ENHANCED DEPTH IMAGING OPTICAL COHERENCE TOMOGRAPHY OF CHOROIDAL METASTASIS IN 14 EYES SAAD A. AL-DAHMASH, MD, CAROL L. SHIELDS, MD, SWATHI KALIKI, MD, TIMOTHY JOHNSON, MD, JERRY A. SHIELDS, MD Purpose: To describe the imaging features of choroidal metastasis using enhanced depth imaging optical coherence tomography (EDI-OCT). Methods: This retrospective observational case series included 31 eyes with choroidal metastasis. Spectral domain EDI-OCT was performed using Heidelberg Spectralis HRA + OCT. The main outcome measures were imaging features by EDI-OCT. Results: Of 31 eyes with choroidal metastasis imaged with EDI-OCT, 14 (45%) eyes displayed image detail suitable for study. The metastasis originated from carcinoma of the breast (n = 7, 50%), lung (n = 5, 36%), pancreas (n = 1, 7%), and thyroid gland (n = 1, 7%). The mean tumor basal diameter was 6.4 mm, and mean thickness was 2.3 mm by B-scan ultrasonography. The tumor location was submacular in 6 (43%) eyes and extramacular in 8 (57%) eyes. By EDI-OCT, the mean tumor thickness was 987 mm. The most salient EDI-OCT features of the metastasis included anterior compression/obliteration of the overlying choriocapillaris (n = 13, 93%), an irregular (lumpy bumpy) anterior contour (n = 9, 64%), and posterior shadowing (n = 12, 86%). Overlying retinal pigment epithelial abnormalities were noted (n = 11, 78%). Outer retinal features included structural loss of the interdigitation of the cone outer segment tips (n = 9, 64%), the ellipsoid portion of photoreceptors (n = 8, 57%), external limiting membrane (n = 4, 29%), outer nuclear layer (n = 1, 7%), and outer plexiform layer (n = 1, 7%). The inner retinal layers (inner nuclear layer to nerve fiber layer) were normal. Subretinal fluid (n = 11, 79%), subretinal lipofuscin pigment (n = 1, 7%), and intraretinal edema (n = 2, 14%) were identified. Conclusion: The EDI-OCT of choroidal metastasis shows a characteristic lumpy bumpy anterior tumor surface and outer retinal layer disruption with preservation of inner retinal layers. RETINA 34:1588–1593, 2014
C
pigment epithelium (RPE), particularly the choroid, due to light scattering from the RPE and the vascular interfaces of the choroid. Further limitations are related to motion artifact with subtle eye movement and occasional inaccurate determination of RPE/choriocapillaris layer by the automated software, which necessitates manual correction for OCT thickness measurements.1,2 Enhanced depth imaging (EDI)-OCT, a method described by Spaide et al,2 was designed to improve imaging of the deeper layers of the eye, primarily the choroid and sclera. The technique of EDI-OCT has been used to characterize the normal features and thickness of the choroid in various age groups and refractive states, as well as the features in selected diseases.3–9 Furthermore, EDI-OCT features of choroidal and scleral tumors have been published, particularly regarding choroidal nevus, melanoma, choroidal lymphoma,
onventional spectral domain (SD) optical coherence tomography (OCT) provides cross-sectional retinal images with an axial resolution of 5 mm.1 The limitations of conventional SD-OCT include poor image resolution of structures deep to the retinal From the Ocular Oncology Service, Wills Eye Institute, Thomas Jefferson University, Philadelphia, Pennsylvania. Dr. Al-Dahmash is now at the Ophthalmology Department, College of Medicine, King Saud University, Riyadh, Saudi Arabia; and Dr. Kaliki is now at the Ocular Oncology Service, L V Prasad Eye Institute, Hyderabad, India. None of the authors have any conflicting interests to disclose. Supported by the Eye Tumor Research Foundation, Philadelphia, PA (C.L.S. and J.A.S.). Carol L. Shields has had full access to all the data in the study and takes responsibility for the integrity of the data and the accuracy of the data analysis. Reprint requests: Carol L. Shields, MD, Ocular Oncology Service, Suite 1440, Wills Eye Institute, Thomas Jefferson University, 840 Walnut Street, Philadelphia, PA 19107; e-mail: [email protected]
1588
EDI-OCT OF CHOROIDAL METASTASIS AL-DAHMASH ET AL
sclerochoroidal calcification, and solitary idiopathic choroiditis.10–15 Distinguishing features are many, but the anterior surface of tumors can vary slightly with a smooth mound shape of nevus, a dome shape of a small melanoma, and a “seasick” surface of advanced lymphoma.11–13 Simulating lesions also show characteristic surfaces such as a “rocky and rolling” surface of sclerochoroidal calcification and a scalloped contour of choroidal hemorrhage.14,16 Herein, we describe the “lumpy bumpy” anterior surface of choroidal metastasis on EDI-OCT and delineate other related features.
