orange pigment is a very important sign of JUM (Almeida et al. 2013).
In this study, the main aim was to evaluate autofluorescent findings in ODM and compare ODM autofluorescence to changes seen in juxtapapillary JCN and JUM.
Materials and Methods Comparative non-interventional case series study of six consecutive eyes with ODM, four eyes with JCN and four
Table 1. Summary of patient demographics and findings. Size (disc d)
Basal diameter (mm)
Thickness (mm)
Configuration
Fluid on OCT
0 0.2 0.5
1.1 0.3 1.2
1.7 0.5 1.8
1 0.5 1.5
Dome-shaped Flat Dome-shaped
No No No
40 66 40 73
0 0 0.2 0
0.5 1 1.5 1.5
0.8 1.5 2.3 1
0.8 0.5 1.3 1
Flat Flat Dome-shaped Dome-shaped
No No No Yes
m
44
0
4
6
0.5
Flat
No
JCN
f
69
0.2
5
7.4
2.2
Dome-shaped
Yes
10
JCN
f
44
0
1.5
2.3
1
Flat
Yes
11
JUM
f
62
0.4
4.5
7
3.6
Dome-shaped
Yes
12
JUM
f
62
0
5
7.5
1.5
Dome-shaped
Yes
13
JUM
m
43
0
3
6
2.2
Flat
Yes
14
JUM
m
59
0.4
5
7.5
1.6
Flat
Yes
ID
Tumour type
Sex
Age
1 2 3
ODM ODM ODM
f f m
50 32 52
4 5 6 7
ODM ODM ODM JCN
m m f f
8
JCN
9
Log MAR
FAF findings Hypofluorescence Hypofluorescence Hypofluorescence over the tumour. Hyper- and hypofluorescent changes in the macula due to non-exudative AMD Hypofluorescence Hypofluorescence Hypofluorescence Larger hyperfluorescent area in areas of leakage. Stippled hyperfluorescence over tumour. Fibrosis (hyperhypofluorescent) Slight, diffuse hyperfluorescence over the tumour. Larger hyperfluorescent area in areas of leakage. Stippled hyperfluorescence over tumour. Fibrosis (hyper- and hypofluorescent) Larger hyperfluorescent area in areas of leakage Stippled hyperfluorescence over tumour Focal hyperfluorescence from orange pigment. Larger hyperfluorescent area in areas of leakage. Stippled hyperfluorescence over tumour Focal hyperfluorescence from orange pigment. Larger hyperfluorescent area in areas of leakage. Stippled hyperfluorescence over tumour Focal hyperfluorescence from orange pigment. Larger hyperfluorescent area in areas of leakage. Stippled hyperfluorescence over tumour Focal hyperfluorescence from orange pigment. Larger hyperfluorescent area in areas of leakage. Stippled hyperfluorescence over tumour
ODM, optic disc melanocytoma; JUM, juxtapapillary uveal melanoma; JCN, juxtapapillary choroidal nevus; f, female; m, male; BCVA, bestcorrected visual acuity; OCT, optical coherence tomography; FAF, fundus autofluorescence; AMD, age-related macular degeneration.
477
Acta Ophthalmologica 2015
eyes with JUM examined with FAF at the Department of Ophthalmology, Oslo University Hospital, in the period September 2012 to December 2013. FAF images were obtained before treatment with proton beam or plaque irradiation in the JUM group. The study was conducted according to the Declaration of Helsinki, and approved written informed consent was obtained from all patients. Work-up included comprehensive ophthalmological examination, ultrasound B-scan examination and was complemented with spectral-domain optical coherence tomography (SD-OCT), fundus imaging (colour imaging and autofluorescence imaging), fluorescein angiography (FA) and indocyanine green angiography (ICG), performed during the same visit. SD-OCT images were captured with the Nidek RS-3000 Advanced OCT/SLO System (Nidek Inc., 47651 Westinghouse Drive, Fremont, CA, USA). Fundus images were obtained using Optomap P200Tx (Optos, Dunfermline, UK), with high-resolution autofluorescent images captured with the 532-nm (green) laser. All FAF images were digitally analysed using the OPTOS V2 VANTAGE REVIEW Software and subsequently transferred to Adobe Photoshop (Adobe Systems, Inc, San Jose, CA) for image processing. Alterations in the autofluorescence signal were defined as either hyperfluorescence or hypofluorescence in comparison with background FAF.
disc in ODM, with no sign of orange pigment or exudative retinal detachment (Fig. 1). Mean tumour basal dimension by ophthalmoscopy was 1.4 mm (range 0.5–2.3 mm), and mean tumour thickness by ultrasonography was 1.0 mm (range 0.5–1.5 mm). All these tumours showed high internal reflectivity with A-scan and acoustic solidity with B-scan (not shown). No vitreous seeds were seen. In cases of JUM, mean tumour thickness was 2.3 mm (range 1.6–3.6 mm) and in cases of JCN, 1.2 mm (range 0.5–2.2 mm). Examination with SD-OCT showed a dome-shaped tumour configuration with gradual transition from normal retina into the mass, hyper-reflectivity at the anterior surface and very limited internal detail of the tumour in all 14 tumours (Fig. 1D). No subretinal or intraretinal fluid was seen in the ODM group, whereas fluid on or adjacent to the tumour area was observed in two patients with JCN and all JUM patients (not shown) (Table 1). Fluorescein angiography revealed a hypofluorescent area on the optic disc throughout the angiogram in all patients with ODM, due to blocking from the pigmented tumour with no sign of leakage (Fig. 1). In contrast, pinpoint and/or diffuse leakage in late phases was found in 3 of the JCN group and all patients with JUM.
