A large body of research has focused on the question of whether the human macular pigments (MPs), lutein (L) and zeaxanthin (Z), protect the retina from age-related changes leading to retinal degeneration.1–3 For example, in vitro studies4,5 have shown that L and Z are potent lipid antioxidants and, as such, would be expected to reduce oxidative insult over time. It also appears clear that the placement of the pigments in the inner retinal layers is optimal for screening the vulnerable foveal cones. The close match between the absorption spectrum of MP and the blue-light hazard function (e.g., as shown in Zimmer and Hammond)6 argues well that the pigments protect the outer retinal layers and retinal pigment epithelium through passive filtration of potentially actinic light.
The more difficult question, of course, is whether the magnitude of MP's protective effects are significant enough over time to meaningfully reduce the risk of acquired retinal disease. Such a question can really only be definitively answered by longitudinal study. This is particularly difficult, however, when considering that degenerative diseases like age-related macular degeneration (AMD) reflect the accumulation of damage accrued over many decades of life. This process likely begins at birth. Both the retina and retinal pigment epithelium age rapidly over the first few years of life due largely to the high oxidative stress associated with the increased metabolic needs of the maturing retina and a crystalline lens that transmits a much higher percentage of actinic short-wave light (for a recent review see Zimmer and Hammond).6 L and Z may provide important protection during this vulnerable period and may even be necessary for proper maturation of the retinal pigment epithelial (e.g., Leung et al.).7 Because it is unlikely that the mechanisms responsible for MP accumulation evolved to protect from a disease that manifests well past reproductive age, a role for protection earlier in life seems feasible. Ultimately, the ability to answer the question of whether MP retards the development of diseases like AMD based on short-term epidemiological study of older subjects is probably inadequate.
Such framing, however, only considers whether L and Z help to prevent retinal disease. A related and potentially more easily addressed question is whether L and Z could actually improve the symptoms of retinal disease (i.e., ameliorate the consequences). MP is, after all, a yellow filter concentrated in an area of the retina, the fovea, that is critical to visual processing. As such, a number of hypotheses have been advanced that MP might improve vision through purely optical mechanisms (for a review see Wooten and Hammond).8 Many symptoms of visual disease are obviously functional. Thus, improving vision through optical means could reduce symptoms and be an important palliative.
Richer et al.9 tested the effects of lutein supplementation on visual performance in a double-blind placebo controlled study of veterans (average age = 65 years) with early stage AMD. He found that the lutein-treated group showed functional improvements (e.g., Snellen acuity) that were consistent with their measured increases in MP density. Olmedilla et al.10 tested the visual effects of lutein supplementation on cataract patients using a double-blind placebo controlled design and also found improvements in visual acuity in only those patients supplemented with L. Similar to Richer et al., the cataract subjects in the Olmedilla et al. study also showed reductions in glare sensitivity after the 2-year L supplementation period. Given the careful experimental design, these studies provided strong evidence that the MP carotenoids could improve visual performance. Both studies, however, used stimuli that were not characterized well enough to determine whether the improvements in vision were due to optical or biological factors.
To disentangle whether the mechanism was optical or biological, Engles et al.11 and Stringham and Hammond12 tested young healthy subjects (reducing the confound of age and disease-state) using stimuli whose optical characteristics were well-defined. Engles et al. tested spatial vision (standard gap- and hyper-acuity) using stimuli illuminated by broadband white light (absorbed by MP) and yellow light (not absorbed by MP). These authors found that MP was not related to acuity when measured under these two conditions. Careful modeling of the effects of chromatic aberration on their stimulus conditions showed that it was unlikely that MP would improve acuity under their conditions. In contrast, Stringham and Hammond found that MP was strongly related to photostress recovery time and glare disability (r's ∼ 0.80) when using broadband light strongly filtered by MP but not when using narrowband light that was not filtered by MP. By varying the wavelength composition of their glare source, Stringham and Hammond showed that the mechanism responsible for their results was based on simple filtration. Of course, the amount of light filtered by MP can vary dramatically across subjects. For example, MP can be very dense [>1.3 optical density (OD) at 460 nm] or essentially transparent (near zero OD) short-wave filter. It is therefore perhaps not surprising that such a specific and variable filter would have equally specific and variable effects on visual performance. Wooten and Hammond,8 for instance, conducted a detailed analysis of how MP might be expected to improve visibility outdoors due to reducing the effects of veiling blue haze. They found that, theoretically, an individual with 1.0 log unit of MPOD could see about 30% farther through the atmosphere compared to someone with little or no MP.
