I want to take a final census of the studies we've looked at before doing the stroke studies from the S-T & K meta-analysis. Adopting the most favorable reading of all studies (studies with no overall significant results) it's 10-6 harmful vs. no harm.
None of the studies of CHD and SFA showed benefit to SFA, and when we read it critically, including all studies that showed harm from a subgroup, we get 14-2 harmful vs. no harm. I want to point out that all the studies excluded people with existing CHD.
This only makes sense if the process that leads to additional cardiac events in a patient with CHD is different than that which leads to the initial event. But there isn't much data to support this. One wonders what a meta-analysis that didn't exclude heart patients would show.
So when we get to the stroke studies we get to a weird choice that the authors made. They double counted lots of studies. The double counted studies on stroke are Honolulu Heart, Israeli Ischemic Heart Disease, Framingham, Nurses Health, HPFUS and Malmo diet and Cancer.
The reasoning for this, I guess, is the assumption that stroke and CHD are caused by the same mechanism, which is true for ischemic stroke, but they included at least one study looking only at hemorrhagic strokes. This is Iso's study on fat and protein. academic.oup.com/aje/article/15…
Now hemorrhagic strokes happen due to a fundamentally different mechanism than heart attacks and ischemic strokes. They are associated with alcohol consumption in almost all studies and the majority of studies do show a protective effect of saturated fat for this condition.
This study showed a clear linkage between hemorrhagic strokes in the Japanese population and alcohol consumption, for example. stroke.ahajournals.org/content/26/5/7…
But what percentage of all strokes are hemorrhagic? Estimates vary, but around 90% of all strokes are ischemic. So why would you include a study that looked only at hemorrhagic stroke by itself in a meta-analysis of this type? Looks like a way to change the overall math to me.
I have looked at a large amount of studies on SFA and hemorrhagic stroke and I have to admit, that it is likely that people who consume lots of alcohol do protect themselves from blowout brain hemorrhages by eating SFA.
Even those individuals are still more likely to suffer an ischemic stroke and would be better off overall reducing SFA intake. It protects against the more common threat. Just as seatbelts can hamper leaving a burning car, but are still overall safer than not wearing one.
Tomorrow we will end the series with Sauvaget's Adult Health Study looking at a similar population with a curious breakdown and outcome.
