Serum Carotenoids and Risk of Age-Related Macular Degeneration in a Chinese Population Sample

Abstract

Purpose.: It has been hypothesized that the macular carotenoids protect against age-related macular degeneration (AMD). In this study, the association between serum concentrations of carotenoids and the presence of AMD was evaluated in a case–control sample of elderly Chinese subjects.

Methods.: Two hundred sixty-three individuals aged between 50 and 88 years enrolled in the study. Subjects included 82 cases with exudative AMD, 92 cases with early AMD, and 89 control individuals. Serum carotenoids (lutein, zeaxanthin, lycopene, α- and β-carotenes, and β-cryptoxanthin) and retinol were measured with reversed-phase high-performance liquid chromatography (HPLC).

Results.: Serum levels of carotenoids and retinol were significantly lower in the cases with exudative AMD than in the controls. Median levels of lutein and zeaxanthin were 0.538 and 0.101 μM, respectively, in the control subjects, and 0.488 and 0.076 μM, respectively, in cases with exudative AMD. After adjustment for age, sex, smoking status, and body mass index (BMI), a significant inverse association was observed for exudative AMD with serum zeaxanthin (relative risk ratio [RRR], 0.04; 95% confidence interval [CI], 0–0.35), lycopene (RRR, 0.22; 95% CI, 0.1–0.48), and α-carotene (RRR, 0.24; 95% CI, 0.12–0.51). Early AMD was inversely associated only with lycopene (RRR, 0.49; 95% CI, 0.28–0.86) but was positively associated with α-carotene (RRR, 2.22; 95% CI, 1.37–3.58). No significant associations were observed between serum lutein and cases with early or exudative AMD.

Conclusions.: The data suggest that higher levels of serum carotenoids, in particular zeaxanthin and lycopene, are associated with a lower likelihood of having exudative AMD. Serum levels of carotenoids were relatively higher in this Chinese cohort than in samples of other ethnicities in previous reports.

Age-related macular degeneration (AMD) is a degenerative disease of the macula characterized by clinical signs ranging from the early stage of macular drusen to the advanced stage of ingrowth of new blood vessels into the subretinal space or geographic atrophy. ^{1 –3} It is the advanced stage of the disease that destroys the macular structure and function, leading to irreversible blindness in the elderly population if untreated or untreatable. ^{4 –6} Because there is no cure for advanced AMD and limited treatment is available to slow its progression, it is important to identify modifiable risk factors that may reduce disease occurrence or prevent progression of the disease from the early to the advanced stage. 

The pathogenesis of AMD is likely to involve a complex interaction of genetic and environmental factors ^7,8 that may be promoted by chronic inflammation, ^8,9 oxidative stress, ¹⁰ and light-induced damage. In recent years, there has been increasing speculation that the xanthophyll carotenoids lutein and zeaxanthin play a role in the pathogenesis of the disease. ^11,12 Lutein and zeaxanthin are the main constituents of the macular pigment, which are primarily located in the inner retinal layers of the macula and may protect the photoreceptors against light-induced damage and oxidative stress by their ability to absorb high-energy blue light and by their antioxidant properties. ¹³ These carotenoids are entirely of dietary origin. Therefore, blood or retinal levels of carotenoids can be modified by dietary change or supplementation. Epidemiologic and clinical studies have suggested that low dietary intake and low blood or retinal levels of carotenoids were associated with an increased risk of AMD, ^{11,14 –18} but not all observational studies could confirm these results. ^{19 –24} In the Chinese population, the prevalence of AMD is lower than among Caucasians. ^25,26 In addition to the differences in genetic susceptibility, ^27,28 a traditional Chinese diet with a higher intake of carotenoids ²⁹ may contribute to the relatively lower prevalence of AMD in the Chinese population. Whether the higher intake of carotenoids can prevent the development of AMD remains unclear. 

In the present study, we sought to further address this question by investigating the relationship between serum levels of carotenoids and the presence of AMD in a case–control cohort of elderly Chinese subjects. To our knowledge, this is the first report on serum levels of carotenoids and their association with AMD in a Chinese population. 

