Year XVI-N.10/2000

Senile or age-related macular degeneration (AMD) causes a severe and irreversible reduction in visual acuity, and is the main cause of blindness in people aged more than 50 years living in industrialised countries. Given that it is a degenerative disease, it is more frequent with advancing age, passing from 1.6% of the population aged 52-64 years to 28% of people older than 75 years (1).

The majority of the patients are affected by the “dry” non-neovascular form, which is characterised by drusen and atrophic alterations of the retina pigment epithelium (RPE). However, 80% of the cases of severe visual impairment are attributable to the “wet” or neovascular form characterised by the onset of choroidal neovascularisation (CNV). These new vessels leak lipids, blood and fluid, which provoke a local inflammatory reaction that leads to the rapid loss of central vision.

The social importance of this problem can be seen from the fact that between 70,000 and 200,000 people in the United States above the age of 65 years develop neovascular AMD every year (2,3).

Current treatments

The natural history of wet AMD leads to a final status of quiescence and involution of the neovascular process, but this is always accompanied by the formation of central connective tissue scarring and consequent vision loss (4,5). Any type of treatment must therefore be started before the formation of the terminal scar.

Over the last few years, the introduction of new diagnostic instruments, such as laser scanning devices for fluoroscein sodium or indocyanine green (ICG) angiography, has made it possible to hypothesise a treatment programme for neovascular AMD that was unimaginable until only a few years ago. The crucial point of the therapeutic strategy is the type of lesion. As we shall see, this means that preferably dynamic high-resolution ICG angiography is fundamental because it allows the filling and emptying of a neovascular lesion to be evaluated and makes it possible to identify its type.

The treatment of neovascular lesions has so far been based on an exquisitely fluoroangiographic classification that distinguishes three main types of lesion:

a) classic extrafoveal CNVs, with a distance from the centre of the fovea of more than 200 microns;

b) classic juxtafoveal or subfoveal CNVs, located at the edge or below the centre of the fovea;

c) type I and II occult CNVs The terms “classic” and “occult” respectively indicated fluoroangiographically well-defined or poorly visualisable lesions.

Classic extrafoveal CNVs

In the case of classic extrafoveal CNVs (i.e. those more than 200 microns from the centre of the fovea), the standard treatment is still direct traditional laser photocoagulation of the lesion according to the dictates of the Macular Photocoagulation Study (MPS) Group (6), the aim of which is to destroy the new vessels by means of the heat generated by the laser: i.e. a hyperthermia of more than 65o at the impact site of the ray.

This thermal effect is obviously not selective but also damages the external retinal layers, including the photoreceptors located above the neovascularisation. The final result will always be an atrophic scar limited to the treatment site and a corresponding absolute scotoma (Fig.1). This result is more than acceptable if the lesion is sufficiently far from the central fixation point.

After five years of follow-up, the MPS demonstrated that the treated patients had lost an average of 5.2 lines of visual acuity as against the 7.1 lines lost by the untreated patients (7); likewise, the mean final visus of the treated patients was 20/125 as against 20/200. In practical terms, the risk of severe visual loss was 1.5 times greater in the untreated eyes (7).

Unfortunately, 54% of the treated eyes were affected by recurrent neovascularisation within a period of five years (7).

Classic juxtafoveal or subfoveal CNVs

In the case of classic juxtafoveal or subfoveal CNVs (i.e. those located at the edge or below the fixation point), or the recurrence of originally extrafoveal membranes that subsequently reach the foveal area, standard thermal photocoagulation treatment has negative implications for the patient.

A parafoveal atrophic scar (as in the case of a juxtafoveal neovascularisation) is rarely compatible with a good functional outcome. This is once again demonstrated by the data coming from the MPS, which show that the laser treatment of juxtafoveal lesions leads to a final median visus of 20/200 as against the 20/250 of the untreated eyes (although very low levels of 20/400 or less were finally observed in only 25% of the treated patients as against 40% of the untreated patients) (8).

The direct photocoagulation of juxtafoveal membranes is at particular risk of relapse because the operator often tries to avoid extensive treatment, especially on the foveal side, as is shown by the fact that persistent or recurrent membranes are observed in 78% of laser-treated cases (8). The “treatment risk” is even more evident in the case of subfoveal neovascularisations, for which the use of direct traditional laser photocoagulation is highly debatable.

The patients complain of a sudden reduction in vision immediately after treatment and the long-term functional results are also poor. The MPS data [9,10] not only show that, two years after being treated, the eyes lose an average of 3.3 lines in comparison with the 4.5 lines lost by untreated eyes but, more significantly, that the final average visual acuity is almost the same in the two groups (20/320 vs. 20/400) (9).

