ENHANCED DEPTH IMAGING OPTICAL COHERENCE TOMOGRAPHY OF CHOROIDAL METASTASIS HAKAN DEMIRCI, MD, ALEXIS CULLEN, CRA, JEFFREY M. SUNDSTROM, MD, PHD Purpose: To evaluate the imaging features of choroidal metastasis using enhanced depth imaging optical coherence tomography (EDI-OCT). Methods: Enhanced depth imaging optical coherence tomography imaging features of 24 choroidal metastatic tumors with ,2.5 mm of thickness on ultrasonography were retrospectively evaluated. Results: Of the 24 tumors, 10 (42%) were located at the macula and 14 (58%) between the macula and equator. Tumor was plateau-shaped in 18 (75%) tumors and dome-shaped in 6 (25%) tumors. On EDI-OCT, choroidal metastasis showed low internal optical reflectivity in 17 (71%) tumors and high internal reflectivity in 7 (29%) tumors. Most common associated features were overlying choriocapillaris thinning in 24 (100%) tumors, shaggy or irregular, elongated photoreceptors in 18 (75%) tumors, subretinal fluid with high-reflective speckles in 16 (67%) tumors, and thickening of retinal pigment epithelium in 9 (37%) tumors. The mean tumor thickness was 854 mm (range, 287–1500 mm) on EDI-OCT and 2064 mm (range, 0–2400 mm) on ultrasonography. After treatment, the mean decrease in tumor thickness was 520 mm (range, 134–917 mm) on EDI-OCT and 714 mm (range, 0–1500 mm) on ultrasonography. Internal optical reflectivity changed from low to high in five tumors. Choriocapillaris thinning and shaggy photoreceptors improved in four and five tumors, respectively. Subretinal fluid resolved in four tumors. Conclusion: On EDI-OCT, choroidal metastasis showed overlying choriocapillaris thinning, plateau-shaped tumor, shaggy photoreceptors, and subretinal fluid with highreflective speckles in more than half of the tumors. Enhanced depth imaging optical coherence tomography was more sensitive than ultrasonography in the evaluation of small metastatic tumors at presentation and after treatment. RETINA 34:1354–1359, 2014
O
introduced a technique called enhanced depth imaging using fourier domain optical coherence tomography (EDI-OCT).3–5 In this technique, improved resolution of the deeper layers of the choroid and sclera are obtained by moving the OCT objective lens closer to the eye. Enhanced depth imaging optical coherence tomography has been used in the imaging of various retinal and choroidal conditions.6–8 By using EDI-OCT, Torres et al9 showed that various choroidal tumors could be imaged and measured including those that were not detectable by ultrasonography. In a review of EDI-OCT features of 104 cases with choroidal nevus, Shah et al10 reported that EDIOCT enabled precise measurement of tumor thickness and found that the overlying choriocapillaris thinning was present in 94% of nevi. In a review of EDI-OCT features of small choroidal melanomas, Shields et al11
ptical coherence tomography (OCT) is a noncontact noninvasive diagnostic imaging device that provides high resolution, real-time, and in situ visualization of the tissue microstructure.1 With axial resolution up to 5 mm, OCT has been widely used in the imaging of the retina and the retinal pigment epithelium (RPE) as well as optic disk and anterior segment structures.2 However, OCT has limitations in imaging the deeper layers of the choroid and sclera. Spaide et al From the Ocular Oncology Clinic, Department of Ophthalmology and Visual Sciences, W.K. Kellogg Eye Center, University of Michigan, Ann Arbor, Michigan. Supported by Mrs. and Mr. Witham. None of the authors have any financial/conflicting interests to disclose. Reprint requests: Hakan Demirci, MD, Department of Ophthalmology and Visual Sciences, Ocular Oncology Clinic, University of Michigan, W.K. Kellogg Eye Center, 1000 Wall Street, Ann Arbor, MI 48105; e-mail: [email protected]
1354
EDI-OCT IN CHOROIDAL METASTASIS DEMIRCI ET AL
reported that subretinal fluid, lipofuscin deposition, and retinal irregularities such as shaggy photoreceptors were common findings. Compared with EDI-OCT, ultrasonography overestimated the thickness of small choroidal melanoma by 55%. In this study, we evaluated the EDI-OCT features of choroidal metastatic tumors and the changes in these features after treatment.