Methods This was a retrospective observational study approved by the Institutional Review Board of Wills Eye Hospital. All patients diagnosed with choroidal metastasis from January 1, 2011 to November 31, 2011 at the Ocular Oncology Service, Wills Eye Hospital, Philadelphia, PA, underwent EDI-OCT using SD imaging with the Heidelberg Spectralis (Heidelberg Engineering, Heidelberg, Germany). Patients with choroidal metastasis measuring ,3.5 mm thickness by ultrasonography were included for the study. The patient demographic data of age, race, and gender were recorded. The presenting symptoms, primary malignancy site, and best-corrected visual acuity were listed. The choroidal metastasis was evaluated for clinical features including quadrantic location of the tumor epicenter, maximal basal diameter (in millimeters), distance of the posterior margin of the metastasis from the foveola and the optic disk (in millimeters), color (pigmented, nonpigmented, or mixed), and overlying RPE alterations (none, atrophy, or hypertrophy). The presence of subretinal fluid, lipofuscin pigment, and drusen was also noted. These findings were documented with a large fundus drawing and confirmed by fundus photography and ultrasonography. The features recorded on standard ultrasonography included the thickness (in millimeters), configuration (plateau, dome, or mushroom), echogenicity (solid or hollow), choroidal excavation (absent or present), and subretinal fluid (absent or present). Spectral domain EDI-OCT was performed using a Heidelberg Spectralis HRA + OCT (Heidelberg Engineering) and acquisition and analysis software (version 5.3.3.0) with automated EDI. The axial resolution was 3.5 mm with a scan speed of 40,000 A-scans per second. The images were captured using a custom scan acquisition protocol of up to 13 raster lines of 9-mm scan length, with 1,536 A-scans per line. Realtime eye tracking by TruTrack Active Eye Tracking was used, and automatic real-time image averaging
1589
was set at 100 images. Enhanced depth imaging optical coherence tomography was performed using a technique similar to that described by Spaide et al.2 The EDI-OCT images were obtained after pupillary dilatation in all patients. They were evaluated for quality (optimal or suboptimal). The EDI-OCT image was considered optimal and suitable for the study when both anterior and posterior margins of the tumor and the overlying retina were visualized, as well as the lateral margins of the tumor. The images were classified as suboptimal when a portion of the tumor or the overlying retina could not be visualized and were excluded from the study. The optimal EDI-OCT images were then reviewed. The recorded features included thinning of the choriocapillaris (absent or present), the contour of the anterior surface of the tumor (smooth or lumpy bumpy), internal quality of the tumor, maximal tumor thickness (in micrometers), and posterior choroidal shadowing (absent or present). The retinal layers were evaluated for abnormalities (absent, atrophy, hypertrophy, irregularity, or thinning) of the RPE, the interdigitation of the cone outer segment tips, ellipsoid portion of photoreceptors, external limiting membrane, outer nuclear layer, outer plexiform layer, inner nuclear layer, inner plexiform layer, ganglion cell layer, and nerve fiber layer. The presence of subretinal fluid, drusen, and orange pigment were noted and correlated with clinical and fundus autofluorescence findings. Results Of 31 eyes with choroidal metastases that underwent EDI-OCT imaging over an 11-month period, the quality of the image was judged to be optimal in 14 eyes (45%) and suboptimal in 17 eyes (55%). The quality of suboptimal images was found with peripherally located tumors (at or anterior to the equator), thick tumors (.2 mm thickness), presence of media opacity, and in elderly or sickly patients uncooperative for proper imaging. This did not correlate to tumor type. The clinical and ultrasonographic features of the choroidal metastasis are summarized in Table 1. Of 14 eyes (13 patients) with optimal EDI-OCT image, the mean age was 63 years (median, 63 years; range, 43–80 years) (Figures 1 and 2). Ocular symptoms were present in 13 (93%) eyes, and 1 (7%) patient was asymptomatic. The primary malignancy was carcinoma of the breast (n = 7, 50%), lung (n = 5, 36%), pancreas (n = 1, 7%), and thyroid gland (n = 1, 7%). The epicenter of the choroidal metastasis was located within the macula (n = 6, 43%) or outside the macula (n = 8, 57%). The mean maximal basal
1590 RETINA, THE JOURNAL OF RETINAL AND VITREOUS DISEASES Table 1. Enhanced Depth Imaging OCT of 14 Eyes (13 Patients) With Choroidal Metastasis: Demographics and Clinical Features Features Age, mean (median, range), years Race White Others Gender Male Female Symptoms None Present Primary malignancy Breast Lung Pancreas Thyroid Visual acuity $20/40 ,20/40–20/100 ,20/100 Location of choroidal metastasis epicenter Macula Extramacular Distance from the foveola, mean (median, range), mm Distance from the optic disk, mean (median, range), mm Maximal basal diameter, mean (median, range), mm Ultrasonographic thickness, mean (median, range), mm Color Melanotic Amelanotic Mixed Subretinal fluid Present Absent RPE changes None Atrophy Hyperplasia Orange pigment Absent Present Drusen Absent Present Ultrasonographic configuration Plateau Dome Mushroom Ultrasonographic hollowness Choroidal excavation
Number (%) (n = 14) 63 (63, 43–80) 13 (100) 0 (0) 2 (15) 11 (85) 1 (7) 13 (93) 7 5 1 1
(50) (36) (7) (7)
4 (29) 7 (50) 3 (22) 6 (43) 8 (57) 2.0 (2.0, 0–4.0) 2.0 (1.0, 0–7.0) 6.0 (5.0, 3.0–12.0) 2.3 (2.3, 1.0–3.3) 1 (7) 13 (93) 0 (0) 11 (79) 3 (21) 9 (64) 3 (22) 2 (14) 13 (93) 1 (7) 14 (100) 0 (0) 1 13 0 4 2
(7) (93) (0) (29) (14)
2014 VOLUME 34 NUMBER 8
diameter of the choroidal metastasis was 6 mm (median, 5.0 mm; range, 3.0–12.0 mm). Subretinal fluid was clinically visible in 79% (n = 11) of the cases. Other clinical findings included RPE atrophy (n = 3, 21%), RPE hyperplasia (n = 2, 14%), and orange pigment (n = 1, 7%). Ultrasonography showed mean thickness of choroidal metastasis at 2.3 mm (median, 2.3 mm; range, 1.0–3.3 mm). The EDI-OCT characteristics of choroidal metastasis are summarized in Table 2. The average thickness of the choroidal metastasis measured by EDI-OCT was 987 mm (median, 1,014 mm; range, 547–1,511 mm). The EDI-OCT features included choriocapillaris compression/obliteration (n = 13, 93%), homogeneous optical reflectivity along the anterior surface, and the anterior tumor surface being lumpy bumpy (n = 9, 64%) or smooth (n = 5, 36%). The internal quality of the tumor could not be evaluated because of EDIOCT light signal attenuation (shadowing) (n = 12, 86%). The numerous RPE and retinal irregularities are listed in Table 2. The inner retinal layers were normal in all eyes.
