Article Text

Download PDFPDF

Update of the vascular model of AMD
  1. E Friedman
  1. Correspondence to: E Friedman Department of Ophthalmology, Harvard Medical School, Retina Service, Massachusetts Eye and Ear Infirmary, Boston, and Massachusetts Eye and Ear Infirmary, 243 Charles Street, Boston, MA 02114, USA;

Statistics from

Request Permissions

If you wish to reuse any or all of this article please use the link below which will take you to the Copyright Clearance Center’s RightsLink service. You will be able to get a quick price and instant permission to reuse the content in many different ways.

Are statins or antihypertensives protective?

The observation that age related macular degeneration (AMD) and atherosclerosis share risk factors and pathogenetic mechanisms1 has led to the development of a hypothesis that is identified as a haemodynamic2 or vascular1 model of the pathogenesis of AMD. It holds that AMD is a vascular disorder characterised by impairment of choroidal perfusion of the retinal pigment epithelium (RPE). This model, evolved over four decades,3 is now updated to incorporate recently reported evidence4 that the changes affecting Bruch’s membrane in age and AMD involve lipoproteins processed by the RPE. The model proposes that these lipoproteins accumulate in drusen and in Bruch’s membrane because the choriocapillaris does not clear them.

Theoretical models of disease are evaluated by how much they explain and by their success in making useful predictions. As the vascular model, by emphasising the pathogenetic importance of atherosclerotic processes and increased intravascular pressure, essentially predicts that statins and antihypertensive agents should have a protective effect on AMD, it is instructive to review the evidence to date.


Figure 1

Schematic representation of the vascular model of the pathogenesis of age related macular degeneration.

  • This model asserts that choroidal vascular resistance is increased by decreased compliance of ocular tissues,5 as a result of progressive infiltration with lipid. It is likely that progressive narrowing of the macular choriocapillaris with age3 also contributes to the increased resistance.6

  • While the systemic circulation is the source of the lipid in the sclera7 and choroidal vasculature, there is evidence that the RPE is the source of lipids in drusen and in Bruch’s membrane.4

  • The vascular model contends that impairment of choroidal perfusion, consisting of decreased blood flow and elevated hydrostatic pressure, compromises the processing of outer segment lipid by the RPE and the clearance of the lipoproteins secreted by the RPE. This results in drusen, pigment changes, and geographic atrophy, as well as calcification and fracture of Bruch’s membrane. Progressive accumulation of hydrophobic lipids in Bruch’s membrane may compromise RPE function further by decreasing its hydraulic conductivity.8

  • The model proposes that the combination of elevated choriocapillary pressure, breaks in Bruch’s membrane, and vascular endothelial growth factor (VEGF) causes choroidal neovascularisation (CNV).


A major strength of the model is that it explains most of what is known about the epidemiology, clinical course, and histopathology of AMD. It explains why atherosclerotic risk factors in general, and systemic hypertension in particular, are risk factors for AMD. It uniquely explains why hyperopia9 is a risk factor in AMD, and myopic eyes are relatively protected.

A weakness of the model is that it is not yet technically feasible to confirm the proposed elevated choriocapillary pressure in AMD by direct measurement. Similarly, the stiffness of Bruch’s membrane is also not measurable, and the methods available to estimate the coefficient of scleral rigidity are notoriously inaccurate.

Another weakness is that the lower incidence of AMD in African Americans can be explained by the model only by proposing that the compliance of ocular tissues may be greater in African Americans than in white people. That has yet to be determined.

Proponents of competing models assert that choroidal vascular changes in AMD are the result,10 rather than the cause, of RPE damage, as proposed by the vascular model. Recent reports,11 however, indicate that the fellow eyes of patients with unilateral neovascular AMD are characterised by impaired choroidal perfusion, in the absence of significant RPE damage. This suggests that the choroidal vascular deficit precedes and may cause the RPE changes.


