Biochemical and Biophysical Research Communications 483 (2017) 425e429

Contents lists available at ScienceDirect

Biochemical and Biophysical Research Communications

journal homepage: www.elsevier.com/locate/ybbrc

Omega-3 fatty acid fish oil dietary supplements contain saturated fats
and oxidized lipids that may interfere with their intended biological
benefits
R. Preston Mason a, b, *, Samuel C.R. Sherratt b
a
Department of Medicine, Cardiovascular Division, Brigham and Women's Hospital, Harvard Medical School, Boston, MA, 02115-6110, USA
b
Elucida Research LLC, Beverly, MA, 01915-0091, USA

a r t i c l e i n f o a b s t r a c t

Article history: Widely available fish oil dietary supplements (DS) may contain fats and oxidized lipids in addition to the
Received 14 December 2016 beneficial omega-3 fatty acids (OM3FAs) for which they are purchased. Little is known about the po-
Accepted 19 December 2016 tential biological effects of these oxidized lipids. The objective of this study was to assess the fatty acid
Available online 21 December 2016
content, oxidation products, and biological effects of leading fish oil DS available in the United States.
Three top-selling fish oil DS in the US were included in this analysis. Fatty acid composition was
Keywords:
measured using gas chromatography. Lipid oxidation (primary and secondary products) was measured
Omega-3 fatty acids
by spectroscopy in both DS and a prescription OM3FA product. OM3FAs were also isolated and
Fish oils
Dietary supplements
concentrated from DS and were tested for the ability to inhibit copper-induced oxidation of human small
Lipid oxidation dense low-density lipoprotein particles (sdLDL) in vitro. Fish oil DS were found to contain more than 30
different fatty acids, including 10 to 14 different saturated species comprising up to 36% of the total fatty
acid content. Levels of OM3FAs also varied widely among DS (33%e79%). Primary (peroxide), secondary
(anisidine), and total oxidation products exceeded maximum levels established by international stan-
dards of quality in the DS but not the prescription OM3FA product. Oxidation of sdLDL was inhibited by
>95% (P < 0.001) with non-oxidized forms of OM3FA but not with OM3FAs isolated from DS, which were
a mixture of oxidized and non-oxidized OM3FAs. These data indicate that levels of saturated fat and
oxidized OM3FAs found in common DS may interfere with their intended/potential biological benefits.
© 2016 The Authors. Published by Elsevier Inc. This is an open access article under the CC BY-NC-ND
license (http://creativecommons.org/licenses/by-nc-nd/4.0/).

1. Introduction
Fish oil dietary supplements (DS) contain the omega-3 fatty
acids (OM3FAs) eicosapentaenoic acid (EPA) and docosahexaenoic
acid (DHA) and are widely available at various pharmacies and
other retail outlets throughout the United States without a pre-
scription. The American Heart Association (AHA)-recommended
intake of EPA þ DHA has been 2e4 g/day for patients with elevated
triglyceride levels [1] and prescription OM3FA products are indi-
cated as an adjunct to diet to reduce triglyceride levels in patients
with severe (500 mg/dL) hypertriglyceridemia at a dose of 4 g/day
[2,3]. Frequently, patients may choose to substitute fish oil DS for
prescription OM3FA products, and some managed care plans may
* Corresponding author. P.O. Box 7100, Beverly, MA, 01915-0091, USA.
E-mail addresses: rpmason@elucidaresearch.com (R.P. Mason), ssherratt@
elucidaresearch.com (S.C.R. Sherratt).
even require prior failure on fish oil DS before covering prescription
OM3FA products. Although DS are widely available, they are not
over-the-counter drugs and are not regulated as drugs; their purity,
chemical integrity, efficacy, and safety remain unverified [4]. In
order to attain prescription-strength or AHA-recommended
amounts of OM3FAs, patients may need to take 10 or more fish
oil DS capsules per day, potentially negating purported cost ad-
vantages over prescription products and increasing intake of
saturated fats [5] and oxidized lipids.