Autofluorescent imaging in the ODM group revealed a totally hypofluorescent mass on or adjacent to the optic disc in 6 of 6 eyes (Figs 1–3, Table 1), and a complete lack of hyperfluorescent areas at the tumour and tumour borders. Mean tumour size using the hypofluorescent borders corresponded to tumour size by ophthalmoscopy with a mean tumour basal diameter of 1.4 mm (range 0.5– 2.3 mm). Thus, no significant discrepancy in the size of the tumour was seen between colour fundus images and FAF images, with typical feathery tumour edges in 4 of 6 patients. No other FAF changes were seen elsewhere, apart from one patient that had stippled hyper- and hypofluorescent changes in the macular area, corresponding to non-exudative agerelated macular degeneration. In contrast, significant FAF changes were seen in the other tumour groups, with focal hyperfluorescence from orange pigment in 4 of 4 in the JUM group, as well as larger hyperfluorescent areas in 3 of 4 patients in the JCN and in all patients in the JUM group, interpreted as areas of subretinal fluid leakage. Stippled hyperfluorescence was also seen in areas over the tumour in 3 of 4 in the JCN and in all JUM group, with no obvious fluid on SD-OCT examination, but with
Results The mean age of the patients was 47 years (range 32–66) in the ODM group, 58 years (range 44–73) in the JCN group, and 57 years (range 43–62) in the JUM group, with similar male/ female ratios in all groups. All patients were Caucasian except for one of Hispanic origin. One patient with ODM was found to have relative afferent pupillary defect. Mean followup time from the first clinical observation and diagnosis was 9 years (range 1–34) for ODM, 6 years (range 1–13) for JCN, and 5 years for JUM (range 1–8). Mean best-corrected logMAR visual acuity was comparable in all groups (Table 1). Clinical examination of all patients with ODM revealed a dome-shaped, darkly pigmented tumour on the optic
478
(A)
(C)
(B)
(D)
Fig. 1. Fundus image of optic disc melanocytoma (patient no. 2). The borders of the optic disc (white circle) and the melanocytoma (red line) are demarcated in this and next image (A). Corresponding autofluorescence image showing the sharply demarcated and feathery edges of the melanocytoma. Notice the correspondence of tumour borders between the composite colour image and fundus autofluorescence (FAF) image (B). Indocyanine green angiography (left) and fluorescein angiography (right) of the posterior pole (C). Spectral-domain optical coherence tomography (SD-OCT) image of the tumour over the optic nerve. Notice the hyper-reflectivity of the anterior tumour surface and the lack of internal tumour detail (D).
Acta Ophthalmologica 2015
(A)
(B)
(C) Fig. 2. Colour and fundus autofluorescence (FAF) images of optic disc melanocytoma (ODM) (patient no. 1). Notice the well-demarcated, feathery edges in both colour and FAF images, as well as the absence of pathologic hyperfluorescence (A). Colour and FAF images of juxtapapillary choroidal nevus (JCN) (patient no. 10). Notice several focal hyperfluorescent areas inside the tumour area, and the larger hyperfluorescent space due to subretinal fluid, obscuring the tumour margins and increasing significantly the risk of tumour growth in this asymptomatic patient (B). Colour and FAF images of juxtapapillary uveal melanoma (JUM) (patient no. 13). Notice the focal hyperfluorescence from orange pigment in the tumour area and the hyperfluorescence from subretinal fluid leakage on and around the tumour (C).
corresponding focal hyperfluorescence in the late phases of FA. Focal hyperfluorescence corresponding to atrophic scars seen with ophthalmoscopy was only observed in 2 of 4 in the JCN group. Tumour margins in FAF images were partly obscured in cases of hyperfluorescent leakage at the tumour border in both the JCN and JUM groups (Figs 2 and 3).