As previously mentioned, the advent of dynamic ICG angiography gave a considerable impulse to the development of alternative treatments for classic juxtafoveal and subfoveal CNVs because its high power of resolution made it possible to treat a larger number of patients with wet AMD than the disconsolate 13% allowed on the basis of the MPS criteria alone (11). Being able to see the arterial origin of a membrane sometimes makes it possible to reclassify its type (from occult to classic) and distinguish whether its site of origin is extrafoveal, juxtafoveal or subfoveal: for example, an apparently subfoveal neovascular extension can be identified as a feeder vessel of extrafoveal origin (Fig. 2).

The treatment of feeder vessels is based on low-intensity (the effect should be almost invisible) traditional direct laser photocoagulation of the extrafoveal origin of the CNV, which allows more selective damage of the new vessels with the preservation of the foveal neurosensory retina and a sensory deficit restricted to the impact area. It has recently been demonstrated that feeder vessels can be visualised by means of dynamic ICG angiography in more than 80% of the cases of subfoveal neovascularisation (12).

Treatment eligibility is high at more than 79%. The percentage of vessel obliteration is 40% after one or more treatments but, if eligibility is limited to the vessels with a diameter of less than 85 microns, the percentage of closure is 75% (12). It is also worth pointing out that final visual acuity has been reported as stable (47%) or improved (33%) in 80% of patients (12).

An even more selective and very different method of treating new vessels is currently in an advanced stage of development (phase IIIb). Photodynamic therapy (PDT) is based on the combined action of two factors: the intravenous administration of a photosensitising substance (verteporfin) followed by the exposure of the area to be treated by means of low-intensity, infra-red (690 nm) laser radiation, which is preferentially absorbed by verteporfin.

The therapeutic concept is based on the fact that the photosensitising substance is borne in the bloodstream by lipoproteins that have specific receptors in the new vessels and, about 15 minutes after being injected, arrives at the site of the subfoveal neovascularisation; low-intensity irradiation of the macular area activates the substance without causing any particular thermal damage, and its activation induces progressive endothelial damage that leads to the final selective occlusion of the new vessel without provoking any alterations in the choriocapillaris or surrounding retina (13) (Fig. 3).

This type of therapy has been used for some time in the treatment of tumours, the growth of which can be arrested or reduced by occluding the tumour neovascularisation (14), and the hypothesis was that it can do the same thing in the case of AMD neovascularisations.

The recently published data (TAP and VIP Study) has been comforting: if the lesion has a >50% classic component, 67% of the treated patients (as against 39% of the controls) show a positive response after one year of follow-up: i.e. unchanged or improved vision, or a visus that has worsened by less than three lines) [13].

After one year of follow-up, 54% of the treated eyes with initial classic neovascularisations were angiographically stable (as against 29% of the control eyes); furthermore, 19% of the treated were angiographically silent (13). In the case of neovascularisations with a 50% occult component, photodynamic therapy does not seem to have any significant benefits (13). The drug is not currently available through the National Health System because it has not yet been registered in Italy, but this should be a matter of only a few months (it was registered in Switzerland in January 2000). We are currently directly involved in a 10-centre Italian study of PTD in AMD (VIT Study), the results of which should accelerate the registration of the drug (15).

The great advantage of this therapeutic approach would therefore seem to be the complete sparing of post-treatment macular function, a result that appears to be confirmed by the microperimetry results in our possession (unpublished data). Despite the fact that more than one treatment (administered every three months) has often been necessary, the side effects are not significant (13).

Occult CNVs of type I and type II

Occult CNVs of type I (vascularised epithelial detachments) and type II (late leakage of undetermined source) represent the vast majority of exudative AMD lesions (11), and their natural history is functionally extremely poor: 41% of the affected eyes show severe vision loss (six lines) within 12 months (16). Nevertheless, the advent of static (but particularly dynamic) ICG angiography has led to major therapeutic advances.

First of all, it has made it possible to reclassify some occult CNVs and thus allowed them to be treated by means of traditional photocoagulation: about 45% type I and 35% type II lesions become treatable after conventional ICG angiography (17-19), which indicates that this diagnostic examination can increase the eligibility of occult CNV patients for standard photocoagulation therapy by about one-third (18).

It is possible that dynamic laser scanning ICG angiography may increase this figure further if it is true that it has enabled the visualisation of 86% of feeder vessels, in comparison with the 27% revealed by fluoroangiography (12).