Studies have raised concerns about the OM3FA content and
chemical integrity of fish oil DS. In an analysis of over 45 different
products, over 70% were shown to contain less EPA and/or DHA
than stated on their labels [6]. Notably, this study was funded by the
US Department of Agriculture. Furthermore, unacceptably high
levels of oxidized lipids, an indication of lipid decomposition, were
found in over 80% of more than 35 fish oil DS from New Zealand,
with only 8% meeting international standards for acceptable
peroxide and total oxidation levels [7]. This was mirrored in a study

http://dx.doi.org/10.1016/j.bbrc.2016.12.127
0006-291X/© 2016 The Authors. Published by Elsevier Inc. This is an open access article under the CC BY-NC-ND license (http://creativecommons.org/licenses/by-nc-nd/4.0/).
426 R.P. Mason, S.C.R. Sherratt / Biochemical and Biophysical Research Communications 483 (2017) 425e429
of over 15 top-selling fish oil DS in North America, in which many
DS were shown to have unacceptably high levels of peroxides, with
more than half not meeting label claims for EPA and DHA content
[8] and in a study of 171 North American fish oil DS available in
Canada wherein 50% exceeded the recommended levels for
markers of oxidation [9].
Oxidized lipids have a central role in atherogenesis and may
play a role in both vascular injury and insulin resistance [10e12].
Ingestion of oxidized lipids from DS can lead to increases in
circulating oxidized lipid levels [13], and elevated oxidized lipid
levels correlate with increased cardiovascular risk in patients with
coronary disease [14,15]. In addition to cell membrane-mediated
damage, oxidized lipids associated with low-density lipoprotein
(LDL) contribute to endothelial dysfunction, inflammation, and
atherosclerotic foam cell formation [16,17]. Small dense LDL
(sdLDL) is especially atherogenic compared with larger LDL par-
ticles [18,19] and are more prone to oxidation than larger LDL
subfractions [20]. A potential mechanism for the observed reduc-
tion in cardiovascular events for the OM3FA EPA [21] may be
through decreasing LDL oxidation as we have recently demon-
strated [22]. The ability of OM3FA to interfere with lipid oxidation
is related to its chemical integrity. Oxidative modification of
OM3FAs in DS may interfere with their intended biological or
clinical benefits [23,24].
The objective of this study was to test the fatty acid content (eg,
saturated fat, EPA, DHA) of three top-selling fish oil DS in the United
States and to examine the extent of oxidative damage in the DS as
compared with chemically characterized standards with respect to
purity and ability to prevent human sdLDL oxidation in vitro.
2. Materials and methods
2.1. Materials
EPA and DHA were purchased from Sigma-Aldrich (Saint Louis,
MO) and solubilized in redistilled ethanol under nitrogen atmo-
sphere. Lipid stock solutions were stored at 20  C until use. Pre-
scription and DS forms of OM3FAs were purchased from retail
stores near Boston, MA, and stored in original packaging under
ambient conditions until tested. Oil was aspirated from individual
capsules just prior to measuring sample oxidation levels and bio-
logical effects.
2.2. Fatty acid sample hydrolysis
DS generally consisted of fatty acids (FAs) in the triglyceride or
ethyl ester forms, which were hydrolyzed to generate free FA forms
for subsequent testing. Liberation of free FAs was verified by thin-
layer chromatography. The reaction mixture was partitioned to
remove water-soluble impurities and subjected to silica chroma-
tography to isolate pure FA potassium salts. These were converted
to protonated species by using a 0.1 M HClesolvent partition and
stored under nitrogen in light-protected containers at 20  C until
use.
2.3. Fatty acid methyl ester analysis
To analyze the FA content of each of the DS tested, individual FAs
were separated by their relative retention times following vapor-
ization. Hydrolysis and methylation occur at room temperature
when reacted with 1 N sodium methoxide. Individual FA methyl
esters were identified and quantitated using capillary gas chro-
matography (Hewlett Packard 5890 Gas Chromatograph) with
flame ionization detection (GC-FID). The FA levels were reported as
weight %.
2.4. Analysis of sample oxidation markers
Peroxide content, quantitated as the sample peroxide value,
predicts the extent to which an OM3FA and other polyunsaturated
FAs have undergone primary oxidation. The peroxide value is
defined as the amount of peroxide oxygen per 1 kg of fat or oil and
is usually expressed in units of milliequivalents (meq) per kg of fat.