Methods Upon institutional review board approval, patients with choroidal metastasis managed in the Ocular Oncology Clinic, Department of Ophthalmology and Visual Sciences, University of Michigan, W.K. Kellogg Eye Center, and who were imaged with EDI-OCT were included in this study. On ultrasonography, choroidal metastatic tumors with ,2.5 mm of thickness whose both posterior and anterior margins could be imaged, were included in this study. Enhanced depth imaging optical coherence tomography images that show both anterior and posterior margins of the tumor and overlying retina were evaluated in the study, and the ones that show a portion of the tumor and the overlying retina were excluded from the study. During the study period, 30 patients with choroidal metastasis were managed. Twelve of them were excluded because the posterior margin of their tumors was not visible on EDI-OCT. In the remaining 18 patients, the posterior margins of the metastatic tumor was visible on ED-OCT. Clinical and imaging features of tumors in these 18 patients were evaluated retrospectively. The EDI-OCT was performed by using Heidelberg Spectralis HRA + OCT (Heidelberg Engineering, Carlsbad, CA) using acquisition and analysis software. The images were obtained using an image acquisition protocol of 13 raster lines of 9-mm image length, with 1536 A-scans per line and automatic real-time averaging set at 100 images. Demographic data included age and gender. Clinical tumor data included the diagnosis of primary systemic cancer, location of the tumor epicenter, tumor basal diameter (millimeters), and associated subretinal fluid and RPE changes. Enhanced depth imaging optical coherence tomography images were assessed for tumor thickness, configuration, optical shadowing, optical reflectivity, choriocapillaris thinning, subretinal fluid with speckles, shaggy photoreceptors, and intraretinal fluid. Overlying retina was also analyzed. Tumor thickness was measured by ultrasonography (micrometers) and EDI-OCT (micrometers). On EDI-OCT, the tumor thickness was measured by two independent observers (A.C., J.M.S.) at the scan showing the greatest tumor thickness by placing the calipers
1355
anteriorly at the base of RPE and posteriorly at the tumor base which was judged to be at the junction of hyperreflective inner sclera. Similarly, on ultrasonography, tumor thickness was measured by two independent observers (A.C., J.M.S.) on cross vector A-scan (Eyecubed, Ellex, Minneapolis, MN) by placing the calipers at the echogenic spikes corresponding to the tumor apex and scleral base. The discrepancies were resolved by a review of the scans by the senior member of the team (H.D.) (Figures 1–3). Results Demographic features of 18 patients are presented in Table 1. Overall, 24 choroidal metastatic tumors in 18 patients were analyzed. Clinical features of these tumors are presented in Table 2. The EDI-OCT features of 24 choroidal metastatic tumors are shown in Table 3. Compared with EDI-OCT, ultrasonography overestimated the tumor thickness by a mean of 59% (range, 15–66%) (Figure 1A-C). The mean difference in tumor thickness measured by ultrasonography versus EDI-OCT was 536 mm (P , 0.05). In five tumors, the thickness of metastatic tumor was only able to be measured by EDI-OCT, whereas they were not detectable by ultrasonography (Figure 2A-C). After therapy, EDI-OCT features were evaluated in seven tumors. Of the seven tumors, four were treated with photodynamic therapy, two with external beam radiotherapy, and one with chemotherapy. The mean decrease in tumor thickness was 520 mm (range, 134– 917 mm) by EDI-OCT and 714 mm (range, 0–1500 mm) by ultrasonography. After therapy, overlying choriocapillaris became more visible, and speckles in subretinal fluid resolved in four tumors (Figure 3A-F). Shaggy photoreceptors improved in five tumors. Internal optical reflectivity changed from low reflective to high reflective in five tumors and remained low reflective in two tumors. Discussion Early studies of OCT provided detailed information about the retinal and RPE features of choroidal tumors including metastasis with limited information from the choroid.12–14 In 2005, Arevalo et al15 reviewed secondary retinal changes associated with choroidal metastasis using time domain OCT findings in seven eyes of four patients. They described the morphologic structures of the retina and RPE overlying the metastatic tumor and commented that OCT was limited by the choroidal location of metastasis. Without providing any thickness measurement, they pointed that it was impossible to have both base and surface of the tumor on the same
1356 RETINA, THE JOURNAL OF RETINAL AND VITREOUS DISEASES
2014 VOLUME 34 NUMBER 7
Fig. 1. A. A 63-year-old woman presents with choroidal metastasis in her left eye from clear cell sarcoma of the tendons and aponeuroses. B. On EDIOCT, metastatic tumor appeared as low internal reflective mass with overlying choriocapillaris thinning and shaggy photoreceptors. C. On ultrasonography, acoustically hollow dome-shaped mass was observed.