If some of the effects of MP on vision are based purely on optical factors, then such effects should be perfectly predictable based on known characteristics of the visual system, the optics of the eye, and the physical characteristics of the stimulus. If, for instance, MP reduces photostress recovery time and improves glare disability due to a simple filtering mechanism, then increasing MP should lead directly to improvements on those measures. In this study, we tested those predictions.
Forty healthy subjects, 23 women and 17 men, participated (mean age = 23.9, SD = 5.1 years; range = 17 to 41). Thirty of the 40 subjects in our sample were white, seven were black, and three were Hispanic. All observers were color normal, and, based on selfreport, none had a history of visual pathology. The subjects were also asked (via questionnaire) about iris color. Based on their response, eight subjects had dark brown, 13 had brown, nine had blue, five had green, four had hazel, and one had gray irises. Only the right eye of each subject was measured Each subject participated in the three experiments detailed below. Subjects were recruited from the population (students and staff) at the University of Georgia. This study was approved by University of Georgia's Institutional Review Board, and the experimental procedures adhered to the tenets of the Declaration of Helsinki. We analyzed whether any personal characteristics (sex, age, diet, or iris color) were related to our results and found that they were not. These variables are therefore not included in the general presentation of our results.
Lutein and Zeaxanthin Supplementation
Forty subjects received the L + Z supplements. The supplement was a single-dose, 500-mg tablet containing 10 mg L and 2 mg Z. The L and Z in these tablets was a beadlet material manufactured by DSM Nutritional Products Inc. (DSM Lutein 5%; DSM OPTISHARP Zeaxanthin 5%). Subjects consumed one tablet daily for 6 months and were instructed to take the supplement with a meal that contained at least a small amount of fat. Subject compliance was checked by weekly phone calls and pill counts at laboratory visits. An independent laboratory (personal communication, Tufts University, Human Nutrition Research Center) determined that the time to dissolve one of the tablets in gastric acid was about 2 min. Because L & Z are lipid soluble, this relatively fast dissolving rate should have provided good bioavailability when taken with food containing a small amount of fat.
Subjects performed each experiment (MPOD measure, disability glare, and photostress recovery) at five time points: baseline, 1, 2, 4, and 6 months after starting supplementation.
Apparatus, Glare Experiments
The apparatus and procedure used for the glare experiments is described in detail elsewhere.12 Briefly, a two-channel standard Maxwellian-view optical system with a 1kW xenon-arc lamp source was used. For the veiling glare experiment we used the standard stimulus configuration, a test target surrounded by an annulus. One channel of the optical system produced the target stimulus, a 1 degree-diameter disk containing a 100% contrast grating stimulus. The spatial frequency of the grating was five cycles per degree. The luminance of the bars within the grating was 0.1 cd/m2 and appeared white (although the actual wavelength composition was broadband, it contained about 20% more mid-to-long wave energy than the surround). The second channel of the optical system produced a xenon-white annulus concentric with the target stimulus, with 11 degree inner and 12 degree outer diameters. The size of the annulus was chosen to spatially obviate absorption by MP, which decreases to optically undetectable levels at about 5 to 7 degree retinal eccentricity.13
For the photostress recovery experiment, the same 1 degree target stimulus described above was used. In the second channel, a 5 degree-diameter, xenon-white disk served as the photostress stimulus. The corneal irradiance of the photostress stimulus was 2.5 μW/cm2, which was determined, via the Westheimer14 method, to provide 5.5 log Trolands of retinal illuminance. The photostress stimulus was presented for 5 s using a Vincent Associates (Rochester, NY) shutter. Two conditions were tested: central and eccentric viewing. For the central viewing condition, subjects fixated the center of the target stimulus prior to photostress. For the eccentric viewing condition, a 10` black fixation point was placed 10 degree in the temporal retina. All photometric calibrations were performed using a PR-650 spectral radiometer (PhotoResearch, Inc., Chatsworth, CA). Wedge and neutral density radiometric calibrations were performed using a Graseby Optronics (Orlando, FL) radiometer. The same radiometer was used before every experimental session to ensure that the total light output of the optical system remained constant.
