Visual Acuity in 3422 Consecutive Eyes with Choroidal Nevus

CLINICAL SCIENCES Visual Acuity in 3422 Consecutive Eyes With Choroidal Nevus

Carol L. Shields, MD; Minoru Furuta, MD; Arman Mashayekhi, MD; Edwina L. Berman, MBBS; Jonathan D. Zahler, DO; Daniel M. Hoberman, BS; Diep H. Dinh, BS; Jerry A. Shields, MD

Objective: To evaluate visual acuity in eyes with cho- foveolar choroidal nevus, respectively. By multivariate roidal nevus. analysis, factors predictive of visual loss of 3 or more log- MAR lines included subfoveolar nevus location (rela- Design: This was an observational case series. Of 3422 tive risk [RR], 15.52), juxtapapillary nevus location (RR, consecutive eyes with choroidal nevus, vision loss at 15 4.52), initial visual acuity of 20/50 or worse (RR, 15.40), years occurred in 2% of eyes with extrafoveolar nevus overlying retinal pigment epithelial detachment (RR, and in 26% of eyes with subfoveolar nevus, particularly 22.16), and foveal edema (RR, 9.02). Factors predictive those with overlying retinal pigment epithelial detach- of poor final visual acuity of 20/200 or worse included ment and foveal edema. A retrospective medical record subfoveolar nevus location (RR, 11.32), overlying or- review was conducted, with evaluation of visual acuity ange pigment (RR, 3.68), overlying retinal pigment at presentation and at final examination. The main out- epithelial detachment (RR, 12.80), and foveal edema (RR, come measure was visual acuity. 18.72). Results: The median visual acuity at presentation was Conclusion: Mild vision loss over many years should be 20/20 for eyes with either extrafoveolar or subfoveolar choroidal nevus. Using Kaplan-Meier estimates, vision anticipated in patients with subfoveolar choroidal ne- loss of 3 or more logarithm of the minimum angle of reso- vus, particularly those with overlying retinal pigment epi- lution (logMAR) lines at 5, 10, and 15 years occurred in thelial detachment, orange pigment, and foveal edema. less than 1%, 1%, and 2% of eyes with extrafoveolar nevus compared with 15%, 20%, and 26% of eyes with sub- Arch Ophthalmol. 2007;125(11):1501-1507

HOROIDAL NEVUS IS THE and found 22 (11%) with visual acuity loss. most common clinically The vision loss was because of subfoveal detected intraocular tu- fluid (50%), presumed photoreceptor de- mor.1,2 In the Blue Moun- generation (42%), and choroidal neovas- tains Eye Study,3 choroi- cularization (8%).8 In 2005, Shields and as- dalC nevi were found in 7% of the white sociates11 evaluated optical coherence population. This benign tumor manifests as tomography (OCT) of the retina overlying a pigmented or nonpigmented mass deep 120 consecutive patients with choroidal ne- to the retina, often with overlying drusen vus to better ascertain the reasons for vi- and retinal pigment epithelial (RPE) alter- sual loss and found overlying retinal edema ations.1-7 Choroidal nevus can produce cen- (15%), photoreceptor attenuation (51%), tral vision loss and peripheral visual field retinal thinning (22%), subretinal fluid loss.4,8-11 Rarely, choroidal nevus can evolve (26%), and RPE detachment (12%). In this into malignant melanoma.7,12-18 report, we analyze a large cohort of 3422 Visual field defects were documented in eyes with stable choroidal nevus to ascer- 38% of 42 eyes with choroidal nevus evalu- tain initial and final visual acuity, loss of vi- ated by Tamler and Maumenee4 and in 85% sual acuity over time, and factors related to of 21 eyes analyzed by Flindall and Drance10 visual acuity outcomes. using static and kinetic techniques. In 1971, Naumann and associates6 found central vi- Author Affiliations: Ocular METHODS Oncology Service, Wills Eye sual acuity loss in 13 of 124 eyes (10%) with 8 Institute, Thomas Jefferson choroidal nevus. Gonder and coworkers University, Philadelphia, later described 206 patients with choroi- A retrospective medical record review was Pennsylvania. dal nevi posterior to the equator of the eye performed on all patients with the clinical