Discussion Optic disc melanocytoma is a rare tumour with typical appearance, and diagnosis is made based on several criteria: no orange pigment, no subre-
tinal fluid and no tumour growth together during long follow-up time. Colour fundus imaging, ultrasonography, OCT, FA, computed tomography (CT) and magnetic resonance imaging (MRI) have been used to characterize ODM (Shields et al. 2008; Finger et al. 2010; Mohmad et al. 2011). To our knowledge, this is the first report on the FAF changes caused by this tumour compared with other juxtapapillary pigmented tumours. Several publications address the characterization of autofluorescent changes caused by nevi and melanomas (Gunduz et al. 2007, 2008, 2009; Albertus et al. 2013; Almeida et al. 2013;
Heimann et al. 2013). Accumulation of lipofuscin-laden macrophages over pigmented tumours, appearing as fine or granular deposits, gives the characteristic orange pigment appearance. This orange pigment can often be subtle and then difficult to detect clinically (Almeida et al. 2013), but gives a strong hyperfluorescent signal when imaged with FAF, making FAF the examination of choice for detection of orange pigment (Hashmi et al. 2012; Kivel€ a 2012). Its presence is a known risk factor of malignancy in a pigmented fundus lesion. FAF showed a smooth hypofluorescent pattern in patients with ODM, in contrast to changes seen in patients with JUM, where focal hyperfluorescence from orange pigment was observed. No focal hyperfluorescence due to orange pigment was observed in the JCN group. The detachment of the neurosensory retina from the RPE, indicating diffuse leakage from the tumour or development of choroidal neovascular membrane in pigmented fundus lesions, increases the metabolic stress of the RPE, with collection of lipofuscin in the cells (Eandi et al. 2005). Tumours with exudative retinal detachment are known to have higher risk of growth (Kivel€ a et al. 2001; Espinoza et al. 2004). It has been shown that FAF is a sensitive tool in the detection of even small amounts of subretinal fluid causing shallow retinal detachment that may be difficult to observe ophthalmoscopically (Salvanos et al. 2013). No evidence of hyperfluorescence on or around was seen in the ODM group, whereas it was an almost constant finding in the other two groups. Shields et al. (2008) have previously reported that SD-OCT examination reveals limited internal details of ODM due to the darkly pigmented mass blocking transmission of light penetration through the tumour. All three tumour groups in this study showed a slightly hyper-reflective signal in SD-OCT examination, without differentiating between the different types. However, it was sensitive in detecting small amounts of subretinal fluid. These findings corresponded to hyperfluorescent areas seen in FAF images in patients with nevi and melanomas. The rationale behind the use of green laser (532 nm) instead of blue (488 nm) for the purpose of FAF imaging in patients with ODM is that the optic
479
Acta Ophthalmologica 2015
(A)
(B)
(C)
Fig. 3. Fundus autofluorescence (FAF) image of optic disc melanocytoma (ODM) (patient no. 3). Notice the stippled hyper- and hypofluorescent changes in the macula due to non-exudative age-related macular degeneration (A). Corresponding colour fundus image (insert). FAF image of juxtapapillary choroidal nevus (JCN) (patient no. 9). Notice the autofluorescent changes due to atrophic scar (asterisk) (B). Corresponding colour fundus image (insert). FAF image of juxtapapillary uveal melanoma (JUM) (patient no. 11) (C). Notice the hyperfluorescent area corresponding to subretinal fluid and orange pigment. Corresponding colour fundus image (insert).
nerve head does not appear darkly hypofluorescent, as it does when using the 488-nm blue laser (Sato et al. 2013). This increases the contrast between the normal optic nerve tissue and the tumour, which makes the hypofluorescent tumour borders easier to delineate. In conclusion, FAF imaging is an important, novel, non-invasive adjuvant tool in the diagnosis of ODM and in the differential diagnosis from other juxtapapillary pigmented lesions. Limitations of the present study are as follows: (1) the small sample size due to the rarity of ODM and (2) the fact that FAF is a dynamic process and quantification of the images is not standardized. Further studies are needed to standardize the image quantification and correlation with the clinical significance of these findings, both in larger ODM, as well as separating small ODM from small JCN.
References Albertus DL, Schachar IH, Zahid S, Elner VM, Demirci H & Jayasundera T (2013): Autofluorescence quantification of benign and malignant choroidal nevomelanocytic tumors. JAMA Ophthalmol 131: 1004–1008. Almeida A, Kaliki S & Shields CL (2013): Autofluorescence of intraocular tumours. Curr Opin Ophthalmol 24: 222–232. Arroyo JG, Yang L, Bula D & Chen DF (2005): Photoreceptor apoptosis in human retinal detachment. Am J Ophthalmol 139: 605–610. Eandi CM, Ober M, Iranmanesh R, Peiretti E & Yannuzzi LA (2005): Acute central serous chorioretinopathy and fundus autofluorescence. Retina 25: 989–993. Espinoza G, Rosenblatt B & Harbour JW (2004): Optical coherence tomography in the evaluation of
480
retinal changes associated with suspicious choroidal melanocytic tumors. Am J Ophthalmol 137: 90–95. Finger PT, Natesh S & Milman T (2010): Optical coherence tomography: pathology correlation of optic disc melanocytoma. Ophthalmology 117: 114–119. Gunduz K, Pulido JS, Bakri SJ & Petit-Fond E (2007): Fundus autofluorescence in choroidal melanocytic lesions. Retina 27: 681–687. Gunduz K, Pulido JS, Pulido JE & Link T (2008): Correlation of fundus autofluorescence with fluorescein and indocyanine green angiography in choroidal melanocytic lesions. Retina 28: 1257–1264. Gunduz K, Pulido JS, Ezzat K, Salomao D & Hann C (2009): Review of fundus autofluorescence in choroidal melanocytic lesions. Eye (Lond) 23: 497–503. Hashmi F, Rojanaporn D, Kaliki S & Shields CL (2012): Orange pigment sediment overlying small choroidal melanoma. Arch Ophthalmol 130: 937– 939. Heimann H, Jmor F & Damato B (2013): Imaging of retinal and choroidal vascular tumours. Eye (Lond) 27: 208–216. Kivel€ a T (2012): Diagnosis of uveal melanoma. Dev Ophthalmol 49: 1–15. Kivel€ a T, Eskelin S, Makitie T & Summanen P (2001): Exudative retinal detachment from malignant uveal melanoma: predictors and prognostic significance. Invest Ophthalmol Vis Sci 42: 2085–2093. Krohn J & Kjersem B (2011): Stereo fundus photography in the diagnosis of optic disc melanocytoma. Acta Ophthalmol 89: e533–e534. Meyerle CB, Smith RT, Barbazetto IA & Yannuzzi LA (2007): Autofluorescence of basal laminar drusen. Retina 27: 1101–1106. Mohmad Z, Kah TA, Yong KC, Abdul Halim WH & Kong Yong T (2011): Melanocytoma of the optic nerve head - a diagnostic dilemma. Clin Pract 1: e60. Salvanos P, Navaratnam J, Ma J, Bragad ottir R & Moe MC (2013): Ultra-widefield autofluorescence imaging in the evaluation of scleral buckling surgery for retinal detachment. Retina 33: 1421–1427. Sato T, Mrejen S & Spaide RF (2013): Multimodal Imaging of Optic Disc Drusen. Am J Ophthalmol 156: e275–e282. Schmitz-Valckenberg S, Holz FG, Bird AC & Spaide RF (2008): Fundus autofluorescence imaging: review and perspectives. Retina 28: 385–409. Sharma PM, Sangal K, Malik P & Mathur MB (2002): Malignant transformation of optic disc melanocytoma? A clinical dilemma at presentation with a review of the literature Ophthalmologica 216: 292–295.
