Clinical and Epidemiologic Research
Circulating Omega-3 Fatty Acids and Neovascular
Age-Related Macular Degeneration
edicte M. J. Merle,
and Eric H. Souied,
for the Nutritional AMD Treatment 2 Study Group
Ophthalmology Department, Hˆ
opital Intercommunal de Cr´
eteil, University Paris Est Cr´
CHRU Lille, Biochemistry and Molecular Biology Institute, Molecular Medicine of Metabolic Diseases (U4M), Lille, France
Lille2 University, School of Medicine, Department of Biochemistry and Molecular Biology, Lille, France
Laboratoire Bausch & Lomb, Montpellier, France
INSERM, Centre INSERM U897-Epidemiologie-Biostatistique, Bordeaux, France
University Bordeaux, ISPED, Bordeaux, France
Merle, Service d’ophtalmologie
eteil, 40 avenue de Verdun,
See the appendix for the members of
the Nutritional AMD Treatment 2
Submitted: January 9, 2014
Accepted: February 12, 2014
Citation: Merle BMJ, Benlian P, Puche
N, Bassols A, Delcourt C, Souied EH.
Circulating omega-3 fatty acids and
neovascular age-related macular de-
generation. Invest Ophthalmol Vis
Sci. 2014;55:2010–2019. DOI:
PURPOSE.We assessed the associations of serum, red blood cell membranes (RBCM) and
dietary long-chain n-3 polyunsaturated fatty acids (LC-PUFAs) with neovascular age-related
macular degeneration (AMD).
METHODS.We included 290 patients of the Nutritional AMD Treatment 2 Study (NAT2) with
neovascular AMD in one eye and early AMD lesions in the other eye, and 144 normal vision
controls without AMD. Dietary intake of seafood was estimated by food frequency
questionnaire. Eicosapentaenoic acid (EPA) and docosahexaenoic acid (DHA) composition
in serum and RBCM were determined by gas chromatography from 12-hour fasting blood
samples and was expressed as percentages of total fatty acids proﬁle. Logistic regressions
estimated associations of neovascular AMD with dietary intake of seafood and circulating n-3
RESULTS.Dietary oily ﬁsh and seafood intake were signiﬁcantly lower in AMD patients than in
controls. After adjustment for all potential confounders (age, sex, CFH Y402H,ARMS2 A69S,
and ApoE4 polymorphisms, plasma triglycerides, hypertension, hypercholesterolemia, and
family history of AMD), serum EPA was associated signiﬁcantly with a lower risk for
neovascular AMD (odds ratio [OR] ¼0.41; 95% conﬁdence interval [CI], 0.22–0.77; P¼
0.005). Analysis of RBCM revealed that EPA and EPAþDHA were associated signiﬁcantly with a
lower risk for neovascular AMD (OR ¼0.25; 95% CI, 0.13–0.47; P<0.0001 and OR ¼0.52;
95% CI, 0.29–0.94; P¼0.03, respectively).
CONCLUSIONS.The RBCM EPA and EPAþDHA, as long-term biomarkers of n-3 dietary PUFA
status, were associated strongly with neovascular AMD and may represent an objective
marker identifying subjects at high risk for neovascular AMD, who may most beneﬁt from
nutritional interventions. (http://www.controlled-trials.com/isrctn number,
Keywords: age-related macular degeneration, omega-3 fatty acids, epidemiology, case-control
Age-related macular degeneration (AMD) is the leading cause
of irreversible vision loss in industrialized countries.
comprises two late forms associated with severe visual
impairment (neovascular and atrophic AMD), generally preced-
ed by early, asymptomatic, retinal abnormalities (drusen,
pigmentary abnormalities). Treatments for neovascular AMD
have been available for a few years. Although they stabilize
vision, they are not curative, supporting the need for a targeted
prevention toward high-risk asymptomatic subjects, identiﬁed
by relevant biomarkers.
The condition of AMD is a multifactorial disease, involving
genetic and environmental factors (in particular smoking and
Omega-3 long-chain polyunsaturated fatty acids (n-3
LC-PUFAs), mainly eicosapentaenoic acid (EPA) and docosahex-
aenoic acid (DHA), have important structural and protective
functions in the retina.
The DHA reaches its highest
concentration in the membranes of photoreceptors, and is
important in photoreceptor differentiation and survival, as well
as in retinal function.
The anti-inﬂammatory properties of EPA
are of particular interest in AMD, since inﬂamma-
tion appears to have a pivotal role in this condition.
n-3 LC-PUFAs may increase the retinal density of macular
pigment, which ﬁlters blue light, and has local antioxidant and
Finally, derivatives of dietary n-3
LC-PUFAs, exhibit antiangiogenic properties in the retina.
In 2008, a meta-analysis
of nine epidemiologic studies
showed a signiﬁcantly reduced risk for AMD in subjects with
high dietary intake of n-3 PUFAs and ﬁsh, the main food source
of n-3 PUFAs. Since then, 10 additional studies have shown
similar and consistent results.
Dietary assessment methods rely on the subjects’ memory
and perceptions, and face the difﬁculties of the extreme day-to-
Copyright 2014 The Association for Research in Vision and Ophthalmology, Inc.
www.iovs.org jISSN: 1552-5783 2010
day variability of human diet, the hidden nature of many fats
used for dressing and cooking, the bias in reporting due to
social standards and nutritional recommendations, and the
estimation of the nutritional content of foods. Because of the
multiple difﬁculties of dietary assessment, circulating biomark-
ers may represent a more objective alternative for the
assessment of nutritional status.
A better assessment of n-3
nutritional status could help identify high-risk subjects, who
may beneﬁt most from nutritional intervention. Such biomark-
ers also might be used to follow the efﬁcacy of nutritional
interventions in restoring adequate nutritional status.
Over the last 20 years, a number of biomarkers have been
developed to assess the nutritional status in fatty acids according
to different source tissues. Because of very limited capacity of
endogenous synthesis, the body status of n-3 LC-PUFA mainly
reﬂects dietary intake of these essential fatty acids. The shortest-
term biomarkers of n-3 LC-PUFA body status are serum or
plasma measurements, reﬂecting dietary intakes of the past few
hours for triglycerides or of the past few days for cholesterol
ester and phospholipid fatty acids carried within circulating
lipoproteins. Red blood cell membranes (RBCM) and platelets
are of particular interest, since they reﬂect longer-term overall
dietary intake of n-3 LC-PUFA, incorporated within membrane
phospholipids of bone marrow–derived cell lines during the
past few months.
Because n-3 fatty acids may undergo variable
interconversion after intestinal absorption, the omega-3 index
(i.e., RBCM EPAþDHA) appears as an interesting long-term
integrator of n-3 LC-PUFA body status.
Circulating n-3 PUFAs have been evaluated in numerous
studies, showing good correlation with dietary intake, and
sensitivity to changes in dietary supplementation studies.
They have been used widely in association studies of n-3 PUFAs
with a variety of health outcomes (cardiovascular diseases,
obesity and diabetes, chronic inﬂammatory or neuro-psychiat-
ric disorders, cancers, and so forth).
However, with regard
to AMD, while many studies have reported associations with
dietary intakes of n-3 PUFAs, very few data are available on
associations of AMD with circulating biomarkers of n-3 PUFA
status. Recently, we have shown that high plasma n-3 LC-PUFAs
were associated signiﬁcantly with a decreased risk for late AMD
in elderly subjects from South of France.
