Received: 4 February 2002 Revised: 20 March 2002Accepted: 28 May 2002Published online: 24 July 2002© Springer-Verlag 2002

Proprietary interest: none

Abstract Background: There isgrowing evidence that the develop-ment of age-related macular degen-eration (AMD) is related to oxidativedamage. This makes macular pig-ment (MP) an interesting target forresearch, as antioxidative quality isrelated to MP. As studies haveshown that MP density (MPD) canbe increased by dietary supplementa-tion, MP may also offer a therapeuticor preventive approach to AMD.Methods: Using scanning laser tech-niques, we quantified MPD in heal-thy subjects and patients with dryAMD. We introduce a modifiedscanning laser ophthalmoscope(SLO) which includes several advan-tages for the use in the clinical rou-tine compared to the standard SLO.Results: We examined 10 healthy

subjects without ocular pathologyand 10 patients with dry AMD (geo-graphic atrophy). Eyes of patientswith AMD had a significantly lowerMPD (mean value 0.094±0.022 den-sity units, DU) than healthy eyes(mean value 0.184±0.023 DU).Conclusions: It is possible to quanti-fy MPD using a scanning laser oph-thalmoscope in a clinical setting.The modified SLO offers several im-provements with regard to the spe-cial requirements of the clinical situ-ation. Using the new instrument, wewere able to detect differences inMPD between patients with dryAMD and young healthy subjects.Further studies are needed to evalu-ate the reliability and reproducibilityof the method.

Graefe’s Arch Clin Exp Ophthalmol(2002) 240:666–671

DOI 10.1007/s00417-002-0515-6

C L I N I C A L I N V E S T I G AT I O N

Henrike WüstemeyerCornelia JahnAndrei NestlerThomas BarthSebastian Wolf

A new instrument for the quantificationof macular pigment density: first results in patients with AMD and healthy subjects

Introduction

Age-related macular degeneration (AMD) is the leadingcause of visual loss in the industrialized world for peo-ple over 65 years of age [7, 8]. Although the exactpathophysiology of AMD remains poorly understood,there is growing evidence that the development of age-related maculopathy (ARM) is related to oxidative dam-age [1, 9, 11, 27]. As many antioxidative properties areattributed to the macular pigment (MP), it has been in-vestigated in respect to its role in the pathological con-cept of AMD [19, 20, 21, 23]. The first investigationsinto the function of MP were performed by MaxSchultze, who concluded in 1866 that there is a func-tional connection between the “yellow spot” in the reti-na and the absorption of blue light. The macular pig-

ment consists of the two hydroxy carotenoids lutein (L)and zeaxanthin (Z). Human MP is detectable throughoutthe retina but the highest concentrations are found in thefovea [1, 17].

The role of MP includes a high capacity to absorbshort-wavelength blue light. This can be considered apassive protection mode [19]. The peak of the MP absor-bance spectrum is at 460 nm and works as a broad-bandfilter for the macula. Two advantages are achieved: (1)the macula’s optical accuracy is improved [10, 16] and(2) the damaging photo-oxidative influence on the neu-rosensory retina is reduced. In the photoreceptor outersegments, the antioxidant effect of L and Z is the essen-tial mechanism [1, 22]. The antioxidant properties enablethe carotenoids to neutralize free radicals. This can bereferred to as an active process.

Presented in part at the annual meeting of the Deutsche Ophthalmologische Gesellschaft (DOG), Berlin, 2001

H. Wüstemeyer · C. Jahn · A. NestlerT. Barth · S. Wolf (✉ )Department of Ophthalmology, University of Leipzig, Liebigstrasse 10–14,04103 Leipzig, Germanye-mail: wolfs@medizin.uni-leipzig.deFax: +49-341-9721509

667

Several previous studies have demonstrated that mea-surement of MP density (MPD) by reflectometry tech-niques using a scanning laser ophthalmoscope [3, 13]can be used to quantify MPD and its distribution. How-ever, currently no objective method for measuring MPD

is available for measurements in clinical routine. Thisprompted us to develop a method that allows fast and re-liable MPD measurement. The purpose of this report isto demonstrate that MPD can be quantified with a modi-fied confocal scanning laser ophthalmoscope in a clinicalsetting. Therefore, we examined a group of healthy sub-jects and a group of patients in whom we expected lowMPD values.

