Age-related macular degeneration (AMD) is the first cause of blindness after 50 years in the Western world.1^–7 AMD is a phenotypically heterogeneous disease, including atrophic (dry) and exudative (wet) forms. More than 80% of severe visual loss cases in AMD are due to choroidal neovascularisation (CNV) and its consequences: exudation, bleeding and disciform scar.8 Well-defined choroidal neovascularisation also called “classic CNV” or “pre-epithelial CNV” represents a small subgroup: 17.6% among all neovascular AMD.9 Clinical aspects of classic CNVs are widely described in literature, including fundus examination, red-free photographs, fluorescein angiography (FA), indocyanine green angiography (ICG) and optical coherence tomography (OCT) features.10^–12

Infrared (IR) imaging is a non-invasive in vivo method. In the literature, IR imaging in AMD is poorly illustrated. Harnett and Elsner (1996) described IR findings in exudative AMD,13 among a heterogeneous group of 10 AMD patients including only one case of classic CNV. In this single case, a whitish ring peculiar aspect was observed by IR imaging. Our purpose was to analyse the infrared features in a homogeneous subgroup of classic CNVs, before any treatment.

PATIENTS AND METHODS

A total of 22 eyes of 22 consecutive patients affected with well-defined extrafoveal or subfoveal CNVs caused by exudative AMD were prospectively included. They were addressed to our department from January to September 2007 because of metamorphopsia or visual acuity loss. Inclusion criteria were classic CNV, age over 55 years old, absence of high myopia (>8 dioptres), absence of angioid streaks, and absence of any treatment in the eye included. Other phenotypes of exudative AMD such as retinal angiomatous proliferation, polypoidal vasculopathy, occult CNV, minimally classic CNV or predominantly classic CNV, were excluded.

For each patient, a complete ophthalmological examination was performed before any treatment. It included a best corrected ETDRS visual acuity (BCVA) measure, fundus examination, retinal colour fundus photographs, red free monochromatic fundus photographs, autofluorescence pictures, fluorescein angiography (FA) (Zeiss FF450IR, Carl Zeiss, San Leandro, California, USA or Canon 60 fundus camera, Tokyo, or Heidelberg Retina Angiograph II, Heidelberg Engineering, Heidelberg, Germany), scanning laser ophthalmoscopy (SLO) ICG (Heidelberg Retina Angiograph II, Heidelberg Engineering, Heidelberg, Germany) and 6 mm sections OCT (Stratus OCT, Carl Zeiss ; Meditech, Dublin, California).

Diagnosis of well-defined CNV caused by AMD was based on the FA international AMD classification.14 The diagnostic criteria for well-defined CNV were: well-demarcated lacy area of early hyperfluorescence, early staining and progressive leakage in the overlying subsensory retinal space during the late phases of the angiogram. CNV margins were well defined on the early phase of angiography.

The OCT confirmed the pre-epithelial localisation of the CNVs in all cases.

Infrared pictures (820 nm) were performed before FA and ICG, using the confocal Heidelberg Retinal Angiograph 2 (Heidelberg Engineering, Germany). The IR pictures were compared with the colour fundus photographs of the retina, the red free frames, the FA, the ICG angiography results at the early and late phases, and the OCT.

Informed consent from all patients was obtained, in agreement with the Declaration of Helsinki and French legislation.

RESULTS

A total of 22 eyes from 22 consecutive patients with newly diagnosed well-defined CNV caused by exudative AMD were included. The mean patient age in this series was 73 (SD 8) years, there were no statistical differences regarding sex: 12 women and 10 men. The localisation of the CNVs was retro- or juxtafoveal in 17 (77.3%) cases and extrafoveal in five cases (22.7%).

On IR pictures of all eyes affected with well-defined CNVs (22/22), a multilayered structure was identified. This structure appeared as a complex with a dark central core surrounded by a whitish reflective ring, giving a halo-like shape. This ring had a closed aspect (O-shape) in 15 cases (68.2%) and a horseshoe aspect (U-shape) in seven cases (31.8%) (fig 1).

• Download figure
• Open in new tab
• Download powerpoint

Figure 1 Infrared frames of well-defined choroidal neovascularisation. This composite frames shows nine examples of eyes affected with well-defined choroidal neovascularisations (CNVs) in exudative AMD. In infrared pictures, the well-defined CNV appear as a complex with a dark central core surrounded by a whitish reflective ring, giving a halo-like shape (black arrows). This whitish ring may have an O-shape (A, C, E, F, H) or a horseshoe aspect (U-shape) (B, D, G, I).