This study used a
single plasma measurement that represented a crude estimate
of body fatty acid status. Measurement of n-3 PUFAs in RBCM
may represent a better biomarker for longer term status, with a
half-life of 120 days.
In the present study, we reported the associations of dietary
intake of seafood, and serum and RBCM n-3 LC-PUFAs with
neovascular AMD in a French case-control study.
Cases. The 290 cases of neovascular AMD were included
from Nutritional AMD Treatment 2 Study (NAT2) baseline
The NAT2 study is a randomized, placebo-
controlled, double-blind, parallel, comparative study. Patients
were enrolled from December 2003 to October 2005 in a
single center at the Department of Ophthalmology, Hˆ
Intercommunal de Creteil, France. The study was reviewed and
approved by the relevant institutional review board (CPP, Paris-
Ile de France 5, Paris, France).
Eligible patients were affected by neovascular AMD in one
eye and early AMD (any drusen or reticular pseudodrusen with
or without pigmentary changes) in the other eye. Neovascular
AMD was deﬁned on the basis of fundus color pictures and
ﬂuorescein angiography examination. Inclusion criteria were
age 55 years or older and younger than 85 years, and visual
acuity better than þ0.4 logarithm of minimum angle of
resolution units in the study period.
The main exclusion
criteria were: choroidal neovascularization (CNV) in both eyes
or no CNV in either eye, wide central subfoveal atrophy of the
study eye, and progressive ocular diseases (severe glaucoma or
other severe retinopathy).
Eye examination included best-corrected visual acuity, slit-
lamp examination, fundus photography, and ﬂuorescein
angiography (Topcon501A; Topcon, Tokyo, Japan). The study
was registered on the International Standard Randomized
Controlled Trial Number Register and was allocated registra-
tion number ISRCTN98246501.
Controls. Controls were enrolled through local-newspa-
pers calls for collaboration. A total of 144 men and women,
aged 55 years or more, with normal visual acuity, no history of
ocular diseases, and normal fundus examination and fundus
photography, was recruited and examined at the Department
of Ophthalmology of Creteil between 2002 and 2008. Controls
were from the same geographical area as the AMD cases.
Written informed consent was obtained for all participants
(cases and controls), as required by the French bioethical
legislation and local ethic committee (CPP Henri Mondor). This
study followed the tenets of the Declaration of Helsinki.
Biological Measurements of Fatty Acids
Overnight fasting blood samples were delivered to a single
clinical chemistry laboratory (Hˆ
opital Saint Antoine, APHP,
Paris, France) within ﬁve hours and processed immediately as
For cases, blood samples collected at baseline
examination (before any supplementation) were used for the
present study. For controls, blood samples were obtained at the
time of eye examination.
Fatty acid composition in serum and RBCM was determined
by gas chromatography after they were transmethylated by
diazomethane following a modiﬁed Dole’s procedure.
for EPA and DHA content were expressed as a percentage of
the total fatty acid proﬁle in serum and RBCM, and were
available for all participants (n¼434).
Biological samples were collected in the same conditions and
at time of fatty acid measurements. They included serum lipids
and lipoproteins, and genetic polymorphisms validated as
genetic markers of exudative AMD.
Serum total, high (HDL) and low (LDL) density lipoprotein-
cholesterol, and triglycerides, were measured by enzymatic
colorimetric and electrophoretic methods as described previ-
Genomic DNA was extracted from 10 mL blood
leukocytes as described previously in AMD patients
using the Illustra kit according to the manufacturer’s protocol
(GE Healthcare, Little Chalfont, Buckinghamshire, UK) in
controls. Genotyping of CFH rs1061170, ARM2/HTRA1
rs10490924, and Apolipoprotein E2, 3, 4 alleles were
performed by quantitative polymerase chain reaction allelic
discrimination using reagents and conditions from Custom
Taqman Single-Nucleotide Polymorphism Genotyping Assays
(Applera, Corp., Saint Aubin, France), using ABI 7900HT
(Applied Biosystems, Carlsbad, CA). Quality control of
genotyping by Sanger sequencing and bioinformatics analysis
were performed as described.
Dietary data were collected using a validated food frequency
questionnaire (FFQ) that recorded the usual food intakes for
Circulating Omega-3 Fatty Acids and AMD IOVS jMarch 2014 jVol. 55 jNo. 3 j2011
the last year.
The interview was conducted by trained
technicians, by telephone, and lasted 45 to 60 minutes. The
FFQ consists of 165 items and portions were estimated using a
validated set of photographs. The set of photographs was given
to the patient before the telephone interview. It was arranged
by food type and meal pattern. In the analysis, the intakes were
expressed in daily consumption in grams. The food composi-
tion table was REGAL
(Ciqual, Edinburgh, UK) expanded
with carotenoid and fatty acid contents from the SU.VI.MAX
Total dietary intake of seafood is the sum of oily ﬁsh,
white ﬁsh, and other seafood, and total dietary intake of ﬁsh is
the sum of oily ﬁsh and white ﬁsh. Dietary data were available
for 423 participants (97.4%).
Socio-demographic factors and medical history were collected
through face-to-face, standardized interviews at the same time
as eye examination. They included age, sex, body mass index
(BMI; weight [kg]/height
]), smoking status (never smoker
or ever smoker), self-reported history of hypercholesterolemia,
hypertension, diabetes, and family history of AMD, circulating
biomarkers (serum total, HDL- and LDL-cholesterol, and
triglycerides), and genetic biomarkers (CFH rs1061170,
ARM2/HTRA1 rs10490924, and Apolipoprotein E2, and E4
alleles). All covariates were available for all participants (n¼
Comparisons between neovascular AMD patients and controls
were performed using the Pearson v
for sex, Student’s t-test
for age, and logistic regression adjusted for age and sex for
Associations of circulating n-3 PUFAs and ﬁsh intake with
socio-demographic factors, medical history, dietary intake of
seafood, and genetic polymorphisms were performed using
Kruskal-Wallis ANOVA and Wilcoxon tests.
Associations of neovascular AMD with dietary intake of
seafood and circulating n-3 PUFAs were estimated using logistic
regression. Potentials confounders retained in the ﬁnal
multivariate model were factors associated signiﬁcantly with
neovascular AMD or n-3 PUFAs in our study (hypercholester-
olemia, hypertension, family history of AMD, plasma triglycer-
ides, and CFH,ARMS2, and ApoE4 polymorphisms; P<0.05).
Dietary intake of seafood and circulating n-3 PUFAs variables
were used as tertiles of distribution, the ﬁrst tertile being the
We also analyzed potential gene-environment interactions,
and potential age- and sex-circulating n-3 PUFAs interactions.
Interactions were introduced independently in the fully
adjusted model and retained if they were signiﬁcant (P<0.05).
For all analyses, differences were considered signiﬁcant at P
<0.05. All statistical analyses were performed using SAS
version 9.3 (SAS Institute, Inc., Cary, NY).
Neovascular AMD patients were older than controls (P<
0.0001), but were not different regarding sex, smoking status,
and BMI (Table 1). After adjustment for age and sex,
neovascular AMD patients declared more frequently a family
history of AMD (P¼0.004), hypercholesterolemia (P¼0.004),
or hypertension (P¼0.001), both latter conditions being under
stable corrective therapy. Frequency of self-declared diabetes
did not differ between neovascular AMD patients and controls.