(REPRINTED) ARCH OPHTHALMOL / VOL 125 (NO. 11), NOV 2007 WWW.ARCHOPHTHALMOL.COM 1501

©2007 American Medical Association. All rights reserved. Downloaded From: https://jamanetwork.com/ on 10/02/2021 Table 1. Choroidal Nevus in 3422 Eyes of 3187 Patients: Comparison of Clinical Features at Initial Examination for Extrafoveolar vs Subfoveolar Nevus

Location of Choroidal Nevusa

Clinical Feature Extrafoveolar Subfoveolar Age, y (n=3187)b 60 (62±15) [4-97] 58 (60±18) [4-93] Race (n=3187) White 2945 (98.5) 196 (99.0) African American 21 (0.7) 0 Hispanic 16 (0.5) 2 (1.0) Asian 6 (0.2) 0 Asian Indian 1 (0.03) 0 Sex (n=3187) Male 1095 (36.6) 75 (37.9) Female 1894 (63.4) 123 (62.1) Symptoms (n=3422) Decreased vision 160 (5.0) 62 (30.2) Visual field defect 33 (1.0) 5 (2.4) Photopsia or floaters 140 (4.4) 11 (5.4) None 2884 (89.6) 127 (62.0) Snellen visual acuity (logMAR) (n=3422) 20/20-20/40 (0-0.3) 2984 (92.8) 154 (75.1) 20/50-20/100 (0.4-0.7) 178 (5.5) 39 (19.0) 20/200 or worse (Ն 1.0) 55 (1.7) 12 (5.9) Tumor base, mm (n=3422)b 5.1 (5.0±2.8) [0.4-20.0] 5.7 (4.5±3.8) [1.0-24.0] Tumor thickness, mm (n=3422)b 1.6 (1.5±0.5) [0.6-4.5] 1.7 (1.6±0.5) [0.7-3.5] Tumor color (n=3422) Pigmented 2460 (76.5) 169 (82.4) Nonpigmented 347 (10.8) 27 (13.2) Mixed 410 (12.7) 9 (4.4) Related retinal or RPE findings (n=3422) Foveal edema 11 (0.3) 14 (6.8) Retinal invasion 5 (0.2) 1 (0.5) Subretinal fluid 285 (9.0) 60 (29.3) Subfoveal fluid 57 (1.8) 37 (18.0) Orange pigment 185 (5.8) 53 (25.9) RPE hyperplasia 240 (7.6) 9 (4.4) RPE detachment 33 (1.0) 9 (4.4) RPE fibrous metaplasia 252 (7.9) 11 (5.4) RPE atrophy 356 (11.2) 12 (5.9) Drusen 1735 (54.7) 84 (41.0) Choroidal neovascularization 13 (0.4) 7 (3.4)

Abbreviations: logMAR, logarithm of the minimum angle of resolution; RPE, retinal pigment epithelium. a Data are given as number (percentage) of each group unless otherwise indicated. Percentages may not total 100 because of rounding. For the extrafoveolar group, n=2989 eyes for race and sex; n=3217 eyes for symptoms, Snellen visual acuity, and tumor color; and n=3172 eyes for related retinal/RPE findings. For the subfoveolar group, n=198 for race and sex; and n=205 for symptoms, Snellen visual acuity, tumor color, and related retinal/RPE findings. b Data are given as mean (median±SD) [range].