scan for lesions .2 mm. Since this study, despite the improvement in OCT technology, OCT features of choroidal metastasis have not been evaluated. By deeper imaging ability, EDI-OCT improved our understanding of the choroid and provided detailed information about the morphologic structure of the choroid in choroidal tumors. In this study, EDI-OCT imaged the whole morphology of the choroidal metastatic tumor when it was ,2.5 mm in thickness. Additionally, it showed the imaging features of the retina, RPE, and choroid in the same scan. In the retina, shaggy photoreceptors (79%), subretinal fluid with speckles (75%), loss of external limiting membrane (50%), abnormality of photoreceptors (27%), loss of inner segment/outer segment junction (27%), irregularity of inner plexiform layer (9%), and irregularity of ganglion cell layer (9%) were the main EDI-OCT features of metastatic tumor. In a comparative study between EDI-OCT imaging features of benign and malignant melanocytic tumors, Shields et al11 reported that some retinal imaging features were significantly different in choroidal melanoma from choroidal nevi. These retinal features have been observed in other conditions with subretinal fluid. Morphologic studies of central serous chorioretinopathy by OCT showed the loss of internal segment/ outer segment junction, and elongated, swollen, and irregular photoreceptors in the presence of subretinal fluid.16,17 These elongated swollen photoreceptors
were called shaggy photoreceptors by Shields et al.11 When the subretinal fluid persisted for several weeks, the outer segments of photoreceptors become granular and shed. We think high-reflective speckles in subretinal fluid observed in this study are shed outer segments of the photoreceptors. These high-reflective spots were also observed in 86% of eyes overlying the RPE–choriocapillaris complex by Arevalo et al.15 In the choroid, thinning of choriocapillaris over the metastatic tumor (100%), plateau-shaped tumor elevation (elevated tumor with flat wavy RPE apical surface) (75%), low internal optical reflectivity (71%), and partial choroidal shadowing deep to the tumor (71%) were the main EDI-OCT features of choroidal metastasis. Although the number of each primary tumor is small, we did not observe any difference in imaging features of EDI-OCT of metastatic tumors from various primary cancers. Similar choroidal EDI-OCT imaging features were reported in malignant melanocytic tumors.11 On EDI-OCT, the plateau-shaped elevation (elevated tumor with wavy RPE apical surface) of metastatic tumors is different from the dome-shaped elevation of melanocytic tumors.11 The involvement of deep, posterior choroidal layers with sparing of choriocapillaris suggests that systemic tumor cells penetrate out of the large, porous choroidal blood vessels much easier and form metastatic tumor in this layer. In this study, tumor thickness was assessed with both EDI-OCT and ultrasonography. Compared with
Fig. 2. A. A 54-year-old woman presents with bilateral choroidal metastasis from breast cancer. B. Enhanced depth imaging optical coherence tomography showed that metastasis had low internal reflectivity and a thickness of 484 mm. C. Ultrasonography could not detect the tumor.
1357
EDI-OCT IN CHOROIDAL METASTASIS DEMIRCI ET AL
Fig. 3. A. A 63-year-old woman presented with solitary choroidal metastasis form lung cancer superior to the optic disk in her left eye. B. On EDIOCT, metastatic tumor appeared as low internal reflective mass with overlying choriocapillaris thinning and shaggy photoreceptors. C. On ultrasonography, dome-shaped mass was observed. D. After photodynamic therapy, metastatic tumor became flat and atrophic. E. On EDI-OCT, the treated tumor appeared flat with high internal reflectivity. Overlying shaggy receptors improved. F. However, ultrasonography could not detect the tumor.
EDI-OCT, ultrasonography overestimated the tumor thickness by a mean of 59% (range, 15–66%). This discrepancy in measurements is consistent with previous reports and seems to be consistent across various choroidal lesions.10,11,18 Shah et al10 and later Shields et al11 reported that choroidal nevus and small melanomas were measured 54% and 55% less on EDI-OCT than ultrasonography. This discrepancy has been attributed to the greater axial resolution of EDI-OCT (3–4 mm) compared with ultrasonography (500–200 mm). This improved resolution allows for a more precise determination of ocular structures and improved placement of calipers while measuring thickness. This improvement in structural resolution will Table 1. Demographic Features of 18 Patients With Choroidal Metastasis Age, years Mean Median Range Sex Male, n (%) Female, n (%) Primary cancer Breast cancer, n (%) Lung cancer, n (%) Renal cell carcinoma, n (%) Thyroid cancer, n (%) Carcinoid tumor, n (%) Pancreas cancer, n (%) Cutaneous melanoma, n (%) Clear cell sarcoma of the knee, n (%)
61 63 42–73 7 (39) 11 (61) 7 4 2 1 1 1 1 1
(39) (22) (11) (6) (6) (6) (6) (6)
likely be particularly beneficial for especially small choroidal lesions. In this study, five choroidal metastatic tumors that were not detectable by ultrasonography could be identified, and measured by EDI-OCT. However, current EDI-OCT technology has the limitation to measure thicker tumors, and ultrasonography is still an important diagnostic tool in these tumors. Table 2. Clinical Features of 24 Choroidal Metastatic Tumors Clinical Features Location of tumor epicenter Macula Superior Inferior Nasal Tumor basal diameter, mm Mean Median Range Tumor thickness on ultrasonography, mm Mean Median Range Configuration on ultrasonography Oblong Dome Flat Associated findings on examination Subretinal fluid Retina pigment epithelium changes