Discussion In 1997, Shields et al17 surveyed 520 eyes with choroidal metastases and noted most tumors of small size with mean thickness of 3 mm and with tumor epicenter in the macular region or between the macula and the equator. The small size and posterior location of most choroidal metastases render these tumors ideal for study with OCT. Early reports on time domain and SD-OCT features of choroidal tumors were mainly focused on overlying retinal features because choroidal imaging was relatively poor.18–29 Schaudig et al18 recognized the limitations of the OCT and suggested that OCT was of little value in the differential diagnosis of choroidal tumors. Arevalo et al20 described time-domain OCT findings in 7 patients with choroidal metastasis and noted subretinal fluid (86%), thickened RPE choriocapillaris complex (71%), and little detail of the choroidal features. Natesh et al21 reviewed time domain OCT findings in 10 cases of choroidal metastases and noted elevation of the RPE and retina, RPE thickening and folds, and subretinal fluid. Sayanagi et al,22 in their general report on SD-OCT of choroidal tumors, described thickening and hyperreflectivity of the RPE, overlying retinal thinning, and subretinal fluid in 4 eyes with choroidal metastasis. They concluded that visualization of the choroidal tumor was limited to the anterior aspect with little deeper detail.22 Others have noted similar findings, particularly with marked “irregularity” in the RPE layer.23
EDI-OCT OF CHOROIDAL METASTASIS AL-DAHMASH ET AL
receptors with intact inner retinal layers. C. Enhanced depth imaging OCT at the fovea demonstrates subretinal fluid.
In our study, using high-resolution SD EDI-OCT, the most salient feature of choroidal metastasis was the irregular or lumpy bumpy anterior tumor surface (64% of cases) despite a smooth-appearing surface on ultrasonography. Unlike the smooth “mound” of choroidal nevus and the smooth “dome” of melanoma, choroidal metastasis often showed characteristic lumpy bumpy surface. Choroidal metastasis is an infiltrative process, which is believed to start at the choriocapillaris level. The resolution of EDI-OCT depicts this malignancy with choriocapillaris compression/ obliteration, similar to nevus and melanoma.11,12 Overlying attenuation of the outer retinal layers and RPE was evident in many cases (Table 2). Retinal pigment epithelial abnormalities were clinically visible in 36% compared with EDI-OCT-visible in 78%. An additional benefit of EDI-OCT is the detection of tiny subclinical choroidal metastasis before they are ophthalmoscopically visible, as highlighted by Torres et al10 and Witkin et al.30 In this series, we imaged tumors as small as 547 mm. In these cases, the tumor was present in the outer choroid with loss of large choroidal vessels and nearly invisible to ophthalmoscopy and ultrasonography.
imaging OCT at the fovea demonstrates subretinal fluid.
1591 Fig. 1. Enhanced depth imaging OCT of choroidal metastasis in a 51-year-old woman. A. The amelanotic choroidal metastasis is visible superotemporal to the optic disk. B. Enhanced depth imaging OCT at the center of the tumor showing slightly irregular lumpy bumpy anterior tumor surface, obliteration/compression of choriocapillaris, and choroidal shadowing. There is overlying RPE thinning, structural loss of the interdigitation of the cone outer segment tips, and the ellipsoid portion of photo-
Previous reports on choroidal nevus and small choroidal melanoma have identified the incongruity between ultrasonography and EDI-OCT measurement of tumor thickness.11,12 Likewise, the same discrepancy was noted for choroidal metastasis thickness by ultrasonography versus EDI-OCT. On ultrasonography, the mean tumor thickness was 2.3 mm compared with 0.987 mm on EDI-OCT, an overestimation by 133% on ultrasonography. This could represent more precise measurement with high-resolution EDI-OCT or could be the result of calibration discrepancy with either technique or could be related to misplacement of the calipers on low-resolution ultrasonography. Witkin et al30 noted that tumors ,1 mm in thickness could be undetectable by ultrasonography and better visualized with EDI-OCT. Although our study includes a few patients, we believe that EDI-OCT has shown an obvious improvement over ultrasound for anatomical depiction and resolution of tumor. This alone contributes to our recognition and understanding of this malignancy and could be instrumental in following tumor regression after treatment. In conclusion, SD EDI-OCT is a promising noninvasive tool for high-resolution imaging of choroidal Fig. 2. Enhanced depth imaging OCT of choroidal metastasis in a 63-year-old woman. A. The amelanotic choroidal metastasis is visible superotemporal to the optic disk. B. Enhanced depth imaging OCT at the center of the tumor showing the typical lumpy bumpy anterior tumor surface. There is overlying RPE atrophy and irregularity, structural loss of the interdigitation of the cone outer segment tips, and the ellipsoid portion of photoreceptors with intact inner retinal layers. C. Enhanced depth
1592 RETINA, THE JOURNAL OF RETINAL AND VITREOUS DISEASES Table 2. Enhanced Depth Imaging OCT of 14 Eyes With Choroidal Metastasis: EDI Characteristics of Choroid and Retinal Structures Features
2014 VOLUME 34 NUMBER 8
Key words: eye, tumor, choroid, metastasis, optical coherence tomography, enhanced depth imaging, EDIOCT.
