744 Goto, Sugita, Okada, Hatano, Ishiguro, Yoshiki, Ohno

found to be CD2, CD3, CD4, CD25, and HLA- lesions and infiltrative lesions by leukaemic DR positive but CD8 negative. It was revealed, cells.2-5 There has been only one report of with the particle agglutination method, that the episcleritis associated with ATL,6 but there has

serum titre for HTLV-I was 25 600 times been no report ofATL that began with recurrent Br J Ophthalmol: first published as 10.1136/bjo.77.11.744 on 1 November 1993. Downloaded from dilution on repeated tests. episcleritis. It has been reported that HTLV-I Also, showed hepato- infects fibroblasts or endothelial cells and megaly and hypercalcaemia. Based on these transforms these membrane antigens.7 In this data, the diagnosis of ATL was established. He case episcleritis recurred many times before was admitted to the department of internal ATL developed, therefore it seems likely that medicine of Yokohama City University Hospital this episcleritis may be related to an immuno- and treated with chemotherapy. Initially he mediated reaction to HTLV-I infected cells. appeared to respond well to the therapy, but Further immunological and molecular biological gradually showed severe complications such as studies are needed to clarify the exact mechanism opportunistic respiratory infection and acute of HTLV-I associated ocular . Recog- renal failure. Hypoxaemia and hypercalcaemia nition by ophthalmologists of the various ocular progressively worsened in spite of intensive inflammatory lesions is important in diagnosing therapy. On 22 November 1991, he died in the ATL. hospital, and necropsy was carried out. Histo- pathological examination ofthe eyes revealed the 1 Uchiyama T, Yodoi J, Sagawa K, Takatsuki K, Uchino H. Adult T-cell leukemia: clinical and hematologic features of slight infiltration of mononuclear cells in the 16 cases. Blood 1977; 50: 481-92. episclera, but ATL cells were not found in the 2 Ohba N, Matsumoto M, Sameshima M, Kabayama Y, Nakao K, Unoki K, et al. Ocular manifestations in patients infected eyes (Fig 3). Interestingly enough, on viral with human T-lymphotropic virus type I. Jpn J Ophthalmol examination the pX region of HTLV-I proviral 1989; 33: 1-12. 3 Roman GC, Vernant JC, Osame M. HTLV-I and the nervous genome was detected from the episclera and the system. Neurology and neurobiology. Vol 51. New York: Alan R with polymerase chain reaction (Fig 4). Liss, 1989. 4 Kohno T, Arita T, Okamura R. Ocular manifestations in human T-lymphotropic virus type I infection. Folia Ophthal- molJpn 1990; 41: 2182-8. 5 Lauer SA, Fischer J, Jones J, Gartner S, Dutcher J, Hoxie JA. Comment Orbital T-cell lymphoma in human T-cell leukemia virus Adult T cell leukaemia is a haematological type I infection. 1988; 95: 110-5. 6 Kabayama Y, Isashiki M, Ohba N, Arima N. Ocular disorders malignancy of high mortality with an increase of associated with adult T-cell leukemia. JpnJ Clin Ophthalmol atypical lymphocytes characterised by multiple 1988; 42: 139-41. 7 Hoxie JA, Matthews DM, Cines DB. Infection of human irregularly lobulated nuclei. Its major ocular endothelial cells by human T-cell leukemia virus type I. Proc manifestations are opportunistic infectious Nat Acad Sci USA 1984; 81: 7591.

BritishlJournal ofOphthalmology 1993; 77: 744-746 Development ofretinal vascular leakage and cystoid macular oedema secondary to central serous http://bjo.bmj.com/ chorioretinopathy

Howard Schatz, Mark D Osterloh, H Richard McDonald, Robert N Johnson on October 1, 2021 by guest. Protected copyright.

