Sunday, May 22, 2011

Too Much of a Good(?) Thing: When Fish Oil Starts Clogging Your Arteries and Fattening Up Your Liver.

If you are no regular visitor of the SuppVersity, I guess, you are religiously taking your (high dose?) fish oil, day in day out. So what? It probably lowers your total and low density cholesterol (LDL-C; it may reduce your triglycerides and thus improve your insulin sensitivity. Yet, in doing all those "great" things, the unmetabolized and peroxidized remainder of everyone's favorite wonder-supplement begin to clog your arteries and liver - at least, if you believe in the validity of that kind of rodent studies which suggested the usefulness of fish oil, in the first place.
Image 1: Micrograph of non-alcoholic fatty liver disease, caused by the same kind of lipid accumulations
Shirazi et al. observed in the rats receiving fish oil treatment (image by Nephron)
Shirazi et al. (Shirazi. 2011) recently published a paper reporting exactly that: "Fish oil increases atherosclerosis and hepatic steatosis, although decreases serum cholesterol in Wistar rat" - The scientists had fed two groups of pregnant rats (and, after birth, their offspring) with either a fish oil containing or a standard, soy oil based diet. Both diets had the same overall fat content of 70 g/kg (according to standard AIN93-G recommended by American Institute of Nutrition). In terms of total calories, the diets thus contained 15.9% of the total energy in form of either fish or soy oil. The overall omega-3 to omega-6 ratios of the diets were 6:10 for the fish oil and 1.1:10 for the soy bean oil groups; where omega-3 in fish oil came from EPA and DHA, while the omega-3 content in the soy bean oil diet came from alpha-linoleic acid (ALA).
Image 2: In the aorta of rats on a diet rich in fish oil, fatty streaks like that
formed in the aortae (image source
After 70 days the rats were killed and hepatic and aortic specimen were analyzed. The results are unsettling:
[...] fatty streak in fish oil fed pups were significantly more than that in the other group. [... liver] ductular cell hyperplasia in pups fed with fish oil was significantly more than that in animals fed with standard diet. There was a positive relationship between fatty streak in aorta and ductular hyperplasia in liver (r = 0.470 and p= 0.037)

Although the animals had free access to food (ad-libitum feeding) the pathological changes cannot be a consequence of differing calorie intakes. Both groups consumed roughly 16g of the respective diet. According to Shirazi et al., a feasible explanation for these observations and their inconsistency with previous studies by Saraswathi et al. (Saraswathi. 2009), Bringhenti et al. (Bringhenti. 2010), Zampolli et al. (Zampolli. 2006),and Casós et al. (Casos. 2008) would be the lower total amount of dietary fish oil, different (more varied) overall fat compositions of the diets and shorter study periods, respectively:
One possible explanation for this discrepancy is that in our study animals faced higher amounts of dietary fish oil; Saraswathi et al. used 209 g/kg of mixed oils (including coconut oil, olive oil, corn oil and soy bean oil) plus 60 g/kg fish oil, while we used 70 g/kg fish oil which was the only dietary fat source. The dosage of fish oil used in Zampolli et al. and Casós et al. studies were 1% and 5%, respectively, which was lower than 15.9% used in the present study. In the study performed by Bringhenti et al. animals were fed with fish oil containing diet from weaning till puberty which is a shorter period comparing to ours.
On the other hand, the results of this study stand in line with those of Ritskes-Hoitinga et al. (Ritskes-Hoitinga. 1998), Verschuren et al. (Verschuren. 1998) and Brenner et al. (Brenner. 1990), which, in parts (e.g. Ritskes-Hoitinga) observed even more severe  aortic atherosclerosis and hepatic steatosis than Shirazi and his colleagues. [ Something to think about: Isn't it telling that all those studies have been published before the fish oil craze? And before GlaxoSmithKline started making big bucks by selling is "pharmaceutical grade fish oil" Lovaza. Add to that the fact that Shirazi et al. obviously did not find an American publisher for their study and make up your own thoughts. ]

So what? In essence these results only confirm what I have been saying before. Supplementation with reasonable amounts of fish oil (~2g) may make sense, especially if your diet is naturally low in omega-3 fatty acids in general and DHA, in particular. Mega-dosing on the other hand, or trying to compensate for fatphobia by overconsumption of fish oil, i.e. consuming a low- to no-fat diet, while supplementing huge amounts of fish oil >5g), falsely believing that you would do your body a favor by providing him exclusively with the "good essential fatty acids", will do more harm than good. After all, the "best" (do we really think low total cholesterol is good) serum cholesterol and triglyceride levels are useless, if you die from clogged arteries and a liver defect.