Comparison between IR pictures and early phases of FA revealed that the whitish ring in IR frames corresponded to the borders of the well-defined CNV (fig 2). More precisely, the whitish ring in IR frames had the exact same position as the hypofluorescent ring observed around well-defined CNV in early phases of FA. Comparison between IR pictures and early phases of ICG revealed similar results. Based on morphological comparisons, this ring was correlated to the neovascular lesion. The external borders of the whitish ring corresponded to the diffusion ring on the FA late frame, and with the dark rim surrounding the neovascular membrane on FA and ICG early phases (fig 3).

• Download figure
• Open in new tab
• Download powerpoint

Figure 2 Comparison between colour pictures, red-free frames, early and late phases of the fluorescein angiography (FA), indocyanine green angiography (ICG), optical coherence tomography (OCT) and infrared (IR) pictures. Comparison between colour pictures (A), red-free frames (B), early (C) and late (D) phases of FA, ICG (E), OCT (G, H) and the IR picture (F). The external borders of the whitish ring (arrows) correlate with the diffusion ring of the late frame of the FA, and with the dark rim surrounding the neovascular membrane at the early phase of the FA and ICG.

• Download figure
• Open in new tab
• Download powerpoint

Figure 3 Comparison between the limits of the infrared (IR) white ring, fluorescein and green indocyanine angiography. The whitish ring in the IR pictures correlates with the dark rim surrounding the choroidal neovascular lesion defined on early phase of fluorescein and green indocyanine angiography (black arrows).

DISCUSSION

Our aim was to describe IR features of classic CNVs associated with exudative AMD in a homogeneous group of patients.

Classic CNV was the first type of exudative AMD lesion described by FA analysis, despite the fact that it is not the most common phenotype. Classic CNVs differ from occult CNVs by their evolution and response to PDT therapy.15^–17 Features of classic CNVs are widely described, including FA, ICG and OCT descriptions. To the best of our knowledge, only one paper has reported the findings of IR imaging in exudative AMD.13 In this paper, the authors analysed a heterogeneous group of 10 patients, presenting various forms of exudative AMD, including treated CNVs, occult CNVs, drusen, subretinal fluid and pigment epithelial detachments. Only one case of well-defined CNV was included in this series. Hartnett and Elsner described a multilayered structure13 in this well-defined CNV single case.

In our series, CNVs appeared in all cases as a whitish ring surrounding a central dark core. The external borders of the whitish ring corresponded to the diffusion ring on the late frame of the FA, and with the dark rim surrounding the neovascular membrane at the early phase of the FA and ICG (fig 3). This structure was constant on the 22 patients included.

IR imaging could be an interesting non-invasive evaluation method of well-defined choroidal neovascularisation because of its high sensitivity. It could help to localise CNV from the fovea in difficult cases and before any contrast product injection. The data are rapidly obtained, with a high reproducibility. Diagnosis of well-defined CNVs is usually based only on FA features. The aim of our study is not to replace the fluorescein angiography but to propose an additional and non-invasive tool in their characterisation.

In their studies, Lopez et al18 and Grossniklaus et al19 correlated clinical, angiographic and ultrastructural features of well-defined subfoveal neovascular membranes. As seen in fluorescein angiography, the neovascular membrane is separated in two distinguished regions. The central core, is a brighter area of hyperfluorescent leakage and is composed of a fibrovascular membrane characterised by endothelium-lined vascular channels with associated retinal pigment epithelium. The second region, the rim, is the hyperfluorescent staining surrounding the core. This rim could correspond to the whitish ring found infrared imaging. The ultrastructural features of this choroidal neovascular membrane differ from the membrane core. It is principally composed of fibrin, photoreceptor outer segments, non-specific proteinaceous material and macrophages. There is no basal laminar deposit, native collagen, endothelium-lined vascular channels, retinal pigment epithelium or fibrocytes found in this rim.

We could hypothesise that the white aspect of the ring in infrared imaging is due to the accumulation of fibrin.

In this description of IR features of classic CNV, a whitish ring, either complete (O-shape) or incomplete (U-shape) was constantly observed. However, the sensitivity of this ring observation has to be confirmed in a larger sample of patients. Specificity of this whitish ring observation associated with classic CNVs could not be analysed by this study because of its initial design. Specificity should also be studied in cases of ill-defined CNVs associated with exudative AMD, high myopia or angioid streaks.

IR is a non-invasive method, probably underestimated, that shows the borders of classic CNVs as a white ring due to exudative AMD.