We present what we believe to be the first case of , nor any sign of subretinal neo- central serous chorioretinopathy to be diagnosed, vascularisation. The left macula showed a few documented angiographically, and followed RPE changes. A fluorescein angiogram con- and which then developed retinal vascular firmed these findings and showed that there was leakage and cystoid macular oedema. We have an RPE leak into the subretinal space from the documented this development with serial fundus shallow RPE detachment in the nasal fovea (Figs photography and fluorescein angiography. IB, IC). A careful stereo examination of the retinal vessels of both fundus photographs and fluorescein angiograms confirmed that the retinal Case report vessels and capillaries were entirely normal and 1 Daniel Burnham Court, A 43-year-old man had reduced vision in the therewas no cystoid macular oedema. A diagnosis Suite 210, San Francisco, right eye for 2 months before being evaluated. of central serous chorioretinopathy was made CA 94109, USA H Schatz On examination, his vision measured 20/30 right and the patient was followed. M D Osterloh eye and 20/15 left eye. The patient had a sensory The patient returned several months later; the H R McDonald of the right macula. There vision was 20/50 right eye. Examination showed R N Johnson was a thin layer of fibrin under the just that cystoid macular oedema had developed in Correspondence to: Howard Schatz, MD. nasal to the fovea, and a small retina pigment the right macula. The subretinal fibrin was still Accepted for publication epithelial (RPE) detachment below that (Fig present. Fluorescein angiography demonstrated 31 May 1993 IA). There was no sign of intraocular the presence of leaking retinal vessels and con- Development ofretinal vascular leakage and cystoid macular oedema secondary to central serous chorioretinopathy 745 Br J Ophthalmol: first published as 10.1136/bjo.77.11.744 on 1 November 1993. Downloaded from

Figure IA Right macula, redfreephotography. Note the Figure 2A Right macula, redfreephotograph. Note thepale turbid sensory retinal detachment ofthe macula. There is also appearance ofthe central macula. The tiny dark dots that are a layer offibrin (pale round area with dark spot in centre) seen in this area are small cystic spaces in the retina. under the retina.

Figure IB Right macula, earlyphase ofthefluorescein angiogram. Note the smallRPE detachmentinferonasal to the Figure 2B Right macula, early arteriovenous phase ofthe macula. Faint hyperfluorescencejust nasal to the centralfovea fluorescein angiogram. Note the retinal capillaries which are was a shallow RPE detachment inferonasal to the macula. beginning to leak (giving thefalse appearance ofdilatation Faint hyperfluorescencejust nasal to the centralfovea was a to shallow RPE detachment that had a smallfocal leaking spot. and telangiectasis). The smallRPE detachment inferonasal thefovea appears the same. http://bjo.bmj.com/ on October 1, 2021 by guest. Protected copyright.

Figure IC Right macula, later arteriovenous phase ofthe fluorescein angiogram. The small RPE detachment inferonasal to the macula remains well demarcated and hyperfluorescent. The RPE leakagejust nasal to thefovea Figure 2C Right macula, late phase ofthe cystic macular showsfuzzy hyperfluorescence indicatingfluorescein leakage oedemafrom the leaking retinal capillaries. The RPE under the sensory retina. detachment inferonasal to thefovea remains the same.

firmed cystoid macular oedema. Because of the greater, and the cystoid macular oedema intense leakage into the retinal vasculature, it appeared more pronounced. A repeat angiogram was not possible to discerd the presence of an confirmed the retinal vascular leakage and RPE leak. marked cystoid macular oedema (Figs 2A-C). Three months later, the vision was still 20/50, The left macula had remained the same. but the sensory retinal detachment was slightly Argon laser photocoagulation was directed to 746 Schatz, Osterloh, McDonald,Johnson

the leaking retinal vessels around the right fovea Malbran and colleagues2 reviewed 3260 retinal in a very light, minimal manner. Over the course detachments and found a 10-7% incidence of of the next year, the sensory retinal detachment retinal vascular changes in that had been