Materials and methods

For quantification of MPD we used a modified confocal scanninglaser ophthalmoscope (SLO). This instrument was developed in col-laboration with Heidelberg Engineering (Heidelberg, Germany). Itis optimized to record reflectance images at wavelengths of 488 nmand 514 nm. It allows a fast switch between the different wave-lengths and digital recording of the images, thus minimizing distur-bances caused by movements of the patient’s eye. We obtain high-resolution images at 488 nm and 514 nm wavelength with an argon-laser. Since the absorption of macular pigment at 488 nm wave-length is very high and at 514 nm wavelength close to zero (Fig. 1),

Fig. 1 Absorbance spectrum of macular pigment. Peak absor-bance at 460 nm. Image recording MPD quantification at 488 nmand 514 nm wavelengths

Fig. 2 A, B Images taken withthe modified SLO from a heal-thy eye at wavelengths of488 nm (B) and 514 nm (A). C Functional image or densitymap following digital process-ing

668

Fig. 3 A, B Images taken froman eye with dry AMD at wave-lengths of 488 nm (B) and514 nm (A). C Functional im-age or density map followingdigital processing. D Fundusphotography of the macular re-gion showing drusen and asmall area of geographic atro-phy in the center

we can determine MPD by comparing foveal and parafoveal reflec-tance at 488 nm and 514 nm. The MPD is given by Eq. 1:

where Cλ is a constant depending on the absorption coefficients ofMP and the Ref values are the measured reflectances.

For easy assessment of MPD we developed a program to createdensity maps. In a first step this program corrects two reflectionimages for eye movements. Thereafter, a density map or function-al image is created by digital subtraction of the log reflectance im-ages (Figs. 2, 3). Since the difference in absorption by pigmentsother than MP (e.g., retinal and choroidal blood, visual pigments,RPE, and choroidal melanin) between the fovea and 14° outsidethe fovea can be neglected, MPD can be determined by comparingfoveal and parafoveal reflectance.

We recruited ten healthy subjects (five males, five females)aged from 16 to 43 years (mean 27±8 years). Subjects were ingood general health and retinal pathology was present neither inthe examined nor in the fellow eye.

Ten patients (three males, seven females) with dry AMD (geo-graphic atrophy present) in the study eye were recruited. Their ageranged from 60 to 77 years (mean 69±6 years). Five patients hadgeographic atrophy bilaterally, two had exudative AMD in the fel-low eye, and three had ARM in the fellow eye. Diagnosis of ad-vanced dry AMD was made according to the criteria published by

the International ARM Epidemiologic Study Group [4]. Geo-graphic atrophy was affecting but not entirely covering the foveain eight of ten cases (Fig. 3D). In two eyes, geographic atrophywas present in the macular area, sparing the fovea. The study fol-lowed the tenets of the Declaration of Helsinki and was approvedby the local ethics committee.

All subjects underwent comprehensive ocular examination.The pupil of the study eye was dilated with eye drops containing0.5% tropicamide and 2.5% phenylephrine. Macular pigment mea-surements were obtained with the modified SLO. The patientswere positioned in front of the tabletop and instructed to lookstraight and steady. We then obtained 20° reflectance images at488 nm and 514 nm wavelengths of the posterior pole. Densitymaps were then processed and maximum MPD was calculated fora 2° diameter circle centered on the fovea.

In the 10 healthy subjects we conducted two separate reflec-tance imaging sessions at wavelengths of 488 nm and 514 nmwithin 30 min in order to obtain reproducibility data for MPDmeasurements. For the assessment of the reproducibility of themethod we calculated the coefficient of variation cvintra from Eq. 2for test-retest variation [26]:

