Intended for healthcare professionals


Risks and benefits of omega 3 fats for mortality, cardiovascular disease, and cancer: systematic review

BMJ 2006; 332 doi: (Published 30 March 2006) Cite this as: BMJ 2006;332:752

Re: Authors' reply - omega 3s and health

One problem with the Hooper analysis and response is the inclusion of
trials without full understanding of what was done. The odd ball MARGARIN
trial was confounded with a truly massive amount of omega-6 yet included,
and Lyon is
characterized in the current eBMJ response as a 'multiple risk factor
intervention' which it was not by any stretch of the imagination.

What really happened in Lyon, this city 500 km from Paris and another
300 km to the Mediterranean? The control group
was left alone under regular doctor's care while the 'experimental' group
of n=302 got high omega-3 canola-based
margarine, 'free for the entire family', as well as yearly [sic] dietary
'advice'. Dietary advice has historically not changed eating patterns
and indeed here is what happened:

Converted into food items (tables 3 & 5 from Lancet (343)1994-6-
11: 1456), in experimentals after 1-4 year followup, per day vs. controls:

* 10% fewer calories but only 0.1 less weight gain (Body Mass Index);

* 0.6 slices more bread but 1/6th less servings of cereals;

* 10 more beans/day and 10% more veggies (the weight of a small

* 20% more fruits [1/2 a small 5 cm (2") apple];

* a sliver (7 g) less 'delicatessen';

* 13% less meat + poultry (14 g); 10% less cheese (3 g);

* 15% more fish (6 grams, an anchovy fillet);

* Zero difference in total fat intake from butter + margarine + cream
+ oil;

* Experimentals ingested 1/2 'egg-equivalents' less cholesterol.

Clearly this was not a 'diet' study considering these tiny global
food intake and zero blood cholesterol changes. One can see why NEJM
refused (private communication) to publish these dramatic results (such as
a 76% reduction in adjusted
cardiac deaths), results that were subsequently published in Lancet 1994;
medline 7911176, AJCN 1995; medline
7754988, and free: Circulation 1999 (medline 9989963, with editorial
[medline 9989956]. Dietitians understandably have
never been able to change the global eating patterns of 300 seniors from
Lyon, Birmingham or anywhere else for that matter, and show clinical
benefit. What people WILL do is take free margarine and replace regular
omega-6 margarine and butter with it; the difference evidently is the
omega-3. Simple comme bonjour!

Indeed, to explain these 'impossible' dogma defying results one has
to look at the plasma fatty acid analysis of 280
subjects at 1 year follow-up (Lancet, as above, table 4) where
experimentals had a 67% higher level of alpha-linolenic acid (ALA, from
canola margarine), a 35% increase of C20:5n3 EPA, the first and main ALA
'daughter', and less than 11% differences in either omega-6, mono or
saturated fatty acids.

Lyon's failing was allowing critique by including once yearly dietary
advise, Hooper's failing was not understanding the fundamentals of this
excellent study and thus excluding it. vos{at}

Competing interests:
None declared

Competing interests: No competing interests

11 April 2006
Eddie Vos
Sutton (Qc) Canada J0E 2K0