Retinal Pigment Epithelium Decompensation I. Clinical Features and Natural Course ALEX E. JALKH, MD,*t NABIL JABBOUR, MD,t MARCOS P. AVILA, MD,* CLEMENT L. TREMPE, MD, * CHARLES L. SCHEPENS, MD*
Abstract: We studied 97 eyes (73 patients) that showed a sharp contrast between the grossly normal appearance of the posterior pole by funduscopy and the fluorescein angiography findings of multiple patches of retinal pigment epithelium (APE) transmission defect in the early transit, associated with focal areas of APE staining in the late transit. The staining was located primarily at the superior edge of the APE defect (63 eyes). The average age of the patients was 52.2 years at the time of diagnosis, and the ratio of men to women was 3.5 to 1. Ocular histories were unremarkable, except for 27 eyes with documented central serous retinopathy. Thirty-two consecutive eyes have been followed for an average of 3.9 years, and 30 of those eyes have shown visual deterioration. [Key words : central serous retinopathy, fluorescein angiography, retinal pigment epithelium, retinal pigment epithelium decompensation, retinal pigment epithelium defects, retinal pigment epithelium staining.] Ophthalmology 91: 1544-1548, 1984
The retinal pigment epithelium (RPE) functions as a pump that maintains a firm RPE-photoreceptor adhesion and a dry subretinal space. I ,2 It is also a barrier to physiologic leakage from the choriocapillaris,l.3 Several diseases affecting the RPE, particularly central serous retinopathy (CSR), can produce a breakdown in this barrier, which leads to plasma leakage from the affected RPE area into the subretinal space.4 CSR is known to be a focal disease of the RPE, and of the junction between the RPE and Bruch's membrane, that causes a serous elevation of the sensory retina in the posterior pole. 5 The diagnosis of CSR is made by fluorescein angiography, when dye leaks through the RPE and pools in the subretinal space. 5- 7 Although an From the Eye Research Institute of Retina Foundation and Retina Associates, Boston, MA,* and the Department of Ophthalmology, American University of Beirut. t Presented at the Eighty·eighth Annual Meeting of the American Academy of Ophthalmology, Chicago, October 3O-November 3, 1983. Supported in part by the Massachusetts Lions Eye Research Fund, Inc. Reprint requests to Library, Eye Research Institute, 20 Staniford Street, Boston, MA 02114.
acute attack of CSR generally resolves spontaneously with subsequent flattening of the retina, there are recurrences. 5,8,9 Over a period of time, recurrences often produce changes at the level of the RPE that are detected by fluorescein angiography as scattered window defects in the posterior pole. 10 CSR is usually regarded as a benign and self-limited disease of the RPE. 11 Recurring CSR, however, can produce over the years extensive degeneration of the RPE in the posterior pole and subsequent visual impairment,8,9,12,13 By fluorescein angiography we detected typical changes in the RPE that were common to some eyes with documented CSR, to eyes suspected from history to have CSR, and to eyes without history of CSR. We have called these changes retinal pigment epithelium decompensation (RPED). We describe the changes typical of RPED in 97 eyes and attempt to determine their natural course in 32 consecutive eyes.
MATERIALS AND METHODS The records of 73 patients (97 eyes) seen by the Retina Associates between 1961 and 1981 were reviewed.
JALKH, et al
EPITHELIUM DECOMPENSATION I
FiR 1. Fluorescein angiography of left eye. Early fluorescein transit shows extensive RPE transmission defects in macula and inferiorly.
FiR 3. Color photographs of same eye as Figure I shows flat retina in posterior pole with minimal RPE changes in macula.
Twenty-four patients had bilateral conditions. All eyes had common clinical features detected by fluorescein angiography either at the initial visit or during followup. There was a sharp contrast between the grossly normal appearance of the posterior pole by funduscopy and the evidence of changes detected by fluorescein angiography. The changes consisted of multiple patches of RPE transmission defect in the early transit associated with focal areas of staining at the level of the RPE in the late transit. The angiograms were reviewed separately by three of us, and the diagnosis of RPED was confirmed in all 97 eyes. There were 57 men and 16 women. The age at which RPED was diagnosed varied from 34 to 79 years (average, 52.2 years). We excluded eyes with trauma, high myopia (6 0 or more), retinal vascular disease, intraocular tumor, senile macular degeneration, retinal or choroidal inflammation, and chorioretinal scars. Thirty-two consecutive eyes, first seen and then followed for at least three years before 1977, were used to determine the natural course of RPED.