Subjects and Methods

Subjects and Clinical Evaluation

The study participants were recruited from June 2007 to December 2008 through a community-based eye disease–screening program in urban Beijing and among the outpatients visiting the Beijing Tongren Hospital. All study participants were unrelated native Chinese from the greater Beijing area. Subjects aged 50 years or older with or without signs of early or late AMD were invited to participate in the study. Exclusion criteria included any contraindications to pupil dilation, advanced cataract or other conditions that impeded a clear view of the retina, or patients with other eye diseases, such as glaucoma, uveitis, and chorioretinal diseases other than AMD. Patients with diabetes were also excluded. The study protocol was reviewed and approved by the Ethics Committee of the Beijing Tongren Hospital. Informed consent was obtained from all participants, and the procedures used conformed to the tenets of the Declaration of Helsinki for research involving human subjects. 

All participants were interviewed with a standardized questionnaire. Information regarding demographic characteristics, smoking habits and medical history was obtained. Height and weight were measured, and the body mass index (BMI) was calculated. All subjects underwent a comprehensive ophthalmic examination, including visual acuity measurement, slit lamp biomicroscopy, and detailed fundus examination, performed by a retinal specialist after pupil dilation with 0.5% tropicamide and 5% phenylephrine. Stereoscopic mydriatic 30° color fundus photographs were taken from both eyes with a digital fundus camera (Carl Zeiss Meditec, Oberkochen, Germany) centered on the fovea. Patients with exudative AMD were examined and the diagnosis confirmed by fluorescein and/or indocyanine green fundus angiography. 

The presence and severity of macular lesions were assessed based on the presence and size of drusen, pigment abnormalities, and other macular lesions. ³⁰ In this study, late AMD was defined as the presence of signs of exudative AMD ²⁷ or geography atrophy. ³⁰ Early AMD was defined as the presence of intermediate (≥63 but <125 μm) or large (≥125μm) drusen with or without pigment abnormalities in the absence of late AMD. Individuals were selected as control subjects based on the presence of a normal fundus or having only a few (<10) small drusen (<63 μm) without pigment abnormalities in both eyes. 

Measurement of Serum Carotenoids

Fasting blood samples were collected from participants in serum separator tubes with no anticoagulant and were protected from exposure to direct light during processing. The samples were allowed to stand at room temperature for 30 minutes for coagulation and then were centrifuged at 1000g for 10 minutes. Serum was collected by pipette and stored at −80°C until analysis for carotenoids and retinol. 

The serum was prepared for extraction by adding 150 μL of sample into 850 μL of 0.9% saline. Echinenone (the kind gift of Kemin Health USA, Des Moines, IA) and retinyl acetate in ethanol were added as internal standards. The mixture was extracted by using 3 mL of chloroform/methanol (2:1, vol/vol). The mixture was vortexed and then centrifuged at 800g for 10 minutes. The chloroform layer was removed and evaporated to dryness under nitrogen. A second extraction was performed on the mixture with 3 mL of hexane. The mixture was vortexed and centrifuged as above. The hexane layer was combined with the first extraction (chloroform layer) and evaporated to dryness under nitrogen. The residue from the serum was redissolved in 200 μL of ethanol, vortexed, and sonicated for 30 seconds. A 20 μL aliquot was used for high-performance liquid chromatography (HPLC). 

The reversed-phase HPLC system consisted of a pump (600E LC; Waters Corp., Milford, MA), a C₃₀ column (Develosil C30-MG; 3.3 μm, 150 × 4.6 mm; Nomura Chemical Co. Ltd., Japan), and a UV detector (model 2487; Waters). The data were collected and analyzed using PC-800 Software (not commercially available). The HPLC mobile phases were methanol:methyl-tert-butyl ether:water (83:15:2, vol/vol/vol, with 1.5% ammonium acetate in H₂O, solvent A) and methanol:methyl-tert-butyl ether:water (8:90:2, vol/vol/vol, with 1% ammonium acetate in H₂O, solvent B). The gradient procedure at a flow rate of 1 mL/min was as follows: The procedure began at 100% solvent A before going to 95% solvent A and 5% solvent B over a 1-minute linear gradient. This step was followed by a 3-minute hold at 95% solvent A; a 1 minute linear gradient to 90% solvent A and a 1 minute hold at 90% solvent A; a 15-minute linear gradient to 45% solvent A and a 1-minute hold at 45% solvent A; an 11-minute linear gradient to 95% solvent B; a 4-minute hold at 95% solvent B; and finally a 2-minute gradient back to 100% solvent A. The system was held at 100% solvent A for 10 minutes for equilibration back to the initial conditions. 