Shields JA (1978): Melanocytoma of the optic nerve head: a review. Int Ophthalmol 1: 31–37. Shields JA, Shields CL, Eagle RC Jr, Singh AD, Berrocal MH & Berrocal JA (2001): Central retinal vascular obstruction secondary to melanocytoma of the optic disc. Arch Ophthalmol 119: 129–133. Shields JA, Demirci H, Mashayekhi A & Shields CL (2004): Melanocytoma of optic disc in 115 cases: the 2004 Samuel Johnson Memorial Lecture, part 1. Ophthalmology 111: 1739–1746. Shields JA, Demirci H, Mashayekhi A, Eagle RC Jr & Shields CL (2006): Melanocytoma of the optic disk: a review. Surv Ophthalmol 51: 93–104. Shields CL, Perez B, Benavides R, Materin MA & Shields JA (2008): Optical coherence tomography of optic disk melanocytoma in 15 cases. Retina 28: 441–446. Shields CL, Kancherla S, Patel J et al. (2012): Clinical survey of 3680 iris tumors based on patient age at presentation. Ophthalmology 119: 407–414. Spaide R (2008): Autofluorescence from the outer retina and subretinal space: hypothesis and review. Retina 28: 5–35. Sparrow JR, Gregory-Roberts E, Yamamoto K, Blonska A, Ghosh SK, Ueda K & Zhou J (2012): The bisretinoids of retinal pigment epithelium. Prog Retin Eye Res 31: 121–135. Zimmerman LE & Garron LK (1962): Melanocytoma of the optic disc. Int Ophthalmol Clin 2: 431–440.
Received on September 2nd, 2014. Accepted on February 21st, 2015. Correspondence: Dr Panagiotis Salvanos, MD FEBOphth PhDcand Department of Ophthalmology Oslo University Hospital Pb 4956 Nydalen 0424 Oslo, Norway Tel: +47 40282762 Fax: +47 22119989 Email: [email protected] Supported by the Norwegian Association of the Blind and Partially Sighted, the Faculty of Medicine, University of Oslo and Oslo University Hospital, Norway.
Autofluorescence imaging in the differential diagnosis of optic disc melanocytoma Panagiotis Salvanos,1,2 Tor P. Utheim,3,4 Morten C. Moe,1,2 Nils Eide1* and Ragnheiður Bragadόttir1,2* 1
Department of Ophthalmology, Oslo University Hospital, Oslo, Norway University of Oslo, Oslo, Norway 3 Department of Medical Biochemistry, Oslo University Hospital, Oslo, Norway 4 Institute of Oral Biology, Faculty of Dentistry, University of Oslo, Oslo, Norway 2
ABSTRACT. Purpose: Optic disc melanocytoma (ODM) is a benign tumour that usually occurs on or adjacent to the optic nerve head. The aim of the study was to evaluate fundus autofluorescence (FAF) imaging as a diagnostic tool in ODM. Methods: Retrospective comparative case series study of six patients with ODM and a comparing group of four patients with juxtapapillary choroidal nevus (JCN) and four with juxtapapillary uveal melanoma (JUM). Clinical examination was supplemented with ultrasound B-scan examination and spectral-domain optical coherence tomography. FAF images were obtained with the 532-nm laser (Optomap P200Tx) from all patients. Results: Clinical examination in the ODM group revealed a dome-shaped, darkly pigmented tumour on or adjacent to the optic disc in all patients, with a mean tumour basal dimension 1.4 mm and mean tumour thickness by ultrasonography of 1.0 mm. FAF revealed a totally hypofluorescent mass with sharply demarcated, feathery edges. No hyperfluorescent changes due to orange pigment or subretinal fluid were seen. In contrast, patients with JCN and JUM manifested focal hyperfluorescence as well as larger hyperfluorescent areas at the tumour and its borders. Conclusion: Fundus autofluorescence imaging is a non-invasive adjuvant tool in the differential diagnosis of ODM characterized by lack of hyperfluorescence compared to JCN and JUM. Key words: autofluorescence – imaging – melanocytoma – nevus – optic disc – uveal melanoma
Acta Ophthalmol. 2015: 93: 476–480 ª 2015 Acta Ophthalmologica Scandinavica Foundation. Published by John Wiley & Sons Ltd
doi: 10.1111/aos.12725
*These two authors are equally contributing senior authors.
