Supplements for macular degeneration: what AREDS and AREDS2 showed

When AREDS was designed, many people in the United States were taking high-dose vitamins and minerals for macular degeneration and for cataract. They were sold freely over the counter, but no solid study had ever measured their effectiveness and safety. It was to fill that gap that the Age-Related Eye Disease Study was born: a large piece of research that combined the study of the natural history of the two diseases with clinical trials of supplements¹.

Two large studies later, the answers are there, and they are more nuanced than the shelves would suggest. Some people with macular degeneration derive a measurable benefit, many others do not^2,7; lutein and zeaxanthin have taken the place of beta-carotene in the formula studied^5,6; and in people without the disease the antioxidant supplements evaluated did not reduce the risk of falling ill⁸.

In this article

1. The AREDS study: vitamins and zinc
2. AREDS2: lutein, zeaxanthin and omega-3
3. The Cochrane review: the absolute numbers
4. Why they are not for everyone
5. Frequently asked questions
6. Sources

The AREDS study: vitamins and zinc

In absolute numbers, the result can be read like this: in five years about 20 people in 100 who took the complete combination progressed to the advanced form, against 28 in 100 with placebo³.

The decisive finding, however, lies in the details. People with the mildest forms, that is, many small drusen, a few intermediate-sized drusen or pigment changes, had at the start a probability of only 1.3% of progressing to the advanced form in five years. Excluding these 1,063 people, the effect in the highest-risk group became clearer, both for zinc and for the combination with antioxidants². Hence the idea that these supplements concern only some of the people with macular degeneration.

AREDS2: lutein, zeaxanthin and omega-3

The second study, AREDS2, enrolled 4,203 people aged 50 to 85 in 82 centres in the United States, all with intermediate macular degeneration in both eyes or advanced in one; the average age was 74 and only 7% smoked⁴. The question was whether adding to the AREDS formula lutein and zeaxanthin, the carotenoids that concentrate in the macula, or the omega-3s from fish, gave an extra advantage; and whether beta-carotene could be removed and zinc reduced⁵. We discuss lutein in foods in our article on what to eat for your eyes.

After a median of five years, the main hypothesis was not confirmed. The probability of progressing to the advanced form was 31% with placebo, 29% with lutein and zeaxanthin, 31% with omega-3 and 30% with both: no statistically significant difference. Neither removing beta-carotene nor reducing zinc changed progression⁵.

The clearest signal concerned safety. Among those who took beta-carotene there were more lung cancers, 2.0% against 0.9%, especially in people who had smoked in the past⁵. After ten years of follow-up, among 3,882 participants, the probability of lung cancer remained higher with beta-carotene, but not with lutein and zeaxanthin⁶.

The long-term analyses added a nuance. In the direct comparison, the formula with lutein and zeaxanthin was associated with a risk of late-stage disease 15% lower than that with beta-carotene; omega-3s, by contrast, showed no effect even at ten years⁶. These are secondary analyses, and have to be read as such; but together with the finding on lung cancers they explain why lutein and zeaxanthin were proposed as substitutes for beta-carotene^5,6,7.

The Cochrane review: the absolute numbers

In 2023 Cochrane updated its review of antioxidant supplements in people who already have macular degeneration: 26 studies, 11,952 people aged 65 to 75, conducted in the United States, Europe, China and Australia; much of the evidence, however, comes from AREDS⁷. People who took antioxidant multivitamins were less likely to progress to the late form, with evidence of moderate certainty. Translated into absolute numbers, the effect depends on how high the starting risk is⁷:

• early forms, at low risk of worsening: about 4 fewer cases of late macular degeneration per 1,000 people, with an estimate ranging from 1 to 6;
• intermediate forms, at higher risk: about 78 fewer cases per 1,000 people, with an estimate ranging from 26 to 126.

The same review reports, again with moderate certainty, a lower risk of losing three or more lines of vision⁷. In an exploratory analysis of the AREDS2 data, replacing beta-carotene with lutein and zeaxanthin was associated with a lower risk of progression to the late form⁷.

Why they are not for everyone

The first reason lies in the numbers just seen: for people with only small drusen or an early form, the expected benefit is very small^2,7. The second concerns people who do not have macular degeneration. Another Cochrane review, of five randomised trials with data for 76,756 people, found that vitamin E or beta-carotene supplements do not reduce the probability of falling ill; vitamin E might even slightly increase the risk of the late form, with evidence of moderate certainty⁸. In one of the studies, moreover, more haemorrhagic strokes were counted in the vitamin E group, a finding of low certainty⁸.

The third reason is safety: the AREDS formulas contained vitamins and minerals at high doses², and the case of beta-carotene in former smokers shows that a supplement can have side effects⁵.

There is also a particular case, geographic atrophy, the advanced dry form. A post hoc analysis of AREDS and AREDS2 found that the supplements studied slowed the advance of the atrophy towards the centre of the macula, although not the overall enlargement of the lesion⁹. We discuss this in our article on geographic atrophy.

In short, the evidence concerns people with specific characteristics. Whether these formulas make sense for someone depends on the stage of the disease, which is established by the examination, and on factors such as smoking: it is an assessment to make with the ophthalmologist.

Frequently asked questions

Are lutein and zeaxanthin of any use to people who do not have macular degeneration?

The AREDS and AREDS2 studies involved people who already had the disease^2,4, so they do not answer this question. In people without macular degeneration, the antioxidant supplements evaluated by Cochrane, that is, vitamin E and beta-carotene, did not reduce the risk of falling ill⁸.

Are omega-3s of any use for macular degeneration?

In AREDS2 the omega-3s added to the formula did not change the probability of progression either after five or after ten years^5,6.

Can eye supplements have side effects?

Yes. In AREDS2 beta-carotene was associated with more lung cancers, especially in former smokers⁵; in the Cochrane reviews vitamin E was associated with a possible small increase in the risk of the late form and, in one study, with more haemorrhagic strokes⁸. In AREDS, by contrast, no formula showed significant serious side effects².

Do supplements bring sight back?

In the studies the effect measured is a lower probability of worsening, in intermediate forms, and a lower risk of losing lines of vision, not a recovery of sight already lost^2,7.

The lesson of AREDS and AREDS2 is precise: a measurable benefit, but for a specific group of people, those with intermediate forms or advanced in one eye, and with particular attention to safety^2,5,7. For everyone else, the evidence does not say that a supplement changes the risk of falling ill⁸.