FiR 2. Fluorescein angiography of same eye as Figure 1. Late fluorescein transit shows focal area of RPE staining at superior edge of patches of RPE defects.
Best corrected visual acuity and central visual fields were measured at each visit using 1- and 6-mm white test stimuli. Visual improvement meant an increase of one line or more on the visual acuity chart between the measurement at the time of RPED diagnosis and the measurement at the last follow-up visit. Worsening of vision meant a loss of one line or more on the visual acuity chart. In the 97 eyes, the average time between visual disturbances and diagnosis of RPED by fluorescein angiography was 5.6 years. In the 32 consecutive eyes first seen before 1977, the average follow-up from initial to last visit was 6.3 years, and the average follow-up from RPED diagnosis to last visit was 3.9 years. P-values were computed using Pearson's chi-square or Fisher's exact test. A finding was considered statistically significant when the P-value was less than 0.05.
CASE REPORT A 54-year-old white man was seen because of progressive decrease of vision in the left eye of seven months duration. The right eye was asymptomatic. Best corrected visual acuity was 20/20 in the right eye and 20/40 in the left. Slit-lamp examination showed normal anterior segments bilaterally. By indirect ophthalmoscopy and biomicroscopy the only abnormality was questionable pigmentary changes in the macular area of the left eye. During the early fluorescein angiography transit of the left eye, extensive RPE defects became apparent in the macular area and inferior to it (Fig 1). A focal area of RPE staining, located at the superior edge of the patches of RPE defect (Fig 2), was noted in the late transit. The stereo color photographs of the left eye revealed a flat retina in the posterior pole with minimal RPE changes in the macula (Fig 3). Central visual field of left eye showed a paracentral scotoma. The right eye was normal. Follow-up over a 26-month period revealed no significant change in the focal area of staining previously detected by fluorescein angiography, but visual acuity decreased from 20/ 40 to 20/200, probably due to degeneration of the RPE and photo receptors in the foveal area.
RESULTS There was a significant predominance of men (57) over women (16), (P < 0.01). In all 97 eyes, the focal RPE staining detected in the late fluorescein transit was located at the edge of the areas of RPE defect, and in 63 eyes it was present at the superior edge. Of the 97 eyes, 51 showed one focal area of late RPE staining, 24 showed two areas, and 23 showed more than two areas. In 54 of the 97 eyes, the sites of RPED were in the posterior pole above the fovea, in 14 eyes they were below the fovea, and in 29 eyes they were in both locations. Therefore, we found a significantly higher incidence of eyes with RPED sites located above the fovea than below the fovea (P < 0.01). CSR was clinically documented by serous elevation of the retina in the posterior pole, with focal leakage through the RPE detected by fluorescein angiography at the initial visit or during follow-up in 27 of the 97 eyes. Eleven eyes had a history of CSR that was diagnosed previously by the referring ophthalmologist. In the remaining eyes, no history of previous ocular disease was found. Thus, 38 of the 97 eyes had a real connection with previous attacks of CSR. At the time of RPED diagnosis, all eyes had central or paracentral scotomata detected by central visual field testing. Figure 4 shows the visual acuity at the time of RPED diagnosis and at the last follow-up visit in the 32 consecutive eyes studied for the natural course of the disease (average follow-up,) 3.9 years. Although visual deterioration was seen in 30 eyes, 14 of these eyes did not show significant change in the fluorescein angiographic picture the follow-up period. Six eyes with CSR developed RPED even in the absence of detectable serous retinal detachment at any of the follow-up visits. In 7 of the 32 eyes the site of RPED was below the fovea, and all 7 maintained a visual acuity of 20/40 or better with either no change or a visual deterioration of only one line during the follow-up period. The 25 remaining eyes with the only site or one of the sites of RPED above the fovea had visual deterioration of two Snellen lines or more. This difference is highly significant (P < 0.001). Enlargement of the scotoma detected by central visual field testing was present in 31 of the 32 eyes.