Regarding genetic polymorphisms, CFH Y402H (P<0.0001),
ARMS2 A69S (P<0.0001), and ApoE4 (P¼0.03) polymor-
phisms were associated signiﬁcantly with neovascular AMD.
Neovascular AMD patients had lower plasma triglycerides than
controls (P¼0.0009), while they had similar plasma total, HDL-
and LDL-cholesterol (Table 1). Neovascular AMD patients had
lower serum EPA (P¼0.03), RBCM EPA (P<0.001), RBCM
DHA (P¼0.03), and omega-3 index (RBCM EPAþDHA, P¼
0.001) than controls, while they had serum DHA and EPAþDHA
similar to controls after adjustment for age and sex (Table 1).
Neovascular AMD patients had lower dietary intake of oily ﬁsh
(P¼0.02) and total seafood (P¼0.03) than controls, but were
not different regarding dietary intake of total ﬁsh, white ﬁsh,
and other seafood (Table 1).
circulating n-3 fatty acids with socio-demographic factors,
medical history, and genetic polymorphisms. Younger partic-
ipants had a higher dietary intake of oily ﬁsh than older
participants (P¼0.0003). Men had a higher dietary intake of
total and oily ﬁsh (respectively, P¼0.002 and P¼0.005).
Participants who declared hypertension had lower dietary
intake of oily ﬁsh (P¼0.003). Participants with at least one
allele E4 for ApoE polymorphism had higher dietary intake of
total ﬁsh and oily ﬁsh (respectively, P¼0.03 and P¼0.03).
Other socio-demographic factors, lifestyle, and AMD-related
genetic polymorphisms were not associated with dietary intake
of ﬁsh or seafood. Remarkably, none of the circulating n-3 LC-
PUFAs appeared inﬂuenced by any of the socio-demographic,
medical, or genetic risk factors for AMD analyzed herein.
As shown in Table 3, serum EPA, DHA, and EPAþDHA were
associated signiﬁcantly with all items of dietary intake of
seafood (total ﬁsh, oily ﬁsh, white ﬁsh, other seafood, and total
seafood). Subjects in the third tertile, for all seafood items had
higher serum EPA, DHA, and EPAþDHA. The same trend was
observed with RBCM EPA, DHA, and EPAþDHA, and reached
statistical signiﬁcance for all items of dietary intake of seafood
except for RBCM DHA and white ﬁsh (P¼0.08). Of note, the
median omega-3 index (i.e., RBCM EPAþDHA) was constantly
>4, in subjects from the third tertile, for all seafood items.
As shown in Table 4, after adjustment for age and sex,
dietary intake of total seafood and of total ﬁsh was associated
inversely with neovascular AMD (respectively, P¼0.05 and P¼
0.04). After adjustment for all potential confounders (age, sex,
CFH Y402H, ARMS2 A69S, and ApoE4 polymorphisms, plasma
triglycerides, hypertension, hypercholesterolemia, and family
history of AMD), these associations were no longer statistically
signiﬁcant. With regard to dietary intake of oily ﬁsh, white ﬁsh,
or other seafood, associations were in the same direction, but
did not reach statistical signiﬁcance.
Associations of neovascular AMD with circulating n-3 PUFAs
are shown in Table 5. After adjustment for age and sex, serum
EPA was signiﬁcantly associated with a lower risk for
neovascular AMD (odds ratio [OR] ¼0.59, P¼0.04), while
serum DHA and EPAþDHA were not signiﬁcantly associated
with neovascular AMD. This association remained signiﬁcant
after adjustment for all potential confounders (P¼0.005).
With regard to RBCM n-3 PUFAs, after adjustment for age
and sex, EPA and EPAþDHA were associated strongly with a
lower risk for neovascular AMD (OR ¼0.33, P<0.0001 and
OR ¼0.44, P¼0.002, respectively) and after adjustment for all
potential confounders, these associations remained signiﬁcant
(OR ¼0.25, P<0.0001 and OR ¼0.52, P¼0.03, respectively).
As in serum, DHA in RBCM was not associated signiﬁcantly
with neovascular AMD.
There was no detectable interaction between dietary intake
of seafood or circulating n-3 PUFAs with CFH,ARMS2 or ApoE
genetic polymorphisms, age, or sex.
Circulating Omega-3 Fatty Acids and AMD IOVS jMarch 2014 jVol. 55 jNo. 3 j2012
TABLE 1. Characteristics of Neovascular AMD Patients and Controls
Characteristics Controls, n¼144 Neovascular AMD Patients, n¼290 Adjusted P*
Age, y, mean 6SD 67.7 68.2 70.8 67.59 <0.0001
Male 55 (38.2) 105 (36.2) 0.69
Female 89 (61.8) 185 (63.8)
Smoking status, n(%)
Never smoker 91 (63.2) 165 (56.9) 0.12
Ever smoker 53 (36.8) 125 (43.1)
, mean 6SD 25.2 63.7 25.7 63.97 0.17
Self–reported medical history
No 102 (70.8) 147 (51.4) 0.0004
Yes 42 (29.2) 143 (49.3)
No 102 (70.8) 149 (51.0) 0.001
Yes 42 (29.2) 141 (48.6)
No 131 (91.0) 266 (91.7) 0.59
Yes 13 (9.0) 24 (8.3)
Family history of AMD, n(%)
No 125 (86.8) 222 (76.6) 0.004
Yes 19 (13.2) 68 (23.5)
TT 56 (38.9) 63 (21.7) <0.0001
CT 68 (47.2) 134 (46.2)
CC 20 (13.9) 93 (32.1)
GG 93 (64.6) 81 (27.9) <0.0001
GT 46 (31.9) 133 (45.9)
TT 5 (3.5) 76 (26.2)
At least 1 allele E2 18 (12.5) 53 (18.3) 0.12
At least 1 allele E4 39 (27.1) 48 (16.6) 0.03
Plasma lipids, mmol/L, median (fifth–95th percentiles) or mean 6SD
Triglycerides 1.14 (0.57–2.30) 0.98 (0.48–2.17) 0.0009
HDL–cholesterol 1.83 60.56 1.79 60.55 0.48
LDL–cholesterol 3.91 (2.51–5.30) 3.64 (2.30–5.59) 0.29
Total cholesterol 5.85 60.93 5.68 61.04 0.16
Circulating omega 3 PUFA, % of fatty acids, median (fifth–95th percentiles)
Serum EPA 0.74 (0.24–1.96) 0.60 (0.30–1.40) 0.03
Serum DHA 1.25 (0.63–2.00) 1.30 (0.60–2.40) 0.1
Serum EPAþDHA 1.99 (1.08–3.53) 1.90 (1.00–3.70) 0.78
Red blood cell membranes EPA 0.78 (0.29–1.47) 0.60 (0.30–1.20) <0.0001
Red blood cell membranes DHA 3.51 (2.13–5.03) 3.20 (1.80–5.10) 0.03
Red blood cell membranes EPAþDHA 4.32 (2.63–6.48) 3.80 (2.10–5.90) 0.001
Dietary intake of seafood, g/d, median (fifth–95th percentiles) n¼139 n¼284
Total fish 19.9 (7.4–51.1) 17.1 (4.9–41.9) 0.05
Oily fish 8.2 (0.0–31.4) 5.5 (0.0–22.9) 0.02
White fish 9.9 (0.0–19.7) 9.9 (0.0–34.0) 0.68
Other seafood 1.8 (0.0–17.1) 0.7 (0.0–15.7) 0.16
Total seafood 22.7 (9.9–64.0) 20.4 (5.3–51.1) 0.03
*PStudent’s t-test for age, Pearson v
for sex, and logistic regression adjusted for age and sex for other variables.