Procedure, Glare Experiments
Before testing, subjects were aligned to the optical system. Careful adjustments were made to ensure that the arc image (1.5 mm diameter) was in focus and in the plane of the subject's pupil. For the veiling glare experiment, subjects first viewed the grating stimulus, and then the annulus was presented. Before each trial, the annulus was set at a level well below that which would cause the target stimulus to be veiled. The subject was then instructed to adjust, via a neutral-density wedge, the intensity of the annulus until the target stimulus was no longer visible. Often, subjects went beyond the threshold of visibility, and then adjusted the wedge to decrease the intensity of the annulus in order to pinpoint their threshold. Five thresholds were determined, and subjects were instructed to carefully maintain their criterion threshold across trials. The duration of the veiling glare experiment (including the alignment procedure) was approximately 20 min.
For the photostress recovery experiment, the same alignment procedure was used. The subject was first instructed to view the grating stimulus. After approximately 30 s the subject was presented with the photostressor for 5 s. The photostressor was intense (2.5 μW/cm2), and in order to control for reduced photopigment bleaching due to eye blinking or closure, subjects were instructed to keep their eyes open during the 5 s exposure. The subjects' eyes were monitored during the exposure (via an infrared camera and monitor), so if the experimenter observed that the test beam was occluded by blinking or eye closure, the trial was repeated. Subjects were instructed to indicate when they could first perceive the grating. The time necessary to recover central visibility of the grating stimulus was measured with a stopwatch. After photostress recovery was achieved, a 2-min waiting period, before the next stimulus presentation, was observed. In order to compare foveal results with data from a retinal location containing no MP, this procedure was also performed for a 7 degree parafoveal location.
Measurement of MPOD
The method and procedure used to measure MPOD was detailed in our article describing the baseline data from this project.12 In brief, MPOD was measured using a macular densitometer that was first described by Wooten et al.15 (Macular Metrics, Rehoboth, MA). This device presents stimuli in free view and uses heterochromatic flicker photometry to derive measures of MPOD from retinal sensitivity values. A 458 nm test stimulus, the radiance of which is adjusted by the subject, was presented in square-wave counterphase with a 570 nm reference field (7.4 cd/m2). The flicker rate was optimized for each subject in order to achieve a relatively narrow null-flicker (i.e., equiluminance between the 458 and 570 nm stimuli) zone of about 0.10 log units. The test was presented near the center of a 470 nm circular background (4.6 cd/m2). Spatial distribution profiles of MPOD were generated, with points at 15′, 30′, 1, 3, 5, and 7 degree and 10 degree eccentricity along the horizontal meridian of the temporal retina. The 15′ and 30′ stimuli were measured using circular targets having those radii. The 1 degree measure was made using a 2 degree-diameter, centrally-fixated annulus (2 degree corresponds to the diameter between two points located in the center of the annular ring). The 3, 5, 7, and 10 degree measures were made using a tiny (5 min) red fixation point that was located the appropriate angular distance to the right of the center of the test stimulus. Subjects fixated this point when making their parafoveal settings. The diameter of the test stimulus for the parafoveal measures was 1 degree (for 3 and 5 degree eccentricity), and 2 degree (for 7 and 10 degree eccentricity). Subtracting the foveal radiance measures (where MP generally is dense) from the 10 degree parafoveal radiance measures (where MP is minimal) yields an OD measure of MP. Subjects made five null flicker settings for each locus. The heterochromatic flicker photometry technique has been validated by measuring several points along the entire spectral absorption band of MP, and it has been determined that this produces a spectrum that corresponds closely to the extinction spectrum of the MPs measured ex vivo.16,17
Measuring a number of spatial points was useful, in that it helped to confirm that our individual MP data were accurate (e.g., a first-degree exponential explained over 90% of the variance in all of the measured profiles). The results for the full profile are provided in Table 1. The MP measure that explained the most variance in our glare measures, however, was the standard value at 30′ eccentricity. This measure was therefore used in reporting the majority of our results. Unless otherwise specified, we use the convention of referring to MPOD at 30′ eccentricity as simply MPOD.
Data from the baseline measures in this study were presented in an earlier paper.12 Briefly, strong, significant correlations (in the direction of benefit) were found between MPOD and glare disability (r = 0.76) and photostress recovery times (r = 0.80). Because we were interested primarily in the effect of changes in MPOD on these glare measures, analyses in the present study are based on changes from baseline at the different time points.
L and Z Supplementation
Baseline MPOD varied widely between subjects (0.08 to 1.04 OD). Nonetheless, all but two of the subjects receiving the supplement responded with increases in MP density. Table 1 provides a summary of the supplementation results for all the subjects at all the measured eccentricities. Table 2 provides the supplementation results for the two subjects that did not show MP increases during the supplementation period. As shown in Table 1, most of our subjects showed increases in MP that generally began past the 1-month time point. At the 2-month time point, average MPOD had increased from 0.41 at baseline to 0.46. MPOD continued increasing at 4-month (p = 0.032) and 6-month (p = 0.003), with increases from baseline of 0.10 and 0.16, respectively.