lengths of light falls exponentially with decreasingwavelength [14]. In primates, carotenoids cannot be syn-thesized de novo and have to be gained from foodstuffs.Studies showed a positive relation between dietary in-take of xanthophylls, blood serum levels and MPD [15,19] in humans. The variability of the MP initiated by nu-tritional supplementation, however, makes it an interest-ing target for future intervention with the aim of prevent-ing or altering the course of AMD. Ex vivo techniques,e.g., high-performance liquid chromatography (HPLC)and microdensitometry, have been used to initiate the re-search on MPD [5, 25]. Today, these are mainly used tovalidate results from in vivo techniques. Bone et al. usedreversed-phase HPLC to quantify retinal L and Z in rela-tion to an internal standard in human donor eyes withand without AMD [6]. There was a significantly higherconcentration of L and Z in eyes without AMD than inthose with AMD [6]. The results imply an inverse asso-ciation between the risk of AMD and MPD, althoughthere is no proof of a causal association. Microdensitom-etry identifies the optical density profile of MP andshows the spatial distribution of L and Z in the retina[24]. Heterochromatic flicker photometry is a widelyused in vivo technique on a psychophysical basis [2].The disadvantage of this method is the fact, that patientsoften need a training program to fulfill the requirementsadequately. This is a barrier for the use in the clinicalroutine. Elsner et al. compared the distribution of conephotopigment and macular pigment using a standardSLO [13]. Other study groups used scanning laser oph-thalmoscopy for the measurement of MPD as well [3].The results were compared to those obtained by thefoveal fundus reflectance measured with a fundus re-flectometer. The results show a high reliability of theSLO measurements with fewer disturbing noises in themeasurements compared to the spectral analysis. Theyfurther showed a substantial and significant increase inMPD in the course of dietary supplementation with caro-tenoids and also a significant correlation between plasmaL levels and MPD. A new approach to in vivo measure-ments has been proposed by Delori et al. [12]. The detec-tion of the autofluorescence of lipofuscin being activatedby two wavelengths, one well absorbed by MP and oneminimally absorbed, enables the examiner to obtain ac-curate single-pass measurements of the MPD. The re-sults correlate highly with the results taken from an alter-native psychophysical method (heterochromatic flickerphotometry) and another optic method (fundus reflecto-metry). The results indicate that the method leads to re-producible data in repeated examinations, at both shortand long intervals.

Previous studies using a SLO for MP measurementswere performed using specially designed research instru-ments [3, 13]. Currently, none of these instruments arecommercially available. The only standard SLO that al-lows recording of reflectance images at 488 nm and

669

where are the values of the first series, the values from thesecond series, and n the number of cases.

Results

Recording of reflectance images at 488 nm and 514 nmwas easy in all subjects. Examination time varied be-tween 3 min and 5 min. Calculation of MPD maps fromthe reflectance images was possible in all cases and wasthe basis for further evaluation.

The mean relative MP in a 2° diameter test field, cen-tered on the fovea, in healthy subjects was 0.184±0.023DU (density units, ± standard deviation, SD). The lowestMPD in this group was 0.152 DU, measured in a womanaged 43 years. The highest MPD in healthy eyes was0.218 DU, measured in a man aged 24 years. The intra-individual variation was calculated and showed a valueof 6.2% for the MPD in the 10 healthy subjects.

In subjects with dry AMD the mean MPD was0.094±0.022 DU. In this group the lowest MPD was0.047 DU (male, 71 years of age), the highest MPD was0.123 DU (female, 77 years of age), both measured inpatients with bilateral geographic atrophy. MPD in thefovea was significantly lower in the patients with dryAMD than in healthy subjects (P<0.001). The results foreach subject are demonstrated in Fig. 4.

Discussion

The role of MP is based on its antioxidative activity anda high capacity to absorb short-wavelength blue light[18, 19]. Since there is growing evidence that oxidativedamage contributes to the development of ARM [1, 9,11, 27] various efforts have been undertaken to quantifyMPD. Studies have examined the effect of cumulativephotochemical damage and it has been shown that thethreshold for retinal damage induced by different wave-

Fig. 4 Results of MPD in healthy eyes and AMD. Single resultsin the two groups. There is no paired association between thegroups. The mean values are indicated by the emphasized lines

670

514 nm is the Rodenstock. However, this instrument hasseveral limitations: (1) it takes a long time (e.g., 30 s) toswitch between the different wavelengths, resulting inartifacts due to eye movements and misalignment; (2)the linearity of the video amplifier cannot be achieved;(3) the signal to noise ratio is poor; and (4) the quality ofthe video signal is poor (e.g., non-standard PAL orNTSC). These limitations prompted us to develope anew instrument for MP measurement. This instrument isbased on the confocal SLO (HRA, Heidelberg Engineer-ing). Compared with previous measurements with a stan-dard SLO (Rodenstock) the new instrument has severaladvantages: (1) The instrument allows a very fast switchbetween the different wavelengths (488 nm and 514 nm);(2) the images are directly digitized, (3) the signal tonoise ratio is improved, and (4) the linearity of the videoamplifier is assured.