Number (%) n = 14
Choriocapillaris Normal 1 (7) Thinning 13 (93) Anterior tumor surface contour Lumpy bumpy 9 (64) Smooth 5 (36) Thickness of metastasis, mean 987 (1,014, 547–1,511) (median, range), mm Choroidal shadowing Absent 2 (14) Present 12 (86) Bruch membrane Intact 14 (100) Ruptured 0 (0) RPE Normal 3 (22) Atrophy 1 (7) Hypertrophy 5 (36) Thinning 5 (36) Interdigitation of the cone outer segment tips Normal 5 (36) Absent 9 (64) Ellipsoid portion of photoreceptors Normal 6 (43) Absent 8 (57) External limiting membrane Normal 9 (64) Irregular 1 (7) Absent 4 (29) Outer nuclear layer Normal 13 (93) Absent 1 (7) Outer plexiform layer Normal 13 (93) Absent 1 (7) Inner nuclear layer and inward Normal 14 (100) Abnormal 0 (0) Associated features Subretinal fluid 11 (79) Subretinal lipofuscin 1 (7) deposition (orange pigment) Intraretinal cystoid edema 1 (7) Intraretinal noncystoid edema 1 (7)
metastasis, which allows visualization of tumor details and overlying retina. The most obvious finding is the lumpy bumpy tumor surface of choroidal metastasis, as well as overlying subretinal fluid and choriocapillaris compression/obliteration. These findings can assist the clinician in the early detection of choroidal metastasis and in the differentiation of choroidal metastasis from simulating conditions.
References 1. Sakata LM, Deleon-Ortega J, Sakata V, Girkin CA. Optical coherence tomography of the retina and optic nerve—a review. Clin Experiment Ophthalmol 2009;37:90–99. 2. Spaide RF, Koizumi H, Pozzoni MC. Enhanced depth imaging spectral-domain optical coherence tomography. Am J Ophthalmol 2008;146:496–500. 3. Margolis R, Spaide RF. A pilot study of enhanced depth imaging optical coherence tomography of the choroid in normal eyes. Am J Ophthalmol 2009;147:811–815. 4. Spaide RF. Enhanced depth imaging optical coherence tomography of retinal pigment epithelial detachment in age-related macular degeneration. Am J Ophthalmol 2009;147:644–652. 5. Fujiwara T, Imamura Y, Margolis R, et al. Enhanced depth imaging optical coherence tomography of the choroid in highly myopic eyes. Am J Ophthalmol 2009;148:445–450. 6. Imamura Y, Fujiwara T, Margolis R, Spaide RF. Enhanced depth imaging optical coherence tomography of the choroid in central serous chorioretinopathy. Retina 2009;29:1469–1473. 7. Maruko I, Iida T, Sugano Y, et al. Subfoveal choroidal thickness after treatment of Vogt- Koyanagi-Harada disease. Retina 2011;31:510–517. 8. Imamura Y, Iida T, Maruko I, et al. Enhanced depth imaging optical coherence tomography of the sclera in dome-shaped macula. Am J Ophthalmol 2011;151:297–302. 9. Chung SE, Kang SW, Lee JH, Kim YT. Choroidal thickness in polypoidal choroidal vasculopathy and exudative age-related macular degeneration. Ophthalmology 2011;118:840–845. 10. Torres VL, Brugnoni N, Kaiser PK, Singh AD. Optical coherence tomography enhanced depth imaging of choroidal tumors. Am J Ophthalmol 2011;151:586–593. 11. Shah SU, Kaliki S, Shields CL, et al. Enhanced depth imaging optical coherence tomography of choroidal nevus in 104 cases. Ophthalmology 2012;119:1066–1072. 12. Shields CL, Kaliki S, Rojanaporn D, et al. Enhanced depth imaging optical coherence tomography of small choroidal melanoma. Comparative analysis to choroidal nevus. Arch Ophthalmol 2012;130:850–856. 13. Arias JD, Kumar N, Fulco EAM, et al. Seasick choroid: a finding on enhanced depth imaging spectral domain optical coherence tomography (EDI-OCT) of choroidal lymphoma. Retina Cases Brief Rep 2013;7:19–22. 14. Fung AT, Arias JD, Shields CL, Shields JA. Sclerochoroidal calcification is primarily a scleral condition based on enhanced depth imaging optical coherence tomography. JAMA Ophthalmol 2013;131:960–963. 15. Fung AT, Kaliki S, Shields CL, et al. Solitary idiopathic choroiditis. Enhanced depth imaging optical coherence tomography in 10 cases. Ophthalmology 2013;120:852–858. 16. Fung AT, Fulco EM, Shields CL, Shields JA. Choroidal hemorrhage simulating choroidal melanoma. Findings on enhanced depth imaging spectral domain optical coherence tomography. Retina 2013;33:1726–1728. 17. Shields CL, Shields JA, Gross NE, et al. Survey of 520 eyes with uveal metastases. Ophthalmology 1997;104:1265–1276. 18. Schaudig U, Hassenstein A, Bernd A, et al. Limitations of imaging choroidal tumors in vivo by optical coherence tomography. Graefes Arch Clin Exp Ophthalmol 1998;236:588–592.
EDI-OCT OF CHOROIDAL METASTASIS AL-DAHMASH ET AL 19. Truong SN, Fern CM, Costa DL, Spaide RF. Metastatic breast carcinoma to the retina: optical coherence tomography findings. Retina 2002;22:813–815. 20. Arevalo JF, Fernandez CF, Garcia RA. Optical coherence tomography characteristics of choroidal metastasis. Ophthalmology 2005;112:1612–1619. 21. Natesh S, Chin KJ, Finger PT. Choroidal metastases fundus autofluorescence imaging: correlation to clinical, OCT, and fluorescein angiographic findings. Ophthalmic Surg Lasers Imaging 2010;41:406–412. 22. Sayanagi K, Pelayes DE, Kaiser PK, Singh AD. 3D Spectral domain optical coherence tomography findings in choroidal tumors. Eur J Ophthalmol 2011;21:271–275. 23. Iuliano L, Scotti F, Gagliardi M, et al. SD-OCT patterns of the different stages of choroidal metastases. Ophthalmic Surg Lasers Imaging 2012;43:e30–e34. 24. Muscat S, Parks S, Kemp E, Keating D. Secondary retinal changes associated with choroidal naevi and melanomas documented by optical coherence tomography. Br J Ophthalmol 2004;88:120–124.