Circulating Omega-3 Fatty Acids and AMD IOVS jMarch 2014 jVol. 55 jNo. 3 j2013
TABLE 2. Variations of Circulating n-3 PUFAs and Dietary Intake of Fish According to Socio-Demographic Factors, Lifestyle, and AMD-Related
(% of Fatty Acids)
(% of Fatty Acids)
<70 203 2.04 (1.15–3.70) 4.10 (2.47–5.83) 199 19.7 (5.3–51.1) 8.2 (0.0–31.4) 9.9 (2.5–19.7)
‡0 231 1.90 (0.90–3.60) 3.86 (2.11–6.02) 224 17.0 (4.9–42.6) 5.0 (0.0–22.9) 9.9 (0.0–38.4)
P* 0.11 0.20 0.05 0.0003 0.84
Men 160 1.91 (1.05–3.70) 4.00 (2.45–5.86) 157 19.9 (4.9–58.3) 7.9 (0.0–31.4) 9.9 (0.0–39.4)
Women 274 1.91 (1.00–3.70) 4.00 (2.10–6.20) 266 15.7 (5.0–41.3) 5.4 (0.0–22.9) 9.9 (0.0–26.4)
P0.61 0.71 0.002 0.005 0.25
Never smoker 256 1.93 (1.11–3.70) 4.00 (2.20–6.40) 248 16.6 (5.0–42.6) 5.7 (0.0–21.4) 9.9 (2.5–24.1)
Ever smoker 178 1.91 (0.90–3.70) 4.00 (2.40–5.80) 175 19.7 (4.9–53.4) 7.9 (0.0–31.4) 9.9 (0.0–39.4)
P0.32 0.72 0.06 0.17 0.31
<25 218 2.00 (1.02–4.10) 4.05 (2.30–5.83) 211 19.7 (4.9–51.1) 5.7 (0.0–25.7) 9.9 (0.0–34.0)
‡25 214 1.90 (1.00–3.53) 4.00 (2.30–6.20) 212 17.8 (4.9–42.6) 7.5 (0.0–25.7) 9.9 (0.0–26.4)
P0.30 0.61 0.71 0.67 0.31
No 245 1.90 (1.00–3.53) 4.03 (2.40–5.90) 237 18.4 (5.3–50.9) 7.9 (0.0–25.7) 9.9 (2.5–38.4)
Yes 189 2.00 (1.05–3.70) 4.00 (2.20–6.00) 186 21.0 (5.0–50.7) 5.5 (0.0–25.7) 19.0 (4.9–42.6)
P0.95 0.62 0.44 0.31 0.70
No 251 2.00 (1.00–3.60) 4.07 (2.40–6.20) 246 18.9 (5.3–45.4) 7.9 (0.0–25.7) 9.9 (0.0–34.0)
Yes 183 1.90 (1.05–3.70) 4.00 (2.20–5.90) 177 17.7 (4.9–47.2) 5.0 (0.0–22.9) 9.9 (0.0–26.9)
P0.26 0.35 0.14 0.003 0.89
No 397 2.00 (1.02–3.70) 4.03 (2.20–6.00) 386 18.9 (5.0–47.2) 7.5 (0.0–25.7) 9.9 (0.0–28.6)
Yes 37 1.60 (0.90–3.20) 3.50 (2.47–6.29) 37 15.7 (2.5–42.6) 5.4 (0.0–31.4) 9.9 (0.0–24.1)
P0.05 0.09 0.47 0.90 0.40
Family history of AMD
No 347 1.90 (1.02–3.60) 4.07 (2.39–5.90) 337 19.5 (4.9–45.4) 7.5 (0.0–25.7) 9.9 (0.0–34.0)
Yes 87 2.00 (1.00–3.90) 3.90 (2.20–6.20) 86 16.5 (7.1–47.3) 5.0 (0.0–25.7) 9.9 (2.5–23.3)
P0.47 0.21 0.51 0.17 0.67
CC 113 1.90 (1.10–4.00) 3.80 (2.20–6.29) 109 19.7 (4.9–42.6) 5.7 (0.0–25.7) 9.9 (2.5–34.0)
CT 202 1.96 (1.08–3.90) 4.10 (2.40–6.02) 198 19.7 (5.7–50.9) 7.9 (0.0–27.9) 9.9 (0.0–39.4)
TT 119 1.90 (0.90–3.00) 4.07 (2.10–5.70) 116 15.6 (3.6–48.3) 5.5 (0.0–22.9) 9.9 (0.0–19.7)
P0.40 0.49 0.13 0.86 0.09
GG 174 1.90 (1.02–3.90) 4.14 (2.50–6.02) 169 19.7 (4.9–51.1) 7.9 (0.0–31.4) 9.9 (0.0–34.0)
GT 179 2.00 (1.00–3.70) 4.03 (2.00–6.20) 176 17.9 (4.9–41.1) 5.7 (0.0–22.9) 9.9 (0.0–19.7)
TT 81 1.90 (1.20–3.10) 3.80 (2.60–5.62) 78 19.5 (5.0–58.0) 6.6 (0.0–31.4) 9.9 (0.0–39.4)
P0.63 0.27 0.66 0.77 0.65
At least 1 E2 allele 71 1.90 (0.90–3.50) 3.75 (2.00–5.80) 69 19.9 (7.1–50.9) 7.9 (0.0–22.9) 9.9 (2.5–39.4)
No E2 allele 363 1.95 (1.10–3.70) 4.07 (2.40–6.00) 354 17.8 (4.9–45.1) 5.7 (0.0–25.7) 9.9 (0.0–26.4)
P0.21 0.10 0.16 0.80 0.10
At least 1 E4 allele 87 1.90 (0.90–3.70) 4.10 (2.00–6.02) 84 19.8 (7.3–58.0) 7.9 (0.0–31.4) 9.9 (2.5–39.4)
No E4 allele 347 1.91 (1.10–3.70) 4.00 (2.30–6.00) 339 17.8 (4.9–42.6) 5.7 (0.0–25.7) 9.9 (0.0–24.1)
P0.88 0.96 0.03 0.03 0.18
*Pfor Wilcoxon test or Kruskal-Wallis ANOVA.
Circulating Omega-3 Fatty Acids and AMD IOVS jMarch 2014 jVol. 55 jNo. 3 j2014
In the present study, a high RBCM EPAþDHA index (omega-3
index) was signiﬁcantly associated with a 48% reduction of the
odds of neovascular AMD. The associations of neovascular
AMD with EPA status also appeared particularly strong (OR ¼
0.25, P<0.0001 for RBCM EPA and OR ¼0.41 P¼0.005 for
In the present study, the results of seafood consumption
are consistent with previous dietary studies. Although AMD
patients had signiﬁcantly lower oily ﬁsh and seafood intake
than controls, associations did not reach statistical signiﬁcance
after adjustment for all potential confounders. Among
published case-control studies reporting associations between
ﬁsh consumption and AMD, one found a signiﬁcant associa-
whereas 3 studies, including the Age-Relate Eye Disease
Study (AREDS), showed no signiﬁcant association.
over, in 2008, a meta-analysis estimated that the risk for late
AMD was reduced by 38% in participants with high dietary
intakes of n-3 LC-PUFAs.
