Arterial stiffening caused by ketosis?


#21

Interesting. Thank you.

It seems to me that there are many studies that show the ketogenic diet as improving cardiovascular health. I would assume that this precludes arterial stiffening?

That being said, and assuming my assumption about precluding arterial stiffening is correct, it would further appear that there are few (I only know of two) studies showing arterial stiffening.

So, would it be safe to say that the (vast?) majority of studies show that the ketogenic diet improves cardiovascular health and does not cause arterial stiffening?


(Todd Allen) #22

Funny how headlines fail to capture the full picture. Actually itā€™s much worse than that they blatantly lie.

If you read the study you find out it was not a controlled experiment testing the effect of coffee but a statistical survey. So it can indicate a correlation but not causation. AND to reach the conclusion that coffee lowers the risk of death mathematical games had to be played. Because the coffee drinkers died younger with more health problems then the non-coffee drinkers. They made coffee look good because the coffee drinkers also engaged in other unhealthy behaviors such as smoking and they then calculated that due to other factors they should have died even younger but coffee saved the day - almost anyway.

I personally give such a study very little credence, do you really believe they managed to identify and account for all the differences between the coffee drinkers and non-drinkers and that they accurately computed the risk of dying without coffee and that it was actually the coffee that kept people from dying at their calculated rate?


(Bacon is a many-splendoured thing) #23

Look at the strength of the statistical correlation. A correlation of 1.36, say, is not really all that strong (@richard feels that itā€™s not significant until it reaches 2.0), and you also need to consider the relative versus absolute risk or benefit. It might be the result of only a one-participant difference between the treatment group and the control group. The statistical significance of the correlation is also important. A correlation with p < 0.05 is so vague that the measured effect might be completely different when the study is repeated. A correlation with p < 0.001, on the other hand, is well worth trying to verify in a randomized, controlled trial.


(Todd Allen) #24

Hereā€™s a link to the study.

From the text, they state they considered p<0.05 significant. Hereā€™s a gem from their results.

In age-adjusted analyses, coffee consumption was associated with increased mortality among both men (Table 2Table 2Association of Daily Coffee Consumption with Total and Cause-Specific Mortality among 229,119 Men.) and women (Table 3Table 3Association of Daily Coffee Consumption with Total and Cause-Specific Mortality among 173,141 Women.). However, after multivariate adjustment for potential confounders, particularly smoking (Table 1 in the Supplementary Appendix), a modest inverse association between coffee drinking and total mortality was observed for both sexes.

Among the factors increased among coffee drinkers were red meat and saturated fat consumption for which the studyā€™s authors attributed an increased risk of mortality that coffee could then fix. However, the science on those isnā€™t firmly settled. There are some people who donā€™t believe saturated fat and red meat are deadly.


(Carpe salata!) #25

ā€œLook at the strength of the statistical correlation. A correlation of 1.36, say, is not really all that strong (@richard feels that itā€™s not significant until it reaches 2.0),ā€

The coefficient of correlation is a number between -1 to +1.


(Bacon is a many-splendoured thing) #26

Not being a statistician, Iā€™d have to review the original conversation to remember what I was talking about, precisely. But I do remember having read one of Richardā€™s posts about interpreting studies, in which he did write about how the correlation needed to be at least 2.0, so Iā€™d also have to go back and try to remember which post it was that I had read.


(Carpe salata!) #27

I heard that podcast and I donā€™t remember exactly, but it was not a correlation but an odds ratio or an effect size of some sort. Maybe @richard could give us a quick reprise?


(Ben) #28

Itā€™s not just @richard saying this.

In observational studies, they will not normally accept an RR of less than 3 as significant and never an RR of less than 2

http://www.numberwatch.co.uk/rr.htm


#29

tl;dr version: Per my doctor: No, the ketogenic diet does not cause arterial stiffening.

Update: Went to the doctor, they said NOPE. The ketogenic diet is not known to cause arterial stiffening.

My assumptions are twofold (helped in development by the fine minds on this site):

  1. The study was on people with health conditions who were put on a ketogenic diet but we donā€™t know what kind of fat they were fed. Perhaps it was all canola oil and trans fats and that, coupled with their existing conditions and medical history, is what caused the stiffening? Who knows.

  2. If the ketogenic diet caused arterial stiffening then we would be seeing new ketoers having heart attacks left and right. This is not happening and so we can conclude that this study is may be relevant only to people with similar medical conditions and histories as those actually in the study.

My doctor did not argue with these assumptions.


(Jonathan Greene) #30

Just consume more L Arginine which gets converted in to Nitric Oxide in the blood and strengthens the cell walls and also keeps them nice and soft, along with preventing any cholestorel or plaque from binding to the arteries.


(Sarah) #31

I would add that these kids were on all kinds of meds, as well.