Restricted Access

You must be logged in to view this content.

AREDS2 review


Gains were modest but real

Dr Jeffrey Anshel
President and founding director, Ocular Nutrition Society, USA

The results of the AREDS2 study are finally in. What do the data ­reveal? Let's take a quick look at the details and the recommendations. The National Eye Institute's (NEI) recommendation adds 10 mg lutein and 2 mg zeaxanthin and eliminates beta carotene that was in the original AREDS supplement. Results from AREDS2 report an 18 per cent reduction in progression to advanced AMD in subjects who received 10 mg lutein and 2 mg zeaxanthin in addition to an AREDS supplement without beta carotene compared to the original AREDS supplement with beta carotene.

This reduction in progression to advanced AMD is even greater in study subjects with the lowest intake of lutein and zeaxanthin in their diet, which is more representative of the general population. In the USA, the dietary intake of lutein and zeaxanthin is typically less than 1 mg per day. This amount is well below the level that science has shown to be needed to be effective in prevention of AMD.

The NEI also recommends improving the safety and efficacy of the AREDS supplement by removing beta carotene and replacing it with 10 mg lutein and 2 mg zeaxanthin. This recommendation is due to potential safety concerns of former smokers who received beta carotene in the original AREDS supplement. About 50 per cent of people with AMD are former smokers.

In summary, the lutein and zeaxanthin made a small change in late-stage disease.

Surprises, observations

Dr Emily Chew, the lead investigator, mentioned during her presentation of the AREDS2 results that lutein supplementation of more than 10 mg per day may be toxic. There are more than 10 peer-reviewed published studies that have supplemented subjects with more than 10 mg of lutein per day, none of which has demonstrated any toxicity.

To the surprise of many, there was no effect of omega-3 essential fatty acids (EPA/DHA). There are speculations why this occurred, ranging from a `well-nourished' study population who took fish oil on their own, to those reaching a saturation point of serum levels of these acids. Considering that the highest levels of fatty acids in the retina are DHA, more evaluation of these results are warranted.

AREDS2 subjects were at a more advanced stage of AMD and were on average five years older (average age 74) than the participants in the original AREDS study (average age 69). The AREDS2 subjects also had double the incidence of diabetes. The authors note that these differences could impact the ability to detect a more significant reduction in the progression to advanced AMD.

Other concerns regarding the study

  • The AREDS2 study design (similar to the original study) was to determine progression to the late stages of AMD—not prevention, reversal or stopping of the disease.
  • In the primary prevention aspect of the study, there were no positive results of any of the intervention nutrients (L/Z or EPA+DHA). Only when secondary outcomes were reviewed were there any significant results.
  • The participants in the AREDS2 study were considered `well-nourished' which is not representative of the US population in general.
  • Evaluations of progression to advanced AMD were based solely on visual acuity and retinal photography. No tests for functional vision (contrast sensitivity, glare recovery, dark adaptation and so on) were performed.
  • Genetic evaluations were conducted on only one-fourth of the study population and those statistics have not been released yet.
  • Macular pigment optical density was measured on only one-fourth of the study population and those statistics have not been released yet.
  • The placebo arm had high intakes of EPA + DHA in their diet that confounded the statistical analysis. Otherwise, it is unclear why this dosage of EFAs did not make a difference. Perhaps a higher intake was needed but a Holman RBC index was not obtained. Perhaps more DHA was required as found within the NAT2 study.1 Folate is required to enhance DHA absorption. Perhaps the participants were taking systemic pharmaceuticals that depleted the B vitamins.
  • Cataracts are a part of a slow developing process and other studies have shown that high levels of certain antioxidants, such as lutein/zeaxanthin, in patients with low habitual dietary intake, can be effective in reducing the incidence and/or progression of cataracts.2,3
  • The amount of zinc recommended in the AREDS2 formula is over twice the recommended daily intake set by the Institute of Medicine board. While this study did not show the negative effects, it might have been due to the type (oxide) of zinc used. Zinc oxide is not well absorbed into the cells. We look forward to more data and discussion.
  • There was no true control group; all study participants received a supplement of some sort.
  • Lutein and zeaxanthin are not a `substitute' for beta-carotene because they are xanthophyll carotenoids and don't convert to vitamin A.
  • In the original AREDS study, about 67 per cent of the participants took Centrum while the AREDS2 participants had about 89 per cent taking Centrum Silver. At the time of AREDS, Centrum did not yet contain lutein.
  • It is unclear why the geographic atrophy patients fared poorer with supplements, compared with AREDS post-op analysis results. Could geographic atrophy be a separate AMD disease state dependent on additional nutrient factors?
  1. Souied EH, Delcourt C et al. Oral docosahexaenoic acid in the prevention of exudative age-related macular degeneration: the Nutritional AMD Treatment 2 study. Ophthalmology 2013; 120: 8: 1619-1631. Epub 2013 Feb 8.
  2. Christen WG, Liu S et al. Dietary carotenoids, vitamins C and E, and risk of cataract in women: a prospective study. Archives of Ophthalmology 2008; 126: 1: 102-109.
  3. Lyle BJ, Mares-Perlman JA et al. Antioxidant intake and risk of incident age-related nuclear cataracts in the Beaver Dam Eye Study. Am J Epidemiol 1999; 149: 9: 801-809.