The carotenoids were quantified by determining peak areas in the HPLC chromatogram and were calibrated against known amounts of standards. The carotenoids and echinenone were detected at 450 nm. Retinol and retinyl acetate were detected at 325 nm. Carotenoid standards were from DSM Nutritionals (DSM Nutritionals, Basel, Switzerland). 

Statistical Analysis

Statistical analysis was performed with the R statistical analysis package. ³¹ Mean ± SD or median (minimum, maximum) were computed for continuous variables. Frequencies and percentages were computed for categorical variables. ANOVA was used to compare means of continuous variables in the controls, the cases with early AMD, and the cases with late AMD. Data that were not normally distributed were transformed by taking the square root. The Kruskal-Wallis test was used if the data were not normally distributed. Tukey multiple comparisons or Wilcoxon rank sum test was used to compare means or median between two groups when a significant difference was shown by ANOVA or the Kruskal-Wallis test. The association between sex and smoking status and early or exudative AMD groups was examined with the χ² test. Polytomous logistic was used for multivariate analysis. The relative risk ratio (RRR) and 95% confidence intervals (CI) were calculated. The statistically significant level was set at P < 0.05. 

Results

A total of 263 subjects participated in the study, including 89 (33.8%) control subjects with normal bilateral fundus, 92 (35%) cases with early AMD in one or both eyes, and 82 (31.2%) cases with exudative AMD in at least one eye. Based on angiograms, 56 (68.3%) of the 82 cases with exudative AMD were classified as having typical AMD with choroidal neovascularization, 17 (20.7%) with polypoidal choroidal vasculopathy (PCV), and 9 (11%) could not be classified. Large drusen (>125μm) were identified in seven typical AMD cases, three PCV cases, and two unclassifiable cases. There were no statistically significant differences in age, sex, BMI, smoking habits, and serum concentrations of carotenoids and retinol between typical AMD and PCV cases (data not shown). Only one case with atrophic AMD was identified during recruitment and was excluded from the study. 

The general characteristics of the study subjects are shown in Table 1. The percentage of men was 32.6% in the controls, 28.3% in the cases with early AMD, and 61% in those with exudative AMD (P < 0.001). More patients with exudative AMD were current or past smokers than were those with early AMD (P < 0.001) or controls (P < 0.001). No statistically significant differences in age and BMI were found among the control subjects and cases with either early or exudative AMD. None of the participants in this study reported the regular use of lutein and/or zeaxanthin supplements. 

Table 1.

 

Distribution of Baseline Characteristics in Control Subjects and Cases with Early or Late AMD

	Controls (n = 89)	Early AMD (n = 92)	Exudative AMD (n = 82)	P *
Age, mean ± SD, y	64.8 ± 8.27	67.6 ± 8.04	66.3 ± 8.98	0.089
Range	50–82	51–82	50–88
Sex, n (%)				<0.001
Male	29 (32.6)	26 (28.3)	50 (61)
Female	60 (67.4)	66 (71.7)	32 (39)
Body mass index mean ± SD, kg/m2	24.67 ± 3.21	24.16 ± 3.40	24.87 ± 3.69	0.37
Smoking status, n (%)				<0.001
Never	65 (73)	80 (87)	42 (51.2)
Current	11 (12.4)	6 (6.5)	20 (24.4)
Former	13 (14.6)	6 (6.5)	20 (24.4)

*  ANOVA (continuous data) or χ² (categorical data) test for comparison among three groups.

The serum concentrations of carotenoids and retinol are shown in Table 2. Serum levels of carotenoids and retinol were not normally distributed, and a statistical comparison was made with the Kruskal-Wallis test. The data showed that serum concentrations of lutein, zeaxanthin, β-carotene, lycopene, and retinol tended to decline with increasing severity of AMD. Serum concentrations of β-cryptoxanthin and α-carotene were lower in the cases with exudative AMD, but higher in the cases with early AMD, than in the control subjects. The serum concentration of lutein was lower in the cases with exudative AMD than in the control subjects (P = 0.038), but there was no significant difference between the cases with exudative AMD and those with early AMD (P = 0.42). The serum concentration of zeaxanthin in the cases with exudative AMD was significantly lower than in the controls (P < 0.001) or in the cases with early AMD (P < 0.001). Serum concentration of β-carotene and retinol in the cases with exudative AMD was also significantly less compared with the controls (P = 0.002 and P < 0.001, respectively), but was similar between the early AMD group and the controls (P = 0.37 and P = 0.13, respectively). Serum concentration of lycopene decreased significantly in cases with exudative AMD compared with the control subjects (P < 0.001) or to the cases with early AMD (P < 0.001). Compared with the control subjects, serum concentrations of β-cryptoxanthin and α-carotene were significantly lower in the cases with exudative AMD (P = 0.003 and P < 0.001, respectively), but were higher in the cases with early AMD (P = 0.071 and P < 0.001, respectively). 