Patient Number (%) 10 9 3 2
(42) (38) (13) (8)
8.2 7.0 1–18 2.0 1.6 0–2.4 13 (4) 6 (38) 5 (21) 15 (63) 6 (21)
1358 RETINA, THE JOURNAL OF RETINAL AND VITREOUS DISEASES Table 3. Enhanced Depth Imaging Optical Coherence Tomography Features of Choroid in 24 Choroidal Metastatic Tumors EDI-OCT Features Tumor thickness, mm Mean Median Range Configuration Dome Plateau Optical shadowing Complete Partial None Optical reflectivity Low optical reflection High optical reflection Associated findings Overlying choriocapillaris thinning Shaggy photoreceptors Subretinal fluid with speckles Intraretinal fluid Associated overlying retinal findings Bruch membrane thickening/atrophy RPE thickening/atrophy Photoreceptor outer segment abnormality Inner segment–outer segment abnormality External limiting membrane abnormality Outer nuclear layer abnormality Outer plexiform layer abnormality Inner nuclear layer abnormality Inner plexiform layer abnormality Ganglion cell layer abnormality Nerve fiber layer abnormality
Patient Number (%) 854 811 287–1500 6 (25) 18 (75) 5 (21) 17 (71) 2 (8) 17 (71) 7 (29) 24 18 16 1
(100) (79) (75) (4)
6 (25) 13 (53) 7 (27)
2014 VOLUME 34 NUMBER 7
Image analysis and measurements are performed by authors and subject to individual interpretation and image quality. As shown previously, the image quality of EDI-OCT may depend on several factors including the location of tumor, tumor thickness, and patient cooperation.10 The diagnosis of choroidal metastasis was established based on the medical history of the patient, clinical features of the tumor, and needle biopsy specimen if available. There might be a bias in the analysis of size and location of small metastatic tumors because we only included tumors whose both anterior and posterior margins could be imaged. In summary, low internal reflectivity, plateau shape with wavy RPE apical surface, overlying choriocapillaris thinning, shaggy photoreceptors, and speckles in the subretinal fluid were common EDI-OCT imaging features of metastatic choroidal tumors. Enhanced depth imaging optical coherence tomography was more sensitive than ultrasonography in the evaluation of small metastatic tumors at presentation and after treatment. Additionally, ultrasonography overestimated the thickness of small metastatic tumors by a mean of 59% when compared with EDI-OCT. Key words: eye, uvea, choroids, metastasis, tumor, EDI-OCT, OCT, imaging.
7 (27)
References 12 (50) 7 6 3 3 3 3
(27) (23) (9) (9) (9) (9)
Given the better resolution noted above, the possibility of using EDI-OCT to determine response to intervention was investigated in small choroidal metastatic tumors. As would be predicted, the longitudinal assessment of EDI-OCT features of small metastatic tumors (,2.5 mm in thickness) can be used to determine response to treatment both in terms of tumor thickness and other imaging characteristics. We observed that EDI-OCT provided longitudinal follow-up of thickness in all treated seven tumors, while ultrasonography could measure thickness in only five tumors after treatment. As the metastatic tumor became inactive, it appeared as a mass with high internal optical reflectivity consistent with an increase in tissue fibrosis. There are several limitations in this study. The relatively small patient number precludes the ability to generalize the current findings to the general population.
1. Huang D, Swanson EA, Lin CP, et al. Optical coherence tomography. Science 1991;254:1178–1181. 2. Gabriele ML, Wollstein G, Ishikawa H, et al. Optical coherence tomography: history, current status, and laboratory work. Invest Ophthalmol Vis Sci 2011;52:2425–2436. 3. Spaide RF, Koizumi H, Pozonni MC. Enhanced depth imaging spectral-domain optical coherence tomography. Am J Ophthalmol 2008;146:496–500. 4. 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. 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. Maruko I, Lida T, Sugano Y, et al. Subfoveal choroidal thickness after treatment of central serous chorioretinopathy. Ophthalmology 2010;117:1792–1799. 7. 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. 8. Dhoot DS, Huo S, Yuan A, et al. Evaluation of choroidal thickness in retinitis pigmentosa using enhanced depth imaging optical coherence tomography. Br J Ophthalmol 2013;97:66–69. 9. Torres VL, Brugnoni N, Kaiser PK, Singh AD. Optical coherence tomography enhanced depth imaging of choroidal tumors. Am J Ophthalmol 2011;151:586–593. 10. 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. 11. Shields CL, Kaliki S, Rojanaporn D, et al. Enhanced depth imaging optical coherence tomography of small choroidal melanoma:
EDI-OCT IN CHOROIDAL METASTASIS DEMIRCI ET AL comparison with choroidal nevus. Arch Ophthalmol 2012;130: 850–856. 12. Shields CL, Mashayekhi A, Materin MA, et al. Optical coherence tomography of choroidal nevus in 120 patients. Retina 2005;25:243–252. 13. 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. 14. Espinoza G, Rosenblatt B, Harbour JW. Optical coherence tomography in the evaluation of retinal changes associated with suspicious choroidal melanocytic tumors. Am J Ophthalmol 2004;137:90–95.