Oils isolated from the DS samples were dissolved in a chloroform-
acetic acid mixture and subjected to an excess of iodide via a
saturated solution of potassium iodide using American Oil Chem-
ists' Society (AOCS) methods (AOCS Official Method Cb 8 b-9) [25].
The amount of iodine produced is directly proportional to the
peroxide value intrinsic to each sample.
We also measured anisidine values using accepted AOCS
methods (AOCS Official Method Cd 18-90) [25]. Anisidine reacts
with secondary oxidation products such as aldehydes and ketones
in fats and oils to form products that absorb at 350 nm. The total
oxidation level for a given fish oil is estimated as PV  2 þ AV
(where PV ¼ peroxide value and AV ¼ anisidine value). Recom-
mended maxima for peroxide values, anisidine values, and total
oxidation values are 5 meq/kg, 20, and 26, respectively [7,26].
2.5. Small dense low-density lipoprotein isolation and oxidation
sdLDL fractions were isolated from the plasma of healthy human
volunteers by iodixanol gradient ultracentrifugation and adjusted
to a final apolipoprotein (apoB100) concentration of 10 mg/mL [27].
Samples were prepared at 100 mg/mL sdLDL (based on apoB100),
treated with vehicle (ethanol) and various OM3FA samples, each at
10.0 mM, and adjusted to a final sample volume of 1.0 mL in phos-
phate buffered saline. Samples were incubated for 30 min at 37  C
in a shaking water bath before initiating lipid oxidation by adding
10.0 mM CuSO4. At different time points (up to 4 h), 100 mL aliquots
were removed from each sample and combined with 1.0 mL thio-
barbituric acid (0.5%), 10 mL trichloroacetic acid (10%), 10 mL butyl-
ated hydroxytoluene (35 mM in ethanol), and 10 mL
ethylenediaminetetraacetic acid (5 mM). Sample aliquots were
incubated at 100  C for 30 min and then assayed for the formation
of thiobarbituric acid-reactive substances (TBARS), which have a
molar absorptivity (ε) value of 1.56  105 M1cm1 at 532 nm and
are derived principally from the reaction of thiobarbituric acid with
malondialdehyde (MDA), a reactive aldehyde produced by oxida-
tion of the lipids in sdLDL [28,29]. Sample TBARS concentrations
were determined spectrophotometrically and expressed as molar
equivalents of MDA.
2.6. Calculations and statistical analyses
Data are presented as mean ± standard deviation (SD) for the
effects on human sdLDL oxidation. The significance of difference
between results from independent experimental conditions was
tested using either the unpaired, two-tailed Student's t-test or one-
way analysis of variance (ANOVA) with Student-Newman-Keuls
multiple comparisons post hoc analysis. Alpha error was set at 0.05.
3. Results
In Fig. 1A, the FA content of three widely used DS was deter-
mined by GC-FID. Thirty-five different FAs were identified in the DS
samples, including 10 to 14 saturated variants, which made up as
much as 36% of total FA content. The levels of saturated fats varied
by six-fold among these top-selling DS. Additionally, total OM3FA
(alpha-linolenic acid [ALA, 18:3], EPA [20:5], docosapentaenoic acid
[DPA, 22:5], and DHA [22:6]) levels varied widely among the DS
(33%e79%), including levels of EPA (21%e52%) and DHA (9%e31%).
 R.P. Mason, S.C.R. Sherratt / Biochemical and Biophysical Research Communications 483 (2017) 425e429 427

undesirable odor and/or color in the oil.

In this study, we measured both primary and secondary prod-
ucts of oxidation associated with FAs containing multiple double
bonds, such as OM3FAs. An elevated peroxide value indicates high
levels of primary oxidation and hydroperoxides. During the reac-
tion process, peroxide values increase over time and then drop as
the primary oxidation products decompose into secondary prod-
ucts upon continued exposure to oxidative conditions. The sec-
ondary oxidation products are measured with the p-anisidine test,
which uses a reagent that reacts with ketones and aldehydes to
produce a measurable chemical complex. This test is an indication
of the amount of aldehydes and ketones in the product. The total
oxidation level is derived from the peroxide and anisidine values.
