So much to blog; so little time! I want to continue with discussing the Plant Positive slide 33 from my last post, but I have more pressing personal concerns that are taking my attention.
I didn’t mention it when I was discussing the lipid results of my last diet experiment, but I was aware of the possibility that my high LDLc (low density lipoprotein cholesterol) could be due to “bad” genetics. The most likely suspect was that I had the ε4 variant of apoE that does better in times of famine and doesn’t do well in regulating LDLc levels in times of feast. Back in December, I had discovered a genome testing service that includes apoE4 as part of their package for cheap ($99) and I jumped on it! After all, for a little more than price of a good dinner for two and movies, it could perhaps shed light on the mystery. Such a no-brainer!
My “23andme” DNA results came back a few days ago, and, yes, as I suspected, I was heterozygous (only one copy of the bad boy) for the apoE4 allele. So, most of my mental energy these last few days has been focused on looking at the implications.
ApoE4 and Alzheimer’s / Cardiovascular Disease!
First, I want to acknowledge the source that galvanized me in getting the DNA testing done: ApoE4 – The Ancestral Allele. There are three very fascinating blog posts over there (you should go read them if you’re interested in apoE4), but the middle one (Jul 2011), has a quote from Dr. William Davis’ Track Your Plaque blog where he talks about the great results he gets putting people on low-carb. But then he says:
But there’s one group of people who can experience unexpected effects with this diet: The 25% of people with apoprotein E4.
Everybody has two genes for apo E; the most common type is apo E 3/3. Around 1 in 4 people have 1, less commonly 2, genes for apo E4.
I hate apo E4. I hate apo E4 because it means I’ve got to dust off the nonsense I used to tell patients about cutting their fat, cutting their saturated fat. But that’s what apo E4 people have to do. But it doesn’t end there.
Apo E4 people also typically have plenty of small LDL particles triggered by carbohydrates. Put fats and carbohydrates together and you get an explosion of small LDL particles. Remove fats, small LDL goes down a little bit, if at all. Remove carbohydrates, small LDL goes down but total LDL (mostly large) goes up. The large LDL in apo E4 does seem to be atherogenic (plaque-causing), though the data are fairly skimpy.
So apo E4 creates a nutritional rock and a hard place: To extract full advantage from diet, people with apo E4 have to 1) go wheat-free, low-carb, then 2) not overdo fats, especially saturated fat.
Aha! So, when I read this back in November, I thought the apoE4 story and (similiar genetic variations) might go a long way towards explaining a lot of dietary controversy. I already knew that apoE4 was implicated in both Alzheimer’s (AD) and cardiovascular disease (CVD). But I didn’t know it was so prevalent – 1 in 4!! I was having trouble with my LDLc being jacked to the roof, and yikes! … I didn’t know apoE4 might produce really bad, “bad” LDL. It seemed vital that I get tested … And so I did; and now, here we are.
Which is the greater risk for me, AD or CVD?
I don’t want to sound like a commercial for it, but the 23andme site is a wealth of information. Here’s what they show for my AD risk:
And here’s what they show for my CVD risk:
But wait! Isn’t that interesting!! According to them, the biggest influence on my CVD risk is the first SNP on the left, 9p21. So my excess AD risk is canceled out by my lower CVD risk???
But what about apoE? It’s not even mentioned by 23andme among the other 14 risk markers! WTF? It is true that apoE4 elevates LDLc, right? Is it untrue that it elevates CVD risk? I sent an email inquiry to 23andme about this, but in the meantime, it’s time to hit PubMed!
Does apoE variant associate with LDLc and CVD risk?
First, the LDLc. Here‘s a full, free 2010 paper verifying the apoE association with lipid variations:
The ε4 allele was associated with
- higher total-cholesterol (p < 0.001),
- (higher) LDL-C (p < 0.001),
- (higher) total-cholesterol/HDL-C ratio (p < 0.001),
- (higher) LDL/HDL-C ratio (p < 0.001),
- lower HDL-C values (p < 0.001)
The actual lipid numbers show a gradient from good (ε2) to bad (ε4), all with strong (p<0.001) significance. I also found a review across 10 studies that agrees with that same good to bad gradient. So there’s no doubt that apoE4 raises TC, LDLc and lowers HDLc.
What about disease outcomes? Now it’s not so clear. In the Framingham Offspring Study (2001), they found significant association for men, but not women:
“After adjustment for all CVD risk factors, the relative odds
- in Group E2 men was 1.94 (P=0.004) and
- in Group E4 men it was 1.51 (P=0.0262).
- in Group E4 women it was 1.48 (P=0.0944).
So, the results for the men were definitely significant and looks like the result for the E4 women would have been significant too, if the study had been bigger.
But wait!! Not so fast there! WTF!! What’s going on with those E2 men? They’re supposed to have less risk according to The Lipid Hypothesis ‘cuz their LDLc is lower!! But is this study saying they had higher risk? Can anyone explain this? [NOTE: Even when the lipid risk is not multivariate removed, E2 still has higher risk than E3 – “Age adjusted period prevalence of CVD in men was 18.6% for Group E4, 18.2% for Group E2 and 12.7% for Group E3 (P=0.004)”]
I looked at the discussion in the Framington study to see how they explained it, and they said:
In men, in contrast to women, the o2 allele signiﬁcantly increases the risk for CVD. This association remains signiﬁcant after adjusting for age and all non-lipid risk factors and after adjusting for all risk factors including lipids. Since the o2 allele has been associated with elevated postprandial lipemia, the inﬂuence of the o2 allele on CVD risk could be due to the higher concentration of atherogenic triglyceride-rich lipoprotein (TRL) present in some carries of the o2 allele [54,55]
Very interesting. So, yes, they have low LDLc, but looks like high triglycerides trumps low LDLc, and guess what’s called for in this situation? Yes, although I haven’t researched it properly yet, I think these are the cases where a low-carb diet is said to work really well. Hmm, I wonder what Plant Positive would say about that??
I started to look further into PubMed at more studies and shortly began to realize that there seemed to be no consistency at all with respect to outcomes from all this elevated LDLc. What to make of this? Too much to continue here – it has to go it it’s own blog post later.