The vascular model proposes that the increase in intravascular pressure results from an increase in the post-capillary resistance of the choroid. Increased resistance was demonstrated by prolonged choroidal filling times12 and by colour6 and laser Doppler imaging.13 The presence of increased intravascular pressure in AMD is suggested by phlebosclerosis of the intrascleral portion of the vortex veins,14 distension and tortuosity of the ciliary arteries15 and vortex veins, and the perivenular distribution of drusen.7

Systemic hypertension, as well as subclinical atherosclerosis, increases the risk of neovascular and non-neovascular AMD in case-control16 as well as population based17 epidemiological studies. Systemic hypertension is associated with narrowing of the terminal arteriole, with the ensuing increase in intravascular pressure upstream of the narrowing, largely sparing the capillary bed. The model proposes that when the increased pressure is superimposed on an elevated capillary pressure vascular decompensation results.


While epidiomological studies have established the risk of AMD attributable to systemic hypertension and elevated blood pressure, there are no reports of controlled clinical trials documenting the effectiveness of antihypertensives in lowering the risk of AMD.

Angiotensin converting enzyme (ACE) inhibitors and angiotensin II (AII) antagonists, two classes of drugs that reduce the activity of the rennin-angiotensin II system, are among the most effective agents in lowering blood pressure, preventing hypertensive end organ damage and mortality, with minimal disturbing side effects.18 Laboratory investigations also suggest that they may have possible advantages in controlling blood pressure in AMD.19–21


Statins reduce serum cholesterol levels by competitively inhibiting an enzyme involved in the synthesis of cholesterol.22 They cause significant reduction of cardiovascular and cerebrovascular morbidity and mortality and are useful in the management of rheumatoid arthritis, multiple sclerosis, and Alzheimer’s disease. Undesirable side effects of statins are offset by its benefits by a wide margin.

The first study to suggest a beneficial effect of statins on the course of AMD is a cross sectional survey in Britain,23 in which participants taking statins had one seventh the risk of developing AMD, compared to those who did not. However, the odds ratio for macular degeneration was 0.14 with an extremely wide confidence interval (0.02 to 0.83). A small study of elderly Australians24 reported a lower rate of progression (3.6% v 13%) of early AMD in those taking statins at baseline. While the effect was large, it was not statistically significant. Two additional reports25,26 at the most recent ARVO meeting suggest a beneficial effect of statins on AMD. Based on retrospective reviews of medical records, they both report that progression of AMD was slowed with statin use in a statistically significant manner. Two population based studies27,28 report that statins showed no protective effect.

The largest and most convincing study to date29 reports that 550 incident cases of early AMD were 70% less likely to have filled a statin prescription than 5500 matched controls.

The authors of most of these studies emphasise the preliminary nature of the reports and recommend that they be followed by larger clinical trials.


Reports of the protective effects of statins, ACE inhibitors, and AII antagonists on AMD are consistent with, even predicted by, the vascular model. However, even if their effectiveness is confirmed, the mechanisms by which they confer protection may remain uncertain.17 The statins can act indirectly by lowering serum cholesterol or directly on the processing of lipids by the RPE. It may not be the lowering of lipid levels by statins nor of blood pressure by hypertensives that offers protection, but rather other properties of these agents. It would be prudent to ascertain whether the preliminary reports of the possible effectiveness of these drugs are confirmed before speculating on the mechanism of their action. Confirmation should take the form of prospective, adequately powered, randomised, controlled clinical trials.


  • The vascular model of AMD is updated. It highlights the roles of the atherosclerotic process and blood pressure in the pathogenesis of the disorder.

  • Epidemiological studies suggest that statins and hypertensives may lower the risk of AMD. These reports are sufficiently promising to warrant adequately designed and properly executed clinical trials.

  • Confirmation of the protective effect of statins and antihypertensives will lend further support to the vascular model.


This work was supported in part by the Solman Friedman Research Fund; The Harold Alfond Research Fund; The Ben Wunsch Research Fund; and The Elinor Bronstein Research Fund, all in Boston, Massachusetts, USA.

Note in Proof

Are statins or antihypertensives protective?