To examine possible effects of baseline MPOD on response to the supplements, we divided our 40 subjects receiving the supplement into three groups: low (MPOD 30′ = 0 ≤ 0.25), n = 13, middle (MPOD 30′ = 0.26 ≤ 0.50), n = 15, and high (MPOD 30′ > 0.50), n = 12. Each of the three MP-level subgroups appeared to respond to the supplement; however, the “middle” group increased significantly more than the “low” and “high” groups (F = 11.52, p = 0.0001). Over the 6-month supplementation period, the overall average increase by the middle group was 0.21 OD, whereas the low group increased 0.14 OD, and the high group 0.11 OD. This is shown graphically in Fig. 1.
As previously noted in Hammond et al.,18 MP increases in response to supplementation were generally lower with increasing distance from the center of the fovea. For example, at the 6-month time point, the group mean MPOD increase was 0.19 for the 15′ eccentricity, 0.16 at 30′, 0.10 at 1 degree, 0.07 at 3 degree, 0.05 at 5 degree, and 0.03 at 7 degree.
MPOD and Veiling Glare
For both the 4- and 6-month time points, glare disability was significantly less than baseline (t = 8.7, p = 0.002; t = 15.7, p < 0.0001, respectively). This change was found to be significantly related to increases in MPOD at the 6-month time point (r = 0.59, p < 0.0001). In other words, subjects could tolerate a significantly greater amount of veiling glare (a more intense annulus), and still detect the central square-wave grating. A graphical representation of this relationship can be seen in Fig. 2. Increases in veiling glare thresholds appeared to be commensurate with increases in MP density (see Fig. 2). For example, the two subjects that did not show increases in MP density also did not show significant changes in their glare disability over the study period (see Table 2).
MPOD and Photostress Recovery
At the 4-month time point, foveal photostress recovery time was found to be significantly shorter than the baseline measure (t = 3.85, p = 0.05). This change increased dramatically by 6 months (t = 12.06, p = 0.0003). Similar to our results for veiling glare, photostress recovery time appears to be a direct function of increases in MPOD. Fig. 3, for example, shows changes in photostress recovery times as a function of changes in MPOD at the 6-month time point; the relationship between the two variables was strongly significant (r = −0.66, p < 0.0001). As with veiling glare, photostress recovery times exhibited a strong, linear relationship with changes in MPOD over the course of the supplementation period (r = −0.988, p = 0.012; Fig. 3). We also measured parafoveal photostress recovery times which did not change significantly with L and Z supplementation (about 1 s over the 6-month period). The slight trend in decreasing parafoveal photostress recovery times over the study period may be attributable to the small increase in MPOD found for parafoveal (i.e., 5 and 7 degree) sites. Similarly, Wenzel et al.19 showed that parafoveal sensitivity to short-wave light was reduced at the 7 degree locus after 120 days of L supplementation, presumably due to increased MPOD. As with our glare disability measures, photostress recovery was stable for the two subjects whose MP density did not change.
The main result in our present study was that increases in MP were directly related to improvements in glare disability and photostress recovery times. Over the 6-month study period, the average MP density at 30` of subjects receiving supplements increased by 39% (0.16 log unit increase). Over this same time period, supplemented subjects could tolerate 58% (0.20 log unit increase) more intense glaring light before losing their ability to detect a central grating target. Similarly, the subjects receiving supplements had, on average, 14% (about 5 s) faster photostress recovery times. Note that these within-subject effects are very nearly what would be predicted based on the across-subject effects originally reported in Stringham and Hammond,12 albeit slightly stronger. In that study, based on the strong cross-sectional relations between glare disability, photostress recovery, and MP density (see Figs. 1 and 2 in Stringham and Hammond), a change in MP of 0.16 OD was related to tolerating an annulus that was about 0.16 log units more intense. Similarly, a change in MP of about 0.16 OD was related to about 3 s faster photostress recovery time. There is good agreement, therefore, between the within-subject data (from the present study) and the across-subject data (from Stringham & Hammond).