diagnosis of choroidal nevus evaluated at the Ocular Oncol- distance of the tumor margin to the optic disc margin and ogy Service at Wills Eye Institute between April 1, 1970, and foveola (in millimeters), largest tumor basal dimension and June 1, 2006. Institutional review board approval was thickness (in millimeters), tumor color (pigmented, mixed, or obtained for this retrospective study. Patients with evidence of nonpigmented), and presence of amelanotic halo around the tumor transformation into melanoma were not included in nevus (halo nevus). Other related data included subretinal the analysis. Clinical data were collected at initial examination fluid, orange pigment, drusen, RPE alterations (hyperplasia, regarding patient age, race, sex, medical history (dysplastic detachment, fibrosis, and atrophy), and choroidal neovascular nevus syndrome; cutaneous, choroidal, or conjunctival mela- membrane. The status of the foveola (involvement with noma; or neurofibromatosis), ocular melanocytosis, symp- underlying choroidal nevus, subretinal fluid, and retinal toms, and best-corrected visual acuity by Snellen charts. edema) was recorded. The final best-corrected visual acuity at Vision data were evaluated using logarithm of the minimum date last seen was recorded. angle of resolution (logMAR) conversion. Data regarding spe- A series of univariate Cox proportional hazards regressions cific features of the tumor were collected from large detailed assessed the degree of relationship of all of the variables previ- fundus drawings that were made of each patient at first exami- ously listed to 3 outcomes, including the initial visual acuity, nation and from fundus photographs. These data included final visual acuity, and visual acuity loss of 3 or more logMAR tumor epicenter quadrantic location (inferior, temporal, supe- lines. All of the variables were analyzed as discrete variables rior, nasal, or macular), tumor epicenter anteroposterior loca- except for patient age at presentation, tumor basal dimension, tion (macular, macular to equator, or equator to ora serrata), tumor thickness, and distance of the tumor to the optic disc

(REPRINTED) ARCH OPHTHALMOL / VOL 125 (NO. 11), NOV 2007 WWW.ARCHOPHTHALMOL.COM 1502

Figure 1. Clinical appearance of pigmented extrafoveolar (A) vs pigmented subfoveolar (B) nevus in eyes without overlying retinal or retinal pigment epithelial changes (the visual acuity was 20/20 in both eyes).

margin and foveola, which were evaluated as continuous variables. Subsequent multivariate models included variables that Table 2. Choroidal Nevus in 2334 Eyes: Comparison were significant on a univariate level (PϽ.05) to identify the of Visual Acuity Loss of 3 or More LogMAR Lines combination of factors best related to the 3 outcomes. Kaplan- in Extrafoveolar vs Subfoveolar Nevus Using Kaplan-Meier Estimates Meier survival estimates were calculated on time to loss of 3 or more lines of logMAR visual acuity. Kaplan-Meier Estimates, % RESULTS Location of the Choroidal Nevus At2y At5y At10y At15y There were 3422 eyes of 3187 patients with choroidal Extrafoveolar (n=2207) Ͻ1 Ͻ11 2 nevus. The mean patient age at presentation was 60 years Subfoveolar (n=127) 7 15 20 26 All choroidal nevi (N=2334) Ͻ112 3 (median, 62 years; range, 4-97 years). The patient and tumor features are listed in Table 1. The tumor was sub- Abbreviation: logMAR, logarithm of the minimum angle of resolution. foveolar in 205 (6.0%) eyes and extrafoveolar in 3217 (94.0%) eyes. The mean tumor base and thickness was 5.7 and 1.7 mm for the subfoveolar nevi, respectively, and 5.1 and 1.6 mm for the extrafoveolar nevi, respec- The multivariate analyses for initial and final visual tively. Eyes with subfoveolar nevus displayed subretinal acuity of 20/50 to 20/100 and 20/200 or worse in the fluid in 60 cases and foveal retinal edema in 14 cases, com- entire group of 3422 eyes is listed in Table 3. The pared with 285 cases and 11 cases, respectively, in eyes most important factors for poor final visual acuity of with extrafoveolar nevus (Table 1). 20/200 or worse included macular location of the nevus The mean initial logMAR visual acuity for extrafoveo- and overlying orange pigment, RPE detachment, and lar and subfoveolar choroidal nevi was 0 (Snellen equiva- foveal edema (Table 3). Factors predictive of loss of 3 lent, 20/20). The initial median±SD logMAR visual acu- or more logMAR lines included reduced initial visual ity for extrafoveolar nevi was 0.09±0.22 (range, 0-3.00); acuity of 20/50 (logMAR, 0.4) or worse, subfoveolar and for subfoveolar nevi, 0.20±0.35 (range, 0-2.00). The location, juxtapapillary location, nevus thickness final median±SD logMAR visual acuity for extrafoveo- greater than 2 mm, related RPE detachment, and foveal lar nevi was 0.14±0.31 (range, 0-4.00); and for subfo- edema (Table 4). Factors predictive of intermediate veolar nevi, 0.34±0.53 (range, 0-3.00). Of the 2334 pa- (20/50-20/100 [logMAR, 0.4-0.7]) or poor (20/200 or tients with stable choroidal nevi who returned for a worse [logMAR, 1.0 or worse]) visual acuity at initial follow-up examination, the mean final logMAR visual acu- and final examination in eyes with subfoveolar choroi- ity for extrafoveolar choroidal nevi was 0.10 (Snellen dal nevus using multivariate analysis are listed in equivalent, 20/25); and for subfoveolar choroidal nevi, Table 5. In the 127 eyes with subfoveolar choroidal 0.18 (Snellen equivalent, 20/30) (Figure 1). The mean nevus for which the patient returned for a follow-up follow-up was 5 years (range, 3 months to 36 years). examination, factors predictive of loss of 3 or more log- Kaplan-Meier estimates of 3 lines of logMAR visual acu- MAR lines included Hispanic race, intermediate (20/50 ity loss at 2, 5, 10, and 15 years are given in Table 2 or worse) initial visual acuity, and overlying orange pig- and in Figure 2 and Figure 3. ment (Table 6).

