Tuesday, July 07, 2009

May 2009: ASTEROID destroys lipid hypothesis!

Here's a nice quote from this paper:

"Atheroma regression occurred in most patients and was not linked to the LDL cholesterol achieved"

Read that line very carefully. Now read it again. That's it, the rest of the post is just rambling on my part.

But is the lipid hypothesis dead? To quote Malcolm Kendrick from 2005 on a different study:

"The great ship Cholesterol-Lowering has ripped its guts out on the harsh rocks of evidence, but still it does not sink"

ASTEROID was a mini version of the J-LIT study (see below) with three modifications. First it was only 501 patients, so there was never any possibility of looking at clinical outcomes, good or bad. Second was it was only run for two years, which repeats point one but with underlining. Third is that they used a crippling dose of statin, sufficient to drop the mean LDL cholesterol level to 60mg/dl.

Luckily there were minimal adverse reactions to this, if you believe Nissen, which I don't:

"Adverse events were infrequent and similar to other statin trials"

Anyway, they went looking for adverse reactions by level of LDL cholesterol, much as did J-LIT. As might be expected there was no graduation of adverse reactions by cholesterol level because, as Nissen might say, there weren't any!

"Adverse events occurred infrequently during the trial, and no pattern appeared relating the frequency of any adverse event to the achieved LDL cholesterol"

Dying of cancer is difficult in 2 years in the USA. Try five years in Japan, or 10 years in Japan, and use over a 40,000 rather than 500 patients. The excess deaths in the lowest LDL groups will be there, just don't expect them to show in 500 patients.

Final quote:

"Similarly, the on-treatment atheroma volume, change in atheroma volume, and high percentage of subjects with atheroma regression did not differ by the achieved LDL cholesterol"

This is a classic. If the lowering of LDL does not correlate with atheroma volume decrease, why do people believe that lowering LDL is what shrinks atheroma? And if the statin per se is causing atheroma shrinkage, which it certainly is, how come long term large studies of the same design show an increasing body count from cardiac causes when statinating to very low levels of LDL? That's what happened in J-LIT and it's what will happen in ASTEROID if they keep going, except 500 people is far to small to separate the occasional death from chance. Is statin induced atheroma shrinkage beneficial? J-LIT suggests not and ASTEROID is too miniscule to look at clinical outcomes. They will be bad.

Executive summary: lower LDL cholesterol has no linkage to shrinkage of plaque.



Ricardo said...

Its always interesting looking at the real world, so have a look at the LDL levels of pople hospitalized with CAD (see links below). Curiously, they don't show very high TC or LDL levels. On the contrary, their TC is lower than the average of healthy people. Also, look at the LDL histogram: there are more people with CAD at admission in hospitals with LDL 90 mg/dl than people with 130 or even 160. If high LDL was a good marker for CAD, than reality would reflect this. But it doesn't! Any good explanation for this phenomena, please?


Dr. B G said...

*haa* THANK GOD Pharma is cannabalizing themselves in obvious fashion now.

--SHARP, IMPROVE-IT (higher cancer)
--Lantus insulin -- increases breast CA 31%

Excellent post Peter -- what a way to follow up on J-LIT.

mark said...

Hey Peter.

Stephan wrote about the Finnish Mental Hospital Study on his blog.



homertobias said...

Dr Davis just posted a new RM Krauss article from J Clin Invest. He's now fooling around with fructose.
Dr B G
My favorite was ENHANCEIT. What a debaucle.

Dr. B G said...


Yeaaahhh ENHANCE my bloody PANTS.

Hey what do you guys think of this wonderful guy...?

Is he a THINCer or what?


mark said...

Dr B G,

That's Eddie Vos. He is a thincs member.

Peter mentioned him here:


Dr. B G said...

I shoulda known... Thanks, G

Dr. Shintani said...

This doesn't destroy the lipid hypothesis at all. Do they have collective data on whether the group lowered their LDL? I imagine that the LDL as a group went down, and atheroma decreased. This would to some extent support the lipid hypothesis.

Also, studies show that MI patients in the hospital show lower cholesterol after their heart attack so you can't use hospital levels of cholesterol to make any conclusions.

In any case, how do you explain Ornish's very low fat diet inducing a low LDL and also reversing atheroma if, as you say, the lipid hypothesis isn't right?

Jim Purdy said...

Nothing I've seen yet has made me regret my decision to stop taking all statins years ago. I just don't see any good science that supports statins.

Peter said...

Hi Dr Shianti,

Of course this doesn't destroy the lipid hypothesis. It appears to be totally indestructable. But from the factual aspect, which part of "Similarly, the on-treatment atheroma volume, change in atheroma volume, and high percentage of subjects with atheroma regression did not differ by the achieved LDL cholesterol" or "Atheroma regression occurred in most patients and was not linked to the LDL cholesterol achieved" do you feel supports your hypothesis?

You have made the classic error of mistaking Ornish's combination of exercise, relaxation, group therapy and a diet which is low in both refined carbohydrate and omega six fats with a simple low fat approach. Don't feel bad about this, you are not the first by a long chalk. It is understandable but not forgivable.

As a lipid hypothesis supporter would you mind taking a peek here

If you would be so good as to download the pdf where it says [PDF 210K] (it's free) and scroll down to page 1092, table six, lines one and four. Why does a TC level of <160mg/dl (LDL around 80mg/dl) result in 6.23 times the cardiac death rate of those people with a TC of 200-219mg/dl?

I look forward to your explanation.