Retinal Pigment Epithelial Depigmented Lesions Associated With Tuberous Sclerosis Complex FREE

Tuberous sclerosis complex (TSC) is a multisystem hamartomatous disorder that can occur in nearly every tissue in the body but primarily affects the skin, brain, and eye. Major and minor diagnostic criteria for TSC have been established by the Tuberous Sclerosis Consensus Conference in 1998¹ (eTable 1). A diagnosis of definite TSC is established by the presence of 2 major features or 1 major feature plus 2 minor features. A diagnosis of probable TSC requires the presence of 1 major feature and 1 minor feature, and a diagnosis of possible TSC requires the presence of either 1 major feature or 2 or more minor features.

The most common eye finding of tuberous sclerosis is the retinal astrocytic hamartoma.² Other features include eyelid angiofibroma, iris atrophy, uveal coloboma, and retinal achromatic patch. The retinal achromatic patch is not clearly described, and this could represent a flat astrocytic hamartoma or retinal pigment epithelial (RPE) “punched-out” depigmentation. The relationship of RPE punched-out depigmentation with TSC is not well recognized. Herein, we analyze the relationship of RPE depigmented lesions with TSC.

The computer-coded records of all patients evaluated on the Ocular Oncology Service at Wills Eye Institute in Philadelphia, Pennsylvania, were reviewed for TSC (eTable 1) or retinal astrocytic hamartoma between July 1974 and November 2010. Each chart was evaluated for patient and tumor features. The affected eye was evaluated for the presence, multiplicity, location, and basal dimension of retinal astrocytic hamartoma(s) and RPE depigmented lesions.

Statistical evaluation was performed to evaluate the relationship of RPE depigmentation with clinical features of TSC using logistic regression analysis (Table). The specific features of RPE lesions were evaluated based on definite vs probable/possible TSC using logistic regression analysis (eTable 2). The demographic and baseline characteristics of the patients were compared using the Fisher exact test, the independent sample t test, or the Wilcoxon test as appropriate. The mean age between groups was compared using the t test. The related number of retinal astrocytic hamartomas per eye and the basal dimensions were compared using the Wilcoxon rank sum test. All proportions were compared using the Fisher exact test. The odds ratios (ORs) were estimated for those factors identified as significant. A multivariate logistic regression analysis using forward stepwise method was performed to further identify the significant factors associated with RPE lesions.

Of 56 patients with definite TSC (23 patients) or probable/possible TSC (33 patients), RPE depigmented lesions were found in 12 patients (12%) (Figures 1 and 2 and eFigure). The clinical features are listed in our Table. Those manifesting RPE lesions demonstrated significant factors of definite TSC (P < .001; OR, 29.3), seizure history (P = .008; OR, 7.8), mental retardation (P = .006; OR, 21.5), cutaneous adenoma sebaceum (P = .008; OR, 7.6), ash leaf macule (P = .002; OR, 10.6), an increased number of retinal astrocytic hamartomas per eye (P < .001; OR, 1.6), and an extramacular location of retinal astrocytic hamartoma (P = .002; OR, 3.2) (Table).

The RPE lesions were found in patients at median of 9 years, compared with patients without RPE lesions who were a median of 25 years (Table). The median number of RPE lesions was 1 per eye (range, up to 7 per eye) (eTable 2). The median RPE lesion diameter was 0.5 mm and did not differ between the definite TSC and the probable/possible TSC groups. Of the 38 RPE lesions in 12 eyes, the lesion sector location was most often superonasal (9 lesions [24%]), inferotemporal (8 lesions [21%]), inferonasal (7 lesions [18%]), and superotemporal (5 lesions [13%]) (eTable 2). There were no lesions in the macular region. The anteroposterior location of the RPE lesions was most commonly in the macula to equator (24 lesions [63%]) or equator to ora serrata region (10 lesions [26%]).

RPE depigmented lesions were not clearly identified as a marker of TSC until recently. In 1975, Shelton³ reported 7 patients with TSC and noted that one had “punched-out” depigmented spots in the peripheral retina labeled (presumed incorrectly) as “ocular histoplasmosis.” Others have since recognized these lesions and speculated a relationship to TSC.^4- 5

In 2001, Rowley and associates⁵ explored the relationship of RPE “punched-out” lesions with TSC. In a population-based study of 100 patients with TSC, they found retinal hamartomas (44%), RPE punched-out areas (39%), eyelid angiofibromas (39%), and other features. The RPE lesions were all less than 1.5 mm and were located in the retina mid-periphery. Compared with age-matched controls, RPE lesions were significantly associated with TSC (P < .001), similar to our results.

In our clinic-based population, RPE lesions appeared in patients diagnosed at a younger age (median, 9 vs 25 years). Significant findings included a clinical diagnosis of definite (vs probable/possible) TSC, history of seizures or mental retardation, and findings of adenoma sebaceum, ash leaf macule, and specific features of retinal astrocytic hamartomas.

In summary, focal RPE depigmentation is an important finding in TSC and could be helpful in establishing the diagnosis. This often-subtle finding could be overlooked during examination or misconstrued as unrelated RPE defect.