Our method to measure MPD noninvasively with amodified confocal SLO utilizes the differential absor-bance of reflected light at 488 nm (well absorbed by MP)and 514 nm (minimally absorbed by MP) and thus obtaina double-pass measure of the optical density of the MP.As with all methods to measure MP, we estimate theMPD from comparison of foveal and parafoveal mea-surements, minimizing the influence of spectral charac-teristics of the underlying tissues, of the ocular media,and of the instrument. The intraindividual coefficient ofvariation of 6.2% demonstrates a good reproducibility ofMPD measurements with our method. This variationshows that our method is at least as stable as the methodused by Delori et al., who found 9–22% test-retest differ-

ences [12]. However, our method for the calculation oftest-retest reproducibility is slightly different form thatused by Delori et al. Using the same calculation our test-retest differences would be 3.1%.

As expected, our measurements show that healthysubjects have a statistically significant higher MPD thanthe studied subgroup of patients with dry AMD. In com-parison with the results Delori et al. obtained with the re-flectance method, our mean MPD for healthy subjects islower, but ranges within a comparable magnitude. Deloriand colleagues found a MPD of 0.23±0.07 DU in a pop-ulation of 159 subjects without ocular pathology aged16–80 years (mean age 52±17 years) [12]. The compari-son between the absolute numbers of Delori et al. andour results presented here should be valid. However, anaccurate comparison is possible only after examining thesame patient with both tools. Points of differentiationcan be the definition of the fovea, resulting in a slightlydifferent positioning of the test field, and subtle differ-ences in the alignment of the instrument.

Our findings demonstrate that measurement of MPdensity with the new instrument leads to results consis-tent with previous studies. The modified confocal SLOhas shown to be a useful device for quantifying MPDand can in future be used in the clinical routine. Notraining sessions to make patients familiar with the tech-nique’s requirements are necessary. In a few minutes weare able to obtain high-quality images to quantify MPD.This will allow us to perform large-scale prospectivestudies in patients with different stages of ARM to eluci-date the role of MPD in the pathogenesis of AMD.

References

1. Beatty S, Boulton M, Henson D,Koh HH, Murray IJ (1999) Macularpigment and age related macular de-generation. Br J Ophthalmol:867–877

2. Beatty S, Murray IJ, Henson DB, Carden D, Koh H, Boulton ME (2001)Macular pigment and risk for age-relat-ed macular degeneration in subjectsfrom a Northern European population.Invest Ophthalmol Vis Sci 42:439–446

3. Berendschot TT, Goldbohm RA, Klopping WA, van de Kraats J, vanNorel J, van Norren D (2000) Influenceof lutein supplementation on macularpigment, assessed with two objectivetechniques. Invest Ophthalmol Vis Sci41:3322–3326

4. Bird AC, Bressler NM, Bressler SB,Chisholm IH, Coscas G, Davis MD, deJong PTVM, Klaver CCW, Klein BEK,Klein R, Mitchell P, Sarks JP,Sarks SH, Soubrane G, Taylor HR,Vingerling JR (1995) An internationalclassification and grading system forage-related maculopathy and age-relat-ed macular degeneration. SurvOphthalmol 39:367–374

5. Bone RA, Landrum JT, Fernandez L,Tarsis SL (1988) Analysis of the macu-lar pigment by HPLC: retinal distributage study. Invest Ophthalmol Vis Sci29:843–849

6. Bone RA, Landrum JT, Mayne ST, Gomez CM, Tibor SE, Twaroska EE(2001) Macular pigment in donor eyeswith and without AMD: a case-controlstudy. Invest Ophthalmol Vis Sci42:235–240

