AMD risk factors

Age-related macular disease is the leading cause of irreversible visual loss in the developed world.1 In the UK, half of those registered as blind or partially sighted every year (approximately 30,000) have the condition.2 Age-related macular disease is the most frequently occurring condition that results in permanent vision loss in those over 65 years of age.3 Information about risk factors for age-related macular disease is limited and there are no treatment options for most people.

Classification of age-related macular disease
Terminology
The International Classification and Grading System for age-related maculopathy (ARM), and age-related macular degeneration (AMD) has been developed in an attempt to standardise terminology. ARM refers to large soft drusen and pigmentary abnormalities of the retinal pigment epithelium (RPE) and the retina (Figure 1), and AMD refers to later stages of the disease (when visual acuity has been affected) such as geographic atrophy (GA) (Figure 2), choroidal neovascularisation (Figure 3), pigment epithelium detachment and fibrous scarring of the macula.4 The term age-related macular disease is often used to encompass ARM and AMD.
Atrophic (or dry or non-exudative) AMD is the most common form of the disease and accounts for around 85-90 per cent of all cases. It is characterised by GA (that is, atrophy of the retina spread over a wide area centred on the macula) and usually associated with a build-up of drusen under the RPE. Atrophic AMD could be described as the chronic form of the disease since its progression is slow and causes the gradual deterioration of central vision over many (up to 20) years.
Exudative (or wet or neovascular) AMD, although less common (10-15 per cent of all cases) than dry AMD, is a more aggressive or 'acute' type of AMD. It can result in severe vision loss in a very short period of time (three to six months). Exudative AMD is characterised by angiogenesis (the growth of new blood vessels). In the case of exudative AMD, the new blood vessels form in the choroid, and are referred to as choroidal neovascularisation (CNV). The new blood vessels that develop are abnormal; they leak causing the accumulation of fluid within and beneath the retina. This results in a sudden loss of central vision. Permanent loss of vision occurs sometime later as the outer retina (including the photoreceptors) becomes atrophic or replaced by fibrous tissues.

Statistical terms
The epidemiological literature uses the terms odds ratio (OR) to quantify association. It is defined as the ratio of the odds of an attribute, such as cigarette smoking, being present in individuals with a disease such as AMD, to the odds of the attribute being present in individuals without the disease. The statistic used to quantify risk is the relative risk (RR), defined as the ratio of the risk of the disease among those with the attribute (eg female gender), or exposure (eg sunlight), of interest, to the risk in those without the attribute or exposure. Estimates of OR and RR are considered to be statistically significant and supportive of a positive association or increased risk if they include values greater than 1.0 and have a confidence interval, typically 95 per cent around the estimate.
Risk factors for age-related macular disease can be divided into two groups (Table 1).

Non-modifiable risk factors
Age
Increasing age is considered to be the most important risk factor for AMD.1 Reported prevalence ranges from 0.2 per cent of the population aged 55 to 64 years,5 up to 25 per cent in those aged 65-75 years.6 GA among the population is reported to increase from 0.04 per cent in those aged between 55 and 64 years, to 4.2 per cent aged over 85 years. Exudative AMD follows a similar pattern, 0.17 per cent in those aged 55 to 64 years and 5.8 per cent for those older then 85 years.7 An 8-10 fold increase in the prevalence of the condition has been recorded in those over the age of 90 years, compared with those aged 50 years.8

Genetic predisposition
A genetic basis for age-related macular disease is supported by the findings of a number of epidemiological studies, for example Klaver C, Wolfs R and Assink J.9 The complex aetiology of the disease has hindered the progress of genetic studies, as it is a multifactorial disorder and is most probably controlled by more than one gene. An autosomal dominant pattern of inheritance has been reported in some patients,10 but no definitive genetic component of AMD has yet been demonstrated. Approximately 25 per cent of AMD in the population has been attributed to a genetic component.9 One study reported that family history of AMD is the strongest risk factor for the disease.11 First-degree relatives of patients with AMD have been reported to develop the disease at an increased rate and at a relatively younger age. Relatives of patients with ARM have a 50 per cent risk of developing the condition.9 In the Rotterdam Eye Study the children of subjects with AMD were 12 times more likely to develop the condition than children of subjects showing no signs of AMD.12 This relationship does not, however, exclude the possibility of an environmental influence.