(REPRINTED) ARCH OPHTHALMOL / VOL 125 (NO. 11), NOV 2007 WWW.ARCHOPHTHALMOL.COM 1503

C D

Figure 2. Clinical appearance of subfoveolar nevus with overlying retinal and retinal pigment epithelial (RPE) changes and good long-term visual acuity. Subfoveolar choroidal nevus with overlying RPE detachment and visual acuity of 20/20 at 18 months’ follow-up (A). Optical coherence tomographic (OCT) scan of the nevus in part A showing the foveola draped on the nasal margin of the RPE detachment (B). Subfoveolar choroidal nevus with overlying orange pigment and a visual acuity of 20/20 at 22 months’ follow-up (C). The OCT scan of the nevus in part C showing the slight elevation of the foveola with optically dense material on the posterior retinal surface correlating with the orange pigment (D).

COMMENT In the current analysis, we specifically analyzed subfoveolar vs extrafoveolar choroidal nevi to appreciate the Choroidal nevus can lead to reduced central and periph- comparative effects of tumor location on central visual acuity. Eyes with subfoveolar nevi composed 205 of the eral visual acuity, depending on its location. Tamler and 8 Maumenee4 demonstrated visual field defects in 38% of 3422 eyes (6.0%). A previous study found subfoveal lo- 42 choroidal nevi, but remarked that it was not clear at cation of choroidal nevus in 28 of 375 patients (7.5%). that time how a choroidal nevus could affect the retinal At initial examination, eyes with subfoveolar choroidal function if the retina was not involved with tumor. Nau- nevus displayed a mean visual acuity of 20/20, similar mann and associates9 later described a patient with para- to those with an extrafoveolar nevus. However, on final central scotoma from a choroidal nevus, found on his- examination, eyes with a subfoveolar choroidal nevus had topathological examination to have loss of the outer retinal a slightly reduced mean visual acuity of 20/30 com- layers with complete loss of the rods and cones overly- pared with 20/25 in those with an extrafoveolar nevus. ing the tumor. These findings sufficiently explained the The difference between these 2 groups was more remark- symptomatic scotoma. able when assessing loss of visual acuity over time, be- Optical coherence tomography has been useful in cause 26% of the subfoveolar nevus group showed loss further delineating the extent of retinal damage overly- of 3 logMAR lines of visual acuity by 15 years, whereas ing choroidal nevus. Shields and associates11 used OCT only 2% of the extrafoveolar group manifested a similar to analyze the retina overlying 120 consecutive choroi- loss. In fact, eyes with a subfoveolar nevus had a 16 times dal nevi and found overlying retinal edema in 15%, higher RR for vision loss compared with eyes with an ex- photoreceptor thinning or complete absence in 51%, trafoveolar nevus (Table 4). and general retinal thinning in 22%.11 These in vivo In the entire group of 2334 eyes with choroidal ne- OCT findings corroborated previous histopathological vus for which the patient returned for follow-up, the most findings.5,6,9 important factors for intermediate (20/50-20/100) or poor