However, it needs to be said that the results of the ICG-guided laser treatment of such lesions are only partially encouraging. Anatomical success has been reported in 43% of cases with type I and 66% of those with type II (20), and the functional results have been similar: 13% of improved cases (16% type II and only 9% type I), and 53% of stable cases. The percentage of relapses is still high: 51% for type I and 35 for type II (20).

The poor results of the traditional laser treatment of type I lesions may be due to the high incidence of abnormal retinal vascular complexes (ARVCs) in the case of senile epithelial detachments, which are angiographically indistinguishable from occult CNVs unless dynamic laser scanning ICG-angiography is used. The very poor anatomical and functional prognosis of these lesions after traditional photocoagulation treatment has been clearly demonstrated (21). An alternative and more selective treatment for occult CNV is trans-pupillar thermotherapy (TTT).

This has been used for some time in the treatment of small choroidal melanomas (22), and is based on the therapeutic concept of reaching a level of hyperthermia of 45-60o in the treated area using an infra-red laser beam of 810 nm. Hypothermia of this type directly occludes the new vessels without causing the effects produced by photocoagulation (>65o), and also optimises heat penetration (3-4 mm as against the 1-1.5 mm offered by traditional photocoagulation) (22). The improvement in the thermal effect is due to the time of exposure (about one minute), the diameter of the spot (between 1.2 and 3 mm), and the variable intensity of 360-1000 mw; the final effect should be virtually invisible biomicroscopically (23).

The results are very encouraging, even though they still need to be confirmed in larger patient series: after one year of follow-up, 94% of the eyes have shown a reduction in exudation (OCT, fluoroangiography, biomicroscopy), an anatomical result that is also reflected in the functional results observed in 75% of the cases (19% showed an improvement and 56% remained stable) (23).

The problem of TTT is the fact that, unlike in the case of PDT, the treatment parameters have not been standardised on an experimental basis.

Alternative treatments for wet AMD

A number of different therapeutical approaches have been proposed over the last few years, but they have proved to be totally inefficacious (radiation or medical therapies) (24-26) or excessively invasive in comparison with new treatments such as PDT (surgical removal or macular translocation (27-29).

It therefore seems to be superfluous to describe in detail therapies that are in any case destined to be replaced by the treatments indicated above.

Conclusions

In conclusion, I think that the new diagnostic and therapeutic developments make it possible to propose the following procedure:

1. ICG angiography (better if performed using dynamic scanning laser); 2. Traditional direct photocoagulation treatment of all extrafoveal CNVs (MPS)

3. Low-intensity direct photocoagulation of the extrafoveal feeder vessels of juxtafoveal or subfoveal neovascularisations (feeder vessels);

4. In the absence of the requirements for point 3, photodynamic therapy of juxtafoveal and subfoveal neovascularisations with a <50% occult component (PDT);

5. Trans-pupillar thermotherapy of juxtafoveal and subfoveal neovascularisations with >50% occult component (TTT)

6. No treatment of epithelial detachments complicated by abnormal retinal vascular complexes

(traduzione dell'autore)

Ferdinando Bottoni

A B

C D

FIG.1 ICG (A) and sodium fluoroscein (B) scanning laser angiography of extrafoveal choroidal neovascularisation; and the same lesion totally obliterated after traditional laser photocoagulation as revealed by ICG angiography (C) and fluoroangiography (D).

A B
C D

E F
G H

FIG.2 A large subfoveal choroidal neovascular extension also visible under black light (A). The large area of temporal paramacular atrophy also allows the fluoroangiographic visualisation of a large supratemporal paramacular choroidal feeder trunk (B, arrow) giving rise to two minor branches that “feed” the two subfoveal neovascular extensions responsible for the marked late exudation (C and D). The patient, who was originally registered for photodynamic therapy, underwent traditional low-intensity direct photocoagulation of the two peripheral feeder vessels and their common trunk. The black light photograph taken 45 days after the treatment (administered twice) showed a marked reduction in the macular edema (E). ICG angiography demonstrated the total obliteration of the common trunk (F, arrow) and the two peripheral vessels (G). Fluoroangiography revealed the absence of late exudation (H).

A B

C D

FIG.3 ICG angiography (A) reveals a large subfoveal feeder vessel peripherically branching into many smaller extensions that are particularly active along the papillomacular area (B, late exudation at fluoroangiography). Given the subfoveal site of the feeder vessel, the patient underwent photodynamic therapy. Angiography performed one week later revealed the complete occlusion of the central feeder vessl (C, late-phase ICG) and the total disappearance of the active and exudative peripheral extensions (D, late-phase fluoroangiography). The late ICG phases also clearly show the diameter of the spot used, which appears as a round hypofluorescent area as large as the entire posterior pole delimited by the vascular arcades.