Gender
The Beaver Dam Eye Study found that, after adjusting for age, women of 75 years or more had twice the incidence ARM, and over seven times the incidence of AMD than men.7 One hypothesis for this increased risk is the loss of protective oestrogen in post-menopausal women. The Eye Disease Case-Control Study Group (EDCCS) reported a higher incidence of ARM in women who had given birth.13 Parity (number of children to which a woman has given birth) greater than one was associated with a 1.8 times increased risk of AMD in a case-control study.14 Data from the Rotterdam Eye Study showed that women who had an early menopause following the removal of one or more ovaries had a significantly increased risk of developing AMD than women who had their menopause at 45 years or later.15

Race
Several studies have provided evidence to suggest that AMD is more common in whites than blacks. In the Baltimore Eye Survey16 AMD accounted for 30 per cent of bilateral blindness in whites and 0 per cent in blacks. AMD occurred in 2.1 per cent of the white population over 70 years, and no cases were detected among black subjects in this age group. Drusen were identified in 20 per cent of individuals in both racial groups, but larger drusen occurred in 15 per cent of whites compared with 9 per cent of blacks. Pigmentary abnormalities were also more common among older whites, occurring in 7.9 per cent of white, and 0.4 per cent of black subjects.16 The Age-related Eye Disease Study (AREDS) found a greater risk of choroidal neovascularisation in whites than non-whites.17 However, AREDS investigators reported that the prevalence of ARM did not differ significantly between whites and blacks, (20.2 per cent and 19.8 per cent respectively).17
It has been hypothesised that racial differences could be related to melanin, which 'acting as a free radical scavenger, or in some other way, may protect the pigment epithelium, Bruch's membrane, choroid, or the outer retina from degenerative changes, predisposing the patient to choroidal neovascularisation'.18 The melanin protects the retina, and in particular the macula, from exposure to direct sunlight, which reduces the risk of oxidative damage.19

Iris colour
A number of studies have reported an increased risk of AMD in people with blue irides and light pigmented ocular fundi, compared with those with darker eyes. Blue iris colour has been significantly associated with an increased risk of ARM (OR 1.45) and AMD (OR 1.69).19 It is thought that the lower risk of AMD in pigmented eyes is due to increased ocular melanin. Several studies found no relationship between iris colour and AMD, for example.20

Refractive error
Hypermetropia was demonstrated to be a risk factor in one case control study in the general population (OR = 1.33), as well as in those with exudative AMD (OR = 1.83).21 This risk factor has been reported in many other studies, for example.22

Cataract
An association between nuclear sclerosis and ARM was found in the Beaver Dam Eye Study.23 The Chesapeake Bay Watermen Study provided further evidence; early AMD was found to be 2.5 times more likely to develop in the presence of nuclear opacities.20 No association was found between AMD and cortical opacities which is inconsistent with the findings of the Framingham Eye Study, where AMD was less likely in those with nuclear sclerosis and more likely in those with cortical opacities.
The relationship between cataract surgery and age-related macular disease has been investigated. In a case-control study there was no increased risk of progression of ARM in patients with mild to moderate degrees of the condition, one year post-operatively.24 Analyses based on three population-based studies25-27 provided a statistically significant OR of 1.7 for an increased risk of AMD with cataract surgery. The Beaver Dam Eye Study found previous cataract surgery to be associated with a higher 10-year incidence of AMD (RR 3.8), as well as a greater progression of ARM (OR 2.0), but not with the incidence of ARM.28
Several studies (for example, Goldberg J, Flowerdew G, Smith et al29) have found a positive association between aphakia and ARM. However, this was not found to be significant in the Oulu study after controlling for age.37 This may be accounted for by easier visualisation of macular lesions after surgery, phototic retinal injuries occurring during surgery, or inflammatory changes.30
It has been proposed that cataract and age-related macular disease may share common risk factors,31 that we may be less able to detect the disease in an eye with lens opacities, that cataract surgery in some way predisposes an eye to develop age-related macular disease,26 or that removal of the cataract exposes the eye to certain wavelengths of light that make the retina more susceptible to it.23