(REPRINTED) ARCH OPHTHALMOL / VOL 125 (NO. 11), NOV 2007 WWW.ARCHOPHTHALMOL.COM 1504

C D

E F

G H

Figure 3. Clinical appearance of subfoveolar nevus with overlying retinal and retinal pigment epithelial (RPE) changes and reduced long-term visual acuity. Subfoveolar choroidal nevus with overlying fibrous metaplasia of the RPE and a visual acuity of 20/40 at 18 months’ follow-up (A). Optical coherence tomographic (OCT) scan of the nevus in part A showing thickening of the RPE layer, particularly under the foveola, slight subretinal fluid, and diffuse optical density in the photoreceptor layer overlying the nevus, suggesting photoreceptor disruption (B). Subfoveolar choroidal nevus with overlying subtle orange pigment and a visual acuity of 20/25 at 51 months (C). The OCT scan of the nevus in part C showing overlying subretinal fluid and debris on the posterior retinal surface consistent with orange pigment (D). Circumpapillary subfoveolar choroidal nevus with subtle overlying orange pigment, mild fibrous metaplasia of the RPE, and a visual acuity of counting fingers at 21 months’ follow-up (E). The OCT scan of the nevus in part E showing extensive confluent cystoid macular edema (F). Subfoveolar choroidal nevus with overlying fibrous metaplasia of the RPE and a visual acuity of 20/100 at 52 months’ follow-up (G). The OCT of the nevus in part G showing dramatic cystoid macular edema (H).

(REPRINTED) ARCH OPHTHALMOL / VOL 125 (NO. 11), NOV 2007 WWW.ARCHOPHTHALMOL.COM 1505

©2007 American Medical Association. All rights reserved. Downloaded From: https://jamanetwork.com/ on 10/02/2021 Table 3. Factors Predictive of Intermediate or Poor Visual Table 4. Factors Predictive of Visual Acuity Loss Acuity at Initial and Final Examination in Eyes of 3 or More LogMAR Lines in 2334 Eyes With Choroidal Nevus Using Multivariate Analysisa With Choroidal Nevus Using Multivariate Analysis