Table 2.

 

Comparison of Serum Carotenoids and Retinol Levels (μmol/L) between Control Subjects and Cases

Serum Carotenoids	Control (n = 89)	Early AMD (n = 92)	Exudative AMD (n = 82)	P *
Lutein	0.538 (0.187, 1.572)	0.506 (0.161, 1.738)	0.488 (0.243, 1.361)*	0.12
Zeaxanthin	0.101 (0.043, 0.358)	0.114 (0.006, 0.428)	0.076 (0.014, 0.211)†	<0.001
β-Cryptoxanthin	0.248 (0.079, 2.861)	0.347 (0.046, 2.79)	0.173 (0.041, 1.22)‡	<0.001
α-Carotene	0.297 (0.02, 2.348)	0.518 (0.076, 1.812)†	0.118 (0.023, 1.116)†	<0.001
β-Carotene	0.946 (0.184, 3.307)	1.01 (0.095, 4.101)	0.63 (0.134, 3.407)‡	<0.001
Lycopene	0.259 (0.01, 1.671)	0.175 (0.029, 0.973)	0.097 (0,0.337)†	<0.001
Retinol	2.775 (1.188, 4.04)	2.574 (0.507, 5.576)	2.376 (0.555, 4.775)†	0.002

 Data are presented as median (minimum, maximum). The Kruskal-Wallis test was used for comparison between groups and the Wilcoxon rank sum test for early or exudative AMD versus controls.

*   P < 0.05.

†   P < 0.001.

‡   P < 0.01.

The relationships between the serum carotenoids were evaluated (Table 3). The highest correlation was found between lutein and zeaxanthin (r = 0.72). The other correlation coefficients ranged from 0.01 for lycopene and β-cryptoxanthin to 0.49 for α- and β-carotene. 

Table 3.

 

Correlation of Serum Carotenoids and Retinol in the Study Subjects

	Lutein	Zeaxanthin	Retinol	β-Cryptoxanthin	α-Carotene	β-Carotene	Lycopene
Lutein	1.0	0.72*	0.19†	0.16†	0.03	0.37*	0.25*
Zeaxanthin		1.0	0.27*	0.34*	0.12‡	0.40*	0.25*
Retinol			1.0	0.11	0.16†	0.15‡	0.26*
β-Cryptoxanthin				1.0	0.28*	0.38*	0.01
α-Carotene					1.0	0.49*	0.35*
β-Carotene						1.0	0.39*
Lycopene							1.0

  n = 263. Data are Pearson's correlation. The date were transformed by taking the square root before the correlation analyses.

*   P < 0.001.

†   P < 0.01.

‡   P < 0.05.

The RRR for AMD was calculated according to the concentration of serum carotenoids and is presented in Table 4. After adjustment for age, sex, smoking status, and BMI, a significant inverse association was observed between exudative AMD and serum levels of zeaxanthin (RRR, 0.04; 95% CI, 0–0.35), α-carotene (RRR, 0.24; 95% CI, 0.12–0.51), and lycopene (RRR, 0.22; 95% CI, 0.1–0.48). However, β-carotene was positively associated with exudative AMD after adjustment (RRR, 2.36; 95% CI, 1.3–4.29). No significant association was observed between the levels of serum lutein and exudative AMD. On comparison of early AMD with the control after adjustment, an association remained significant only with α-carotene and lycopene. The association with early AMD was observed to be inverse with lycopene (RRR, 0.49; 95% CI, 0.28–0.86) but positive with α-carotene (RRR, 2.22; 95% CI, 1.37–3.58). No association was identified between early AMD and serum levels of lutein or zeaxanthin. 

Table 4.