1593
25. Shields CL, Mashayekhi A, Dai VV, et al. Optical coherence tomographic findings of combined hamartoma of the retina and retinal pigment epithelium in 11 patients. Arch Ophthalmol 2005;123:1746–1750. 26. Shields CL, Materin MA, Shields JA. Review of optical coherence tomography for intraocular tumors. Curr Opin Ophthalmol 2005;16:141–154. 27. Shields CL, Benevides R, Materin MA, Shields JA. Optical coherence tomography of retinal astrocytic hamartoma in 15 cases. Ophthalmology 2006;113:1553–1557. 28. Shields CL, Perez B, Materin MA, et al. Optical coherence tomography of choroidal osteoma in 22 cases: evidence for photoreceptor atrophy over the decalcified portion of the tumor. Ophthalmology 2007;114:53–58. 29. Say EAT, Shah SU, Ferenczy S, Shields CL. Optical coherence tomography of retinal and choroidal tumors. J Ophthalmol 2012;2012:385058. 30. Witkin AJ, Fischer DH, Shields CL, et al. Enhanced depth imaging spectral-domain optical coherence tomography of a subtle choroidal metastasis. Eye (Lond) 2012;26:1598–1599.
C
pigment epithelium (RPE), particularly the choroid, due to light scattering from the RPE and the vascular interfaces of the choroid. Further limitations are related to motion artifact with subtle eye movement and occasional inaccurate determination of RPE/choriocapillaris layer by the automated software, which necessitates manual correction for OCT thickness measurements.1,2 Enhanced depth imaging (EDI)-OCT, a method described by Spaide et al,2 was designed to improve imaging of the deeper layers of the eye, primarily the choroid and sclera. The technique of EDI-OCT has been used to characterize the normal features and thickness of the choroid in various age groups and refractive states, as well as the features in selected diseases.3–9 Furthermore, EDI-OCT features of choroidal and scleral tumors have been published, particularly regarding choroidal nevus, melanoma, choroidal lymphoma,
onventional spectral domain (SD) optical coherence tomography (OCT) provides cross-sectional retinal images with an axial resolution of 5 mm.1 The limitations of conventional SD-OCT include poor image resolution of structures deep to the retinal From the Ocular Oncology Service, Wills Eye Institute, Thomas Jefferson University, Philadelphia, Pennsylvania. Dr. Al-Dahmash is now at the Ophthalmology Department, College of Medicine, King Saud University, Riyadh, Saudi Arabia; and Dr. Kaliki is now at the Ocular Oncology Service, L V Prasad Eye Institute, Hyderabad, India. None of the authors have any conflicting interests to disclose. Supported by the Eye Tumor Research Foundation, Philadelphia, PA (C.L.S. and J.A.S.). Carol L. Shields has had full access to all the data in the study and takes responsibility for the integrity of the data and the accuracy of the data analysis. Reprint requests: Carol L. Shields, MD, Ocular Oncology Service, Suite 1440, Wills Eye Institute, Thomas Jefferson University, 840 Walnut Street, Philadelphia, PA 19107; e-mail: [email protected]
1588
EDI-OCT OF CHOROIDAL METASTASIS AL-DAHMASH ET AL
sclerochoroidal calcification, and solitary idiopathic choroiditis.10–15 Distinguishing features are many, but the anterior surface of tumors can vary slightly with a smooth mound shape of nevus, a dome shape of a small melanoma, and a “seasick” surface of advanced lymphoma.11–13 Simulating lesions also show characteristic surfaces such as a “rocky and rolling” surface of sclerochoroidal calcification and a scalloped contour of choroidal hemorrhage.14,16 Herein, we describe the “lumpy bumpy” anterior surface of choroidal metastasis on EDI-OCT and delineate other related features.
Methods This was a retrospective observational study approved by the Institutional Review Board of Wills Eye Hospital. All patients diagnosed with choroidal metastasis from January 1, 2011 to November 31, 2011 at the Ocular Oncology Service, Wills Eye Hospital, Philadelphia, PA, underwent EDI-OCT using SD imaging with the Heidelberg Spectralis (Heidelberg Engineering, Heidelberg, Germany). Patients with choroidal metastasis measuring ,3.5 mm thickness by ultrasonography were included for the study. The patient demographic data of age, race, and gender were recorded. The presenting symptoms, primary malignancy site, and best-corrected visual acuity were listed. The choroidal metastasis was evaluated for clinical features including quadrantic location of the tumor epicenter, maximal basal diameter (in millimeters), distance of the posterior margin of the metastasis from the foveola and the optic disk (in millimeters), color (pigmented, nonpigmented, or mixed), and overlying RPE alterations (none, atrophy, or hypertrophy). The presence of subretinal fluid, lipofuscin pigment, and drusen was also noted. These findings were documented with a large fundus drawing and confirmed by fundus photography and ultrasonography. The features recorded on standard ultrasonography included the thickness (in millimeters), configuration (plateau, dome, or mushroom), echogenicity (solid or hollow), choroidal excavation (absent or present), and subretinal fluid (absent or present). Spectral domain EDI-OCT was performed using a Heidelberg Spectralis HRA + OCT (Heidelberg Engineering) and acquisition and analysis software (version 5.3.3.0) with automated EDI. The axial resolution was 3.5 mm with a scan speed of 40,000 A-scans per second. The images were captured using a custom scan acquisition protocol of up to 13 raster lines of 9-mm scan length, with 1,536 A-scans per line. Realtime eye tracking by TruTrack Active Eye Tracking was used, and automatic real-time image averaging
1589