Since then, 4 large prospec-
and 4 large cross-sectional
studies published consistent and similar results.
The present results for serum EPAþDHA are consistent
with the only published study on plasma n-3 LC-PUFAs in
AMD, from the population-based Alienor Study.
showed a 33% decreased risk for neovascular AMD in subjects
with high plasma n-3 LC-PUFAs; however, not reaching
statistical signiﬁcance (OR ¼0.67, P¼0.08).
AMD risk was found here, in a new and independent sample
of the French population, in the same range (OR ¼0.74, P¼
0.35) for serum EPAþDHA. In the Alienor study, plasma EPA
was not associated with neovascular AMD (P¼0.51), while
plasma DHA was borderline with neovascular AMD (P¼0.06).
In the present study, we found a signiﬁcant association with
serum EPA (P¼0.005), but not with serum DHA (P¼0.81).
To our knowledge, the present study is the ﬁrst case-
control study reporting associations of RBCM n-3 long-chain
fatty acids with neovascular AMD. We showed signiﬁcant and
strong associations of neovascular AMD with RBCM EPA and
RBCM EPAþDHA. As expected, association with AMD was
stronger for RBCM than serum measurements, because EPA or
DHA measured in RBCM are more stable and longer-term
biomarkers of body LC-PUFAs homeostasis and less inﬂuenced
by lifestyle or other endogenous factors than EPAþDHA in
serum or plasma.
In the present study, associations of neovascular AMD with
circulating EPA (in serum and RBCM) were markedly stronger
than with circulating DHA. This could reﬂect differences in
endogenous metabolism of n-3 LC-PUFA, which could be
visible more readily through circulating EPA than through
circulating DHA. For example, there is high interindividual
variability with different tissue-speciﬁc rates of EPA/DHA
interconversion, depending on age, sex, nutritional, or
Moreover, although DHA is quantita-
tively more abundant than EPA in serum or cell membranes,
changes in serum and RBCM EPA are more pronounced than
serum or RBCM DHA, with changes in dietary intakes of
EPAþDHA, even in subjects taking n-3 LC-PUFA oral supple-
ments exclusively enriched in DHA.
Alternately, the protec-
tive role of EPA is supported by oxidative metabolism by
cyclooxygenases and lipoxygenases to produce eicosanoids
with vasoregulatory and anti-inﬂammatory properties in the
The EPA also is the precursor of docosapentaenoic
acid (DPA), which is known to be the potential precursor of n-
3 very long chain PUFAs (VLC-PUFAs), including 24:5 n-3 fatty
acid, the most abundant VLC-PUFA present in the retina.
recent study has observed a decreased of some n-3 VLC-PUFAs
(notably 24:5 n-3) in early and intermediate AMD retinas as
TABLE 3. Variations of Circulating n-3 PUFAs According to Dietary Intake of Seafood
Seafood Tertile (Range, g/d)
SERUM (% of Fatty Acids) Median (Fifth–95th Percentiles) RBCM (% of Fatty Acids) Median (Fifth–95th Percentiles)
EPA P* DHA PEPAþDHA PEPA PDHA PEPAþDHA P
Total fish 1, n¼151 (0–12.8) 0.60 (0.22–1.20) <0.0001 1.20 (0.60–2.20) 0.0004 1.77 (0.90–3.10) <0.0001 0.60 (0.29–1.00) <0.0001 3.00 (1.70–5.10) <0.0001 3.70 (1.90–5.70) <0.0001
2, n¼147 (12.8–23.0) 0.70 (0.20–1.60) 1.30 (0.63–2.40) 2.00 (1.00–3.52) 0.60 (0.30–1.18) 3.22 (2.00–4.90) 3.92 (2.40–5.83)
3, n¼125 (23.0–139.0) 0.76 (0.40–2.20) 1.40 (0.73–2.38) 2.20 (1.20–4.77) 0.80 (0.40–1.60) 3.70 (2.37–5.30) 4.50 (2.90–6.68)
Oily fish 1, n¼198 (0–5.4) 0.60 (0.23–1.34) 0.0008 1.20 (0.60–2.20) 0.02 1.80 (1.00–3.40) 0.002 0.60 (0.24–1.12) <0.0001 3.00 (1.60–5.10) 0.0001 3.70 (2.00–5.77) <0.0001
2, n¼125 (5.4–12.0) 0.75 (0.20–2.00) 1.30 (0.60–2.60) 2.00 (0.90–4.00) 0.79 (0.40–1.40) 3.40 (2.20–5.00) 4.29 (2.60–6.20)
3, n¼100 (12.0–100.0) 0.70 (0.30–2.10) 1.37 (0.80–2.31) 2.20 (1.24–4.65) 0.71 (0.31–1.60) 3.71 (2.28–5.30) 4.55 (2.81–6.70)
White fish 1, n¼156 (0–9.0) 0.60 (0.22–1.23) 0.004 1.20 (0.60–2.10) 0.002 1.82 (0.90–3.10) <0.0001 0.60 (0.30–1.10) 0.0002 3.20 (1.81–5.10) 0.08 3.86 (2.20–5.80) 0.01
2, n¼135 (9.0–14.0) 0.70 (0.24–1.70) 1.30 (0.70–2.40) 1.90 (1.00–3.70) 0.60 (0.29–1.20) 3.30 (1.90–4.80) 3.90 (2.20–5.80)
3, n¼132 (14.0–69.0) 0.70 (0.25–2.15) 1.40 (0.70–2.38) 2.20 (1.10–4.00) 0.70 (0.40–1.60) 3.55 (1.80–5.30) 4.30 (2.39–6.40)
1, n¼254 (0–2.6) 0.60 (0.20–1.40) 0.05 1.27 (0.63–2.20) 0.01 1.90 (1.0–3.41) 0.002 0.60 (2.29–1.16) 0.008 3.20 (1.80–5.32) 0.03 3.80 (2.10–6.29) 0.003
2, n¼86 (2.6–7.0) 0.67 (0.29–1.82) 1.23 (0.60–2.30) 1.90 (1.00–4.13) 0.61 (0.33–1.40) 3.32 (2.00–4.96) 4.10 (2.60–5.70)
3, n¼83 (7.0–62.9) 0.73 (0.30–2.00) 1.40 (0.80–2.40) 2.20 (1.20–4.00) 0.70 (0.33–1.56) 3.67 (2.20–4.90) 4.50 (2.60–5.80)
1, n¼142 (0–15.7) 0.60 (0.25–1.10) <0.0001 1.17 (0.60–2.20) <0.0001 1.70 (1.00–3.10) <0.0001 0.57 (0.28–0.98) <0.001 3.00 (1.80–5.10) 0.001 3.65 (2.10–5.70) <0.0001
2, n¼142 (15.7–26.0) 0.60 (0.18–1.42) 1.29 (0.60–2.10) 1.90 (0.90–3.41) 0.60 (0.30–1.12) 3.29 (1.80–4.94) 3.91 (2.30–5.83)
3, n¼139 (26.0–155.4) 0.80 (0.40–2.20) 1.40 (0.71–2.40) 2.26 (1.20–4.40) 0.80 (0.40–1.60) 3.70 (2.20–5.10) 4.50 (2.60–6.40)
*Pfor Kruskal-Wallis ANOVA.
Circulating Omega-3 Fatty Acids and AMD IOVS jMarch 2014 jVol. 55 jNo. 3 j2015
compared to age-matched control.