Relative Risk P Relative Risk P Clinical Factor (95% Confidence Interval) Value Clinical Factor (95% Confidence Interval) Value Initial visual acuity of 20/50-20/100 Initial visual acuity of 20/50 15.40 (6.27-37.84) Ͻ.001 Symptoms (decreased vision 12.59 (4.89-32.43) Ͻ.001 or worse (present vs absenta) vs noneb) Nevus distance to optic nerve 4.52 (1.48-13.76) .008 Anteroposterior location of nevus 6.80 (2.66-17.38) Ͻ.001 (0 vs Ͼ0mma) epicenter (macula vs equatorb) Nevus distance to foveola 15.52 (5.49-43.88) Ͻ.001 Subretinal fluid over nevus (0 vs Ͼ0mma) Minimal vs noneb 7.43 (2.21-25.03) .001 Nevus thickness (Ͼ2vsՅ2mma) 3.89 (1.41-10.70) .009 Moderate vs noneb 4.57 (1.49-14.01) .008 RPE detachment over nevus 22.16 (4.40-111.65) Ͻ.001 Initial visual acuity of 20/200 (present vs absenta) or worse Foveal edema (present vs absenta) 9.02 (2.19-37.13) .002 Ultrasonographic acoustic 6.98 (1.54-31.67) .01 quality (hollow vs flatb) Abbreviations: logMAR, logarithm of the minimum angle of resolution; Anteroposterior location of 7.68 (1.40-44.04) .02 RPE, retinal pigment epithelium. nevus epicenter (macula a Reference. vs equatorb) Subretinal fluid over nevus 15.16 (2.86-80.32) .001 (moderate vs noneb) RPE detachment over nevus 16.84 (1.62-175.35) .02 Table 5. Factors Predictive of Intermediate (present vs absentb) or Poor Visual Acuity at Initial and Final Examination Foveal edema (present 26.32 (4.01-172.60) .001 in Eyes With Subfoveolar Choroidal Nevus vs absentb) Using Multivariate Analysisa Final visual acuity of 20/50-20/100 Subretinal fluid over nevus Relative Risk P Minimal vs noneb 7.66 (2.46-23.86) Ͻ.001 Clinical Factor (95% Confidence Interval) Value Moderate vs noneb 5.19 (1.79-15.08) .002 Extensive vs noneb 25.60 (3.95-165.82) .001 Initial visual acuity of 20/50-20/100 Nevus location (subfoveal 20.51 (7.39-56.90) Ͻ.001 Subretinal fluid over nevus vs extrafovealb) Minimal vs noneb 16.24 (1.46-180.48) .02 Final visual acuity of 20/200 Moderate vs noneb 19.87 (1.87-211.25) .01 or worse Initial visual acuity of 20/200 Nevus distance to foveola 11.32 (1.30-98.81) .03 or worse (Յ3vsϾ3mmb) RPE fibrous metaplasia 6.96 (1.23-39.45) .03 Nevus quadratic location (present vs absentb) Macula vs inferiorb 8.92 (1.06-75.27) .04 Final visual acuity of 20/50-20/100 Macula vs temporalb 12.53 (2.44-64.27) .002 Age (Ն65yvsϽ65 yb) 4.35 (1.35-13.95) .01 Orange pigment over nevus 3.68 (1.31-10.35) .01 Initial visual acuity of 20/50 21.23 (2.74-164.44) .003 (present vs absentb) or worse (present vs absentb) RPE detachment over nevus 12.80 (2.44-67.06) .003 Nevus thickness 4.93 (1.43-16.98) .01 (present vs absentb) (Ͼ2vsՅ2mmb) Foveal edema (present vs absentb) 18.72 (5.45-64.33) Ͻ.001 Choroidal neovascularization 16.13 (2.54-102.50) .003 over nevus (present b Abbreviation: RPE, retinal pigment epithelium. vs absent ) a There were 3422 eyes for the initial visual acuity data and 2334 eyes for Final visual acuity of 20/200 the final visual acuity data. Intermediate visual acuity was 20/50 to 20/100 or worse and poor visual acuity was 20/200 or worse. Decreased vision vs noneb 10.12 (1.19-86.13) .03 b Reference. Flashes or floaters vs noneb 23.36 (1.37-399.58) .03 Initial visual acuity of 20/50 6.60 (1.72-25.33) .006 or worse (present vs absentb) (20/200 or worse) final visual acuity included subfoveo- Nevus distance to optic nerve 4.13 (1.08-15.83) .04 b lar nevus location and retinal or RPE changes overlying (0 vs Ͼ0mm ) the nevus, such as foveal edema, subretinal fluid, or- Abbreviation: RPE, retinal pigment epithelium. ange pigment, and RPE detachment (Table 5). Similar a There were 205 eyes for the initial visual acuity data and 127 eyes for the to the previous results, subfoveolar location imparted a final visual acuity data. Intermediate visual acuity was 20/50 to 20/100 and poor visual acuity was 20/200 or worse. 21 times higher RR for intermediate final visual acuity b and an 11 times higher RR for poor final visual acuity Reference. compared with extrafoveolar nevus (Table 3). Some of the related retinal and RPE findings, such as subretinal fluid and orange pigment, that were associated with the Based on published OCT data, retinal and RPE alter- visual outcome have also been shown in previous re- ations are fairly common overlying choroidal nevi and ports13,14,16,18 to predict tumor growth into melanoma. In these features could translate to visual field loss if the ne- this analysis, only stable choroidal nevi without growth vus is extrafoveal and central vision loss if the nevus is were included. subfoveal. In the current analysis of all 2334 eyes with