flattened, the retinal vessels stopped leaking, detached for 6 months or more. This supports Br J Ophthalmol: first published as 10.1136/bjo.77.11.744 on 1 November 1993. Downloaded from and the cystoid oedema resorbed. Twelve years the idea that prolonged separation ofthe sensory later, the patient had 20/15 vision in the right retina from the RPE could cause the subsequent eye. There were small RPE laser scars, but the development ofretinal vascular changes, perhaps sensory retina was flat and there was no retinal by prolonged hypoxia and deprivation of vascular leakage. nutrients to the detached retina. In our patient the-separation ofthe outer retina from the RPE may have caused retinal hypoxia of Comment the outer retinal layers by separating the outer Wecarefully re-examined all fundus photographs retina from its normal oxygen supply, the and fluorescein angiograms ofthis patient. At no .34 time was there any evidence of subretinal The subretinal fluid may be directly toxic to neovascularisation. Although the fluorescein the outer retinal layers apart from any hypoxic angiogram appeared to show retinal vascular effect. The breakdown ofthe photoreceptors and telangiectatic changes, this was only in appear- their toxic byproducts may alter the retinal ance (false impression), caused by the rapidly capillaries. leaking fluorescein through the capillaries. The type ofsubretinal fluid does not appear to There were no retinal vascular telangiectatic be significant, because fluid of choroidal origin, channels or dilated retinal vessels seen on slit- as in central serous chorioretinopathy,2 or lamp contact examination, stereo fundus vitreous fluid, as in rhegmatogenous retinal photography, or red free photography. The fine, detachment,5 both appear to be associated with tiny, dark spots seen in the macula were small retinal vascular changes. The common factor is cystic spaces. detachment. This tends to support the theory Yannuzzi et al' reported that retinal that direct hypoxia to the outer retinal layers telangiectases and cystoid macular oedema were causes their breakdown, since the two different associated with the sensory retinal detachment of subretinal fluids produce the same effect. central serous chorioretinopathy. They did not It may be that the subretinal fluid is directly show, however, that retinal telangiectases (and toxic to the retinal capillaries, and any leakage of cystoid macular oedema) developed after central fluid causes retinal vascular leakage. Experi- serous chorioretinopathy had occurred and mental studies have shown that large molecules had been diagnosed. They noted the retinal like horseradish peroxidase can diffuse easily telangiectases only apparent on fluorescein through all layers of the retina.5 angiography; no telangiectatic vessels were seen on slit-lamp biomicroscopic examination. Three mechanisms are possible causes of the Supported by the Retina Research Fund ofSt Mary's Hospital and breakdown ofthe inner blood-retinal barrier that Medical Center, San Francisco, and by a Grant from the Wayne lead to the development of retinal vascular and Gladys Valley Foundation, San Ramon, California. leakage in our case: (1) subretinal fluid damages the outer retinal layers causing toxic breakdown 1 Yannwzzi LA, Shakin JL, Fisher YL, Altomonte MA. http://bjo.bmj.com/ products in the retina that result in retinal Peripheral retinal detachments and retinal pigment epithelial atrophic tracts secondary to central serous pigment epitheli- vascular leakage; (2) the subretinal fluid is opathy. Ophthalmology 1984; 91: 1554-72. directly toxic to the retinal capillaries; (3) the 2 Malbran ES, Dodds RA, Huslbus R. Retinal telangiectasias associated with longstanding retinal detachment as prog- detached retina is hypoxic, and this leads to nostic sign. ModProbi Ophthalmol 1979; 20: 96. retinal vascular leakage in a compensatory effect 3 Collier RH. Experimental embolic ischemia of the choroid. Arch Ophthalmol 1967; 77: 683-92. to provide more oxygen to the outer retinal 4 Gay AJ, Goldor H, Smith M. Chorioretinal vascular occlusions

layers. The fibrin that appeared attached to the with latex spheres. Invest Ophthalmol VisSci 1964; 3: 647-56. on October 1, 2021 by guest. Protected copyright. 5 Peyman GA, Bok D. Peroxidase diffusion in the normal and detached photoreceptors might also have led to laser-coagulated primate retina. Invest Ophthalmol Vis Sci hypoxia ofthe retina. 1972; 11:35.