Finally, two randomized,
prospective, placebo-controlled, clinical trials have tested the
efﬁciency of oral n-3 LC-PUFAs supplementation on late AMD
First, the NAT2 study found no effect of a
three-year oral EPAþDHA (1:3, EPA:DHA [mg/mg ratio] from
ﬁsh-oil) on progression from early AMD to neovascular AMD, in
the second eye of patients with unilateral neovascular AMD at
Second, AREDS2 primary analyses showed that
addition of luteinþzeaxanthin, EPAþDHA (2:1, EPA:DHA [mg/
mg ratio] from ethyl esters) or both to the AREDS formulation
did not further reduce the 5-year risk of progression from early
to late AMD (geographic or neovascular AMD).
in placebo groups from both trials, incidence of late AMD at
follow-up was lower than that expected from observational
studies, suggesting that trial-effects (e.g., healthy lifestyle,
unreported self-supplementation in LC-PUFA, and so forth)
might have reduced statistical study power in both randomized
trials. Therefore, these two recent clinical trials, may not
challenge more than one decade of observational studies in
favor of a protective effect of dietary n-3 PUFAs on AMD. The
AREDS study recently published that 5 years after the clinical
trial end, the beneﬁcial effects of the AREDS formulation
persisted for development of neovascular AMD, suggesting a
potential long-term effect of nutritional factors involved in
TABLE 5. Associations of Circulating n-3 PUFAs With Neovascular AMD.
% of Fatty Acids
Model 1* Model 2†
OR 95% CI Pfor Trend OR 95% CI Pfor Trend
EPA 1 0–0.5 1.00 Ref 0.04 1.00 Ref 0.005
2 0.5–0.9 0.61 0.37–1.00 0.50 0.27–0.91
3 0.9–3.7 0.59 0.36–0.98 0.41 0.22–0.77
DHA 1 0–1.1 1.00 Ref 0.46 1.00 Ref 0.81
2 1.1–1.5 0.66 0.40–1.07 0.69 0.39–1.24
3 1.5–3.9 1.23 0.74–2.04 1.10 0.60–2.01
EPAþDHA 1 0–1.7 1.00 Ref 0.87 1.00 Ref 0.35
2 1.7–2.4 1.10 0.67–1.80 0.95 0.53–1.72
3 2.4–7.5 0.96 0.58–1.59 0.74 0.40–1.38
EPA 1 0–0.5 1.00 Ref <0.0001 1.00 Ref <0.0001
2 0.5–0.8 0.63 0.37–1.09 0.46 0.24–0.87
3 0.8–3.4 0.33 0.20–0.55 0.25 0.13–0.47
DHA 1 0–2.9 1.00 Ref 0.09 1.00 Ref 0.37
2 2.9–3.9 0.51 0.31–0.83 0.59 0.33–1.07
3 3.9–7.3 0.64 0.38–1.07 0.76 0.41–1.39
EPAþDHA 1 0–3.5 1.00 Ref 0.002 1.00 Ref 0.03
2 3.5–4.6 0.53 0.32–0.89 0.60 0.33–1.10
3 4.6–9.3 0.44 0.27–0.74 0.52 0.29–0.94
* Model 1, OR estimated using logistic regression adjusted for age and sex; AMD patients, n¼290; controls, n¼144.
† Model 2, OR estimated using logistic regression adjusted for age, sex, CFH Y402H,ARMS2 A69S, and ApoE4 polymorphisms, plasma
triglycerides, hypertension, hypercholesterolemia and family history of AMD; AMD patients, n¼290; controls, n¼144.
TABLE 4. Associations of Dietary Intake of Seafood With Neovascular AMD
of Seafood Tertile Range, g/d
Model 1* Model 2†
OR 95% CI Pfor Trend OR 95% CI Pfor Trend
Total fish 1 0–12.8 1.00 Ref 0.04 1.00 Ref 0.21
2 12.8–23.0 0.63 0.38–1.05 0.55 0.30–1.00
3 23.0–139.0 0.57 0.34–0.97 0.69 0.37–1.29
Oily fish 1 0–5.4 1.00 Ref 0.13 1.00 Ref 0.56
2 5.4–12.0 0.85 0.52–1.39 0.99 0.55–1.80
3 12.0–100.0 0.67 0.40–1.12 0.82 0.44–1.53
White fish 1 0–9.0 1.00 Ref 0.34 1.00 Ref 0.17
2 9.0–14.0 1.00 0.60–1.67 1.25 0.68–2.29
3 14.0–69.0 0.79 0.47–1.29 0.63 0.34–1.15
Other seafood 1 0–2.6 1.00 Ref 0.10 1.00 Ref 0.64
2 2.6–7.0 0.60 0.36–1.01 0.59 0.32–1.11
3 7.0–62.9 0.71 0.42–1.20 0.98 0.52–1.86
Total seafood 1 0–15.7 1.00 Ref 0.05 1.00 Ref 0.22
2 15.7–26.0 0.60 0.36–1.01 0.50 0.27–0.92
3 26.0–155.4 0.59 0.35–0.99 0.68 0.36–1.28
* Model 1, OR estimated using logistic regression adjusted for age and sex; AMD patients, n¼284; controls, n¼139.
† Model 2, OR estimated using logistic regression adjusted for age, sex, CFH Y402H,ARMS2 A69S, and ApoE4 polymorphisms, plasma
triglycerides, hypertension, hypercholesterolemia, and family history of AMD; AMD patients, n¼284; controls, n¼139.
Circulating Omega-3 Fatty Acids and AMD IOVS jMarch 2014 jVol. 55 jNo. 3 j2016
Moreover, in the NAT2 study, the 3-year
incidence of CNV was reduced signiﬁcantly (hazard ratio [HR],
0.32; 95% conﬁdence interval [CI], 0.10–0.99; P¼0.047) in
patients achieving the highest RBCM EPAþDHA (omega-3 index
>8) over 3 years.
From these combined results, it seems to
be relevant to analyze n-3 RBCM EPAþDHA status in AMD.
Biological status of n-3 PUFAs could help identify those
subjects at risk for AMD, and RBCM n-3 PUFAs appear more
relevant as a biomarker of AMD.
Strength of our study was the combined use of biological
data, mainly EPAþDHA RBCM measurements with dietary
assessment of n-3 PUFA status, in the same groups of
individuals affected or not with AMD. Indeed, from differences
in well-established risk factors (age, medical history, CFH,
ARMS2, and APOE polymorphisms) found with a group of
normal vision/normal fundus individuals, the AMD group
seemed as typical of a population of patients with exudative
AMD. Although apparently paradoxical, that triglycerides were
found signiﬁcantly lower in AMD patients despite them being
more numerous with dyslipidemia, may be somewhat expect-
ed since the whole population had plasma triglyceride
concentrations within the normal range, including AMD
patients regularly taking lipid-lowering medications. Finally,
the omega-3 index (EPAþDHA index) measured in RBCM is a
very good biomarker of n-3 PUFAs status in humans and
recognized as a risk factor in cardiovascular diseases.
future, it may prove useful in the clinical setting, for the
identiﬁcation of AMD patients deﬁcient in n-3 LC-PUFAs, which
may beneﬁt the most from nutritional intervention.
Selection of controls always is a concern in case-control
studies, selection bias being difﬁcult to avoid.
