We observed two vitiligo patients with extensive ocular involvement. In contrast to previous observations, our vitiligo patients show only hypopigmentation of the choroid, while the RPE cells are normal. Case 1 is a 59-year-old Chinese man who had vitiligo over 30 years and was referred to us because of visual problems. Except for vitiligo, he had no history of general or ocular disease. The ocular examination was normal except for presbyopia and an abnormal fundus appearance. He had no signs of past or present uveitis. Both fundi showed wedge-shaped areas of choroidal hypopigmentation (Figure 1). Fluorescein angiography showed no abnormalities (Figure 2). Additional Humphrey 30-2 and 30/60-2 visual fields and electrodiagnostic tests showed no evidence of retinal dysfunction. Case 2 is a 48-year-old Chinese women with vitiligo who was referred to us because of an abnormal fundus appearance. She had no previous ocular disorders or complaints. The ocular examination was completely normal with the exception of the fundus. Her fundi showed large areas of choroidal hypopigmentation very similar to case 1. Humphrey 30-2 visual fields and electrodiagnostic tests were normal. Large studies suggest that choroidal pigmentary changes occurring in the presence of vitiligo can be ascribed to uveitis in many cases, although in some no evidence of inflammation could be found.1, 2 Only a small number of cases have been reported with widespread hypopigmentation of both choroid and retinal pigment epithelium (RPE).1, 2, 3 Our two patients with vitiligo had extensive choroidal hypopigmentation without signs or a previous history of ocular inflammation. They both had normal visual function, and additional fluorescein angiogram and electrodiagnostic tests in case 1 were normal. These findings suggest that the lack of pigment was confined to the choroid and that the RPE was not involved. The hypopigmentation of the choroid may be congenital and the coexistence of cutaneous vitiligo in our patients may have occurred by chance. Alternatively, this observation may be explained on the basis of the embryological origin of the affected cells. Both choroidal and skin melanocytes are derived from neural crest cells, differing in this respect from RPE cells that are derived from neuroectoderm.4 A disease related to cells of this origin could therefore affect dermal and choroidal melanocytes without involvement with the RPE cells. The pathogenesis of vitiligo is unclear but neural, autoimmune, and autocytotoxic mechanisms have been implicated. In favour of the first is the observation that the skin lesions sometimes follow the borders of dermatomes, and abnormalities of peripheral nerves have been described in areas of vitiligo.5 Thus, the innervation of the choroid may play a role in the pathogenesis of loss of choroidal pigment. Extensive choroidal involvement in vitiligo patients has rarely been reported, and is clearly uncommon.2 However, it may not be detected since it may not always be evident clinically and it gives rise to no functional deficit. Owing to their Asian origin, the normal dark choroidal pigmentation could easily be distinguished from hypopigmented choroid.

Figure 1
figure 1

Case 1: Colour fundus photograph, right eye. Note the cone-shaped area of choroidal hypopigmentation in the area inferior to the optic disc. A similar area was seen superotemporal in the left eye.

Figure 2
figure 2

Case 1: Fluorescein angiogram of the same area as in Figure 1. The areas corresponding to choroidal hypopigmentation show no abnormalities.