1359
15. Arevalo JF, Fernandez CF, Garcia RA. Optical coherence tomography characteristics of choroidal metastasis. Ophthalmology 2005;112:1612–1619. 16. Fujimoto H, Gomi F, Wakabayashi T, et al. Morphologic changes in acute central serous chorioretinopathy evaluated by fourier-domain optical coherence tomography. Ophthalmology 2008;115:1494–1500. 17. Matsumoto H, Kishi S, Sato T, Mukai R. Fundus autofluorescence of elongated photoreceptor outer segments in central serous chorioretinopathy. Am J Ophthalmol 2011;151:617–623. 18. Mrejen S, Fung AT, Silverman RH, et al. Potential pitfalls in measuring the thickness of small melanocytic tumors with ultrasonography. Retina 2013;33:1293–1299.
O
introduced a technique called enhanced depth imaging using fourier domain optical coherence tomography (EDI-OCT).3–5 In this technique, improved resolution of the deeper layers of the choroid and sclera are obtained by moving the OCT objective lens closer to the eye. Enhanced depth imaging optical coherence tomography has been used in the imaging of various retinal and choroidal conditions.6–8 By using EDI-OCT, Torres et al9 showed that various choroidal tumors could be imaged and measured including those that were not detectable by ultrasonography. In a review of EDI-OCT features of 104 cases with choroidal nevus, Shah et al10 reported that EDIOCT enabled precise measurement of tumor thickness and found that the overlying choriocapillaris thinning was present in 94% of nevi. In a review of EDI-OCT features of small choroidal melanomas, Shields et al11
ptical coherence tomography (OCT) is a noncontact noninvasive diagnostic imaging device that provides high resolution, real-time, and in situ visualization of the tissue microstructure.1 With axial resolution up to 5 mm, OCT has been widely used in the imaging of the retina and the retinal pigment epithelium (RPE) as well as optic disk and anterior segment structures.2 However, OCT has limitations in imaging the deeper layers of the choroid and sclera. Spaide et al From the Ocular Oncology Clinic, Department of Ophthalmology and Visual Sciences, W.K. Kellogg Eye Center, University of Michigan, Ann Arbor, Michigan. Supported by Mrs. and Mr. Witham. None of the authors have any financial/conflicting interests to disclose. Reprint requests: Hakan Demirci, MD, Department of Ophthalmology and Visual Sciences, Ocular Oncology Clinic, University of Michigan, W.K. Kellogg Eye Center, 1000 Wall Street, Ann Arbor, MI 48105; e-mail: [email protected]
1354
EDI-OCT IN CHOROIDAL METASTASIS DEMIRCI ET AL
reported that subretinal fluid, lipofuscin deposition, and retinal irregularities such as shaggy photoreceptors were common findings. Compared with EDI-OCT, ultrasonography overestimated the thickness of small choroidal melanoma by 55%. In this study, we evaluated the EDI-OCT features of choroidal metastatic tumors and the changes in these features after treatment.
Methods Upon institutional review board approval, patients with choroidal metastasis managed in the Ocular Oncology Clinic, Department of Ophthalmology and Visual Sciences, University of Michigan, W.K. Kellogg Eye Center, and who were imaged with EDI-OCT were included in this study. On ultrasonography, choroidal metastatic tumors with ,2.5 mm of thickness whose both posterior and anterior margins could be imaged, were included in this study. Enhanced depth imaging optical coherence tomography images that show both anterior and posterior margins of the tumor and overlying retina were evaluated in the study, and the ones that show a portion of the tumor and the overlying retina were excluded from the study. During the study period, 30 patients with choroidal metastasis were managed. Twelve of them were excluded because the posterior margin of their tumors was not visible on EDI-OCT. In the remaining 18 patients, the posterior margins of the metastatic tumor was visible on ED-OCT. Clinical and imaging features of tumors in these 18 patients were evaluated retrospectively. The EDI-OCT was performed by using Heidelberg Spectralis HRA + OCT (Heidelberg Engineering, Carlsbad, CA) using acquisition and analysis software. The images were obtained using an image acquisition protocol of 13 raster lines of 9-mm image length, with 1536 A-scans per line and automatic real-time averaging set at 100 images. Demographic data included age and gender. Clinical tumor data included the diagnosis of primary systemic cancer, location of the tumor epicenter, tumor basal diameter (millimeters), and associated subretinal fluid and RPE changes. Enhanced depth imaging optical coherence tomography images were assessed for tumor thickness, configuration, optical shadowing, optical reflectivity, choriocapillaris thinning, subretinal fluid with speckles, shaggy photoreceptors, and intraretinal fluid. Overlying retina was also analyzed. Tumor thickness was measured by ultrasonography (micrometers) and EDI-OCT (micrometers). On EDI-OCT, the tumor thickness was measured by two independent observers (A.C., J.M.S.) at the scan showing the greatest tumor thickness by placing the calipers
1355