Modifiable risk factors
Cigarette smoking
Smoking depresses serum antioxidant levels and is associated with alteration of choroidal blood flow. These mechanisms may be involved with the development of age-related macular disease.32 Men who smoke one pack of cigarettes per day have 72 per cent of the plasma antioxidant levels of non-smokers.33 The reduced availability of antioxidant components may result in oxidative insults to the outer retina, leading to an increased risk of developing the condition. Tobacco is a vasoconstrictor and the damage it causes to choroidal blood vessels and blood flow may promote atherosclerotic and hypoxic changes in choroidal vessels.17 Greater number of cigarettes per day, greater number of smoking years and lower ages at which smoking started are all associated with a higher risk of developing ARM.34 Another study found that subjects who were younger than 85 years and current smokers had a 6.6 times increased risk of exudative AMD compared with those who had never smoked.35 This increased risk remained in subjects who had stopped smoking less than 20 years before examination.36
The risk of AMD has been shown to increase with longer duration of smoking, with up to a five-fold increased risk among people who had smoked for 40 years or more.36 These results provide evidence for a dose-response relationship of lifetime exposure that reflects the duration and amount of current and past smoking with AMD.36 A number of small studies have shown no relationship between smoking and AMD, for example, Hirvela H, Luukinen H, Laara E et al.37

Alcohol
The NHANES I29 found that red wine was associated with lower risk of AMD, which may be related to the antioxidant properties of phenolic compounds within red wine.38 In the Beaver Dam Eye Study, consumption of beer was related to an increased risk of retinal pigmentation (OR 1.13) and exudative AMD (OR 1.41).39 Compared with non-drinkers, high alcohol intake has been positively related with AMD. Men (RR 2.16) and women (RR 2.20) in the highest category of wine intake (two or more glasses of wine per day) have been shown to have a two fold increased risk. This positive association was shown to be strongest with white wine. A high spirits intake was also associated with a significantly increased risk in both men and women (RR 2.03).40 The high correlation between smoking and drinking may contribute to the positive association between AMD and heavy drinking.40 Several studies have found no relation between alcohol consumption and age-related macular disease, for example, Cho EY, Hankinson SE, Willett WC et al.40

Socio-economic factors
Compared with subjects who had less than nine years of education, those with nine to 12 years had a 36 per cent reduced risk, and those with more than 12 years had a 39 per cent reduced risk of developing age-related macular disease.29 The AREDS17 found that number of years of education was inversely related to the presence of one or more large drusen, GA, extensive intermediate drusen and exudative. Several other studies found no relationship between AMD and education.11,41 This conflicting evidence could be attributed to selection or location bias.

Nutrition
It is hypothesised that people with low antioxidant levels may be more prone to oxidative damage of the retina and thus to AMD.42 This hypothesis received support from another study that showed higher serum levels of antioxidant micronutrients, particularly carotenoids, to be protective against AMD. Subjects with the highest serum levels of carotenoids had a 43 per cent lower risk of developing advanced AMD, compared to those with lower levels.13
Recent studies have indicated that dietary changes implemented early in life have an important role in preventing and slowing the development of AMD.43 Fruits and vegetables are rich sources of protective carotenoids, in particular beta-carotene, lutein and zeaxanthin.44 People who consume fruits and vegetables, in particular dark, leafy greens, at least once a day have a significantly lower risk of age-related macular disease compared to those who consume such foods less than once per week. Spinach is a particularly rich source of lutein and zeaxanthin and its consumption is associated with an 86 per cent lower risk of developing AMD.45
Three randomised controlled trials have been carried out to investigate the effect of nutritional supplements on the progression of age-related macular disease. The Visaline and ATBC trials,46,47 found no relationship between supplementation and the development and progression of AMD. However, the AREDS, the largest of the three studies, determined that a combination of high dose vitamin C, vitamin E, beta-carotene and zinc was moderately effective in preventing progression to advanced AMD. This effect was only seen in those subjects with extreme intermediate drusen, large drusen or non-central GA without advanced AMD.48 At present, people with age-related macular disease should be advised to follow the dietary guidelines that promote an increased consumption of fruit and vegetables.