Introduction Optic disc melanocytoma (ODM) is a rare, benign, dark brown- to blackcoloured tumour that usually occurs on or adjacent to the optic nerve head (Zimmerman & Garron 1962; Shields 1978) and is unilateral in almost all cases
476
(Shields et al. 2004, 2012). Although most tumours are asymptomatic, some patients may experience visual symptoms related to neural or vascular compression or tumour necrosis (Shields et al. 2001). In addition, it has been estimated that 1–2% of ODM can undergo malignant transformation
(Sharma et al. 2002; Shields et al. 2004, 2006). This intraocular tumour must be clinically differentiated from juxtapapillary uveal melanoma (JUM), juxtapapillary choroidal nevus (JCN), hyperplasia of the retina pigment epithelium (RPE), hypertrophy of the RPE, combined hamartoma of the retina and RPE and adenoma of the RPE (Shields et al. 2006; Krohn & Kjersem 2011). Standard fundus photography has been used for follow-up to document any changes of the tumour. Optical coherence tomography (OCT) has been utilized to characterize ODM, showing a gradual transition from normal retina to nodular tumour, and the mass displays a bright anterior border layer with optically empty internal details (Shields et al. 2008; Finger et al. 2010). Fundus autofluorescence (FAF) imaging is a non-invasive modality that is highly correlated to the lipofuscin content at the RPE level (Arroyo et al. 2005). Formation of lipofuscin is an indirect marker of metabolic activity between the photoreceptor outer segment turnover and RPE phagocytic ability. The FAF signal can thus indicate the balance between the formation and disposal mechanisms, and can be used as a hallmark of metabolic stress on the RPE (Schmitz-Valckenberg et al. 2008; Sparrow et al. 2012). However, dysfunction in the outer retina and subretinal space may also lead to FAF abnormalities (Spaide 2008). The distribution and intensity of FAF may thus help elucidate pathophysiological mechanisms
Acta Ophthalmologica 2015
(Meyerle et al. 2007). The fluorescence properties are often altered in intraocular malignancy, and the detection of orange pigment is a very important sign of JUM (Almeida et al. 2013).
In this study, the main aim was to evaluate autofluorescent findings in ODM and compare ODM autofluorescence to changes seen in juxtapapillary JCN and JUM.
Materials and Methods Comparative non-interventional case series study of six consecutive eyes with ODM, four eyes with JCN and four
Table 1. Summary of patient demographics and findings. Size (disc d)
Basal diameter (mm)
Thickness (mm)
Configuration
Fluid on OCT
0 0.2 0.5
1.1 0.3 1.2
1.7 0.5 1.8
1 0.5 1.5
Dome-shaped Flat Dome-shaped
No No No
40 66 40 73
0 0 0.2 0
0.5 1 1.5 1.5
0.8 1.5 2.3 1
0.8 0.5 1.3 1
Flat Flat Dome-shaped Dome-shaped
No No No Yes
m
44
0
4
6
0.5
Flat
No
JCN
f
69
0.2
5
7.4
2.2
Dome-shaped
Yes
10
JCN
f
44
0
1.5
2.3
1
Flat
Yes
11
JUM
f
62
0.4
4.5
7
3.6
Dome-shaped
Yes
12
JUM
f
62
0
5
7.5
1.5
Dome-shaped
Yes
13
JUM
m
43
0
3
6
2.2
Flat
Yes
14
JUM
m
59
0.4
5
7.5
1.6
Flat
Yes
ID
Tumour type
Sex
Age
1 2 3
ODM ODM ODM
f f m
50 32 52
4 5 6 7
ODM ODM ODM JCN
m m f f
8
JCN
9
Log MAR
FAF findings Hypofluorescence Hypofluorescence Hypofluorescence over the tumour. Hyper- and hypofluorescent changes in the macula due to non-exudative AMD Hypofluorescence Hypofluorescence Hypofluorescence Larger hyperfluorescent area in areas of leakage. Stippled hyperfluorescence over tumour. Fibrosis (hyperhypofluorescent) Slight, diffuse hyperfluorescence over the tumour. Larger hyperfluorescent area in areas of leakage. Stippled hyperfluorescence over tumour. Fibrosis (hyper- and hypofluorescent) Larger hyperfluorescent area in areas of leakage Stippled hyperfluorescence over tumour Focal hyperfluorescence from orange pigment. Larger hyperfluorescent area in areas of leakage. Stippled hyperfluorescence over tumour Focal hyperfluorescence from orange pigment. Larger hyperfluorescent area in areas of leakage. Stippled hyperfluorescence over tumour Focal hyperfluorescence from orange pigment. Larger hyperfluorescent area in areas of leakage. Stippled hyperfluorescence over tumour Focal hyperfluorescence from orange pigment. Larger hyperfluorescent area in areas of leakage. Stippled hyperfluorescence over tumour
ODM, optic disc melanocytoma; JUM, juxtapapillary uveal melanoma; JCN, juxtapapillary choroidal nevus; f, female; m, male; BCVA, bestcorrected visual acuity; OCT, optical coherence tomography; FAF, fundus autofluorescence; AMD, age-related macular degeneration.