anteriorly at the base of RPE and posteriorly at the tumor base which was judged to be at the junction of hyperreflective inner sclera. Similarly, on ultrasonography, tumor thickness was measured by two independent observers (A.C., J.M.S.) on cross vector A-scan (Eyecubed, Ellex, Minneapolis, MN) by placing the calipers at the echogenic spikes corresponding to the tumor apex and scleral base. The discrepancies were resolved by a review of the scans by the senior member of the team (H.D.) (Figures 1–3). Results Demographic features of 18 patients are presented in Table 1. Overall, 24 choroidal metastatic tumors in 18 patients were analyzed. Clinical features of these tumors are presented in Table 2. The EDI-OCT features of 24 choroidal metastatic tumors are shown in Table 3. Compared with EDI-OCT, ultrasonography overestimated the tumor thickness by a mean of 59% (range, 15–66%) (Figure 1A-C). The mean difference in tumor thickness measured by ultrasonography versus EDI-OCT was 536 mm (P , 0.05). In five tumors, the thickness of metastatic tumor was only able to be measured by EDI-OCT, whereas they were not detectable by ultrasonography (Figure 2A-C). After therapy, EDI-OCT features were evaluated in seven tumors. Of the seven tumors, four were treated with photodynamic therapy, two with external beam radiotherapy, and one with chemotherapy. The mean decrease in tumor thickness was 520 mm (range, 134– 917 mm) by EDI-OCT and 714 mm (range, 0–1500 mm) by ultrasonography. After therapy, overlying choriocapillaris became more visible, and speckles in subretinal fluid resolved in four tumors (Figure 3A-F). Shaggy photoreceptors improved in five tumors. Internal optical reflectivity changed from low reflective to high reflective in five tumors and remained low reflective in two tumors. Discussion Early studies of OCT provided detailed information about the retinal and RPE features of choroidal tumors including metastasis with limited information from the choroid.12–14 In 2005, Arevalo et al15 reviewed secondary retinal changes associated with choroidal metastasis using time domain OCT findings in seven eyes of four patients. They described the morphologic structures of the retina and RPE overlying the metastatic tumor and commented that OCT was limited by the choroidal location of metastasis. Without providing any thickness measurement, they pointed that it was impossible to have both base and surface of the tumor on the same
1356 RETINA, THE JOURNAL OF RETINAL AND VITREOUS DISEASES
2014 VOLUME 34 NUMBER 7
Fig. 1. A. A 63-year-old woman presents with choroidal metastasis in her left eye from clear cell sarcoma of the tendons and aponeuroses. B. On EDIOCT, metastatic tumor appeared as low internal reflective mass with overlying choriocapillaris thinning and shaggy photoreceptors. C. On ultrasonography, acoustically hollow dome-shaped mass was observed.
scan for lesions .2 mm. Since this study, despite the improvement in OCT technology, OCT features of choroidal metastasis have not been evaluated. By deeper imaging ability, EDI-OCT improved our understanding of the choroid and provided detailed information about the morphologic structure of the choroid in choroidal tumors. In this study, EDI-OCT imaged the whole morphology of the choroidal metastatic tumor when it was ,2.5 mm in thickness. Additionally, it showed the imaging features of the retina, RPE, and choroid in the same scan. In the retina, shaggy photoreceptors (79%), subretinal fluid with speckles (75%), loss of external limiting membrane (50%), abnormality of photoreceptors (27%), loss of inner segment/outer segment junction (27%), irregularity of inner plexiform layer (9%), and irregularity of ganglion cell layer (9%) were the main EDI-OCT features of metastatic tumor. In a comparative study between EDI-OCT imaging features of benign and malignant melanocytic tumors, Shields et al11 reported that some retinal imaging features were significantly different in choroidal melanoma from choroidal nevi. These retinal features have been observed in other conditions with subretinal fluid. Morphologic studies of central serous chorioretinopathy by OCT showed the loss of internal segment/ outer segment junction, and elongated, swollen, and irregular photoreceptors in the presence of subretinal fluid.16,17 These elongated swollen photoreceptors
were called shaggy photoreceptors by Shields et al.11 When the subretinal fluid persisted for several weeks, the outer segments of photoreceptors become granular and shed. We think high-reflective speckles in subretinal fluid observed in this study are shed outer segments of the photoreceptors. These high-reflective spots were also observed in 86% of eyes overlying the RPE–choriocapillaris complex by Arevalo et al.15 In the choroid, thinning of choriocapillaris over the metastatic tumor (100%), plateau-shaped tumor elevation (elevated tumor with flat wavy RPE apical surface) (75%), low internal optical reflectivity (71%), and partial choroidal shadowing deep to the tumor (71%) were the main EDI-OCT features of choroidal metastasis. Although the number of each primary tumor is small, we did not observe any difference in imaging features of EDI-OCT of metastatic tumors from various primary cancers. Similar choroidal EDI-OCT imaging features were reported in malignant melanocytic tumors.11 On EDI-OCT, the plateau-shaped elevation (elevated tumor with wavy RPE apical surface) of metastatic tumors is different from the dome-shaped elevation of melanocytic tumors.11 The involvement of deep, posterior choroidal layers with sparing of choriocapillaris suggests that systemic tumor cells penetrate out of the large, porous choroidal blood vessels much easier and form metastatic tumor in this layer. In this study, tumor thickness was assessed with both EDI-OCT and ultrasonography. Compared with
Fig. 2. A. A 54-year-old woman presents with bilateral choroidal metastasis from breast cancer. B. Enhanced depth imaging optical coherence tomography showed that metastasis had low internal reflectivity and a thickness of 484 mm. C. Ultrasonography could not detect the tumor.