Body mass index
Excessive caloric intake may increase the risk of AMD because of an increased risk of oxidative damage.37 In the AREDS population, higher body mass index (BMI), a measure of obesity, was associated with exudative AMD.17 These results were consistent with those determined in the Oulu study.37 The Blue Mountains Eye Study reported an increased risk for the presence of ARM in obese patients compared to those with a normal BMI (OR 1.78).49 Increasing waist-hip ratio was associated with an increased risk of pigmentary abnormalities in the Colorado Study (OR 1.87).50

Dietary fat intake
Dietary fat intake may influence the risk of developing AMD in two ways. First, the increased risk of atherosclerosis caused by higher levels of cholesterol in the bloodstream may affect the choroidal circulation. Second, the supply and removal of waste products from the RPE could be affected by deposition of fat in Bruch's membrane.51
The Beaver Dam Eye Study found a weak relationship between higher dietary fat intake and increased risk of ARM (RR 1.6).52 However, in the EDCCS,13 participants with high serum cholesterol had an increased risk of exudative AMD compared with people with low serum cholesterol values (RR 4.1). This finding was not repeated in NHANES I where subjects with high cholesterol intake were less likely to develop AMD than those with low cholesterol (OR 0.51).29

Cardiovascular disease
A predisposition to cardiovascular disease may affect blood circulation in the retina and contribute to the development of AMD.40 The narrowing of the choriocapillaris caused by atherosclerosis may result in increased haemodynamic resistance and a decreased supply of nutrients and oxygen to the retina.40 This contributes to the deterioration of the RPE, either due to the development of ischaemia or the build up of waste product.16 Risk factors for cardiovascular disease such as smoking, hypertension, high serum cholesterol levels and high BMI have all been adversely related to age-related macular disease.53 Some protective factors for cardiovascular disease such as antioxidant vitamin intake54 and postmenopausal hormone use have also been reported to provide protection.40

Hypertension
It has been suggested that the choroidal blood vessels are affected by hypertension, leading to an increased risk of signs of ARM and that the atherosclerotic process which affects the cardiovascular system may also affect the choroidal vasculature.55
The Beaver Dam Eye Study reported that higher pulse pressure was associated with a 30 per cent increase in the five year incidence of AMD, and a 25 per cent increased progression of exudative AMD in persons aged 65 years or more.49 The AMD Risk Factors Study Group found a relationship between exudative AMD and moderate to severe hypertension, particularly among patients being treated with antihypertensive medication such as diuretics and vasodilators. All of the findings relating hypertension to AMD were observed in patients with the exudative form of disease. Atrophic AMD was not found to be related to hypertension.56
However, no relationship was found in several other studies, for example, Hirvela H, Luukinen H, Laara E et al.37 Differences in selection criteria, number of subjects, location, or timescale between the studies may be responsible for the inconsistency between results. Individuals with moderate to severe hypertension have been identified as being at high risk for development of exudative AMD, and require careful monitoring and follow up.17