477
Acta Ophthalmologica 2015
eyes with JUM examined with FAF at the Department of Ophthalmology, Oslo University Hospital, in the period September 2012 to December 2013. FAF images were obtained before treatment with proton beam or plaque irradiation in the JUM group. The study was conducted according to the Declaration of Helsinki, and approved written informed consent was obtained from all patients. Work-up included comprehensive ophthalmological examination, ultrasound B-scan examination and was complemented with spectral-domain optical coherence tomography (SD-OCT), fundus imaging (colour imaging and autofluorescence imaging), fluorescein angiography (FA) and indocyanine green angiography (ICG), performed during the same visit. SD-OCT images were captured with the Nidek RS-3000 Advanced OCT/SLO System (Nidek Inc., 47651 Westinghouse Drive, Fremont, CA, USA). Fundus images were obtained using Optomap P200Tx (Optos, Dunfermline, UK), with high-resolution autofluorescent images captured with the 532-nm (green) laser. All FAF images were digitally analysed using the OPTOS V2 VANTAGE REVIEW Software and subsequently transferred to Adobe Photoshop (Adobe Systems, Inc, San Jose, CA) for image processing. Alterations in the autofluorescence signal were defined as either hyperfluorescence or hypofluorescence in comparison with background FAF.
disc in ODM, with no sign of orange pigment or exudative retinal detachment (Fig. 1). Mean tumour basal dimension by ophthalmoscopy was 1.4 mm (range 0.5–2.3 mm), and mean tumour thickness by ultrasonography was 1.0 mm (range 0.5–1.5 mm). All these tumours showed high internal reflectivity with A-scan and acoustic solidity with B-scan (not shown). No vitreous seeds were seen. In cases of JUM, mean tumour thickness was 2.3 mm (range 1.6–3.6 mm) and in cases of JCN, 1.2 mm (range 0.5–2.2 mm). Examination with SD-OCT showed a dome-shaped tumour configuration with gradual transition from normal retina into the mass, hyper-reflectivity at the anterior surface and very limited internal detail of the tumour in all 14 tumours (Fig. 1D). No subretinal or intraretinal fluid was seen in the ODM group, whereas fluid on or adjacent to the tumour area was observed in two patients with JCN and all JUM patients (not shown) (Table 1). Fluorescein angiography revealed a hypofluorescent area on the optic disc throughout the angiogram in all patients with ODM, due to blocking from the pigmented tumour with no sign of leakage (Fig. 1). In contrast, pinpoint and/or diffuse leakage in late phases was found in 3 of the JCN group and all patients with JUM.
Autofluorescent imaging in the ODM group revealed a totally hypofluorescent mass on or adjacent to the optic disc in 6 of 6 eyes (Figs 1–3, Table 1), and a complete lack of hyperfluorescent areas at the tumour and tumour borders. Mean tumour size using the hypofluorescent borders corresponded to tumour size by ophthalmoscopy with a mean tumour basal diameter of 1.4 mm (range 0.5– 2.3 mm). Thus, no significant discrepancy in the size of the tumour was seen between colour fundus images and FAF images, with typical feathery tumour edges in 4 of 6 patients. No other FAF changes were seen elsewhere, apart from one patient that had stippled hyper- and hypofluorescent changes in the macular area, corresponding to non-exudative agerelated macular degeneration. In contrast, significant FAF changes were seen in the other tumour groups, with focal hyperfluorescence from orange pigment in 4 of 4 in the JUM group, as well as larger hyperfluorescent areas in 3 of 4 patients in the JCN and in all patients in the JUM group, interpreted as areas of subretinal fluid leakage. Stippled hyperfluorescence was also seen in areas over the tumour in 3 of 4 in the JCN and in all JUM group, with no obvious fluid on SD-OCT examination, but with
Results The mean age of the patients was 47 years (range 32–66) in the ODM group, 58 years (range 44–73) in the JCN group, and 57 years (range 43–62) in the JUM group, with similar male/ female ratios in all groups. All patients were Caucasian except for one of Hispanic origin. One patient with ODM was found to have relative afferent pupillary defect. Mean followup time from the first clinical observation and diagnosis was 9 years (range 1–34) for ODM, 6 years (range 1–13) for JCN, and 5 years for JUM (range 1–8). Mean best-corrected logMAR visual acuity was comparable in all groups (Table 1). Clinical examination of all patients with ODM revealed a dome-shaped, darkly pigmented tumour on the optic
478
(A)
(C)
(B)
(D)
Fig. 1. Fundus image of optic disc melanocytoma (patient no. 2). The borders of the optic disc (white circle) and the melanocytoma (red line) are demarcated in this and next image (A). Corresponding autofluorescence image showing the sharply demarcated and feathery edges of the melanocytoma. Notice the correspondence of tumour borders between the composite colour image and fundus autofluorescence (FAF) image (B). Indocyanine green angiography (left) and fluorescein angiography (right) of the posterior pole (C). Spectral-domain optical coherence tomography (SD-OCT) image of the tumour over the optic nerve. Notice the hyper-reflectivity of the anterior tumour surface and the lack of internal tumour detail (D).