 

The Relative Risk for AMD According to Serum Carotenoid and Retinol Levels

	Early AMD	Exudative AMD
Lutein	0.64 (0.23–1.8)	2.68 (0.79–9.09)
Zeaxanthin	2.39 (0.37–15.3)	0.04 (0–0.35)*
β-Cryptoxanthin	1.11 (0.73–1.71)	0.54 (0.27–1.07)
α-Carotene	2.22 (1.37–3.58)*	0.24 (0.12–0.51)†
β-Carotene	1.02 (0.64–1.62)	2.36 (1.3–4.29)*
Lycopene	0.49 (0.28–0.86)‡	0.22 (0.1–0.48)†
Retinol	0.78 (0.46–1.32)	0.84 (0.46–1.52)

 Data are the RRR (95% CI) adjusted for age, sex, smoking status, and body mass index. The original data were transformed by taking square root before correlation analyses and the control subjects were used as reference group.

*   P < 0.01.

†   P < 0.001.

‡   P < 0.05.

Discussion

We conducted a case–control study to assess the association between serum levels of carotenoids and the risk of AMD in a Chinese population. Carotenoids are known antioxidants ³² and results from the Age-Related Eye Disease Study (AREDS) suggest that supplemental antioxidants delay the progression of AMD. ^33,34  

Our data showed that the serum levels of all carotenoids measured, including lutein, zeaxanthin, lycopene, α-carotene, β-carotene, and β-cryptoxanthin, were significantly lower in cases with exudative AMD than in control subjects. Serum concentration of retinol was also significantly lower in the exudative AMD patients than in the controls. After adjustment for age, sex, smoking status, and BMI, a significant inverse relationship was found between serum concentrations of zeaxanthin, lycopene, α-carotene, and exudative AMD. However, serum β-carotene showed a twofold increased risk of exudative AMD after adjustment. In the cases with early AMD, lycopene continued to be a protective factor, but α-carotene became a risk factor after adjustment. These data suggest that high circulating levels of carotenoids, in particular zeaxanthin and lycopene, are associated with reduced risk of AMD. In agreement with a previous report, ¹⁶ we found that serum carotenoids were highly intercorrelated, with the highest correlation found between lutein and zeaxanthin. 

Evidence regarding the association between carotenoids and AMD has been inconsistent. For example, a multicenter eye disease case–control study with 356 cases of advanced AMD and 520 control subjects found that higher dietary intake of lutein and zeaxanthin was strongly associated with a reduced risk of AMD. ¹⁷ The Blue Mountains Eye Study, a population-based cohort study of 2454 subjects, also found that higher dietary intake of lutein and zeaxanthin reduced the risk of the long-term incidence of AMD. ³⁵ Other studies, however, showed no correlation between serum levels of lutein and/or zeaxanthin and the risk of AMD. ^{19 –24} Consistent with the findings of Gale et al., ¹¹ in the present study we found a statistically significant protective effect of serum zeaxanthin against exudative AMD after adjustment. Such a protective effect against AMD was not observed for lutein although the serum levels of lutein were relatively lower in AMD groups than in the controls. Our data suggest that zeaxanthin may have a more powerful protective role against AMD than does lutein. Although both are the main constitutes of macular pigment, lutein and zeaxanthin may have unique functions in the retina. It has been demonstrated, for example, that the distribution of these two carotenoids differs, with zeaxanthin dominant in the central fovea and lutein dominant in the peripheral macula. ¹¹ The ratio of zeaxanthin to lutein is approximately 2:1 in the fovea, 1:1 in the peripheral macula, and 1:5 in the serum, showing that the fovea preferentially accumulates zeaxanthin. ¹⁶ It should be noted, however, that approximately one half of the zeaxanthin found in the fovea is meso-zeaxanthin which is derived from lutein. ^36,37 Lutein and zeaxanthin also differ in their orientation within biological membranes. ¹¹ Although both protect the lipid membranes from light-induced oxidative stress and free radical attack, zeaxanthin has been shown to be a better photoprotector during prolonged exposure to ultraviolet radiation and to be particularly effective in protecting lipid membranes against oxidation by peroxyl radicals. ¹⁶ Our finding that the risk of AMD was significantly associated with low levels of zeaxanthin, but not lutein, suggests that the use of the sum of these xanthophyll carotenoids in previous investigations may have obscured evidence of a protective role for zeaxanthin. More studies are needed on the specific functions of lutein and zeaxanthin and their relation with AMD on both the biological and epidemiologic levels. 