A corollary result of this study worth noting is the very strong retinal response of our subjects to our 12 mg/d L and Z supplement. Only 2 out of 40 subjects (5%) showed no substantial MP increases. As can be seen in Fig. 3, after the first month, this change was relatively linear. This increase is about twice as strong as that generally seen in analogous studies. For example, Hammond et al.18 supplemented 11 subjects with spinach and corn (containing a similar LZ content) and found an MP increase of about 12% at 3 months (our increase was about 18%). Berendschot et al.20 supplemented eight subjects with 10 mg of L and found an MP increase of about 10% at 3 months. Larger studies (e.g., Schalch et al., n = 108)21 have found increases of about 15% over 12 months with similar supplementation. In general, however, there have been large inconsistencies across studies in how subjects respond to L and Z supplementation. Very few studies have examined what factors are related to the large individual differences that are typically seen. When combined with methodological limitations (e.g., most studies depend on a peripheral reference that could also change with supplementation; Schalch et al.),21 it makes interpretation of differences across studies purely speculative. It is clear, however, that how subjects respond to L and Z supplementation, which is widely available, is an important issue that needs to be carefully and systematically studied.
When our interventional data are considered with the strong cross-sectional correlations, we feel that our results provide strong evidence that there is a causal connection between MP, glare disability, and photostress recovery, when measured using these stimulus conditions. This latter point raises an important, and common, caveat: like all studies, our results are specific to our experiment. For example, the magnitude of the effect we found for our photostress measurements is strongly tied to both the adaptive state of the subject (e.g., percent of photopigment that can be bleached) and our actual bleaching conditions. For example, when using intense lights that approach visual saturation, for short-duration exposures (say, <1 s), MP would be a very effective filter, because a relatively large proportion of the light would be absorbed. However, as the bleaching light exposure time increases, the absolute amount of light that reaches the photoreceptors would increase, and, despite the existence of MP, the visual system would approach saturation. Therefore, for very intense lights, the effect of MP level would be gradually minimized with increased exposure duration. Although our stimulus intensity (5.5 log td) was below visual system saturation levels (∼7.6 log td), our 5-s exposure time was relatively long, and probably reduced the magnitude of the photostress-MP relation. We also carefully chose our stimulus conditions in order to minimize possible confounds. For example, confounding effects due to pupil size were removed because the glare stimuli were delivered in Maxwellian-view. We also preselected our sample to be young healthy observers with a wide range of MP density in order to minimize aging or disease confounds. Whether our results generalize to an older or diseased sample (the group most negatively affected by glare) is therefore not clear.
Given the specificity of the results, what can be concluded? Certainly anytime MP strongly absorbs a bleaching light, it will reduce the photostress of the receptors it screens by preventing photopigment isomerization. This is a situation that is common in everyday life because many broadband light sources contain significant proportions of short-wave light. Similarly, any time a surround or background has more short-wave energy than a centrally-viewed target, MP will absorb more of the surround than the target. This effect can be produced with any yellow filter with similar properties.22 The selective reduction of a background relative to a target will increase the contrast between the two. In regular lighting, this should lead to contrast enhancement and better visibility of a target relative to its background or surround.8 As noted by Wooten and Hammond,8 this is a situation that is probably also common in everyday vision. In bright light, absorption of veiling luminance by MP should decrease disability due to a glaring peripheral light source. Ultimately, photoreceptors simply respond only to quanta and MP is simply a yellow filter. Any influences that MP has on visual performance must therefore follow the same predictions as any filter with similar characteristics.
It seems likely then that our results are both large enough and sufficiently general to be meaningful in real life. This conclusion, however, should be tested. Ecological validity is always a concern when considering results from laboratory studies, and we caution against premature generalization of our results. Nonetheless, acute effects of MP on improving vision would not be surprising. As noted previously, it is doubtful that the mechanisms responsible for accumulating MP evolved to protect the retina from damage that would ultimately manifest past reproductive age. Rather, MP would be likely to confer a selective advantage earlier in life. Effects of MP on glare would clearly manifest early as our data on younger subjects show.
If MP does improve some aspects of vision as a direct function of increases in MP, then supplementing L and Z could, indeed, be palliative. Because many elderly and AMD patients suffer disability due to glare, further study of whether L and Z supplementation could reduce these problems is warranted.
Billy R. Hammond
Vision Science Laboratory
University of Georgia
Athens, GA 30602-3013
1. Snodderly DM. Evidence for protection against age-related macular degeneration by carotenoids and antioxidant vitamins. Am J Clin Nutr 1995;62:1448S–1461S.
2. Beatty S, Boulton M, Henson D, Koh HH, Murray IJ. Macular pigment
and age related macular degeneration. Br J Ophthalmol 1999;83:867–77.