Acta Ophthalmologica 2015
(A)
(B)
(C) Fig. 2. Colour and fundus autofluorescence (FAF) images of optic disc melanocytoma (ODM) (patient no. 1). Notice the well-demarcated, feathery edges in both colour and FAF images, as well as the absence of pathologic hyperfluorescence (A). Colour and FAF images of juxtapapillary choroidal nevus (JCN) (patient no. 10). Notice several focal hyperfluorescent areas inside the tumour area, and the larger hyperfluorescent space due to subretinal fluid, obscuring the tumour margins and increasing significantly the risk of tumour growth in this asymptomatic patient (B). Colour and FAF images of juxtapapillary uveal melanoma (JUM) (patient no. 13). Notice the focal hyperfluorescence from orange pigment in the tumour area and the hyperfluorescence from subretinal fluid leakage on and around the tumour (C).
corresponding focal hyperfluorescence in the late phases of FA. Focal hyperfluorescence corresponding to atrophic scars seen with ophthalmoscopy was only observed in 2 of 4 in the JCN group. Tumour margins in FAF images were partly obscured in cases of hyperfluorescent leakage at the tumour border in both the JCN and JUM groups (Figs 2 and 3).
Discussion Optic disc melanocytoma is a rare tumour with typical appearance, and diagnosis is made based on several criteria: no orange pigment, no subre-
tinal fluid and no tumour growth together during long follow-up time. Colour fundus imaging, ultrasonography, OCT, FA, computed tomography (CT) and magnetic resonance imaging (MRI) have been used to characterize ODM (Shields et al. 2008; Finger et al. 2010; Mohmad et al. 2011). To our knowledge, this is the first report on the FAF changes caused by this tumour compared with other juxtapapillary pigmented tumours. Several publications address the characterization of autofluorescent changes caused by nevi and melanomas (Gunduz et al. 2007, 2008, 2009; Albertus et al. 2013; Almeida et al. 2013;
Heimann et al. 2013). Accumulation of lipofuscin-laden macrophages over pigmented tumours, appearing as fine or granular deposits, gives the characteristic orange pigment appearance. This orange pigment can often be subtle and then difficult to detect clinically (Almeida et al. 2013), but gives a strong hyperfluorescent signal when imaged with FAF, making FAF the examination of choice for detection of orange pigment (Hashmi et al. 2012; Kivel€ a 2012). Its presence is a known risk factor of malignancy in a pigmented fundus lesion. FAF showed a smooth hypofluorescent pattern in patients with ODM, in contrast to changes seen in patients with JUM, where focal hyperfluorescence from orange pigment was observed. No focal hyperfluorescence due to orange pigment was observed in the JCN group. The detachment of the neurosensory retina from the RPE, indicating diffuse leakage from the tumour or development of choroidal neovascular membrane in pigmented fundus lesions, increases the metabolic stress of the RPE, with collection of lipofuscin in the cells (Eandi et al. 2005). Tumours with exudative retinal detachment are known to have higher risk of growth (Kivel€ a et al. 2001; Espinoza et al. 2004). It has been shown that FAF is a sensitive tool in the detection of even small amounts of subretinal fluid causing shallow retinal detachment that may be difficult to observe ophthalmoscopically (Salvanos et al. 2013). No evidence of hyperfluorescence on or around was seen in the ODM group, whereas it was an almost constant finding in the other two groups. Shields et al. (2008) have previously reported that SD-OCT examination reveals limited internal details of ODM due to the darkly pigmented mass blocking transmission of light penetration through the tumour. All three tumour groups in this study showed a slightly hyper-reflective signal in SD-OCT examination, without differentiating between the different types. However, it was sensitive in detecting small amounts of subretinal fluid. These findings corresponded to hyperfluorescent areas seen in FAF images in patients with nevi and melanomas. The rationale behind the use of green laser (532 nm) instead of blue (488 nm) for the purpose of FAF imaging in patients with ODM is that the optic
479
Acta Ophthalmologica 2015
(A)
(B)
(C)
Fig. 3. Fundus autofluorescence (FAF) image of optic disc melanocytoma (ODM) (patient no. 3). Notice the stippled hyper- and hypofluorescent changes in the macula due to non-exudative age-related macular degeneration (A). Corresponding colour fundus image (insert). FAF image of juxtapapillary choroidal nevus (JCN) (patient no. 9). Notice the autofluorescent changes due to atrophic scar (asterisk) (B). Corresponding colour fundus image (insert). FAF image of juxtapapillary uveal melanoma (JUM) (patient no. 11) (C). Notice the hyperfluorescent area corresponding to subretinal fluid and orange pigment. Corresponding colour fundus image (insert).
nerve head does not appear darkly hypofluorescent, as it does when using the 488-nm blue laser (Sato et al. 2013). This increases the contrast between the normal optic nerve tissue and the tumour, which makes the hypofluorescent tumour borders easier to delineate. In conclusion, FAF imaging is an important, novel, non-invasive adjuvant tool in the diagnosis of ODM and in the differential diagnosis from other juxtapapillary pigmented lesions. Limitations of the present study are as follows: (1) the small sample size due to the rarity of ODM and (2) the fact that FAF is a dynamic process and quantification of the images is not standardized. Further studies are needed to standardize the image quantification and correlation with the clinical significance of these findings, both in larger ODM, as well as separating small ODM from small JCN.