1357
EDI-OCT IN CHOROIDAL METASTASIS DEMIRCI ET AL
Fig. 3. A. A 63-year-old woman presented with solitary choroidal metastasis form lung cancer superior to the optic disk in her left eye. B. On EDIOCT, metastatic tumor appeared as low internal reflective mass with overlying choriocapillaris thinning and shaggy photoreceptors. C. On ultrasonography, dome-shaped mass was observed. D. After photodynamic therapy, metastatic tumor became flat and atrophic. E. On EDI-OCT, the treated tumor appeared flat with high internal reflectivity. Overlying shaggy receptors improved. F. However, ultrasonography could not detect the tumor.
EDI-OCT, ultrasonography overestimated the tumor thickness by a mean of 59% (range, 15–66%). This discrepancy in measurements is consistent with previous reports and seems to be consistent across various choroidal lesions.10,11,18 Shah et al10 and later Shields et al11 reported that choroidal nevus and small melanomas were measured 54% and 55% less on EDI-OCT than ultrasonography. This discrepancy has been attributed to the greater axial resolution of EDI-OCT (3–4 mm) compared with ultrasonography (500–200 mm). This improved resolution allows for a more precise determination of ocular structures and improved placement of calipers while measuring thickness. This improvement in structural resolution will Table 1. Demographic Features of 18 Patients With Choroidal Metastasis Age, years Mean Median Range Sex Male, n (%) Female, n (%) Primary cancer Breast cancer, n (%) Lung cancer, n (%) Renal cell carcinoma, n (%) Thyroid cancer, n (%) Carcinoid tumor, n (%) Pancreas cancer, n (%) Cutaneous melanoma, n (%) Clear cell sarcoma of the knee, n (%)
61 63 42–73 7 (39) 11 (61) 7 4 2 1 1 1 1 1
(39) (22) (11) (6) (6) (6) (6) (6)
likely be particularly beneficial for especially small choroidal lesions. In this study, five choroidal metastatic tumors that were not detectable by ultrasonography could be identified, and measured by EDI-OCT. However, current EDI-OCT technology has the limitation to measure thicker tumors, and ultrasonography is still an important diagnostic tool in these tumors. Table 2. Clinical Features of 24 Choroidal Metastatic Tumors Clinical Features Location of tumor epicenter Macula Superior Inferior Nasal Tumor basal diameter, mm Mean Median Range Tumor thickness on ultrasonography, mm Mean Median Range Configuration on ultrasonography Oblong Dome Flat Associated findings on examination Subretinal fluid Retina pigment epithelium changes
Patient Number (%) 10 9 3 2
(42) (38) (13) (8)
8.2 7.0 1–18 2.0 1.6 0–2.4 13 (4) 6 (38) 5 (21) 15 (63) 6 (21)
1358 RETINA, THE JOURNAL OF RETINAL AND VITREOUS DISEASES Table 3. Enhanced Depth Imaging Optical Coherence Tomography Features of Choroid in 24 Choroidal Metastatic Tumors EDI-OCT Features Tumor thickness, mm Mean Median Range Configuration Dome Plateau Optical shadowing Complete Partial None Optical reflectivity Low optical reflection High optical reflection Associated findings Overlying choriocapillaris thinning Shaggy photoreceptors Subretinal fluid with speckles Intraretinal fluid Associated overlying retinal findings Bruch membrane thickening/atrophy RPE thickening/atrophy Photoreceptor outer segment abnormality Inner segment–outer segment abnormality External limiting membrane abnormality Outer nuclear layer abnormality Outer plexiform layer abnormality Inner nuclear layer abnormality Inner plexiform layer abnormality Ganglion cell layer abnormality Nerve fiber layer abnormality
Patient Number (%) 854 811 287–1500 6 (25) 18 (75) 5 (21) 17 (71) 2 (8) 17 (71) 7 (29) 24 18 16 1
(100) (79) (75) (4)
6 (25) 13 (53) 7 (27)
2014 VOLUME 34 NUMBER 7
Image analysis and measurements are performed by authors and subject to individual interpretation and image quality. As shown previously, the image quality of EDI-OCT may depend on several factors including the location of tumor, tumor thickness, and patient cooperation.10 The diagnosis of choroidal metastasis was established based on the medical history of the patient, clinical features of the tumor, and needle biopsy specimen if available. There might be a bias in the analysis of size and location of small metastatic tumors because we only included tumors whose both anterior and posterior margins could be imaged. In summary, low internal reflectivity, plateau shape with wavy RPE apical surface, overlying choriocapillaris thinning, shaggy photoreceptors, and speckles in the subretinal fluid were common EDI-OCT imaging features of metastatic choroidal tumors. Enhanced depth imaging optical coherence tomography was more sensitive than ultrasonography in the evaluation of small metastatic tumors at presentation and after treatment. Additionally, ultrasonography overestimated the thickness of small metastatic tumors by a mean of 59% when compared with EDI-OCT. Key words: eye, uvea, choroids, metastasis, tumor, EDI-OCT, OCT, imaging.