(REPRINTED) ARCH OPHTHALMOL / VOL 125 (NO. 11), NOV 2007 WWW.ARCHOPHTHALMOL.COM 1506

©2007 American Medical Association. All rights reserved. Downloaded From: https://jamanetwork.com/ on 10/02/2021 Retina Society, Capetown, South Africa (Dr C. L. Shields); Table 6. Factors Predictive of Visual Acuity Loss Mellon Charitable Giving from the Martha W. Rogers of 3 or More LogMAR Lines in 127 Eyes Charitable Trust, Philadelphia, Pennsylvania (Dr C. L. With Subfoveolar Choroidal Nevus Shields); the LuEsther Mertz Retina Research Founda- tion, New York, New York (Dr C. L. Shields); a dona- Relative Risk P Clinical Factorsa (95% Confidence Interval) Value tion from Michael, Bruce, and Ellen Ratner, New York (Drs C. L. Shields and J. A. Shields); the Eye Tumor Re- Race (Hispanic vs whitea) 55.04 (4.49-674.85) .002 Initial visual acuity of 20/50 5.70 (1.56-20.79) .008 search Foundation, Philadelphia (Drs C. L. Shields and or worse (present vs absenta) J. A. Shields); and the Paul Kayser International Award Orange pigment (present 4.36 (1.46-13.01) .008 of Merit in Retina Research, Houston, Texas (Dr J. A. vs absenta) Shields). Previous Presentation: This study was presented at the Abbreviation: LogMAR, logarithm of the minimum angle of resolution. International Congress of Ocular Oncology; June 29, 2007; a Reference. Siena, Italy. Additional Contributions: Rishita Nutheti, Interna- choroidal nevus, the most important factor for poor tional Centre for Advancement of Rural Eye Care, L. V. final visual acuity was foveal edema, which imparted a Prasad Institute, Hyderabad, India, performed the sta- 19 times RR (Table 3), and the most important factor for tistical analysis. visual acuity loss was overlying RPE detachment, which imparted a 22 times RR (Table 4). Most of the informa- tion on the foveal features in our study, which extends REFERENCES over 4 decades, was gathered by clinical examination; and only a few patients underwent OCT imaging of the fo- 1. Shields JA, Shields CL. Choroidal nevus. In: Shields JA, Shields CL, eds. vea. Related clinical features, such as RPE detachment, Intraocular Tumors: A Text and Atlas. Philadelphia, PA: WB Saunders Co; can be difficult to visualize clinically. In an analysis of 1992:85-100. 2. Shields JA, Shields CL. Choroidal nevus. In: Shields JA, Shields CL, eds. Atlas OCT findings of 120 patients with choroidal nevi, RPE of Intraocular Tumors. Philadelphia, PA: Lippincott Williams & Wilkins; 1999: detachment overlying the nevus was found on OCT in 53-59. 12% of cases, whereas it was visualized clinically in only 3. Sumich P, Mitchell P, Wang JJ. Choroidal nevi in a white population: the Blue 2%.11 Optical coherence tomography could be benefi- Mountains Eye Study. Arch Ophthalmol. 1998;116(5):645-650. 4. Tamler E, Maumenee AE. A clinical study of choroidal nevi. AMA Arch Ophthalmol. cial in estimating risks for poor visual outcome. 1959;62(2):196-202. This analysis focused on visual results in a large co- 5. Naumann G, Yanoff M, Zimmerman LE. Histogenesis of malignant melanomata hort of 3422 consecutive patients with choroidal nevus. of the uvea: nevi of choroid and ciliary body. Arch Ophthalmol. 