We reviewed 28 eyes with PCV in 26 patients who visited our clinic due to decreased vision, metamorphopsia or scotoma, from December 1999 to January 2003. Each of these patients underwent standardized, clinical, fluorescein and indocyanine green angiography and OCT examinations.

The diagnosis of PCV was made according to evidence of a branching network of inner choroidal vessels with polypoidal vascular dilatations at the border of the network of vessels via indocyanine green angiography,1 and the presence of a serous and/or hemorrhagic pigment epithelium detachment and retinal detachment in the posterior pole. Exclusion criteria were as follows: other diseases such as exudative age-related macular degeneration, high myopia, angioid streaks, and any previous ocular surgery. Indocyanine green angiography was performed with a fundus camera and the IMAGEnet system (IMAGEnet for Windows, version 1.55; Topcon, Tokyo, Japan) after the injection of 50 mg of indocyanine green dye in the antecubital vein. All patients underwent serial vertical and horizontal OCT scans (OCT 2000, software version A 7.0; Carl-Zeiss ophthalmic systems, San Leandro, California) through the lesion. An infrared light with a wavelength of 840 nm and energy of 750 µW was incident on the retina. The location and direction of the scan were determined to display the most verifiable image of the orange-red nodular elevation. All OCT images were evaluated to find the RPED with attenuation of internal reflectivity. It exhibits the dome-like focal protrusion of a highly reflective band corresponding to the detached RPE, and sometimes shows a cluster of multiple protrusions. There is moderate focal reflection just beneath the RPE protrusion and the characteristic attenuation of the reflectivity within the RPE detachment (Fig. 1A). When the RPED with attenuation of internal reflectivity was detected in the OCT images, we measured the base diameter and height of each lesion after center of the RPED, as described by Iijima et al.9 (Fig. 2). The base diameter was measured as the longest distance between the two points of the detached RPE's elevated upper margin. The height was measured as the longest distance between the extension line of the upper margin of the attached RPE and the upper margin of the detached RPE. The base diameter and height of each lesion were determined by averaging the two values from the vertical and horizontal scans.

To evaluate whether the RPED with attenuation of internal reflectivity is a typical OCT finding for PCV, we compared it with the OCT findings of RPEDs presented in exudative age-related macular degeneration. Criteria for exudative age-related macular degeneration included: (1) the funduscopic appearance of subretinal fibrovascular membranes, serous and/or hemorrhagic neurosensory detachment, and serous and/or hemorrhagic RPEDs; (2) the presence of CNV revealed with fluorescein and indocyanine green angiography; and (3) no evidence of polypoidal dilatations at the border of a branching network of inner choroidal vessels via indocyanine green angiography.

The 112 eyes with exudative age-related macular degeneration examined in our clinic from April 2002 to March 2003 met the above inclusion criteria. The mean age of the patients was 67.3 years (range, 54-88) and included 65 men and 26 women. In the review of OCT images of exudative age-related macular degeneration, we focused on the prevalence and size of the RPED with attenuation of internal reflectivity.

The Mann-Whitney test was used for statistical analysis and conducted using SPSS statistical software (version 11.0; SPSS for Windows, 2001). A p value of less than 0.05 was considered significant.

Previous reports suggested that elevated RPE and moderate reflection beneath the RPE are characteristic OCT findings of PCV.7,8 Little data exists regarding the sensitivity and specificity of the OCT findings and the difference between the RPED in PCV and RPED in exudative age-related macular degeneration. Since the serous and/or hemorrhagic RPED typically presents in the exudative age-related macular degeneration and PCV, our study compared the OCT images of patients with PCV and those with exudative age-related macular degeneration.

The RPED detected in PCV typically showed attenuation of internal reflectivity on OCT examination, and was distinguishable from serous or hemorrhagic RPED in eyes with exudative age-related macular degeneration (Fig. 1B, 1C). The optical reflectivity of the sub-RPED fluid in PCV is presumed to be between that of pure blood and serosanguineous fluid, which causes the RPED attenuation of internal reflectivity on OCT typical in PCV. Figure 1B shows an example of a serous RPED containing an optically empty space due to low-reflective serous fluid. An OCT image of hemorrhagic RPED (Fig. 1C) can be distinguished from a serous RPED by the presence of optical backscattering arising from blood beneath the detached RPE and the shadowing of deeper structures. The sub-RPED fluid in PCV may have the optical reflectivity anywhere between serous RPED and hemorrhagic RPED, and show moderate reflectivity directly beneath the detached RPE, but increasing attenuation of reflectivity in deeper space. The characteristic results from OCT examination appear to be correlated with those from ICGA, in which small polyps become increasingly hyperfluorescent in the middle phase rather than early phase. The polyps in PCV are considered to have a slow blood flow and are clinically visible as a pinkish rather than dark red structure. The above findings may be correlated with the typical RPED of PCV having attenuated internal reflectivity on OCT.