Peroxide, anisidine, and total oxidation levels are shown in

Fig. 1. FA content of three top-selling (in the United States) fish oil dietary sup-
plements (DS1, DS2, and DS3). A) The content of individual FAs within each DS was
determined by GC-FID analysis and are shown using the carbon nomenclature. Each DS
contained more than 30 different FAs, including 10 to 14 different saturated fat species
comprising up to 36% of the total FA content. Levels of total OM3FAs (ALA [18:3], EPA
[20:5], DPA [22:5], and DHA [22:6]) also varied widely among the DS (33%e79%). Data
are presented as % of total FA (for a given DS sample) by weight. B) EPA, DHA, saturated
fats, and other fats (consisting of mono- and polyunsaturated FAs). Data are presented
as % of total FA (for a given DS sample) by weight. ALA, alpha linolenic acid; DHA,
docosahexaenoic acid; DPA, docosapentaenoic acid; DS, dietary supplement; EPA,
eicosapentaenoic acid; FA, fatty acid; GC-FID, gas chromatography with flame ioniza-
tion detection; OM3FA, omega-3 fatty acid.

In one product, a single species of saturated fat (palmitic acid, 16:0)

accounted for 19% of total fat content, exceeding DHA levels (9%)
two-fold. Another saturated fat, myristic acid (14:0), contributed 9%
and when combined with palmitic acid, these two FAs accounted
for 28% of the total FA content, which was similar to the amount of
total OM3FA in that product. The relative levels of EPA, DHA,
saturated, and unsaturated FAs expressed as percent of total FA
content by weight for these same DS are shown in Fig. 1B.
Oxidation of OM3FAs and other unsaturated FAs is a multistep
process that occurs after exposure to atmospheric oxygen during
Fig. 2. Products of oxidation of three top-selling (in the United States) fish oil
the manufacturing process. The initial target of oxidation results in
dietary supplements (DS1, DS2, DS3) and a prescription product. A) peroxide, B)
chemical formation of peroxides and dienes associated with the anisidine, and C) TOTOX values measured directly in total oil extract and then
multiple double bonds of the OM3FAs [30]. If exposure to oxygen normalized to a per-capsule prescription level of omega-3 fatty acid (1 g). Dashed lines
continues, the secondary oxidation products carbonyl and aldehyde in each panel represent recommended international thresholds for each marker
are produced [30]. These oxidation products may cause an (including those recommended by the US Council for Responsible Nutrition [26]). DS,
dietary supplement; Rx, prescription; TOTOX, total oxidation.
428 R.P. Mason, S.C.R. Sherratt / Biochemical and Biophysical Research Communications 483 (2017) 425e429
Fig. 2AeC. All three DS exceeded the recommended maxima for
peroxide and total oxidation values (5 meq/kg and 26, respectively)
when normalized to 1 g OM3FA (based on the number of capsules
needed to achieve 1 g of OM3FA). The prescription product of pure
OM3FA did not contain significant levels of these oxidation prod-
ucts under identical test conditions.
Finally, we tested the effects of OM3FAs isolated from a DS on
sdLDL oxidation and compared these effects with those of a non-
oxidized control of pure EPA þ DHA. We also deliberately
oxidized preparations of the pure EPA þ DHA mixture by exposure
to atmospheric oxygen for 5 h at room temperature. As shown in
Fig. 3, sdLDL oxidation was inhibited by >90% (P < 0.001) with non-
oxidized pure OM3FA but not with the isolate from a top-selling DS
containing a mixture of oxidized and non-oxidized OM3FAs. The
oxidized mixture of pure EPA þ DHA inhibited oxidation by 33 ± 4%,
while the OM3FA extracted from the DS inhibited oxidation by
21 ± 4%.