was set at 100 images. Enhanced depth imaging optical coherence tomography was performed using a technique similar to that described by Spaide et al.2 The EDI-OCT images were obtained after pupillary dilatation in all patients. They were evaluated for quality (optimal or suboptimal). The EDI-OCT image was considered optimal and suitable for the study when both anterior and posterior margins of the tumor and the overlying retina were visualized, as well as the lateral margins of the tumor. The images were classified as suboptimal when a portion of the tumor or the overlying retina could not be visualized and were excluded from the study. The optimal EDI-OCT images were then reviewed. The recorded features included thinning of the choriocapillaris (absent or present), the contour of the anterior surface of the tumor (smooth or lumpy bumpy), internal quality of the tumor, maximal tumor thickness (in micrometers), and posterior choroidal shadowing (absent or present). The retinal layers were evaluated for abnormalities (absent, atrophy, hypertrophy, irregularity, or thinning) of the RPE, the interdigitation of the cone outer segment tips, ellipsoid portion of photoreceptors, external limiting membrane, outer nuclear layer, outer plexiform layer, inner nuclear layer, inner plexiform layer, ganglion cell layer, and nerve fiber layer. The presence of subretinal fluid, drusen, and orange pigment were noted and correlated with clinical and fundus autofluorescence findings. Results Of 31 eyes with choroidal metastases that underwent EDI-OCT imaging over an 11-month period, the quality of the image was judged to be optimal in 14 eyes (45%) and suboptimal in 17 eyes (55%). The quality of suboptimal images was found with peripherally located tumors (at or anterior to the equator), thick tumors (.2 mm thickness), presence of media opacity, and in elderly or sickly patients uncooperative for proper imaging. This did not correlate to tumor type. The clinical and ultrasonographic features of the choroidal metastasis are summarized in Table 1. Of 14 eyes (13 patients) with optimal EDI-OCT image, the mean age was 63 years (median, 63 years; range, 43–80 years) (Figures 1 and 2). Ocular symptoms were present in 13 (93%) eyes, and 1 (7%) patient was asymptomatic. The primary malignancy was carcinoma of the breast (n = 7, 50%), lung (n = 5, 36%), pancreas (n = 1, 7%), and thyroid gland (n = 1, 7%). The epicenter of the choroidal metastasis was located within the macula (n = 6, 43%) or outside the macula (n = 8, 57%). The mean maximal basal
1590 RETINA, THE JOURNAL OF RETINAL AND VITREOUS DISEASES Table 1. Enhanced Depth Imaging OCT of 14 Eyes (13 Patients) With Choroidal Metastasis: Demographics and Clinical Features Features Age, mean (median, range), years Race White Others Gender Male Female Symptoms None Present Primary malignancy Breast Lung Pancreas Thyroid Visual acuity $20/40 ,20/40–20/100 ,20/100 Location of choroidal metastasis epicenter Macula Extramacular Distance from the foveola, mean (median, range), mm Distance from the optic disk, mean (median, range), mm Maximal basal diameter, mean (median, range), mm Ultrasonographic thickness, mean (median, range), mm Color Melanotic Amelanotic Mixed Subretinal fluid Present Absent RPE changes None Atrophy Hyperplasia Orange pigment Absent Present Drusen Absent Present Ultrasonographic configuration Plateau Dome Mushroom Ultrasonographic hollowness Choroidal excavation
Number (%) (n = 14) 63 (63, 43–80) 13 (100) 0 (0) 2 (15) 11 (85) 1 (7) 13 (93) 7 5 1 1
(50) (36) (7) (7)
4 (29) 7 (50) 3 (22) 6 (43) 8 (57) 2.0 (2.0, 0–4.0) 2.0 (1.0, 0–7.0) 6.0 (5.0, 3.0–12.0) 2.3 (2.3, 1.0–3.3) 1 (7) 13 (93) 0 (0) 11 (79) 3 (21) 9 (64) 3 (22) 2 (14) 13 (93) 1 (7) 14 (100) 0 (0) 1 13 0 4 2
(7) (93) (0) (29) (14)
2014 VOLUME 34 NUMBER 8
diameter of the choroidal metastasis was 6 mm (median, 5.0 mm; range, 3.0–12.0 mm). Subretinal fluid was clinically visible in 79% (n = 11) of the cases. Other clinical findings included RPE atrophy (n = 3, 21%), RPE hyperplasia (n = 2, 14%), and orange pigment (n = 1, 7%). Ultrasonography showed mean thickness of choroidal metastasis at 2.3 mm (median, 2.3 mm; range, 1.0–3.3 mm). The EDI-OCT characteristics of choroidal metastasis are summarized in Table 2. The average thickness of the choroidal metastasis measured by EDI-OCT was 987 mm (median, 1,014 mm; range, 547–1,511 mm). The EDI-OCT features included choriocapillaris compression/obliteration (n = 13, 93%), homogeneous optical reflectivity along the anterior surface, and the anterior tumor surface being lumpy bumpy (n = 9, 64%) or smooth (n = 5, 36%). The internal quality of the tumor could not be evaluated because of EDIOCT light signal attenuation (shadowing) (n = 12, 86%). The numerous RPE and retinal irregularities are listed in Table 2. The inner retinal layers were normal in all eyes.
Discussion In 1997, Shields et al17 surveyed 520 eyes with choroidal metastases and noted most tumors of small size with mean thickness of 3 mm and with tumor epicenter in the macular region or between the macula and the equator. The small size and posterior location of most choroidal metastases render these tumors ideal for study with OCT. Early reports on time domain and SD-OCT features of choroidal tumors were mainly focused on overlying retinal features because choroidal imaging was relatively poor.18–29 Schaudig et al18 recognized the limitations of the OCT and suggested that OCT was of little value in the differential diagnosis of choroidal tumors. Arevalo et al20 described time-domain OCT findings in 7 patients with choroidal metastasis and noted subretinal fluid (86%), thickened RPE choriocapillaris complex (71%), and little detail of the choroidal features. Natesh et al21 reviewed time domain OCT findings in 10 cases of choroidal metastases and noted elevation of the RPE and retina, RPE thickening and folds, and subretinal fluid. Sayanagi et al,22 in their general report on SD-OCT of choroidal tumors, described thickening and hyperreflectivity of the RPE, overlying retinal thinning, and subretinal fluid in 4 eyes with choroidal metastasis. They concluded that visualization of the choroidal tumor was limited to the anterior aspect with little deeper detail.22 Others have noted similar findings, particularly with marked “irregularity” in the RPE layer.23
EDI-OCT OF CHOROIDAL METASTASIS AL-DAHMASH ET AL
receptors with intact inner retinal layers. C. Enhanced depth imaging OCT at the fovea demonstrates subretinal fluid.