7 (27)
References 12 (50) 7 6 3 3 3 3
(27) (23) (9) (9) (9) (9)
Given the better resolution noted above, the possibility of using EDI-OCT to determine response to intervention was investigated in small choroidal metastatic tumors. As would be predicted, the longitudinal assessment of EDI-OCT features of small metastatic tumors (,2.5 mm in thickness) can be used to determine response to treatment both in terms of tumor thickness and other imaging characteristics. We observed that EDI-OCT provided longitudinal follow-up of thickness in all treated seven tumors, while ultrasonography could measure thickness in only five tumors after treatment. As the metastatic tumor became inactive, it appeared as a mass with high internal optical reflectivity consistent with an increase in tissue fibrosis. There are several limitations in this study. The relatively small patient number precludes the ability to generalize the current findings to the general population.
1. Huang D, Swanson EA, Lin CP, et al. Optical coherence tomography. Science 1991;254:1178–1181. 2. Gabriele ML, Wollstein G, Ishikawa H, et al. Optical coherence tomography: history, current status, and laboratory work. Invest Ophthalmol Vis Sci 2011;52:2425–2436. 3. Spaide RF, Koizumi H, Pozonni MC. Enhanced depth imaging spectral-domain optical coherence tomography. Am J Ophthalmol 2008;146:496–500. 4. 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. 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. Maruko I, Lida T, Sugano Y, et al. Subfoveal choroidal thickness after treatment of central serous chorioretinopathy. Ophthalmology 2010;117:1792–1799. 7. 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. 8. Dhoot DS, Huo S, Yuan A, et al. Evaluation of choroidal thickness in retinitis pigmentosa using enhanced depth imaging optical coherence tomography. Br J Ophthalmol 2013;97:66–69. 9. Torres VL, Brugnoni N, Kaiser PK, Singh AD. Optical coherence tomography enhanced depth imaging of choroidal tumors. Am J Ophthalmol 2011;151:586–593. 10. 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. 11. Shields CL, Kaliki S, Rojanaporn D, et al. Enhanced depth imaging optical coherence tomography of small choroidal melanoma:
EDI-OCT IN CHOROIDAL METASTASIS DEMIRCI ET AL comparison with choroidal nevus. Arch Ophthalmol 2012;130: 850–856. 12. Shields CL, Mashayekhi A, Materin MA, et al. Optical coherence tomography of choroidal nevus in 120 patients. Retina 2005;25:243–252. 13. 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. 14. Espinoza G, Rosenblatt B, Harbour JW. Optical coherence tomography in the evaluation of retinal changes associated with suspicious choroidal melanocytic tumors. Am J Ophthalmol 2004;137:90–95.
1359
15. Arevalo JF, Fernandez CF, Garcia RA. Optical coherence tomography characteristics of choroidal metastasis. Ophthalmology 2005;112:1612–1619. 16. Fujimoto H, Gomi F, Wakabayashi T, et al. Morphologic changes in acute central serous chorioretinopathy evaluated by fourier-domain optical coherence tomography. Ophthalmology 2008;115:1494–1500. 17. Matsumoto H, Kishi S, Sato T, Mukai R. Fundus autofluorescence of elongated photoreceptor outer segments in central serous chorioretinopathy. Am J Ophthalmol 2011;151:617–623. 18. Mrejen S, Fung AT, Silverman RH, et al. Potential pitfalls in measuring the thickness of small melanocytic tumors with ultrasonography. Retina 2013;33:1293–1299.