3. Whitehead AJ, Mares JA, Danis RP. Macular pigment
: a review of current knowledge. Arch Ophthalmol 2006;124:1038–45.
4. Stahl W, Sies H. Antioxidant effects of carotenoids: implication in photoprotection in humans. In: Cadenas E, Packer L, eds. Handbook of Antioxidants, 2nd ed. New York: Marcel Dekker; 2002:223–33.
5. Sujak A, Gabrielska J, Grudzinski W, Borc R, Mazurek P, Gruszecki WI. Lutein
and zeaxanthin as protectors of lipid membranes against oxidative damage: the structural aspects. Arch Biochem Biophys 1999;371:301–7.
6. Zimmer JP, Hammond BR. Possible influences of lutein
and zeaxanthin on the developing retina. Clin Ophthalmol 2007;1:181–9.
7. Leung IY, Sandstrom MM, Zucker CL, Neuringer M, Snodderly DM. Nutritional manipulation of primate retinas. II. Effects of age, n-3 fatty acids, lutein
, and zeaxanthin on retinal pigment epithelium. Invest Ophthalmol Vis Sci 2004;45:3244–56.
8. Wooten BR, Hammond BR. Macular pigment
: influences on visual acuity and visibility. Prog Retin Eye Res 2002;21:225–40.
9. Richer S, Stiles W, Statkute L, Pulido J, Frankowski J, Rudy D, Pei K, Tsipursky M, Nyland J. Double-masked, placebo-controlled, randomized trial of lutein
and antioxidant supplementation in the intervention of atrophic age-related macular degeneration: the Veterans LAST study (Lutein
Antioxidant Supplementation Trial). Optometry 2004;75:216–30.
10. Olmedilla B, Granado F, Blanco I, Vaquero M. Lutein
, but not alpha-tocopherol, supplementation improves visual function in patients with age-related cataracts: a 2-y double-blind, placebo-controlled pilot study. Nutrition 2003;19:21–4.
11. Engles M, Wooten B, Hammond B. Macular pigment
: a test of the acuity hypothesis. Invest Ophthalmol Vis Sci 2007;48:2922–31.
12. Stringham JM, Hammond BR Jr. The glare hypothesis of macular pigment
function. Optom Vis Sci 2007;84:859–64.
13. Trieschmann M, van Kuijk FJ, Alexander R, Hermans P, Luthert P, Bird AC, Pauleikhoff D. Macular pigment
in the human retina: histological evaluation of localization and distribution. Eye 2007. (epub March 30, 2007).
14. Westheimer G. The Maxwellian view. Vision Res 1966;6:669–82.
15. Wooten BR, Hammond BR Jr, Land RI, Snodderly DM. A practical method for measuring macular pigment
optical density. Invest Ophthalmol Vis Sci 1999;40:2481–9.
16. Hammond BR Jr, Wooten BR, Smollon B. Assessment of the validity of in vivo methods of measuring human macular pigment
optical density. Optom Vis Sci 2005;82:387–404.
17. Wooten BR, Hammond BR Jr. Spectral absorbance and spatial distribution of macular pigment
using heterochromatic flicker photometry. Optom Vis Sci 2005;82:378–86.
18. Hammond BR Jr, Johnson EJ, Russell RM, Krinsky NI, Yeum KJ, Edwards RB, Snodderly DM. Dietary modification of human macular pigment
density. Invest Ophthalmol Vis Sci 1997;38:1795–801.
19. Wenzel AJ, Sheehan JP, Gerweck C, Stringham JM, Fuld K, Curran-Celentano J. Macular pigment
optical density at four retinal loci during 120 days of lutein
supplementation. Ophthalmic Physiol Opt 2007;27:329–35.
20. Berendschot TT, Goldbohm RA, Klopping WA, van de Kraats J, van Norel J, van Norren D. Influence of lutein
supplementation on macular pigment
, assessed with two objective techniques. Invest Ophthalmol Vis Sci 2000;41:3322–6.
21. Schalch W, Cohn W, Barker FM, Kopcke W, Mellerio J, Bird AC, Robson AG, Fitzke FF, van Kuijk FJ. Xanthophyll accumulation in the human retina during supplementation with lutein
or zeaxanthin—the LUXEA (LUtein
Xanthophyll Eye Accumulation) study. Arch Biochem Biophys 2007;458:128–35.
22. Wolffsohn JS, Cochrane AL, Khoo H, Yoshimitsu Y, Wu S. Contrast is enhanced by yellow lenses because of selective reduction of short-wavelength light. Optom Vis Sci 2000;77:73–81.