7. Bressler NM, Bressler SB (1995) Pre-ventative Ophthalmology Age-relatedmacular degeneration. Ophthalmology102:1206–1211

8. Bressler NM, Bressler SB, West SK,Fine SL, Taylor HR (1989) The grad-ing and prevalence of macular degener-ation in Chesapeake Bay watermen.Arch Ophthalmol 107:847–852

9. Challa JK, Gillies MC, Penfold PL(1998) Exsudative degeneration an dit-ravitreal triamcinolone: 18 month fol-low up. Aust N Z J Ophthalmol26:277–281

10. Dagnelie G, Zorge IS, McDonald TM(2000) Lutein improves visual functionin some patients with retina degenera-tion: a pilot study via the internet. Op-tometry 71:147–164

11. De La Paz MA, Anderson RE (1992)Lipid peroxidation in rod outer seg-ments. Role of hydroxyl and lipid hy-droperoxides. Invest Ophthalmol VisSci 33:2091–2096

671

12. Delori FC, Goger DG, Hammond BR,Snodderly DM, Burns SA (2001) Mac-ular pigment density measured by auto-fluorescence spectrometry: comparisonwith reflectometry and heterochromaticflicker photometry. J Opt Soc Am18:1212–1230

13. Elsner AE, Burns SA, Beausencourt E,Weiter JJ (1998) Foveal cone photopig-ment distribution: small alterations as-sociated with macular pigment distri-bution. Invest Ophthalmol Vis Sci39:2394–2404

14. Ham WTJ, Mueller HA, Sliney DH(1976) Retinal sensitivity to damagefrom short wavelength light. Nature260:153–155

15. Hammond BR, Wooten BR,Snodderly DM (1996) Cigarette smok-ing and retinal carotenoids: implica-tions for age-related macular degenera-tion. Vision Res 36:3003–3009

16. Hammond BR, Jr. Wooten BR,Snodderly DM (1997) Density of thehuman crystalline lens is related to themacular pigment carotenoids, luteinand zeaxanthin. Optom Vis Sci74:499–504

17. Handelmann GJ, Dratz EA, Reay CC,van Kuijk JG (1988) Carotenoids in thehuman macula and whole retina. InvestOphthalmol Vis Sci 29:850–855

18. Khachik F, Bernstein PS, Garland DL(1997) Identification of lutein and zea-xanthin oxidation products in and mon-key retinas. Invest Ophthalmol Vis Sci38:1802–1811

19. Landrum JT, Bone RA (2001) Lutein,zeaxanthin, and the macular pigment.Arch Biochem Biophys 385:28–40

20. Landrum JT, Bone RA, Joa H, KilburnMD, Moore LL, Sprague KE (1997) A one year study of the macular pigment: the effect of 140 d lutein supplement. Exp Eye Res 65:57–62

21. Newsome DA, Miceli MV, Liles MR,Tate DA, Oliver PD (1994) Antioxi-dants in the retinal pigment epithelium.Prog Ret Eye Res 13:101–123

22. Rapp LM, Maple SS, Choi JH (2000)Lutein and zeaxanthin concentrationsin rod outer segment membranes fromperifoveal and peripheral human retina.Invest Ophthalmol Vis Sci41:1200–1209

23. Snodderly DM (1995) Evidence forprotection against age-related maculardegeneration by carotenoids and anti-oxidant vitamins. Am J Clin Nutr 62 [6Suppl]:1448S–1461S

24. Snodderly DM, Brown PK, Delori FC,Auran JD (1984) The macular pig-ment. I. Absorbance spectra, localiza-tion and discrimination from other yel-low pigments in primate retinas. InvestOphthalmol Vis Sci 25:660–673

25. Snodderly DM, Handelman GJ, AdlerAJ (1991) Distribution of individualmacular pigment carotenoids in centralretina of macaque and squirrel mon-keys. Invest Ophthalmol Vis Sci32:268–279

26. Wolf S, Arend O, Reim M (1994) Mea-surement of retinal hemodynamicswith scanning laser ophthalmoscopy:reference values and variation. SurvOphthalmol [Suppl] 38:95–100

27. Young RW (1987) Pathophysiology ofage-related macular degeneration. SurvOphthalmol 31:291–306