Statins
Atherosclerosis and abnormalities in lipid metabolism are associated with an increased risk of AMD.57 Statins are drugs that can modify the lipid profile and help to lower the risk of coronary heart disease, and may even induce regression.58 Among men and women aged 66 to 75 not taking statins, 22 per cent showed signs of AMD, compared with only 4 per cent of treated participants. There are two suggested mechanisms. First, statins may prevent accumulation of deposits in Bruch's membrane, which is thought to occur with high concentrations of plasma cholesterol. Second, antioxidant properties of statins might protect the outer retina from oxidative damage.58
The small sample size and cross-sectional design of this study was addressed in a case-control study were a significantly lower frequency of age-related macular disease was reported in statin users compared with non-users (OR 0.45).59 Two studies investigating the effects of lipid-lowering agents and hypocholesterolaemic medications however, found no association with ARM.60,61

Aspirin
AMD is consistently associated with a reduced number of capillaries in the macular region, and atrophy of the choriocapillaris. A corresponding age-related decrease in choroidal blood flow has also been reported in studies investigating AMD.62
If vascular disease is a contributing factor in AMD, then agents that enhance systemic circulation and reduce the risk of vascular events could potentially have a beneficial effect on choroidal circulation. One possible protective agent is aspirin, which irreversibly inhibits platelet aggregation, resulting in a marked reduction in platelet function.63
The Physician's Health Study was the first to investigate the role of aspirin in development of AMD.64 Results indicated that there was no beneficial effect of five years of low dose aspirin treatment on the development of AMD. However, aspirin treatment was associated with a reduced risk of AMD among those who reported systemic hypertension. No beneficial effect of aspirin was reported by other studies, for example, Hirvela H, Luukinen H, Laara E et al.37
Sunlight exposure
Individuals who have received excessive light exposure and have low levels of protective antioxidant nutrients may be particularly vulnerable to the age-related macular disease. The Beaver Dam Eye study showed that the amount of outdoor leisure time was significantly associated with the development of ARM. Outdoor leisure time during teenage years (13 to19 years) and in the 30s has been significantly associated with the risk of ARM, and five or more daily outdoor hours has been shown to double the chance of developing ARM compared with two daily outdoor hours. A slight, but significant protective effect was associated with the use of hats and sunglasses during the teens and 30s.65 The POLA (Pathologies Oculaires Lies l'Age) Study found a reduced risk of soft drusen with use of sunglasses.66
The Chesapeake Bay Study concluded that cumulative exposure to blue light or all wavelengths of light during the 20 years previous to examination was associated with an increased prevalence of AMD.67 Several studies, however, found no such relationship, for example, Hirvela H, Luukinen H, Laara E et al.37

Management of age-related macular disease
Optometry
Non-exudative disease is often monitored by the optometrist and advice given to the patient regarding lighting and adjustment of their viewing habits. The slowly progressive disease is unlikely to warrant any treatment by an ophthalmologist and referral is only necessary via the GP for confirmation of the diagnosis and for registration purposes. As Primary Care Trusts start to develop a more efficient LVA service, it is likely that such patients may be directly referred to an LVA specialist centre to allow for provision of LVAs or registration. The best approach for non-exudative AMD has to be one of prevention. Advice about giving up smoking, avoiding blue light sources, leading a healthy lifestyle with plenty of exercise and antioxidant rich food stuffs is all useful.
Thermal laser photocoagulation
Laser photocoagulation uses the thermal energy produced by an argon laser to destroy the choroidal neovascularisation (CNV). Randomised trials have shown:

<25C6> Photocoagulation can significantly reduce the rate of severe visual loss and preserve contrast sensitivity in selected patients with exudative AMD where the lesions are well demarcated. Where the lesions are hidden, due to haemorrhage or multiple new vessel sites confusing the view (occult neovascularisation), the technique is not appropriate
<25C6> CNV recurs within two years in about half of those treated
<25C6> Photocoagulation may reduce visual acuity initially, and in cases of subfoveal CNV, the resultant loss of visual acuity due to the laser scar means the technique is ineffective and contraindicated.