DISCUSSION Several authors have pointed out that CSR, particularly when recurrent, can produce extensive changes at the level of the RPE with subsequent visual impairment.8.9.11-13 The follow-up of27 eyes with CSR enabled us to observe long-term changes in the RPE. Similar findings were also present in 11 eyes in which CSR was suspected by history and in 59 eyes not known to have had CSR. In the latter cases the possibility of previous 1546
CSR is not ruled out because if the macula is not involved by the serous retinal elevation during acute attacks of CSR, the patient may remain asymptomatic and the disease undiagnosed. Findings in the posterior pole that were common to the 97 eyes included: (1) absence of elevation of the RPE or retina, and minimal RPE changes barely detectable by ophthalmoscopy, biomicroscopy, or color photography of the posterior pole; (2) fluorescein angiography showing patches of hyperfluorescence in the early transit, caused by defects in the RPE; and (3) focal areas of late staining at the level of the RPE. The changes detected by fluorescein angiography sharply contrasted with the grossly normal appearance of the posterior pole when studied by other techniques. Similar fluorescein angiographic findings in CSR were described by Gass l4 as tear drop or irregular zones of hyperfluorescence in the posterior pole, which he attributed to long-standing serous detachment of the sensory retina. However we, as well as other,15 have seen these RPE changes after documented, previous CSR, even in the absence of serous retinal detachment and of clinically detectable attacks of CSR during the follow-up period. Long-standing serous retinal detachment in the posterior pole may lead to secondary RPE changes. However, the RPE changes we observed may also be explained by a mechanism that is effective in the absence of serous retinal detachment. In eyes where CSR had been previously documented by fluorescein angiography, the area of staining coincided with the site (or one of the sites) of leakage detected during the acute attacks of CSR. In spite of the fact that the retina flattens and visual acuity recovers after each CSR attack, the RPE, in the site of CSR, may gradually become incompetent, and probably allow some fluid to ooze continually from the choriocapillaris through the RPE. Even though the RPE pump cannot contain choriocapillaris leakage,I,2 it can still function well enough to keep the subretinal space dry. The oozing may be so minimal that it is detected by fluorescein angiography not as dye leakage or pooling but as staining at the level of the RPE that appears flat. RPE staining is characterized by an increase in intensity, but not in size of the area of hyperfluorescence, from the early to the late fluorescein transit. Over the years this slow oozing causes degeneration of the overlying photoreceptors, and the adjacent RPE atrophies. This degeneration will provide a scotoma (central or paracentral), and if the fovea is involved, decreased central vision. Degeneration of the RPE adjacent to the site of late staining is detected by fluorescein angiography as a transmission defect. Patches of such a defect are usually located inferior to the focal area of late staining because the chronic leakage from the decompensated RPE gravitates inferiorly. A result of this observation is that there is more marked visual impairment when the site of late focal staining is above the fovea than when it is below the fovea. The relation between RPED and CSR is supported by several observations in our series: (1) the significant predominance of men over women, which is also present
JALKH, et al
.... . .... .. .. . .. .... ... .. .. .. .. .... .. .
EPITHELIUM DECOMPENSATION I
10/100 20/200 20/300 20/400 CF
Vision ot diagnosis
Fig 4. Progressive worsening of visual acuity in 32 consecutive eyes with RPE decompensation.
in CSR;14 (2) 21 eyes followed for CSR (mostly recurrent) subsequently developed RPED; and (3) six eyes initially seen with RPED had one or more acute attacks of CSR during the follow-up period. The fact that the average age of our patients at the time of diagnosis of RPED (52.2 years) was higher than the age of patients with CSR (less than 45 years)5 supports the concept that RPED is likely to be a late complication of CSR. We also found that RPED occurred an average of 5.6 years after the first attack ofCSR. Wessing l6 similarly observed extensive RPE changes occurring about 5 years after the first CSR attack. All eyes with RPED had a central or paracentral scotoma usually corresponding to the focal area of late staining detected by fluorescein angiography. At the time of RPED diagnosis the visual impairment was rarely severe, since only 2 of 32 eyes had a visual acuity of 20/70 or worse (Fig 4). However, the natural course of RPED in the 32 eyes followed for an average of 3.9 years showed that visual acuity deteriorated in 30 eyes, and 20 of those eyes experienced a visual loss of three lines or more during the follow-up period (Fig 4). This visual deterioration associated with the natural course of RPED can be explained by the progressive degeneration of the RPE and photoreceptors in the areas adjacent to the site of RPED, which also led to worsening of the central visual fields in 31 of 32 eyes. However, this progressive degeneration is not always demonstrable by fluorescein angiography. Indeed, 14 of the 30 eyes with visual deterioration showed no significant fluorescein angiographic changes during the follow-up period after RPED diagnosis, but their corresponding central visual fields were worse.