4. Discussion
Our study found that top-selling OM3FA DS purchased in the
United States contained more than 30 FAs, including significant
levels of saturated fat in addition to desirable OM3FAs. We also
observed levels of peroxides and secondary oxidation products that
exceeded international industry standards. The combination of
oxidized and non-oxidized lipids interfered with biological anti-
oxidant activity of the DS. These data indicate that levels of satu-
rated fat and oxidized lipids found in commonly available fish oil DS
may interfere with their potential/intended biological benefits. This
suggests that use of fish oil DS should be approached with caution
in healthy individuals and that it is inappropriate to substitute a fish
oil DS for a prescription OM3FA product in patients with
hypertriglyceridemia.
There is a perception among the public that consumption of fish
oil DS capsules is a healthy dietary habit [31], even though low-
dose supplementation with 1 g/day of OM3FAs has not demon-
strated cardiovascular benefit in recent large clinical trials [32e35].
This public perception contributed to a greater than 3-fold increase
Fig. 3. Effects of a DS fatty acid extract versus non-oxidized and partially oxidized
preparations of EPA and DHA on human sdLDL oxidation. Oxidized forms of EPA and
DHA were prepared by exposing the OM3FAs to atmospheric oxygen for 5 h. Each
agent was tested separately at 10.0 mM or in combination at 5.0 mM against vehicle-
treated controls. Values are mean ± SD (N ¼ 3). *P < 0.001 versus vehicle alone;
y
P < 0.001 versus DS; zP < 0.001 versus oxEPA þ oxDHA (Student-Newman-Keuls
multiple comparisons test; overall ANOVA: P < 0.001, F ¼ 993.26). ANOVA, analysis of
variance; DHA, docosahexaenoic acid; DS, dietary supplement; EPA, eicosapentaenoic
acid; MDA, malondialdehyde; ox, oxidized; SD, standard deviation; sdLDL, small dense
low-density lipoprotein; TOTOX, total oxidation.
in the use of these DS in the United States between 2005 and 2011
[36]. According to the National Health Statistics Report in 2008 and
an analysis of National Health and Nutrition Examination Survey
data from 2007 to 2010, fish oil DS are the most commonly used
nonvitamin/nonmineral DS among US adults [37,38]. In 2012, the
National Health Interview Survey showed that 7.8% of adults (19
million) had taken a fish oil DS in the previous 30 days [39].
Patients may be asked by physicians, pharmacists, or repre-
sentatives of managed care plans to take fish oil DS instead of
prescription products for very high triglyceride levels despite the
fact that they are not regulated by the FDA. However, patients may
not be aware of the need to take a large number of DS to try to
reproduce a prescription therapeutic dose of 4 g of OM3FA. Given
the variance in the amount of OM3FAs stated on DS labels versus
measured levels, patients cannot confidently know the actual dose
they are taking; surveys have shown substantial patient confusion
regarding OM3FA DS dose and content [40,41]. In this study, we
observed highly variable amounts of OM3FAs and other FAs in
leading DS. The top-selling DS examined in this study contained
more than 30 FAs, including significant levels of saturated fat in
addition to desirable OM3FAs. A diet high in saturated fat is asso-
ciated with increased risk for cardiovascular disease [42]. We also
observed elevated levels of peroxides and secondary oxidation
products, which interfered with biological antioxidant activity.
The results of this study are consistent with previously pub-
lished analyses of DS with respect to content and oxidative damage.
The human health effects of partially oxidized DS are not fully
known; however, oxidized lipids have demonstrated negative ef-
fects on atherogenic lipids and other biomarkers of cardiovascular
disease and may be predictive of clinical events in patients with
cardiovascular risk or disease [14,15,23,24,30]. Collectively, these
earlier studies combined with our own findings highlight the need
for pharmacists and health care providers to be aware that
substituting fish oil DS for prescription products in patients with
very high triglycerides is not appropriate due to potential dosing
issues, as well as the presence of unwanted ingredients such as
saturated fats and oxidized lipids, which may interfere with the
intended biological effects. There is evidence that pharmacists may
need additional education and training about DS to be better sit-
uated to offer counseling or recommendations about prescribed
medications and DS [43].
A limitation of this study is that, although we evaluated the top-
selling DS in the United States, we only assessed three products
given the scope of this initial investigation. There are dozens of
other fish oil DS on the market that may also vary in quality and
purity and thus further study is warranted with a larger sample
size. Further study is also needed for measurements of heavy
metals and cholesterol content.