In our study, using high-resolution SD EDI-OCT, the most salient feature of choroidal metastasis was the irregular or lumpy bumpy anterior tumor surface (64% of cases) despite a smooth-appearing surface on ultrasonography. Unlike the smooth “mound” of choroidal nevus and the smooth “dome” of melanoma, choroidal metastasis often showed characteristic lumpy bumpy surface. Choroidal metastasis is an infiltrative process, which is believed to start at the choriocapillaris level. The resolution of EDI-OCT depicts this malignancy with choriocapillaris compression/ obliteration, similar to nevus and melanoma.11,12 Overlying attenuation of the outer retinal layers and RPE was evident in many cases (Table 2). Retinal pigment epithelial abnormalities were clinically visible in 36% compared with EDI-OCT-visible in 78%. An additional benefit of EDI-OCT is the detection of tiny subclinical choroidal metastasis before they are ophthalmoscopically visible, as highlighted by Torres et al10 and Witkin et al.30 In this series, we imaged tumors as small as 547 mm. In these cases, the tumor was present in the outer choroid with loss of large choroidal vessels and nearly invisible to ophthalmoscopy and ultrasonography.
imaging OCT at the fovea demonstrates subretinal fluid.
1591 Fig. 1. Enhanced depth imaging OCT of choroidal metastasis in a 51-year-old woman. A. The amelanotic choroidal metastasis is visible superotemporal to the optic disk. B. Enhanced depth imaging OCT at the center of the tumor showing slightly irregular lumpy bumpy anterior tumor surface, obliteration/compression of choriocapillaris, and choroidal shadowing. There is overlying RPE thinning, structural loss of the interdigitation of the cone outer segment tips, and the ellipsoid portion of photo-
Previous reports on choroidal nevus and small choroidal melanoma have identified the incongruity between ultrasonography and EDI-OCT measurement of tumor thickness.11,12 Likewise, the same discrepancy was noted for choroidal metastasis thickness by ultrasonography versus EDI-OCT. On ultrasonography, the mean tumor thickness was 2.3 mm compared with 0.987 mm on EDI-OCT, an overestimation by 133% on ultrasonography. This could represent more precise measurement with high-resolution EDI-OCT or could be the result of calibration discrepancy with either technique or could be related to misplacement of the calipers on low-resolution ultrasonography. Witkin et al30 noted that tumors ,1 mm in thickness could be undetectable by ultrasonography and better visualized with EDI-OCT. Although our study includes a few patients, we believe that EDI-OCT has shown an obvious improvement over ultrasound for anatomical depiction and resolution of tumor. This alone contributes to our recognition and understanding of this malignancy and could be instrumental in following tumor regression after treatment. In conclusion, SD EDI-OCT is a promising noninvasive tool for high-resolution imaging of choroidal Fig. 2. Enhanced depth imaging OCT of choroidal metastasis in a 63-year-old woman. A. The amelanotic choroidal metastasis is visible superotemporal to the optic disk. B. Enhanced depth imaging OCT at the center of the tumor showing the typical lumpy bumpy anterior tumor surface. There is overlying RPE atrophy and irregularity, structural loss of the interdigitation of the cone outer segment tips, and the ellipsoid portion of photoreceptors with intact inner retinal layers. C. Enhanced depth
1592 RETINA, THE JOURNAL OF RETINAL AND VITREOUS DISEASES Table 2. Enhanced Depth Imaging OCT of 14 Eyes With Choroidal Metastasis: EDI Characteristics of Choroid and Retinal Structures Features
2014 VOLUME 34 NUMBER 8
Key words: eye, tumor, choroid, metastasis, optical coherence tomography, enhanced depth imaging, EDIOCT.
Number (%) n = 14
Choriocapillaris Normal 1 (7) Thinning 13 (93) Anterior tumor surface contour Lumpy bumpy 9 (64) Smooth 5 (36) Thickness of metastasis, mean 987 (1,014, 547–1,511) (median, range), mm Choroidal shadowing Absent 2 (14) Present 12 (86) Bruch membrane Intact 14 (100) Ruptured 0 (0) RPE Normal 3 (22) Atrophy 1 (7) Hypertrophy 5 (36) Thinning 5 (36) Interdigitation of the cone outer segment tips Normal 5 (36) Absent 9 (64) Ellipsoid portion of photoreceptors Normal 6 (43) Absent 8 (57) External limiting membrane Normal 9 (64) Irregular 1 (7) Absent 4 (29) Outer nuclear layer Normal 13 (93) Absent 1 (7) Outer plexiform layer Normal 13 (93) Absent 1 (7) Inner nuclear layer and inward Normal 14 (100) Abnormal 0 (0) Associated features Subretinal fluid 11 (79) Subretinal lipofuscin 1 (7) deposition (orange pigment) Intraretinal cystoid edema 1 (7) Intraretinal noncystoid edema 1 (7)
metastasis, which allows visualization of tumor details and overlying retina. The most obvious finding is the lumpy bumpy tumor surface of choroidal metastasis, as well as overlying subretinal fluid and choriocapillaris compression/obliteration. These findings can assist the clinician in the early detection of choroidal metastasis and in the differentiation of choroidal metastasis from simulating conditions.