1966;76(6): The results, however, could be biased toward larger or 784-796. thicker nevi because many patients were referred to this 6. Naumann GOH, Hellner K, Naumann LR. Pigmented nevi of the choroid: clinical study of secondary changes in the overlying tissue. Trans Am Acad Ophthalmol ocular oncology center with the diagnosis of possible small Otolaryngol. 1971;75:110-123. choroidal melanoma vs choroidal nevus. In this cohort, 7. Ganley JP, Comstock GW. Benign nevi and malignant melanomas of the choroid. the mean nevus diameter was 5.1 mm and the mean ne- Am J Ophthalmol. 1973;76(1):19-25. vus thickness was 1.6 mm. In comparison, Sumich and 8. Gonder JR, Augsburger JJ, McCarthy EF, Shields JA. Visual loss associated with 3 choroidal nevi. Ophthalmology. 1982;89(8):961-965. associates analyzed 264 choroidal nevi and found a mean 9. Naumann G, Zimmerman LE, Yanoff M. Visual field defect associated with cho- nevus diameter of 1.25 mm (thickness not recorded). roidal nevus. Am J Ophthalmol. 1966;62(5):914-917. In summary, central visual acuity can be diminished 10. Flindall RJ, Drance SM. Visual field studies of benign choroidal melanomata. Arch in eyes with choroidal nevus, particularly if the nevus is Ophthalmol. 1969;81(1):41-44. subfoveolar and there are related retinal and RPE alter- 11. Shields CL, Mashayekhi A, Materin MA, et al. Optical coherence tomography of choroidal nevus in 120 patients. Retina. 2005;25(3):243-252. ations affecting the foveola. Patients with subfoveolar cho- 12. Gass JDM. Problems in the differential diagnosis of choroidal nevi and malig- roidal nevi should be forewarned that visual acuity could nant melanomas: the XXXIII Edward Jackson Memorial Lecture. Am J Ophthalmol. decrease slowly over time. 1977;83(3):299-323. 13. Butler P, Char DH, Zarbin M, Kroll S. Natural history of indeterminate pigmented choroidal tumors. Ophthalmology. 1994;101(4):710-716. Submitted for Publication: March 1, 2007; final revi- 14. Shields CL, Shields JA, Kiratli H, De Potter P, Cater JR. Risk factors for growth sion received April 17, 2007; accepted April 19, 2007. and metastasis of small choroidal melanocytic lesions. Ophthalmology. 1995; Correspondence: Carol L. Shields, MD, Ocular Oncol- 102(9):1351-1361. ogy Service, Wills Eye Institute, 840 Walnut St, Ste 1440, 15. Desjardins L, Lumbroso L, Levy C, Plancher C, Asselain B. Risk factors for the Philadelphia, PA 19107 (carol.shields@shieldsoncology degeneration of the choroid naevi: a retrospective study of 135 cases [in French]. J Fr Ophtalmol. 2001;24(6):610-616. .com). 16. Shields CL, Cater J, Shields JA, Singh AD, Santos MC, Carvalho C. Combination Author Contributions: Dr C. L. Shields had full access to of clinical factors predictive of growth of small choroidal melanocytic tumors. all the data in the study and takes responsibility for the in- Arch Ophthalmol. 2000;118(3):360-364. tegrity of the data and the accuracy of the data analysis. 17. Shields CL, Shields JA. Clinical features of small choroidal melanoma. Curr Opin Ophthalmol. 2002;13(3):135-141. Financial Disclosure: None reported. 18. The Collaborative Ocular Melanoma Study Group. Factors predictive of growth Funding/Support: This study was supported by the Retina and treatment of small choroidal melanoma: COMS report No. 5. Arch Ophthalmol. Research Foundation, Charles L. Schepens Lecture of the 1997;115(12):1537-1544.

(REPRINTED) ARCH OPHTHALMOL / VOL 125 (NO. 11), NOV 2007 WWW.ARCHOPHTHALMOL.COM 1507