In the differential diagnosis of RPED, one must consider other clinical entities affecting primarily the RPE, particularly macular dystrophies, RPE inflammations, drug toxicity, and choroidal neovascularization . Differentiation of RPED from Stargardt's disease, old vitelliform dystrophy, and certain types of pattern dystrophies is easily made by the absence of bilateral symmetrical macular involvement in RPED, and by a sharp contrast between the normally appearing posterior pole by funduscopy and the typical RPE changes by fluorescein angiography. Similarly, the effect of toxic substances on the RPE can be ruled out easily by the patient's history and the typical fluorescein pattern of RPED. Retinal pigment epitheliitis is a discrete focal inflammation of the RPE; a picture similar to RPED decompensation has not been reported in this disease. Choroidal neovascularization can be easily ruled out by the presence of serous detachment of the retina or of the RPE, and by leakage from choroidal new vessels on fluorescein angiography. It is interesting to note that some patients were referred with the diagnosis of optic nerve disease because of decreased vision without detectable chorioretinal abnormalities by fundus examination. When fluorescein angiography was performed, however, RPED was present.
CONCLUSION Our study demonstrates that retinal pigment epithelium decompensation is a clinical entity closely related to central serous retinopathy but not always preceded by clinically documented central serous retinopathy. Retinal pigment epithelium decompensation has a typical fluorescein angiographic pattern that distinguishes it from other diseases of the retinal pigment epithelium. The natural course of retinal pigment epithelium decompensation has been determined in 32 eyes in which it is characterized by progressive visual deterioration, in spite of the absence of significant changes in the fluorescein angiographic picture in some of these eyes during follow-up.
ACKNOWLEDGMENTS The authors thank Drs. J.W. McMeel, H.M. Freeman, R.c. Pruett, F.l. Tolentino, T. Hirose, J.J. Weiter, M.A. Mainster, and S.M. Buzney for their assistance.
REFERENCES 1. Records RE. Retina: metabolism and photochemistry. In Duane TD, Jaeger EA, eds. Biomedical Foundations of Ophthalmology. Philadelphia: Harper & Row, 1982; Vol 2, Chapter 12. 2. Gass JDM. Stereoscopic Atlas of Macular Diseases; Diagnosis and Treatment, 2nd ed. St Louis: CV Mosby, 1977; 10. 3. Gass JDM. Pathogenesis of disciform detachment of the neuroepi-
DECEMBER 1984 •
thelium. I. General concepts and classification. Am J Ophthalmol 1967; 63:573-85. Archer DB. Fluorescein angiography. In: Krill AE. Hereditary Retinal and Choroidal Diseases. Vol I. Evaluation. Hagerstown: Harper & Row, 1972; 155-60. Gass JDM. Pathogenesis of disciform detachment of the neuroepithelium. II. Idiopathic central serous choroidopathy. Am J Ophthalmol 1967; 63:587-615. Maumenee AE. Fluorescein angiography in the diagnosis and treatment of lesions of the ocular fundus. Trans Ophthalmol Soc UK 1968; 88:529-56. Norton EWD, Gass JD, Smith JL, et al. Diagnosis: fluorescein in the study of macular disease. Trans Am Acad Ophthalmol Otolaryngol 1965; 69:631-42. Bennett G. Central serous retinopathy. Br J Ophthalmol 1955; 39:605-18. Wessing A. Changing concept of central serous retinopathy and its treatment. Trans Am Acad Ophthalmol Otolaryngol1973; 77:0P27580.
10. Bonnet M, Pingault C, Bakri M. La retinopathie sereuse centrale est-elle une maladie unilaterale et Ires limitee de I'epithelium pigmentaire? Bull Soc Ophtalmol Fr 1974; 74:233-6. 11 . Klein ML, Van Buskirk EM, Friedman E, et al. Experience with nontreatment of central serous choroidopathy. Arch Ophthalmol 1974; 91 :247-50. 12. Dellaporta A. Central serous retinopathy. Trans Am Ophthalinol Soc 1976; 74:144-53. 13. Watzke RC, Burton TC, Leaverton PE. Ruby laser photocoagulation therapy of central serous retinopathy. Part I: A controlled clinical study. Part II: Factors affecting prognosis. Trans Am Acad Ophthalmol Otolaryngol 1974; 78:0P205-11. 14. Gass JDM. Stereoscopic Atlas of Macular Diseases; Diagnosis and Treatment, 2nd ed. St Louis: CV Mosby, 1977; 28-38. 15. Nanjiani M. Long-term follow-up of central serous retinopathy. Trans Ophthalmol Soc UK 1977; 97:656-61. 16. Wessing A, Meyer-Schwickerath G. Lichtchirurgische Behandlung und sonstige chirurgische Massnahmen bei Maculaaffektionen. Ber Dtsch Ophthalmol Ges 1973; 73:585-94.