Photodynamic therapy (PDT)
This technique, recently approved by NICE for appropriate patients, involves the injection of verteporfin (trade name Visudyne) into a patient. This photosensitive dye binds to low density lipoproteins (LDLs) present in the blood stream. Areas of vasoproliferation, such as in the site of CNV, have a high concentration of LDL receptors. The drug will tend to accumulate in the CNV site and being light sensitive, is denatured when exposed to a specific light source. On exposure to red light (689nm) the verteporfin initiates a number of photochemical reactions that result in the formation of free radicals and singlet oxygen; these are cytotoxic and induce occlusion of the neovasculature. Other nearby vessels remain intact, while minimal damage occurs to adjacent retinal structures such as the overlying photoreceptors and RPE.
PDT with verteporfin reduced the risk of moderate and severe vision loss in selected people with exudative AMD (those with predominantly classic as opposed to occult lesions as defined after fluorescein angiography). Results indicate patients treated with Visudyne had, after three years, better vision than untreated patients after only three months. Figure 4 shows an eye after PDT.
NICE guidelines now recommend that all patients with exudative AMD who have a confirmed diagnosis of classic with no occult subfoveal CNV should be considered for PDT. As far as optometrists are concerned, it is clear that the two treatments, which have been proven to have some clinical benefit, are suitable for a very small proportion of AMD. Optometrists are in a vitally important position to be able to detect and refer exudative AMD at an early enough stage to allow assessment and possible treatment if subsequently found to be suitable.

Radiotherapy
A focused beam of radioactivity has been used to destroy CNV tissue. As yet results have proven inconclusive.

Surgery
Submacular surgery involves a procedure whereby the CNV plaque is physically excised from under the retina. At present there is insufficient evidence on beneficial effects of sub-macular surgery. Rates of recurrence of choroidal neovascularisation are high, and there is a clinically significant risk of ocular complications resulting in visual loss and further surgical intervention.
Retinal translocation may be partial or total. In the former, an incision is made near the damaged macular and the retina stretched so that the atrophic area is no longer on the visual axis. The latter requires a complete circle of retina, with the atrophic zone towards the edge of the circle, to be cut and rotated. Both result in peripheral retina now being on axis. The reorientation of the image is minimised by surgery to the extraocular muscles in an attempt to induce torsion to compensate from the image deviation.
RPE transplantation attempts to insert healthy RPE cells into the atrophic area and much further work needs to be done before the technique can be properly appraised.
Transpupillary thermal therapy
Several ad hoc trials have tested the effects of low energy light on CNV. Some studies have claimed success in treating occult CNV, though other researchers dispute these claims and, as such, the treatment is not yet assumed to be effective.

Antiangiogenesis
New vessel growth (angiogenesis) is a complex phenomenon. The influence of certain drugs in inhibiting such growth has, however, been known for a long time. When, in the 1960s, thalidomide was introduced as a pain relief for pregnant women, little was known about its anti-angiogenetic properties until events took a tragic turn. Such drugs have been at the forefront of recent research aimed at selective intervention by agents that prevent or inhibit new vessel growth and, obviously, the implications for CNV and proliferative diabetic retinopathy are great. Intravitreal triameinolone is showing some promise, and much research is directed at influencing the build-up of vessel endothelial growth factor (or VEGF), which many believe to be the vasoactive substance at the heart of vasoproliferative disease in the eye.

Conclusion
The identification of risk factors for AMD is particularly important, as treatment options for the condition are limited. The main factor in the development of the condition is age, but smoking and genetic predisposition are also linked. Although factors such as alcohol intake, sunlight exposure, BMI, dietary fat intake and hypertension have not been consistently linked with increased risk of developing the condition, it may be that in those people with genetic predisposition for AMD, the condition manifests itself according to exposure to these, and other relevant risk factors.

References
A full list of references is available from the optician clinical editor - email: william.harvey@rbi.co.uk

<25C6> Dr Frank Eperjesi is a lecturer in the Dept of Optometry and a research optometrist in the Neurosciences Research Institute at Aston University. Hannah Bartlett is a postgraduate at in the Neurosciences Research Institute and is funded by the College of Optometrists