I seem to be constantly under fire by a throng of vegetarian/vegan acquaintances who think my lifestyle is wrong wrong wrong. I can usually defend myself fairly intelligently, but I'm not exactly sure what to make of their most recent round of ammo:
Any takers? Thoughts? Opinions?
For a fixed level of omega 3 intake, once over a certain threshold (about 4% of calories), more omega 6 PUFAs don't significantly increase inflammation, including vascular inflammation.
PUFAs do lower LDL cholesterol, however. The end-result seems to be that more omega 6 PUFAs does reduce heart disease slightly (among people over-consuming omega 6 oils and under-consuming omega 3 oils) when compared to other kinds of fat, including monounsaturated and saturated fats. Both of these other fats, however, likely reduce CVD risk when compared to calories from sugar.
On the other hand, getting PUFA intake balanced between omega 3 and omega 6, seems to produce much more dramatic results (huge reductions in CVD and cancer risk). Since most Americans get about 10% or more of their calories form omega 6 PUFAs, this is much more safely done while limiting omega 6 PUFA consumption rather than taking massive doses of fish oil. See the Lyon Diet Heart Study.
By the way, the reason you are better off reducing total PUFA consumption is because of PUFAs tendency to oxidize. It's unclear whether this damage can be mitigated by consuming polyphenol/antioxidant rich omega 6s (like almonds or walnuts or olive oil or avocados) but I would guess it can be to some extent. Modern vegetable oils are often lacking in proper antioxidants.
I try to limit overall PUFA and sugar intake since I know that saturated fat, monounsaturated fat, and starch are safe and I'm not sure I want to gamble on antioxidants preserving my fats to keep me healthy. I also take a small quantity of fish oil (about 1 gram).
Well, there is one good thing about this study -- the full text is actually available online for free!
I'll read the full thing later but initial impression is that "meta-analysis" throws off all sorts of alarm bells. They took a bunch of randomized controlled trials (the gold standard for studies) that showed no connection between saturated fat and CVD and then did a meta-analysis to show a connection. Meta-analysis is a useful but potentially dubious method. If you've read GCBC, you know Taubes's negative opinion of meta-analysis. The Wikipedia article on meta-analysis does a good job of summing up the problems:
In any case, this isn't the first meta-analysis to be done on saturated fat and I'm not sure why your vegetarian/vegan acquaintances would think this one is more valuable than say these two...
A meta-analysis of prospective epidemiologic studies showed that there is no significant evidence for concluding that dietary saturated fat is associated with an increased risk of CHD or CVD. - http://www.ajcn.org/cgi/content/abstract/ajcn.2009.27725v1?papetoc
No significant differences in fat intake were noted in six case-control studies of CVD patients and CVD-free controls; and neither total or CHD mortality were lowered in a meta-analysis of nine controlled, randomized dietary trials with substantial reductions of dietary fats, in six trials combined with addition of PUFA. The harmful effect of dietary SFA and the protective effect of dietary PUFA on atherosclerosis and CVD are questioned. - http://www.ncbi.nlm.nih.gov/pubmed/9635993
I guess it's back to the old game of pick the study that supports whatever viewpoint you already have.
The other obvious thing that stands out is that this involved replacing one kind of fat with another kind of fat. In the context of the past research on SF, this study seems to be more about the potential protective quality of PUFA than harm from SF. Maybe getting more PUFA is a good thing but it seems unrealistic that someone could get the bulk of their calories from PUFA so the rest have to come from somewhere else. It seems clear that getting those calories from saturated fat is vastly preferable to getting them from carbs and nothing about this particular meta-analysis suggests otherwise.
For starters, this is a
review meta-analysis article of RCT data- potentially the highest quality of evidence we can come across. So I am glad to see these kinds of things being cited and used as evidence instead of the usual non-causal evidence. But the problem with review articles and meta-analysis is the bias in applying the inclusion/exclusion criteria. I can almost guarantee they found a way to exclude the studies that showed margarine was killing people. These studies have been gone over in GCBC by Gary Taubes.
The Cochrane Collaboration was specifically setup to provide unbiased review articles, and they have already done these reviews- they never find a protective effect on total mortality
This is the glaringly obvious problem with this article- the "hard endpoint" that was chosen (heart attacks). A much better endpoint is total mortality, and they would have used it if they could have. Do you care if you die of cancer instead of a heart attack?
There may well be other flaws with the studies used- I haven't read past the abstract. Hopefully Dr. Eades or someone else (that is not brainwashed) will review this article more thoroughly.
Stephan from WholeHealthSource will be taking this study apart soon in a blog post near you. Here's what he said today:
"OK, this review is total BS. They included the Finnish trial in their analysis, really unbelievable!!
The title of their study says "Meta-Analysis of Randomized Controlled Trials". The Finnish trial was neither randomized, nor blinded, nor controlled!
I'm looking through the data, there are so many problems with this review it's ridiculous. That's probably why it went to PLoS and not NEJM. I may have to post on it."
Living on SF only: is it possible? 4 Answers
Bad Reaction to SATURATED FAT 0 Answers
Replacing Sat Fat with PUFA reduces CHD? 2 Answers