Claims don't match results

Associate Professor Peter Keller
BAppSc(Optom) MBA MHEth PGCertOcTher PhD(UWA) FACO

Primarily, AREDS2 was designed to determine whether adding oral supplements (lutein and zeaxanthin) and/or long-chain omega-3 fatty acids (DHA and EPA) to the AREDS cocktail of vitamins and minerals decreases the risk of progression to advanced AMD for those with intermediate disease. Its secondary aim was to evaluate the effect of eliminating beta carotene, lowering zinc doses, or both in the AREDS formulation.

The results of AREDS2 were eagerly anticipated by patients, eye-care practitioners, researchers and the purveyors of dietary supplements. But what did the study chairwoman, Dr Emily Chew, present at the Association for Research in Vision and Ophthalmology conference in May, and how does it compare to what is contained within the JAMA article published online at the same time, and how does that measure up to what has been claimed by various third-party commentators?

As with many oral presentations, Dr Chew's discussion of the AREDS2 results was nuanced, and can be used to support a number of different positions when taken in isolation from the published results. Despite the possible differences in interpretation, there seems little disagreement with her conclusion that the addition of lutein and zeaxanthin, DHA and EPA, or both, did not significantly reduce risk of progression to advanced AMD.

The apparent contradiction with her conclusion that the addition of lutein and zeaxanthin showed a significant additional decrease in risk of progression to advanced AMD is explained by the difference between primary and secondary analyses. Put simply, the small positive result for adding lutein and zeaxanthin was found only through lower-order exploratory analyses and cannot be used as the basis for a clinical recommendation.

Dr Chew also stated that there was no effect due to eliminating beta carotene or lowering zinc dose. Although used by some to argue for the retention of high-dose zinc, the application of Ockham's razor would suggest the opposite: that beta-carotene could be removed and the zinc dose lowered without a loss of benefit.

Dr Chew did suggest in her presentation that removal of beta carotene from the AREDS formulation could reduce the risk of lung cancer in smokers and former smokers, but it should be noted that AREDS2 did not randomise any smoker to receive beta carotene, and her suggestion is based on exploratory analyses of lung cancer rates among former and non-smokers taking all formulations without beta carotene versus all formulations including beta carotene.

Once again, the conclusion is based not on primary analyses but exploratory lower-order analyses of the AREDS2 data and combined with evidence from other sources. Nonetheless, it is a reasonable suggestion based on minimising potential harm. Overstating Dr Chew's conclusion as a clear recommendation to remove beta carotene and replace it with lutein zeaxanthin is not what was presented and is not supported by the evidence from AREDS2.


One could argue that it would be better to counsel individuals to increase their dietary uptake of lutein and zeaxanthin, something Dr Chew hinted at in her ARVO presentation: `[AREDS2] confirms the basic recommendation that people should have good diets; first of all, eat a healthy diet of vegetables'.

For those who have not yet read the JAMA article the language is unequivocal: `addition of lutein and zeaxanthin, DHA and EPA, or both, to the AREDS formulation in primary analyses did not further reduce risk of progression to advanced AMD' and `there was no apparent effect of beta carotene elimination or lower dose zinc on progression to advance AMD'.1 Essentially, there is strong evidence of no benefit in adding to the formulation and less evidence of no (loss of) effect by removing beta carotene or reducing the zinc dose.

It is instructive to note that the authors also stated that `exploratory analyses demonstrated results that suggest the role of lutein and zeaxanthin needs to be examined further' and that `lutein and zeaxanthin may play a role for reducing risk of progression to advanced AMD when given without beta carotene' but `this hypothesis requires further study'.1 Hardly a clear recommendation.

Finally, it is interesting to note that all four primary randomisation groups in AREDS2 (each receiving a variation of the AREDS formulation) progressed to advanced AMD at rates higher than the placebo (no treatment, natural history, high risk categories 3 and 4) sub-group in the original AREDS clinical trial (29 to 31 per cent compared to 28 per cent) and much higher than the same high risk category receiving the AREDS formulation (20 per cent) in the original AREDS.

It has been stated that the AREDS2 study population was older, with a median age being five years older than the full AREDS study population, but that does not compare the median age for participants in the same high risk group (categories 3 and 4) across the two studies.

Whatever the median age, it does demonstrate the importance of comparing similarly aged groups when making sense of risk of progression to advanced AMD, and in combination with the absence of a true placebo group in AREDS2, disallows the claim that AREDS2 confirms the benefits of dietary supplements found in AREDS.

Unfortunately, as Emily Chew said in her ARVO presentation, AREDS and AREDS2 will never be repeated, so a number of outstanding questions will never be answered, but no doubt we will read more about this and interesting exploratory analyses of the data in future AREDS2 reports.

  1. Age-Related Eye Disease Study 2 Research Group. Lutein + zeaxanthin and omega-3 fatty acids for age-related macular degeneration: the Age-Related Eye Disease Study 2 (AREDS2) randomised clinical trial. JAMA 2013; 309: 19: 2005-2015.

Like us on Facebook

Subscribe to our News RSS Feed

Latest Tweets

Recent Comments