In the present study, lycopene was the only carotenoid showing a statistically significant protective effect against both early and exudative AMD, but with a more powerful protection against exudative AMD. This result is consistent with the report from the Beaver Dam Eye Study. ²¹ Cardinault et al. ²² also reported a significant inverse association between serum level of lycopene and the occurrence of AMD in a relatively small study sample. In addition, lycopene intake was associated with reduced risk of prostate cancer and cardiovascular disease. ³⁸ Unlike lutein and zeaxanthin, lycopene is not found in the retina, ³⁹ suggesting that the protective role of lycopene may be systemic rather than local, although lutein and zeaxanthin may also have a systemic effect. Being a strong antioxidant, lycopene protects lipids, lipoproteins, proteins, and DNA from oxidative processes and thus has potential anticarcinogenic and antiatherogenic effects. ³⁸ The strong protective role of lycopene against both early and exudative AMD, as demonstrated in our study and others, should be investigated further. 

It has been reported that β-carotene may have potentially harmful effects among smokers, with an increased risk of lung cancer among smokers taking daily β-carotene supplements. ^40,41 After adjustment for age, sex, smoking status, and BMI, our data showed that serum level of β-carotene was associated with a 2.36-fold increased risk of exudative AMD but not with early AMD, suggesting that the observed association between β-carotene and exudative AMD is not fully explained by smoking. The finding of an association between higher serum levels of β-carotene and an increased risk of exudative AMD is in agreement with the report of the Blue Mountain Eye Study. ³⁵ The AREDS and the Rotterdam Study, ^34,42 however, reported reduced risk of AMD with higher dietary intakes of β-carotene. Most other studies on serum carotenoids showed no statistically significant association between β-carotene and the risk of AMD. ^16,21,43 Therefore, our result for β-carotene should be regarded with caution. Increased serum concentrations of α-carotene showed a protective effect against exudative AMD but had a 2.2-fold increased risk for early AMD. We do not have a reasonable explanation for this result. Further studies are needed to confirm the observations in regard to the effect of α- and β-carotene on AMD. Serum levels of retinol and β-cryptoxanthin did not show significant association with the risk of AMD after adjustment. 

It is interesting to note that the average serum levels of carotenoids in this study, in particular lutein and zeaxanthin, were higher in comparison with those of the previous reports in other ethnicities (Table 5). ^{11,19 –23,38,43 –52} Differences in serum carotenoids between populations have been reported in other studies. ^38,47 Many factors affect the serum levels of carotenoids, the most important one being dietary intake. ⁵³ Other dietary factors such as dietary fat, fiber, food matrix, and food preparation or processing may also affect the carotenoid bioavailability or serum response and thus could influence the relationship between intake and serum concentrations. ⁴⁸ Serum levels of carotenoids are also influenced by other factors such as age, sex, ethnicity, smoking, education, body fat, BMI, and exercise. ^48,50 In the present study, all participants consumed the traditional Chinese foods throughout their lives, which contain high amounts of lutein- and zeaxanthin-rich vegetables. ²⁹ This explanation is the most likely one for the higher serum levels of carotenoids in the studied subjects compared with those in ethnicities evaluated in other studies. The higher serum levels of lutein and zeaxanthin may contribute to the relatively lower prevalence of AMD in the Chinese population. 

Table 5.

 