In the present
study, controls were selected from the general population, in
the same geographic area as cases. They were not aware of the
speciﬁc objectives of the study, before the interview and blood
sample. When we compared cases and controls, they were not
different for sex, smoking, BMI, diabetes, and plasma
cholesterol. However, cases were older than controls. Also,
hypercholesterolemia and hypertension were more frequent in
cases, which is partially consistent with previous studies.
Our two groups also were comparable for dietary intakes. To
limit the potential bias due to differences in age, hypertension,
or hypercholesterolemia, we used multivariate modeling.
However, despite that we adjusted our analyses for these
potential confounders, as well as major AMD-related genes, we
cannot exclude residual confounding as in all epidemiologic
Also, as our study focused on neovascular AMD cases only,
our results can be generalized only to this type of AMD.
In conclusion, from the present report, elderly individuals
with high RBCM levels of EPAþDHA, a long-term marker of
intracellular LC-PUFAs, have a strongly reduced risk for
neovascular AMD. This suggests the RBCM EPAþDHA index
to be considered as added to the list of clinically relevant
biomarkers of AMD.
The authors thank all physicians, nurses, and patients from Creteil
University Eye Clinic; all biologists, in particular Claude Wolf, for
scientiﬁc advice and support, all laboratory technicians, particu-
larly Dominique Farabos, Myriam Mahe, and Dominique Labaud,
for excellent technical assistance, from Biochimie B Laboratory
from Saint Antoine Hospital.
Supported by Laboratoire Bausch & Lomb, Clinical Research,
Montpellier, France, and by a grant from F´
eration des aveubles et
handicaps visuels de France and from Fondation Nestl´
Disclosure: B.M.J. Merle,F
eration des aveugles et handicap´
visuels de France (F), Fondation Nestl´
e France (F), Laboratoires
ea (R), Bausch & Lomb (R); P. Benlian, Bausch & Lomb (R); N.
Puche, None; A. Bassols, Bausch & Lomb (E); C. Delcourt,
ea (C), Bausch & Lomb (C, R), Novartis (C),
ea (R); E.H. Souied, Bausch & Lomb (F, C, R),
ea (C), Laboratoires Th´
1. Lim LS, Mitchell P, Seddon JM, Holz FG, Wong TY. Age-related
macular degeneration. Lancet. 2012;379:1728–1738.
2. SanGiovanni JP, Chew EY. The role of omega-3 long-chain
polyunsaturated fatty acids in health and disease of the retina.
Prog Retin Eye Res. 2005;24:87–138.
3. Bazan NG. Neuroprotectin D1-mediated anti-inﬂammatory and
survival signaling in stroke, retinal degenerations, and
Alzheimer’s disease. J Lipid Res. 2009;50(suppl):S400–S405.
4. Donoso LA, Kim D, Frost A, Callahan A, Hageman G. The role
of inﬂammation in the pathogenesis of age-related macular
degeneration. Surv Ophthalmol. 2006;51:137–152.
5. Delyfer MN, Buaud B, Korobelnik JF, et al. Association of
macular pigment density with plasma omega-3 fatty acids: the
PIMAVOSA study. Invest Ophthalmol Vis Sci. 2012;53:1204–
6. Connor KM, SanGiovanni JP, Lofqvist C, et al. Increased dietary
intake of omega-3-polyunsaturated fatty acids reduces patho-
logical retinal angiogenesis. Nat Med. 2007;13:868–873.
7. Chong EW, Kreis AJ, Wong TY, Simpson JA, Guymer RH.
Dietary omega-3 fatty acid and ﬁsh intake in the primary
prevention of age-related macular degeneration: a systematic
review and meta-analysis. Arch Ophthalmol. 2008;126:826–
8. Chua B, Flood V, Rochtchina E, Wang JJ, Smith W, Mitchell P.
Dietary fatty acids and the 5-year incidence of age-related
maculopathy. Arch Ophthalmol. 2006;124:981–986.
9. Cho E, Hung S, Willett WC, et al. Prospective study of dietary
fat and the risk of age-related macular degeneration. Am J Clin
10. Arnarsson A, Sverrisson T, Stefansson E, et al. Risk factors for
ﬁve-year incident age-related macular degeneration: the
Reykjavik Eye Study. Am J Ophthalmol. 2006;142:419–428.
11. Seddon JM, Rosner B, Sperduto RD, et al. Dietary fat and risk
for advanced age-related macular degeneration. Arch Oph-
12. Seddon JM, George S, Rosner B. Cigarette smoking, ﬁsh
consumption, omega-3 fatty acid intake, and associations with
age-related macular degeneration: the US Twin Study of Age-
Related Macular Degeneration. Arch Ophthalmol. 2006;124:
13. SanGiovanni JP, Chew EY, Clemons TE, et al. The relationship
of dietary lipid intake and age-related macular degeneration in
a case-control study: AREDS Report No. 20. Arch Ophthalmol.
14. Mares-Perlman JA, Brady WE, Klein R, VandenLangenberg GM,
Klein BE, Palta M. Dietary fat and age-related maculopathy.
Arch Ophthalmol. 1995;113:743–748.
15. Heuberger RA, Mares-Perlman JA, Klein R, Klein BE, Millen AE,
Palta M. Relationship of dietary fat to age-related maculopathy
in the Third National Health and Nutrition Examination
Survey. Arch Ophthalmol. 2001;119:1833–1838.
16. Delcourt C, Carriere I, Cristol JP, Lacroux A, Gerber M. Dietary
fat and the risk of age-related maculopathy: the POLANUT
study. Eur J Clin Nutr. 2007;61:1341–1344.
17. Robman L, Vu H, Hodge A, et al. Dietary lutein, zeaxanthin,
and fats and the progression of age-related macular degener-
ation. Can J Ophthalmol. 2007;42:720–726.
Circulating Omega-3 Fatty Acids and AMD IOVS jMarch 2014 jVol. 55 jNo. 3 j2017
18. Augood C, Chakravarthy U, Young I, et al. Oily ﬁsh
consumption, dietary docosahexaenoic acid and eicosapen-
taenoic acid intakes, and associations with neovascular age-
related macular degeneration. Am J Clin Nutr. 2008;88:398–
19. SanGiovanni JP, Chew EY, Agron E, et al. The relationship of
dietary omega-3 long-chain polyunsaturated fatty acid intake
with incident age-related macular degeneration: AREDS report
no. 23. Arch Ophthalmol. 2008;126:1274–1279.
20. Tan JS, Wang JJ, Flood V, Mitchell P. Dietary fatty acids and the
10-year incidence of age-related macular degeneration: the
Blue Mountains Eye Study. Arch Ophthalmol. 2009;127:656–
21. SanGiovanni JP, Agron E, Clemons TE, Chew EY. Omega-3 long-
chain polyunsaturated fatty acid intake inversely associated
with 12-year progression to advanced age-related macular
degeneration. Arch Ophthalmol. 2009;127:110–112.
22. Parekh N, Voland RP, Moeller SM, et al. Association between
dietary fat intake and age-related macular degeneration in the
Carotenoids in Age-Related Eye Disease Study (CAREDS): an
ancillary study of the Women’s Health Initiative. Arch
23. Merle B, Delyfer MN, Korobelnik JF, et al. Dietary omega-3 fatty
acids and the risk for age-related maculopathy: the Alienor
Study. Invest Ophthalmol Vis Sci. 2011;52:6004–6011.