To explore the potential biological effects of fish oil DS, we
tested their activity in isolated sdLDL in vitro. Our findings suggest
that further study in laboratory animals is needed to better un-
derstand the effects of ingestion of oxidized lipids on the biological
activity of the DS. Ironically, it would be difficult to justify testing
these widely available DS in human subjects given that we now
know they contain significant levels of oxidized lipids and satu-
rated fat. There would likely be ethical concerns for a clinical study
involving long-term administration of DS that are known to contain
oxidized lipids and saturated fat due to possible adverse health
effects.
Conflict of interest and funding
These experimental studies were designed and conducted by
Elucida Research without any specific grant from funding agencies
in the public, commercial, or not-for-profit sectors. Editorial and
 R.P. Mason, S.C.R. Sherratt / Biochemical and Biophysical Research Communications 483 (2017) 425e429 429

medical writing assistance was provided by Peloton Advantage, T. Sakata, K. Shimada, K. Shirato, Effects of eicosapentaenoic acid on major
coronary events in hypercholesterolaemic patients (JELIS): a randomised
Parsippany, NJ and funded by Amarin Pharma, Inc.
open-label, blinded endpoint analysis, Lancet 369 (2007) 1090e1098.
[22] R.P. Mason, R.F. Jacob, Eicosapentaenoic acid inhibits glucose-induced mem-
Acknowledgments brane cholesterol crystalline domain formation through a potent antioxidant
mechanism, Biochim. Biophys. Acta 1848 (2015) 502e509.
[23] V.M. Garcia-Hernandez, M. Gallar, J. Sanchez-Soriano, V. Micol, E. Roche,
The authors wish to thank Dr. Robert F. Jacob, PhD for scientific E. Garcia-Garcia, Effect of omega-3 dietary supplements with different
discussions and editorial assistance. oxidation levels in the lipidic profile of women: a randomized controlled trial,
Int. J. Food Sci. Nutr. 64 (2013) 993e1000.
[24] M.S. Nogueira, M.C. Kessuane, A.A. Lobo Ladd, F.V. Lobo Ladd, B. Cogliati,
Transparency document I.A. Castro, Effect of long-term ingestion of weakly oxidised flaxseed oil on
biomarkers of oxidative stress in LDL-receptor knockout mice, Br. J. Nutr.
Transparency document related to this article can be found (2016) 1e12.
[25] D. Firestone, Official Methods and Recommended Practices of the American
online at http://dx.doi.org/10.1016/j.bbrc.2016.12.127 Oil Chemists' Society, fourth ed., American Oil Chemists' Society, Champaign,
IL, 1997.
References [26] Oxidation in Omega-3 Oils: an Overview, A white paper prepared by the
Global Organization for EPA and DHA Omega-3s and the Council for
Responsible Nutrition, 2015. Available at: http://crnusa.org/pdfs/
[1] P.M. Kris-Etherton, W.S. Harris, L.J. Appel, Fish consumption, fish oil, omega-3
GOEDþCRNWhitePaper-Omega-3oxidation.pdf. Accessed November 30, 2016.
fatty acids, and cardiovascular disease, Circulation 106 (2002) 2747e2757.
[27] I.G. Davies, J.M. Graham, B.A. Griffin, Rapid separation of LDL subclasses by
[2] Lovaza [package insert], GlaxoSmithKline, Research Triangle Park, NC, 2014.
iodixanol gradient ultracentrifugation, Clin. Chem. 49 (2003) 1865e1872.
[3] Vascepa [package insert], Amarin Pharma Inc., Bedminster, NJ, 2016.
[28] I.T. Mak, W.B. Weglicki, Antioxidant activity of calcium channel blocking
[4] D. Hilleman, A. Smer, Prescription omega-3 fatty acid products and dietary
drugs, Methods Enzymol. 234 (1994) 620e630.
supplements are not interchangeable, Manag. Care Jan. (2016) 46e52B.
[29] K. Yagi, Assay for blood plasma or serum, Methods Enzymol. 105 (1984)
[5] A. Zargar, M.K. Ito, Long chain omega-3 dietary supplements: a review of the
328e331.