Serum Concentration of Lutein and Zeaxanthin from Previous Publications

Data Source	Study Subjects	Lutein	Zeaxanthin	Lutein + Zeaxanthin	Year of Report
Beaver Dam Eye Study United States*	n = 160; AMD and controls	AMD: 0.204 ± 0.101; controls: 0.201 ± 0.103	AMD: 0.055 ± 0.032; controls: 0.051 ± 0.026	AMD: 0.294 ± 0.128; controls: 0.291 ± 0.138	1995 21
United States	n = 7; normal	0.37 ± 0.05	0.058 ± 0.012		2000 44
Netherlands	n = 8; normal	0.18 ± 0.08			2000 45
United States	n = 278; normal	0.28 ± 0.13	0.091 ± 0.044		2001 46
NHANES III United States	n = 8222; early and late ARM			Age 40–59: 0.38; 60–79: 0.40; ≥80: 0.40	2001 23
Netherlands	n = 367; normal	M: 0.16 ± 0.07; F: 0.19 ± 0.08	M: 0.05 ± 0.02; F: 0.05 ± 0.02		2002 47
OPMSS United States	n = 2786; normal	0.226 ± 0.120	0.071 ± 0.039		2002 48
Costa Rica Mestizo*	n = 531; normal			M: 0.316 ± 0.143; F: 0.328 ± 0.165;	2002 49
United Kingdom*	n = 380; normal and ARM	Normal: 0.177 (0.13, 0.232); ARM: 0.153 (0.118, 0.215)	Normal: 0.036 (0.024, 0.055); ARM: 0.03 (0.02, 0.046)	Normal: 0.213 (0.16–0.287) ARM: 0.193 (0.147, 0.256)	2003 11
Ireland†	n = 100; normal	M: 0.16 ± 0.07; F: 0.18 ± 0.08	M: 0.037 ± 0.023; F: 0.046 ± 0.03		2004 50
Massachusetts Hispanic Elders Study, United States*	n = 584;	Hispanic: M: 0.212 ± 0.00004; F: 0.21 ± 0.00003; non-Hispanic white: M: 0.185 ± 0.00009; F: 0.189 ± 0.00007	Hispanic: M: 0.069 ± 0.00004; F: 0.067 ± 0.00003; non-Hispanic white: M: 0.058 ± 0.0001 F: 0.065 ± 0.00008		2005 38
MARS Germany†	n = 1060; normal and ARM	Normal: 0.218 ± 0.011; early ARM: 0.197 ± 0.007; late ARM: 0.23 ± 0.011	Normal: 0.033 ± 0.002; Early ARM: 0.035; ± 0.001; late ARM: 0.039 ± 0.002		2005 20
United States	n = 108; normal	F: 0.381 ± 0.05; M: 0.331 ± 0.06	F: 0.069 ± 0.00; M: 0.070 ± 0.00	F: 0.450 ± 0.02; M: 0.401 ± 0.02	2005 51
France	n = 61; AMD and controls	AMD: 0.178 ± 0.013; controls: 0.187 ± 0.023;	AMD: 0.044 ± 0.004; controls: 0.058 ± 0.009		2005 22
United States‡	n = 56; nonsmokers and smokers	Nonsmokers: 0.14 ± 0.0; smokers: 0.12 ± 0.0	Nonsmokers: 0.86 ± 0.0; smokers: 0.23 ± 0.0		2006 52
CAREDS United States	n = 1787; women	Low intake: 0.21; high intake: 0.31	Low intake: 0.054; high intake: 0.064		2006 19
Takasaki City Japanese	n = 722; normal and AMD			Normal: 0.48 ± 1.67; early AMD: 0.47 ± 1.94 Late AMD: 0.36 ± 1.70	2009 43

 Data are presented as the mean ± SD or median (min, max) in μmol/L. ARM, age-related maculopathy; M, male; F, female.

*  Original data in nmol/L.

†  Original data in μg/mL.

‡  Original data in μg/dL.

There are limitations of our study. The cross-sectional design of the study was only able to detect associations, but not the temporal sequence or causality, between serum levels of carotenoids and the risk of AMD. These results therefore need to be confirmed in prospective or interventional studies. In addition, this study was not population based, and a selection bias of the study subjects may have occurred. All participants, however, were from the greater Beijing area and presumably had similar dietary patterns. Furthermore, confounding factors are always concerns in observational studies. We performed multivariate adjustments to take into account some of the known risk factors for AMD, but not all known risk factors are included, which may lead to confounding bias. Finally, serum levels of carotenoids reflect the more recent nutritional intake and provide only indirect information about the macular pigment density, the latter has a noticeably slower biological turnover. However, most studies found a positive and significant correlation between serum levels and macular pigment density, corroborating the use of blood-based biomarkers. ^45,53  

In summary, we have shown an inverse association between serum levels of carotenoids and the presence of AMD in a sample of the Chinese population. These results suggest that higher levels of serum carotenoids, in particular zeaxanthin and lycopene, may be associated with a lower likelihood of having exudative AMD. The observed association of carotenoids with AMD in this study requires confirmation in longitudinal studies or clinical trials. 

Footnotes

 Supported by the National Basic Research Program of China (Program 973) Grant 2007CB512201 and the Beijing Municipal Health Bureau Grant 2009208.

Footnotes

 Disclosure: H. Zhou, None; X. Zhao, None; E.J. Johnson, None; A. Lim, None; E. Sun, None; J. Yu, None; Y. Zhang, None; X. Liu, None; T. Snellingen, None; F. Shang, None; N. Liu, None