  1. Prof. Andro, when you say "2 g", are you referring to 2 pills of 2 grams of fat, of which 1.2 g are from fish oil and 1 g from EPA + DHA (for example), or are you saying 2 g of EPA + DHA?.

  2. Also, what are the optimal ratio of EPA/DHA?

  3. This comment has been removed by the author.

  4. Grams of fish oil always refers to EPA/DHA as the active compounds of the pills (so I take like 5 pills of 1000mg fish oils each, but only 1.5g EPA/DHA total). EPA and DHA can interconvert (well, I think?), and so just take 1-2g total.

  5. That doesn't explain the discrepancy with this study, which is likely the most relevant since the subjects were a primate model, the African green monkey.

    Atherosclerosis. 1990 Oct;84(2-3):83-94.
    Effect of fish oil on atherosclerosis and lipoprotein metabolism.

    LDL and HDL were about 1/3 lower in the fish oil group, as was apo A-I concentration, but apo B concentration was not different. The fish oil group had plasma LDL particles that were smaller, contained fewer cholesteryl ester molecules and had lower cholesteryl ester transition temperatures due to a relative enrichment of n-3 fatty acids in the CE fraction. In addition, hepatic cholesterol and cholesteryl ester concentrations were significantly lower in the animals fed fish oil. Liver perfusion was used to show that hepatic secretion of cholesterol and triglyceride was lower in the fish oil group, although the number of cholesterol and triglyceride enriched apo B-containing particles secreted was not different. We also demonstrated a lower plasma LCAT reactivity for the plasma phospholipids of the animals fed fish oil. Taken together, these findings clearly demonstrate important effects of n-3 fatty acids on cholesterol metabolism in a primate model that have not been previously recognized. In addition, the monkeys fed fish oil had less atherosclerosis in the coronary arteries and in the aorta. "

  6. I have been on a high/very high fish oil regimen for about 3 months, after reading the following articles:

    I started at about 24-36 grams (2-3tbs) of Carlson Labs Norweigen Cod Liver Oil (
    After about one month, I noticed my appetite changing -- the junk foods I normally craved just didn't appeal as much. It was noticeable, but I wondered if it was just a phase (I've had a few short-lived periods of low appetite). Now, after 3 months, some days without any fish oil, and some days with 12-24 grams (1-2tbs), it's even more noticable. I've been so impartial to food in general that I've wondered if I'm getting some kind of toxicity, so I backed off for a week -- no difference.

    I was over 300lbs as a teen and have never really felt like the fat guy's appetite would go away -- until now. It's not that McDonald's and Taco Bell doesn't taste good, I just feel like something has normalized and I'd rather get some quality macros. Both taste good, so why not stick with the substance? Logical enough, and I THOUGHT the same way before, but I'd still CRAVE crap food like that and end up binging (I could drop $20 at Taco Bell and eat 10+ burritos, including XL Grilled Stuffed, Crunch Wrap, etc). Not only has the appeal for those foods gone down, but my overall capacity for ALL food has gone down.

    I started 1 mile hill runs 4 days a week, and fat has been coming off much more quickly than before, and my cheat days don't even remotely compare to my old binges. Alot could contribute to the fat loss...fewer calories, the intensity of runs, the heat, etc., but I can really pinpoint my change in appetite to about 4 weeks after I started heavy fish oil supp.

    The change is so profound that I'm still not certain it's healthy, though I feel nice and lean, and my gym numbers are stabilizing after some other irresponsible behavior. I've always loved food. I've always been able to keep eating. Not anymore...something is wrong!...Or right?

    I've recently started adding in some high quality olive oil and coconut oil. I regularly eat lots of eggs and plenty of (conventional, usually) beef and pork, so I don't think my fat source is too singular, if that's what caused some of the issues in this study. Anyhow, just a few of my cents...interesting if anyone has any insight or shared experience.