Friday, August 3, 2012

Cholesterol: My View

An anonymous commenter on my post "Bill Clinton: A Sucker or Smart?" left a link to Robb Wolf's blog post "Cholesterol: We Are Dumb". My Clinton post was about his diet change to a plant-based diet to lower his cholesterol and reverse his heart disease. I also shared my concerns about my own cholesterol levels at the time. The commenter only left the link, so I can only assume by the context of the link that their intention was to reassure me that it's ok to have high cholesterol.

Wolf made comments in his post like:
"most people who suffer a heart attack have...low cholesterol!"
and then in the next sentence, just to make his point:
"most heart attacks are in folks with...low cholesterol!"
Nowhere in his post does he describe what low cholesterol is? I'm a number's guy, so this bugs me. Wolf was referencing an article written by a UCLA newsroom which referenced an article in the American Heart Journal. The researchers measured lipid levels from about 137,000 (59%) people out 232,000 people hospitalized for coronary artery disease in over 500 hospitals from 2000 to 2006. They reported the following LDL cholesterol levels from those measurements.

from American Heart Journal January 2009
What were the cholesterol levels from the other 41% of the people hospitalized for coronary heart disease? Nearly half of the population wasn't measured for lipid levels. The researchers state that there is a potential for selection bias because not all patients were measured for lipid levels.Why would so many patients not have their blood lipids measured? Is it reasonable to assume that patients with higher cholesterol were resistant to getting their blood lipids tested?

Notice the shape of the LDL distribution of the population of patients that was measured. It's not a bell shaped curve. There is a tail on the high end of the curve. The results are skewed to higher cholesterol. This skewness of data can indicate that the sample of data (137,000 patients) doesn't accurately represent the total population of data (232,000 patients). It can also indicate that there is a natural limit for coronary artery disease at the low end of the distruibution (low cholesterol) that is not observed at the high end of the distribution (high cholesterol).

The results reported in this graph show that about half of the sampled population in the study had LDL cholesterol levels below 100 mg/dL which is considered optimal. Assuming there was no selection bias, the researchers concluded that these findings provide support for guideline revisions with even lower LDL goals. That's not what Wolf concluded, he stated:
"it completely calls into question the notion that we need to reduce cholesterol levels"
Wolf referenced a second article on statins. This study measured the effect of 20 mg daily of Crestor on people with LDL cholesterol below 130 mg/dL. In about 2 years, the participants lowered their LDL cholesterol by 50% and CRP levels by 37%. And compared to participants who took placebos, the statin users had half as many heart attacks, strokes and deaths from cardiovascular causes. Wolf suggested in his blog post that the decrease in heart attack rates was due to the lower CRP levels, and not the lower cholesterol levels.

Rather than dismissing cholesterol as a risk factor for heart disease as Wolf apparently does, I think cholesterol has to be considered as a relative risk factor. Nothing is absolute. In other words, multiple problems contribute to heart disease. Cholesterol is not a reliable predictor by itself. That doesn't mean cholesterol is meaningless.

At least this is how I interpret the guidelines presented by the NIH (National Institutes of Health) referred to as the ATP III Guidelines (Adult Treatment Panel). They present three risk categories for heart disease: low, intermediate and high. In addition to LDL, a person's risk category depends also on issues like smoking, high blood pressure, low HDL cholesterol, family history of heart disease, diabetes, age and being overweight. The LDL goal for the low risk category is less than 160 mg/dL, for the intermediate risk category is less than 130 mg/dL and for the high risk group is less than 100 mg/dL, but consideration should be given for less than 70 mg/dL.

However, there are folks found on the internet like Wolf who don't see cholesterol this way. Cholesterol is necessary they say. OK. But how much is necessary? It seems to me that some of the most ardent supporters of cholesterol eat a diet consisting of a high portion of meat. Have you every heard someone eating a high carb plant based diet say that it was no big deal to have high cholesterol? It was a year ago that I was convinced to give up a low carb diet. I replaced a lot of that meat with carbs from fruits and veggies. Here's what has happened to my blood lipids since then.

Just so you don't think that I'm stuck on just getting low cholesterol numbers, let me share a few other improved risk factors with my diet change to more whole plant carbs. My blood pressure reduced from 130s over 80s to 110s over 60s. My body weight has dropped by about 25 lbs. And, my resting pulse has dropped from the 60s to the 50s.

Meat provides protein (plants provide enough too), but what else did I do to my body consuming so much meat? I'm not willing to continue that self experiment. Meat certainly did not help my risk factors for heart disease.


  1. Looking at your blood test results my question is are you concerned that your HDL-C is now <45 and your triglycerides are now >100??

    Trigs above 100 are usually associated with particles that are small and dense as opposed to large and buoyant

    Perhaps you should research these?

    Your thoughts

  2. HDL decreased and triglycerides increased, but actual levels don't concern me. HDL > 40 and triglycerides < 110 isn't too bad.

  3. Anthony Colpo is a complete fraud who follows the money. Here are examples of his fraud that came back to bite his buttocks:

  4. Tota cholesterol is meaningless. But, LDL is still important- especially the small dense particles. There is good evidence this contributes to CAD.

    There is NEW research on HDL. It seems the level per se does not matter, it's the FUNCTION.

    "HDL efflux capacity"

    Dr. Daniel Rader is leading the way on this.