Chris Kresser on Fish Oil
Chris Kresser is apparently going to keep at
this, so here's another look at his 2015 fish oil critique,
as revised by him on 2017-01-24:
You Really Be Taking Fish Oil?.
I remarked on
the original 2015 article when a reader asked
about it on the WB Blog. The cite numbers have since changed,
as reflected below (and now linked, which I couldn't do on the
blog, due to link count limits).
Chris Kresser's 2017 version of the article embeds this tweet-bait:
Is it possible that fish oil is beneficial for one person
and harmful for another? #fishoil
to which I would reply : absolutely.
If someone stays on a standard
diet, or remains on a statin, uses a sub-clinical dose of ω3,
or uses an inferior product,
they might well expect no benefit. An otherwise therapeutic dose might even
be harmful in some of those scenarios, particularly if they also follow some of the
crazy protocols in the reported trials (margarine, seriously?).
#1 is a meta, which concludes “Dietary supplementation
with omega-3 fatty acids should be considered in the secondary
prevention of cardiovascular events.” Bottom line:
study does not conflict with Cureality/Wheat Belly (C/WB) recommendations.
#2 was a “placebo controlled”
trial of B-vitamins and ω3,
but used only 600 mg of DHA+EPA daily. Statin use was not
accounted for. There was apparently
no control of diet. The placebo was not disclosed, and it's
often a confounder, as the majority of nutritional researchers
are either clueless about what is really inert, or don't know the
difference between “placebo” and “control”.
Bottom line: at 18% of
WB recommendation, dose is too low to provide insight.
#3 was another placebo controlled trial, in a post-MI population
“…receiving state-of-the-art antihypertensive,
antithrombotic, and lipid-modifying therapy…”, with even
less DHA+EPA (376 mg), delivered in 18 grams of PUFA (margarine)!
The paper itself admitted that this was “low dose”.
Bottom line: dose too low, statins in use, and no chance of driving ω6:3 ratio
to below 2.
References #4, 5 and 6 were focused on afib. I didn't look at
those because C/WB makes no claims about fish oil and afib.
#7 was a meta, relying on trials using doses ranging
from 400 to 4,800 mg day. There was no focus on subject diets.
Placebos were all over the place, or poorly defined, often
adverse fats (a negative confounder). I didn't re-dig for
statin confounding. Papers (only 14, and
the list appears to include the “negatives” above)
were selected strictly by keyword search of databases. The keyword
list was rather narrow. “Our meta-analysis showed
insufficient evidence…”, but they had to ignore
(exclude) one trial even to say that. Also, “Among
the trials, there was no significant difference in the
occurrence of adverse events, such as gastrointestinal
troubles or gastrointestinal bleeding, between the intervention
and placebo groups.”, which is what happens on
otherwise adverse [consensus] diets. Bottom line: iffy
analysis of iffy original research.
#8 (another meta) is pay-walled.
Bottom line: no way to know, but the abstract itself was
non-inferior: “Overall, omega-3 PUFA supplementation
was not associated with a lower risk of all-cause mortality,
cardiac death, sudden death, myocardial infarction, or stroke
based on relative and absolute measures of association.”.
Reference #9 (another meta) - can't even find full-text.
Bottom line: no way to know.
#10 is new for the article (a 2016 paper) and another meta.
I didn't drill down all the cites, because they were all 2013 or
earlier, and likely covered by the other meta mastications here.
This meta concluded inconclusively with another non-inferiority:
“There is currently a lack of evidence to support the
routine use of omega-3 PUFAs in the primary and secondary
prevention of CVD.”
#11 (another meta) concludes
“our results supply evidence that long-term
effect of high dose omega-3 fatty acid supplementation
may be beneficial for the onset of cardiac death, sudden
death and myocardial infarction among patients with a
history of cardiovascular disease.” Bottom
line: study does not conflict with C/WB recommendations.
#12 was a controlled trial that had odd
intervention groups. The fish oil amounted to 3 grams
daily, but the FO-only group got no other dietary advice.
The other groups got pretty bizarre dietary advice. One
chart shows that the fish-only advice group had better
survival, but these clueless researchers decided it was
better to conclude “This result is
unexplained…”. Bottom line: seems to
support C/WB recommendation despite inept design.
#13 was a clinical trial of FO on people with
metabolic syndrome. The full text is available, but is
in Portuguese. The abstract (English) captures the essence
of it “We conclude that intake of fish oil
resulted in decreased levels of triglycerides and
increased the TRAP (total antioxidant capacity) of
patients with MS (metsyn); however, increased LDL
levels and insulin resistance, were observed.”
Bottom line: the FO improved a metric that matters (TG),
but violated LDL-C dogma, and they were pretty
clueless about when apparent IR is really a problem.
#14: “The present study provides
evidence that increased dietary ω-3 polyunsaturated
fatty acids and extra virgin olive oil have beneficial
synergistic effects on lipid metabolism and oxidative
stress in patients with MetS.”, so does
not contradict C/WB.
#15 is pay-walled, so I can only see the
abstract. It was really a trial of soy, and sailed
into the trial with a bias against fish oil. No other
dietary intervention was done, and I can't draw any
real conclusions from the abstract. Bottom line: no way
#16 was a test of krill oil, which C/WB
discourages. Bottom line: no insight on C/WB fish oil.
#17 used only 900 mg of DHA+EPA and was
“restricted to patients with impaired
fasting glucose, impaired glucose tolerance, or
diabetes”, plus a bunch of CVD diagnoses.
Half the subjects were on statins. There was no apparent other
dietary intervention. Bottom line: dose too low
to be insightful on C/WB recommendation, plus
statin confounding (admitted by the article).
References #18-23 were in respect to cancer. Chris himself
concludes “research on omega-3 and fish oil
supplementation on cancer is decidedly mixed.”
Based on what I found by following the earlier ones,
I didn't bother to look. Omega 3 doesn't strike me at
the moment as being a huge factor one way or the other
in cancer. The things to focus on in cancer may be:
very low carb, grain free, low inflammatory, immune
system support (microbiome) and generally avoid
References #24-27 were all about oxidation levels in
actual ω3 products. These papers have a point: namely
that all ω3 trials need to specify quality/potency
of the ω3. None of these oxy-caution papers specified brands,
alas. ConsumerLab does have
oxidation as one of the quality criteria they focus on
when testing ω3 supplements.
Getting your FO from actual fish is a great idea, as
long as you have some assurance that the fish is
fresh (or fresh-frozen), is wild-caught (not farmed,
nor sea-farmed), and is non-GMO. A CAFO fish is going
to be low in ω3 and high in ω6.