National Library of Medicine Herbal supplement Database, Metab. Syndr.
[30] B.B. Albert, D. Cameron-Smith, P.L. Hofman, W.S. Cutfield, Oxidation of marine
Relat. Disord. 9 (2011) 255e271.
omega-3 supplements and human health, Biomed. Res. Int. 2013 (2013)
[6] A.C. Kleiner, D.P. Cladis, C.R. Santerre, A comparison of actual versus stated
464921.
label amounts of EPA and DHA in commercial omega-3 dietary supplements
[31] J. Burger, M. Gochfeld, Perceptions of the risks and benefits of fish con-
in the United States, J. Sci. Food Agric. 95 (2015) 1260e1267.
sumption: individual choices to reduce risk and increase health benefits,
[7] B.B. Albert, J.G. Derraik, D. Cameron-Smith, P.L. Hofman, S. Tumanov,
Environ. Res. 109 (2009) 343e349.
S.G. Villas-Boas, M.L. Garg, W.S. Cutfield, Fish oil supplements in New Zealand
[32] B. Rauch, R. Schiele, S. Schneider, F. Diller, N. Victor, H. Gohlke, M. Gottwik,
are highly oxidised and do not meet label content of n-3 PUFA, Sci. Rep. 5
G. Steinbeck, C.U. Del, R. Sack, H. Worth, H. Katus, W. Spitzer, G. Sabin,
(2015) 7928.
J. Senges, OMEGA, a randomized, placebo-controlled trial to test the effect of
[8] J.C. Ritter, S.M. Budge, F. Jovica, Quality analysis of commercial fish oil prep-
highly purified omega-3 fatty acids on top of modern guideline-adjusted
arations, J. Sci. Food Agric. 93 (2013) 1935e1939.
therapy after myocardial infarction, Circulation 122 (2010) 2152e2159.
[9] S.A. Jackowski, A.Z. Alvi, A. Mirajkar, Z. Imani, Y. Gamalevych, N.A. Shaikh,
[33] D. Kromhout, E.J. Giltay, J.M. Geleijnse, n-3 fatty acids and cardiovascular
G. Jackowski, Oxidation levels of North American over-the-counter n-3
events after myocardial infarction, N. Engl. J. Med. 363 (2010) 2015e2026.
(omega-3) supplements and the influence of supplement formulation and
[34] ORIGIN Trial Investigators, n-3 fatty acids and cardiovascular outcomes in
delivery form on evaluating oxidative safety, J. Nutr. Sci. 4 (2015) e30.
patients with dysglycemia, N. Engl. J. Med. 367 (2012) 309e318.
[10] J.A. Berliner, A.D. Watson, A role for oxidized phospholipids in atherosclerosis,
[35] The Risk and Prevention Study Collaborative Group, n-3 fatty acids in patients
N. Engl. J. Med. 353 (2005) 9e11.
with multiple cardiovascular risk factors, N. Engl. J. Med. 368 (2013)
[11] M. Bertelsen, E.E. Anggard, M.J. Carrier, Oxidative stress impairs insulin
1800e1808.
internalization in endothelial cells in vitro, Diabetologia 44 (2001) 605e613.
[36] D.M. Qato, J. Wilder, L.P. Schumm, V. Gillet, G.C. Alexander, Changes in pre-
[12] J.W. Baynes, Role of oxidative stress in development of complications in
scription and over-the-counter medication and dietary supplement use
diabetes, Diabetes 40 (1991) 405e412.
among older adults in the United States, 2005 vs 2011, JAMA Intern. Med. 176
[13] R. Turner, C.H. McLean, K.M. Silvers, Are the health benefits of fish oils limited
(2016) 473e482.
by products of oxidation? Nutr. Res. Rev. 19 (2006) 53e62.
[37] P.M. Barnes, B. Bloom, R.L. Nahin, Complementary and Alternative Medicine
[14] M.F. Walter, R.F. Jacob, R.E. Bjork, B. Jeffers, J. Buch, Y. Mizuno, R.P. Mason,
Use Among Adults and Children: United States, 2007, 2008, pp. 1e23. Na-
Circulating lipid hydroperoxides predict cardiovascular events in patients
tional health statistics reports.
with stable coronary artery disease: the PREVENT study, J. Am. Coll. Cardiol.