References Albertus DL, Schachar IH, Zahid S, Elner VM, Demirci H & Jayasundera T (2013): Autofluorescence quantification of benign and malignant choroidal nevomelanocytic tumors. JAMA Ophthalmol 131: 1004–1008. Almeida A, Kaliki S & Shields CL (2013): Autofluorescence of intraocular tumours. Curr Opin Ophthalmol 24: 222–232. Arroyo JG, Yang L, Bula D & Chen DF (2005): Photoreceptor apoptosis in human retinal detachment. Am J Ophthalmol 139: 605–610. Eandi CM, Ober M, Iranmanesh R, Peiretti E & Yannuzzi LA (2005): Acute central serous chorioretinopathy and fundus autofluorescence. Retina 25: 989–993. Espinoza G, Rosenblatt B & Harbour JW (2004): Optical coherence tomography in the evaluation of
480
retinal changes associated with suspicious choroidal melanocytic tumors. Am J Ophthalmol 137: 90–95. Finger PT, Natesh S & Milman T (2010): Optical coherence tomography: pathology correlation of optic disc melanocytoma. Ophthalmology 117: 114–119. Gunduz K, Pulido JS, Bakri SJ & Petit-Fond E (2007): Fundus autofluorescence in choroidal melanocytic lesions. Retina 27: 681–687. Gunduz K, Pulido JS, Pulido JE & Link T (2008): Correlation of fundus autofluorescence with fluorescein and indocyanine green angiography in choroidal melanocytic lesions. Retina 28: 1257–1264. Gunduz K, Pulido JS, Ezzat K, Salomao D & Hann C (2009): Review of fundus autofluorescence in choroidal melanocytic lesions. Eye (Lond) 23: 497–503. Hashmi F, Rojanaporn D, Kaliki S & Shields CL (2012): Orange pigment sediment overlying small choroidal melanoma. Arch Ophthalmol 130: 937– 939. Heimann H, Jmor F & Damato B (2013): Imaging of retinal and choroidal vascular tumours. Eye (Lond) 27: 208–216. Kivel€ a T (2012): Diagnosis of uveal melanoma. Dev Ophthalmol 49: 1–15. Kivel€ a T, Eskelin S, Makitie T & Summanen P (2001): Exudative retinal detachment from malignant uveal melanoma: predictors and prognostic significance. Invest Ophthalmol Vis Sci 42: 2085–2093. Krohn J & Kjersem B (2011): Stereo fundus photography in the diagnosis of optic disc melanocytoma. Acta Ophthalmol 89: e533–e534. Meyerle CB, Smith RT, Barbazetto IA & Yannuzzi LA (2007): Autofluorescence of basal laminar drusen. Retina 27: 1101–1106. Mohmad Z, Kah TA, Yong KC, Abdul Halim WH & Kong Yong T (2011): Melanocytoma of the optic nerve head - a diagnostic dilemma. Clin Pract 1: e60. Salvanos P, Navaratnam J, Ma J, Bragad ottir R & Moe MC (2013): Ultra-widefield autofluorescence imaging in the evaluation of scleral buckling surgery for retinal detachment. Retina 33: 1421–1427. Sato T, Mrejen S & Spaide RF (2013): Multimodal Imaging of Optic Disc Drusen. Am J Ophthalmol 156: e275–e282. Schmitz-Valckenberg S, Holz FG, Bird AC & Spaide RF (2008): Fundus autofluorescence imaging: review and perspectives. Retina 28: 385–409. Sharma PM, Sangal K, Malik P & Mathur MB (2002): Malignant transformation of optic disc melanocytoma? A clinical dilemma at presentation with a review of the literature Ophthalmologica 216: 292–295.
Shields JA (1978): Melanocytoma of the optic nerve head: a review. Int Ophthalmol 1: 31–37. Shields JA, Shields CL, Eagle RC Jr, Singh AD, Berrocal MH & Berrocal JA (2001): Central retinal vascular obstruction secondary to melanocytoma of the optic disc. Arch Ophthalmol 119: 129–133. Shields JA, Demirci H, Mashayekhi A & Shields CL (2004): Melanocytoma of optic disc in 115 cases: the 2004 Samuel Johnson Memorial Lecture, part 1. Ophthalmology 111: 1739–1746. Shields JA, Demirci H, Mashayekhi A, Eagle RC Jr & Shields CL (2006): Melanocytoma of the optic disk: a review. Surv Ophthalmol 51: 93–104. Shields CL, Perez B, Benavides R, Materin MA & Shields JA (2008): Optical coherence tomography of optic disk melanocytoma in 15 cases. Retina 28: 441–446. Shields CL, Kancherla S, Patel J et al. (2012): Clinical survey of 3680 iris tumors based on patient age at presentation. Ophthalmology 119: 407–414. Spaide R (2008): Autofluorescence from the outer retina and subretinal space: hypothesis and review. Retina 28: 5–35. Sparrow JR, Gregory-Roberts E, Yamamoto K, Blonska A, Ghosh SK, Ueda K & Zhou J (2012): The bisretinoids of retinal pigment epithelium. Prog Retin Eye Res 31: 121–135. Zimmerman LE & Garron LK (1962): Melanocytoma of the optic disc. Int Ophthalmol Clin 2: 431–440.
Received on September 2nd, 2014. Accepted on February 21st, 2015. Correspondence: Dr Panagiotis Salvanos, MD FEBOphth PhDcand Department of Ophthalmology Oslo University Hospital Pb 4956 Nydalen 0424 Oslo, Norway Tel: +47 40282762 Fax: +47 22119989 Email: [email protected] Supported by the Norwegian Association of the Blind and Partially Sighted, the Faculty of Medicine, University of Oslo and Oslo University Hospital, Norway.