As a result of our study, this typical appearance of RPED in OCT examination proved to be highly sensitive (84%) and specific (94%) for PCV when compared with exudative age-related macular degeneration. This OCT manifestation therefore strongly supports the diagnosis of PCV. The result was compatible with one report insisting that 85% of the cases presenting large hemorrhagic and exudative pigment epithelial detachments in the absence of drusen were PCV.10

Iijima et al. reported that OCT images in eyes with PCV show a highly reflective band of remarkable anterior protrusion that is thicker and steeper than that of the serous RPED.9 Our study results indicate that the steepness of the RPED, which was calculated by the ratio of its height to base diameter, cannot differentiate the RPED in PCV from the RPED in exudative age-related macular degeneration. Instead, the dimensions of the typical appearance of RPED on OCT examination were different between the two disorders. We found that the base diameter and the product of the base diameter and height of the characteristic RPED on OCT examination was statistically smaller in eyes with PCV than in eyes with age-related macular degeneration. Although the value calculated by multiplying the base diameter by the height provides the area of a rectangle, it can be interpreted as the area of RPED because of its low height-to-base ratio (0.2).

RPED with attenuation of internal reflectivity on OCT examination represents a dome-like focal protrusion of a highly reflective band corresponding to the detached RPE with moderate reflections below the protrusion and shadowing underneath. The detachment of RPE appears to be accompanied by the detachment of Bruch's membrane. The moderate reflections and shadowing within the focal protrusion may correspond to the fibrovascular proliferation, dilated vessels and voluminous accumulation of extravasated blood and exudate per diapedesis11 in the sub-RPE space. Marked attenuation of incident light through the RPE produces moderate reflections under the detached RPE. The shadowing under moderate reflections is derived not only from the increased attenuation of the incident light due to a high light scattering coefficient of blood, but also from the enhanced reflectivity of the overlying layer.5,6

Although indocyanine green angiography is usually indispensable for the diagnosis of PCV,12 it is not always diagnostic of PCV. In a recent report on the differential diagnosis of hot spots detectable by indocyanine green angiography, the majority of the lesions were categorized into PCV, retinal angiomatous proliferation and pure occult CNV.13 When only one hot spot is detected on indocyanine green angiography, or when an interconnecting vessel is not detected in an apparent PCV case, the diagnosis of PCV imposes challenges even with the indocyanine green angiography. Notably in our case studies, only one polyp was detected in 2 eyes (7.1%) and an interconnecting vessel was not detected in 6 eyes (21.4%). The detection of the characteristic RPEDs on OCT examination may provide a firm basis for the diagnosis of PCV in these cases.

Little is known of the histopathology of PCV, especially as to the spatial relation between polyp and the interconnecting vessel, and as to whether PCV is a peculiar type of CNV or just a vascular dilation. One study disclosed extensive fibrovascular proliferation in the subretinal space and within the Bruch's membrane, and the presence of many choroidal vessels that transversed defects in the Bruch's membrane.14 The subsequent histopathological studies of PCVs noted sub-RPE and intra-Bruch's fibrovascular membrane.15,16 PCV has been regarded as a peculiar type of choroidal neovascularization in all of the above reports. Whether CNV was secondary to the RPED in PCV or not, it is remarkable that the coincidence of the RPED and CNV was identified in 12 of 28 eyes with PCV in our study. In contrast, a recent histopathological study on eyes with PCV suggested the lesion as a degenerated RPE-Bruch's membrane-choriocapillaris complex and inner choroids rather than an intra-Bruch's fibrovascular membrane.11 The study proposed that PCV might be caused by the stasis of the venule and arteriole at the arteriovenous crossing site associated with arteriolar sclerosis. Although the systemic condition of patients with PCV is generally healthy,3,17 the correlation between systemic hypertension and PCV has been suggested in a few reports.10 Some authors insist that hypertensive insult to the choroidal vascular beds involves the pathogenesis of PCV.18 Histopathological data from one study suggested that the arteriosclerotic change might also be involved in the pathogenesis.19 Since the correlation between PCV and systemic disorders including hypertension and diabetes was not noted in our study, we believe further data is necessary to clarify the role of systemic disorders in PCV.

Twenty-four (92%) of 26 patients in our study showed unilateral involvement of PCV, although the rate of bilaterality was about 70% in original reports on PCV.2,3 Recent large case studies indicate that the rate of bilateral involvement is around 10%.20,21 The dominance of unilaterality in our study may be attributed to the fact that the data from each case was mainly obtained from the patient's first visit to our clinic, implying that our data was collected from the acute symptomatic phases of PCV. Since the involvement in the contralateral eye may develop after a dozen years,2 the proportion of bilateral cases in our studies would increase with follow-up for an extended period. Other possibilities may include racial differences.

In conclusion, an RPED with attenuation of internal reflectivity on OCT examination is a sensitive and specific finding for PCV. Detection of this finding along with a small cross sectional area on OCT examination strongly supports the diagnosis of PCV.