[38] R.L. Bailey, J.J. Gahche, P.E. Miller, P.R. Thomas, J.T. Dwyer, Why US adults use
51 (2008) 1196e1202.
dietary supplements, JAMA Intern. Med. 173 (2013) 355e361.
[15] M.F. Walter, R.F. Jacob, B. Jeffers, M.M. Ghadanfar, G.M. Preston, J. Buch,
[39] T.C. Clarke, L.I. Black, B.J. Stussman, P.M. Barnes, R.L. Nahin, Trends in the Use
R.P. Mason, Serum levels of thiobarbituric acid reactive substances predict
of Complementary Health Approaches Among Adults: United States, 2002-
cardiovascular events in patients with stable coronary artery disease: a lon-
2012, 2015, pp. 1e16. National health statistics reports.
gitudinal analysis of the PREVENT study, J. Am. Coll. Cardiol. 44 (2004)
[40] J. Backes, B. Melton, J. Ruisinger, C. Burkhardt, P. Moriarty, A comparison of
1996e2002.
patients' prescribed, self-reported, and actual intake of supplemental EPA/
[16] N. Lamharzi, C.B. Renard, F. Kramer, S. Pennathur, J.W. Heinecke, A. Chait,
DHA [abstract 161], J. Clin. Lipidol. 10 (2016) 699e700.
K.E. Bornfeldt, Hyperlipidemia in concert with hyperglycemia stimulates the
[41] E.B. Hawkins, H. Ling, T.L. Burns, D.E. Hilleman, Perceptions, knowledge, and
proliferation of macrophages in atherosclerotic lesions: potential role of
patterns of use of “fish oil” products in cardiac patients [abstract 23], Phar-
glucose-oxidized LDL, Diabetes 53 (2004) 3217e3225.
macotherapy 32 (2012) e184.
[17] P. Libby, Inflammation in atherosclerosis, Nature 420 (2002) 868e874.
[42] N.J. Stone, J. Robinson, A.H. Lichtenstein, C.N. Bairey Merz, D.M. Lloyd-Jones,
[18] M. Rizzo, K. Berneis, Low-density lipoprotein size and cardiovascular risk
C.B. Blum, P. McBride, R.H. Eckel, J.S. Schwartz, A.C. Goldberg, S.T. Shero,
assessment, QJM 99 (2006) 1e14.
D. Gordon, S.C. Smith Jr., D. Levy, K. Watson, P.W. Wilson, 2013 ACC/AHA
[19] S. Koba, T. Hirano, Y. Ito, F. Tsunoda, Y. Yokota, Y. Ban, Y. Iso, H. Suzuki,
guideline on the treatment of blood cholesterol to reduce atherosclerotic
T. Katagiri, Significance of small dense low-density lipoprotein-cholesterol
cardiovascular risk in adults: a Report of the American College of cardiology/
concentrations in relation to the severity of coronary heart diseases,
American heart association task force on practice guidelines, J. Am. Coll.
Atherosclerosis 189 (2006) 206e214.
Cardiol. 63 (2014) 2889e2934.
[20] J. de Graaf, H.L. Hak-Lemmers, M.P. Hectors, P.N. Demacker, J.C. Hendriks,
[43] S.A. Coon, V.W. Stevens, J.E. Brown, S.E. Wolff, M.J. Wrobel, Comparison of
A.F. Stalenhoef, Enhanced susceptibility to in vitro oxidation of the dense low
dietary supplement product knowledge and confidence between pharmacists
density lipoprotein subfraction in healthy subjects, Arterioscler. Thromb. 11
and health food store employees, J. Am. Pharm. Assoc. 55 (2003) (2015)
(1991) 298e306.
161e168.
[21] M. Yokoyama, H. Origasa, M. Matsuzaki, Y. Matsuzawa, Y. Saito, Y. Ishikawa,
S. Oikawa, J. Sasaki, H. Hishida, H. Itakura, T. Kita, A. Kitabatake, N. Nakaya,