24. Christen WG, Schaumberg DA, Glynn RJ, Buring JE. Dietary
omega-3 fatty acid and ﬁsh intake and incident age-related
macular degeneration in women. Arch Ophthalmol. 2011;129:
25. Swenor BK, Bressler S, Caulﬁeld L, West SK. The impact of ﬁsh
and shellﬁsh consumption on age-related macular degenera-
tion. Ophthalmology. 2010;117:2395–2401.
26. Reynolds R, Rosner B, Seddon JM. Dietary omega-3 fatty acids,
other fat intake, genetic susceptibility, and progression to
incident geographic atrophy. Ophthalmology. 2013;120:1020–
27. Serra-Majem L, Nissensohn M, Overby NC, Fekete K. Dietary
methods and biomarkers of omega 3 fatty acids: a systematic
review. Br J Nutr. 2012;107(suppl 2):S64–S76.
28. Arab L. Biomarkers of fat and fatty acid intake. J Nutr. 2003;
29. Arterburn LM, Hall EB, Oken H. Distribution, interconversion,
and dose response of n-3 fatty acids in humans. Am J Clin
30. Feart C, Peuchant E, Letenneur L, et al. Plasma eicosapentae-
noic acid is inversely associated with severity of depressive
symptomatology in the elderly: data from the Bordeaux
sample of the Three-City Study. Am J Clin Nutr. 2008;87:
31. Samieri C, Feart C, Letenneur L, et al. Low plasma
eicosapentaenoic acid and depressive symptomatology are
independent predictors of dementia risk. Am J Clin Nutr.
32. Wilk JB, Tsai MY, Hanson NQ, Gaziano JM, Djousse L. Plasma
and dietary omega-3 fatty acids, ﬁsh intake, and heart failure
risk in the Physicians’ Health Study. Am J Clin Nutr. 2012;96:
33. Kroger E, Verreault R, Carmichael PH, et al. Omega-3 fatty
acids and risk of dementia: the Canadian Study of Health and
Aging. Am J Clin Nutr. 2009;90:184–192.
34. Djousse L, Biggs ML, Lemaitre RN, et al. Plasma omega-3 fatty
acids and incident diabetes in older adults. Am J Clin Nutr.
35. Merle BM, Delyfer MN, Korobelnik JF, et al. High concentra-
tions of plasma n3 fatty acids are associated with decreased
risk for late age-related macular degeneration. J Nutr. 2013;
36. Souied EH, Delcourt C, Querques G, et al. Oral docosahex-
aenoic acid in the prevention of exudative age-related macular
degeneration: the Nutritional AMD Treatment 2 Study.
37. Dole VP, Meinertz H. Microdetermination of long-chain fatty
acids in plasma and tissues. J Biol Chem. 1960;235:2595–
38. Benlian P, Cansier C, Hennache G, et al. Comparison of a new
method for the direct and simultaneous assessment of LDL-
and HDL-cholesterol with ultracentrifugation and established
methods. Clin Chem. 2000;46:493–505.
39. Leveziel N, Souied EH, Richard F, et al. PLEKHA1-LOC387715-
HTRA1 polymorphisms and exudative age-related macular
degeneration in the French population. Mol Vis. 2007;13:
40. Bonifacj C, Gerber M, Scali J, Daures JP. Comparison of dietary
assessment methods in a southern French population: use of
weighed records, estimated-diet records and a food-frequency
questionnaire. Eur J Clin Nutr. 1997;51:217–231.
41. Carriere I, Delcourt C, Lacroux A, Gerber M. Nutrient intake in
an elderly population in southern France (POLANUT):
deﬁciency in some vitamins, minerals and omega-3 PUFA. Int
J Vitam Nutr Res. 2007;77:57–65.
42. Favier J, Ireland-Ripert J, Toque C, Feinberg M. R´
eral des Aliments. Table de Composition, 2nd ed. Paris,
France: Editions Tec et Doc Lavoisier et INRA ´
43. Hercberg S. Table de Composition des Aliments SU.VI.MAX.
Paris, France: Editions INSERM; 2005.
44. Liu A, Chang J, Lin Y, Shen Z, Bernstein PS. Long-chain and
very long-chain polyunsaturated fatty acids in ocular aging and
age-related macular degeneration. J Lipid Res. 2010;51:3217–
45. Age-Related Eye Disease Research Group. Lutein þzeaxanthin
and omega-3 fatty acids for age-related macular degeneration:
the Age-Related Eye Disease Study 2 (AREDS2) Randomized
Clinical Trial. JAMA. 2013;1–11.
46. Chew EY, Clemons TE, Agron E, et al. Long-term effects of
vitamins C and E, beta-carotene, and zinc on age-related
macular degeneration: AREDS Report No. 35. Ophthalmology.
47. von Schacky C. The Omega-3 Index as a risk factor for
cardiovascular diseases. Prostaglandins Other Lipid Mediat.
48. Rothman K, Greenland S. Modern Epidemiology, 2nd ed.
Philadelphia, PA: Lippincott Williams & Wilkins; 1998.
49. Chakravarthy U, Wong TY, Fletcher A, et al. Clinical risk factors
for age-related macular degeneration: a systematic review and
meta-analysis. BMC Ophthalmol. 2010;10:31.
Nutritional AMD Treatment 2 Study Group (alphabetic
Catherine Allaire, MD, Laboratoires Bausch & Lomb, Montpel-
lier, France; Ana Bassols, MD, Laboratoires Bausch & Lomb,
Montpellier, France; Khaldia Belabbas, APHP, Hˆ
Antoine, Laboratoire de Biochimie B, Paris, France; Dominique
Brault, Laboratoires Bausch & Lomb, Montpellier, France; Yves
Brouquet, Laboratoires Bausch & Lomb, Montpellier, France;
ephanie Castagnet, APHP, Hˆ
opital Saint Antoine, Laboratoire
de Biochimie B, Paris, France; Antoine Cri´
e, APHP, Hˆ
Antoine, Laboratoire de Biochimie B, Paris, France; Isabelle
Gaudino, APHP, Hˆ
opital Saint Antoine, Laboratoire de Bio-
chimie B, Paris, France; Patricia Gawrilow, MD, Department of
Ophthalmology, NYU School of Medicine, New York, NY;
ele Lablache-Combier, PhD, Laboratoires Bausch & Lomb,
Montpellier, France; Nicolas Leveziel, MD, PhD, Service
Circulating Omega-3 Fatty Acids and AMD IOVS jMarch 2014 jVol. 55 jNo. 3 j2018
opital Intercommunal de Cr´
Paris Est Cr´
eteil, France; Nadja Mechai, PhD, MSc,
Laboratoires Bausch & Lomb, Montpellier, France; Gilles
Morineau, PharmD, PhD, Laboratoires Bausch & Lomb,
Montpellier, France; Natasa Orlic-Pleyer, MD, Laboratoires
Bausch & Lomb, Montpellier, France; Brigitte Paccou, Service
opital Intercommunal de Cr´
Paris Est Cr´
eteil, France; Nicole Pumariega, Depart-
ment of Ophthalmology, NYU School of Medicine, New York,
NY; Giuseppe Querques, Service d’Ophtalmologie, Hˆ
Intercommunal de Cr´
e Paris Est Cr´
ele Siou-Mermet, MD, MS, Laboratoires Bausch &
Lomb, Montpellier, France; Isabelle Turquois, Laboratoires
Bausch & Lomb, Montpellier, France.
Circulating Omega-3 Fatty Acids and AMD IOVS jMarch 2014 jVol. 55 jNo. 3 j2019