References 1. Sakata LM, Deleon-Ortega J, Sakata V, Girkin CA. Optical coherence tomography of the retina and optic nerve—a review. Clin Experiment Ophthalmol 2009;37:90–99. 2. Spaide RF, Koizumi H, Pozzoni MC. Enhanced depth imaging spectral-domain optical coherence tomography. Am J Ophthalmol 2008;146:496–500. 3. Margolis R, Spaide RF. A pilot study of enhanced depth imaging optical coherence tomography of the choroid in normal eyes. Am J Ophthalmol 2009;147:811–815. 4. Spaide RF. Enhanced depth imaging optical coherence tomography of retinal pigment epithelial detachment in age-related macular degeneration. Am J Ophthalmol 2009;147:644–652. 5. Fujiwara T, Imamura Y, Margolis R, et al. Enhanced depth imaging optical coherence tomography of the choroid in highly myopic eyes. Am J Ophthalmol 2009;148:445–450. 6. Imamura Y, Fujiwara T, Margolis R, Spaide RF. Enhanced depth imaging optical coherence tomography of the choroid in central serous chorioretinopathy. Retina 2009;29:1469–1473. 7. Maruko I, Iida T, Sugano Y, et al. Subfoveal choroidal thickness after treatment of Vogt- Koyanagi-Harada disease. Retina 2011;31:510–517. 8. Imamura Y, Iida T, Maruko I, et al. Enhanced depth imaging optical coherence tomography of the sclera in dome-shaped macula. Am J Ophthalmol 2011;151:297–302. 9. Chung SE, Kang SW, Lee JH, Kim YT. Choroidal thickness in polypoidal choroidal vasculopathy and exudative age-related macular degeneration. Ophthalmology 2011;118:840–845. 10. Torres VL, Brugnoni N, Kaiser PK, Singh AD. Optical coherence tomography enhanced depth imaging of choroidal tumors. Am J Ophthalmol 2011;151:586–593. 11. Shah SU, Kaliki S, Shields CL, et al. Enhanced depth imaging optical coherence tomography of choroidal nevus in 104 cases. Ophthalmology 2012;119:1066–1072. 12. Shields CL, Kaliki S, Rojanaporn D, et al. Enhanced depth imaging optical coherence tomography of small choroidal melanoma. Comparative analysis to choroidal nevus. Arch Ophthalmol 2012;130:850–856. 13. Arias JD, Kumar N, Fulco EAM, et al. Seasick choroid: a finding on enhanced depth imaging spectral domain optical coherence tomography (EDI-OCT) of choroidal lymphoma. Retina Cases Brief Rep 2013;7:19–22. 14. Fung AT, Arias JD, Shields CL, Shields JA. Sclerochoroidal calcification is primarily a scleral condition based on enhanced depth imaging optical coherence tomography. JAMA Ophthalmol 2013;131:960–963. 15. Fung AT, Kaliki S, Shields CL, et al. Solitary idiopathic choroiditis. Enhanced depth imaging optical coherence tomography in 10 cases. Ophthalmology 2013;120:852–858. 16. Fung AT, Fulco EM, Shields CL, Shields JA. Choroidal hemorrhage simulating choroidal melanoma. Findings on enhanced depth imaging spectral domain optical coherence tomography. Retina 2013;33:1726–1728. 17. Shields CL, Shields JA, Gross NE, et al. Survey of 520 eyes with uveal metastases. Ophthalmology 1997;104:1265–1276. 18. Schaudig U, Hassenstein A, Bernd A, et al. Limitations of imaging choroidal tumors in vivo by optical coherence tomography. Graefes Arch Clin Exp Ophthalmol 1998;236:588–592.
EDI-OCT OF CHOROIDAL METASTASIS AL-DAHMASH ET AL 19. Truong SN, Fern CM, Costa DL, Spaide RF. Metastatic breast carcinoma to the retina: optical coherence tomography findings. Retina 2002;22:813–815. 20. Arevalo JF, Fernandez CF, Garcia RA. Optical coherence tomography characteristics of choroidal metastasis. Ophthalmology 2005;112:1612–1619. 21. Natesh S, Chin KJ, Finger PT. Choroidal metastases fundus autofluorescence imaging: correlation to clinical, OCT, and fluorescein angiographic findings. Ophthalmic Surg Lasers Imaging 2010;41:406–412. 22. Sayanagi K, Pelayes DE, Kaiser PK, Singh AD. 3D Spectral domain optical coherence tomography findings in choroidal tumors. Eur J Ophthalmol 2011;21:271–275. 23. Iuliano L, Scotti F, Gagliardi M, et al. SD-OCT patterns of the different stages of choroidal metastases. Ophthalmic Surg Lasers Imaging 2012;43:e30–e34. 24. Muscat S, Parks S, Kemp E, Keating D. Secondary retinal changes associated with choroidal naevi and melanomas documented by optical coherence tomography. Br J Ophthalmol 2004;88:120–124.
1593
25. Shields CL, Mashayekhi A, Dai VV, et al. Optical coherence tomographic findings of combined hamartoma of the retina and retinal pigment epithelium in 11 patients. Arch Ophthalmol 2005;123:1746–1750. 26. Shields CL, Materin MA, Shields JA. Review of optical coherence tomography for intraocular tumors. Curr Opin Ophthalmol 2005;16:141–154. 27. Shields CL, Benevides R, Materin MA, Shields JA. Optical coherence tomography of retinal astrocytic hamartoma in 15 cases. Ophthalmology 2006;113:1553–1557. 28. Shields CL, Perez B, Materin MA, et al. Optical coherence tomography of choroidal osteoma in 22 cases: evidence for photoreceptor atrophy over the decalcified portion of the tumor. Ophthalmology 2007;114:53–58. 29. Say EAT, Shah SU, Ferenczy S, Shields CL. Optical coherence tomography of retinal and choroidal tumors. J Ophthalmol 2012;2012:385058. 30. Witkin AJ, Fischer DH, Shields CL, et al. Enhanced depth imaging spectral-domain optical coherence tomography of a subtle choroidal metastasis. Eye (Lond) 2012;26:1598–1599.
