Re: Re: ApoB/ApoA1 ratio and its significance to CHD - Chris M.

[Guest guest]

On 3/1/08, chriskjezp <chriskresser@...> wrote:

> Look at me - I'm having a conversation with myself!

You are doing pretty good with it so far, too. :-)

> I was mistaken about the meaning of the ApoB/ApoA1 ratio. ApoB is a rough

> marker for the

> number of LDL, IDL and VLDL particles in circulation, and ApoA1 is a marker

> for HDL. So this

> is measuring the ratio between LDL-IDL-VLDL and HDL.

Right.

> Still, I wonder about the correlation between a high ApoB/ApoA1 ratio and

> small, dense LDL.

I didn't read the study, but I'll say that first, don't forget that

the ApoB:LDL ratio is the inverse indicator of particle size whereas

the ApoB is the indicator of total LDL (plus IDL and VLDL). So the

reason you could have two independent effects is that one is saying,

all things being equal, the more LDL you have, the greater the risk

(which is true if all things are equal, such as the proportion of

oxidized LDL, because it would mean you'd have more oxidized LDL), and

the other is saying, all things being equal, the smaller and denser

your LDL is the greater your risk. And the combined effect is saying

that when a greater proportion of your LDL is small and dense AND you

have more total LDL and thus even more small, dense, LDL, your risk is

the highest. I'm not sure why a high HDL/LDL ratio would be

correlated with pattern A, but it could result from the basic balance

of lipid transfer among lipoproteins that leads to each.

> I also wonder whether a high ApoB/ApoA1 ratio is actually a causative factor

> for CHD, or if it

> is simply an association - as could be argued in the case of high LDL

> levels.

> What do you think, Chris?

I'm not sure. There is a protective effect of HDL on LDL oxidation,

it delivers vitamin E to the endothelium, which is

anti-atherosclerotic, and it apparently engages in some reverse

cholesterol transport, so there are various mechanisms by which it

could be causally protective.

Chris

[Guest guest]

Hi

> There's a lot of contradictory data out there. Ravnskov et al. have

> presented some pretty

> convincing evidence that LDL isn't a risk factor when it comes to CHD

> mortality or total

> mortality.

I haven't evaluated all the information out there, but usually when

the study is negative, as far as I've seen, there is a positive

association that doesn't reach statistical significance. This is

especially true in the older groups, where you go from a pretty strong

and significant association in young people, to a weaker and

non-significant association in older people, and then no association

for people in their 70s and an inverse association for people in their

80s. I think there are a lot of folks who tend to treat

non-significant findings as the same thing as negative findings, when

in actuality it could just mean the study didn't have the statistical

power to pick up the association.

Comparatively far fewer studies have looked at ox-LDL, and this

association is much stronger. The proportion of oxidized

phospholipids in the LDL particle multiplied by the total LDL level

(and thus the total concentration of ox-LDL) is eight times more

predictive than the LDL level itself.

> And there's a lot of data which shows that high cholesterol in

> the elderly is

> protective, whereas low cholesterol increases chances of dying from a heart

> attack.

Right, though this is in the very old -- like in the 80s.

Chris

[Guest guest]

On 3/1/08, chriskjezp <chriskresser@...> wrote:

>

> > I haven't evaluated all the information out there, but usually when

> > the study is negative, as far as I've seen, there is a positive

> > association that doesn't reach statistical significance. This is

> > especially true in the older groups, where you go from a pretty strong

> > and significant association in young people, to a weaker and

> > non-significant association in older people, and then no association

> > for people in their 70s and an inverse association for people in their

> > 80s. I think there are a lot of folks who tend to treat

> > non-significant findings as the same thing as negative findings, when

> > in actuality it could just mean the study didn't have the statistical

> > power to pick up the association.

> But again, is this an association or a cause?

I agree this question remains, but I was responding to your statement

that Ravnksov presents a compelling case that the association does not

exist. And I'm saying that I think part of this is because he treats

associations that lack statistical significance the same as negative

associations, so I think the association is actually there (in age

groups under 70 or so), just fairly weak in the older age groups.

> As you pointed out in your

> Rho activation

> article, with inflammation we might expect to see higher LDL because of

> increased

> synthesis of mevalonate but inhibition of squalene synthase. Likewise,

> cholesterol is the

> precursor to all stress hormones so perhaps the elevated levels of LDL are

> simply a marker

> for stress - and it is actually stress that is causing the heart disease,

> not LDL.

I agree that these are confounding variables that probably account for

much of the association with LDL.

> Maybe that's why we see a decreasing correlation between LDL and CHD as

> people get

> older. Young men tend to have much more stress in their lives, so we'd

> expect to see

> higher levels of LDL in those populations. As men & women get older and

> retire from the

> workforce, it's likely that their stress levels decline. Of course their

> LDL is still elevated,

> but perhaps for reasons other than stress - reasons that are less

> atherogenic and

> causative for CHD.

I don't think so, at least that isn't all or most of it. Young people

with familial hypercholesterolemia are at a HUGE increased risk of

heart disease mortality, while older people are at only a mild

increased risk, and very old people are at a decreased risk. Here,

the cholesterol levels are not secondary to stress or inflammation,

but are a direct result of insufficient cholesterol uptake, and so the

disease risk is probably directly mediated through either a higher LDL

level per se or a longer time spent in the blood for each LDL particle

(both of which probably increase ox-LDL).

> > Comparatively far fewer studies have looked at ox-LDL, and this

> > association is much stronger. The proportion of oxidized

> > phospholipids in the LDL particle multiplied by the total LDL level

> > (and thus the total concentration of ox-LDL) is eight times more

> > predictive than the LDL level itself.

> Yes, I'm pretty convinced of this now. If you happen to remember where the

> 8x statistic

> came from I'd love to see that reference.

http://www.ncbi.nlm.nih.gov/pubmed/16000355?dopt=Abstract

Looks like I may have been misremembering the findings a little bit --

most importantly this is limited to age 60 and younger (non-sig

inverse association above age 60) and also this is looking at

angiography rather than CHD mortality or MI.

> Yes, but still significant in my mind. Why would something that is

> supposedly a risk factor

> in younger people suddenly stop being a risk factor in older people? No one

> has

> explained this adequately to me yet. It's like saying that smoking causes

> lung cancer in

> young people but not in old people.

Perhaps because native LDL is helpful and ox-LDL is harmful. Since

LDL and ox-LDL correlate with each other, LDL might act as a marker

for ox-LDL. However, the proportion of LDL that is ox-LDL might

change with age, or, the relative protective effect of native LDL and

harmful effect of ox-LDL might vary with age. In other words, the

protective effect of LDL might become more important with age, and so

might overshadow the correlation between LDL and ox-LDL.

To give an example from something other than heart disease, this basic

finding was shown with mobility in the elderly:

http://www.ncbi.nlm.nih.gov/pubmed/16217006?dopt=Abstract

Check the full text if it doesn't say it in the abstract, but there

was an inverse correlation (not statistically significant, but fairly

substantial) between LDL and mobility limitation, while there was a

positive correlation between ox-LDL and the proportion of LDL that was

ox-LDL. Also, there was a correlation between ox-LDL and LDL. This

seems to suggest that ox-LDL is harmful and native LDL is helpful, but

that the helpful effect is difficult to tease out because ox-LDL and

LDL correlate with one another.

Chris