Before we get into the big job of interpreting cholesterol numbers, let's review what cholesterol actually is.

Cholesterol is cholesterol: a waxy steroid of fat that serves as an essential structural component of cellular membranes and in the production of steroid hormones, vitamin D, and bile acids. Contrary to what the terminology indicates, there's actually only one "type" of cholesterol in the human body, and it's called, quite simply, cholesterol. What we think of when we use the word "cholesterol" is actually a lipoprotein - a fatty conglomerate of protein and lipids that delivers cholesterol and fat and fat-soluble nutrients to different parts of the body. It's not just free cholesterol floating around in your blood; it's cholesterol bound up by lipoproteins.

So LDL, HDL, VLDL, all those (in)famous measurements we get at the doctor's office are just different types of lipoproteins. They're not actually cholesterol. I discussed this briefly a couple years back, and there's always Griff's big primer in the forum, so take the time to go check out both. And also take a peak at The Definitive Guide to Cholesterol for review.

Okay, let's talk about the most commonly bandied-about cholesterol numbers: LDL-C and HDL-C. What do they really mean? What are they actually measuring?

To understand what these numbers mean, let's play the freeway analogy game. Both LDL-C and HDL-C, the standard, basic readings you get from the lab, do not reflect the number of LDL or HDL particles - the number of lipoproteins - in your serum. Instead, they reflect the total amount of cholesterol contained in your LDL and HDL particles. Hence, the "C" in LDL/HDL-C, which stands for "cholesterol." Measuring the LDL/HDL-C and then making potentially life-changing health decisions based on the number is like counting the number of people riding in vehicles on a freeway to determine the severity of traffic. It's data, and it might give you a rough approximation of the situation, but it's not as useful as actually counting the number of vehicles. A reading of 100 could mean you're dealing with a hundred compact cars, each carrying a single driver, or it could mean you've got four buses carrying 25 passengers each. Or it could be a couple buses and the rest cars. You simply don't know how bad (or good) traffic is until you get a direct measurement of LDL and HDL particle number.

Say you go ahead and get those particle numbers directly measured. You're still limited, because that is just a single datapoint from a specific time in your life/day/week. Analogies are fun and helpful, I think, so let's take this traffic and freeway stuff further. To get an accurate idea of traffic, you need constant updates, right? Imagine you counted the number of cars on the freeway at 12:05 on a Saturday afternoon four weeks ago. That's great, but what does it tell you about traffic at 5 PM on a Thursday? Even though it's the same stretch of asphalt/artery, we can't divine much at all from that single measurement. You need more data points. That traffic fluctuates wildly is entirely uncontroversial. Any southern Californian could tell you that. But did you know that LDL, HDL, and total cholesterol readings in the same person can fluctuate just as wildly, oftentimes enough to move that person from "desirable" to "high risk" and back to "desirable" lipid status without any nutritional or lifestyle changes in the span of a few mere weeks?

In biology, a single snapshot rarely, if ever, tells the whole story. Who woulda known?

But just because the standard cholesterol test is but a snapshot of a dynamic system in flux doesn't negate the potential usefulness of getting your cholesterol checked. As much as Conventional Wisdom has gotten things wrong when it comes to cholesterol and heart disease, the two do have a relationship together. There is a connection; contrary to what the AHA might think, we just don't have it ironed out yet. In my opinion, the most persuasive hypothesis about the real causes of atherosclerosis and heart disease comes from Chris Masterjohn and is highlighted in his recent AHS talk, "Heart Disease and Molecular Degeneration," and on his blog. It's a synthesis of the two prevailing notions regarding cholesterol and heart disease - the one which says elevated blood cholesterol plays no causal role in heart disease and the one which says elevated blood cholesterol is the primary cause of heart disease - and it goes something like this:

LDL receptors normally "receive" LDL particles and remove them from circulation so that they can deliver nutrients and cholesterol to cells, and fulfill their normal roles in the body.

If LDL receptor activity is downregulated, LDL particles clear more slowly from and spend more time in the blood. Particles accumulate.

When LDL particles hang out in the blood for longer stretches of time, their fragile polyunsaturated fatty membranes are exposed to more oxidative forces, like inflammation, and their limited store of protective antioxidants can deplete.

When this happens, the LDL particles oxidize.

Once oxidized, LDL particles are taken up by the endothelium - a layer of cells that lines the inside of blood vessels - to form atherosclerotic plaque so they don't damage the blood vessel. This sounds bad (and is), but it's preferable to acutely damaging the blood vessels right away.

So it's the oxidized LDL that gets taken up into the endothelium and precipitates the formation of atherosclerotic plaque, rather than regular LDL. OxLDL, poor receptor activity, and inflammation are the problems. But since measuring oxidized LDL in serum is difficult (oxidized LDL gets taken up out of serum and into the endothelium rather quickly) and expensive, we need other, more realistic, more obtainable methods. We need to work with what we've got. It would be great if a doctor could quickly order up an "LDL receptor activity" test, but I don't see that happening anytime soon.

Enter the various lipid panels.

First up is your basic lipid panel, the standard test the average doctor is going to order for a patient. If you go this route, you'll typically get four measurements: total cholesterol (TC); high density lipoprotein cholesterol (HDL-C); low density lipoprotein cholesterol (LDL-C); and triglycerides.

Total cholesterol

What they say: Get that TC below 200, or else (you'll have a heart attack or you'll have to pay a higher health insurance premium, if we take you on at all).

My take: Mostly meaningless. Even though the epidemiological evidence suggests a TC between 200 and 240 mg/dl is best for all-cause mortality, we can't hang our hats on it. First off, total cholesterol is limited because it's only telling us the amount of cholesterol contained in all our lipoproteins without saying anything about what kind of lipoproteins we have or how many there are. Second, total cholesterol is limited because it's determined by a bizarre formula - HDL-C+LDL-C+(Triglycerides/5) - that reduces various types of blood lipids, each with a different role in the body and a unique impact on our risk for illness, to mere numbers. Someone with low HDL and high triglycerides could easily have the same TC as someone with high HDL and low triglycerides, so long as the numbers work out. Whether it's being used to predict wellness or disease, total cholesterol by itself is mostly meaningless.


What they say: "Good" cholesterol. It's the "garbage truck" that cleans up "excessive" cholesterol and fat from tissues, so the higher the better! Though men and women should strive for levels exceeding 60 mg/dl, above 40 is acceptable for the former and above 50 is acceptable for the latter.

My take: Higher HDL-Cs correlate strongly with better cardiovascular health. No real argument here. Higher HDLs are desirable. Just remember, it's only a snapshot of a glimpse into the cholesterol content of your HDL particles. Among most groups tested, the TC:HDL ratio is actually a strong indicator of heart disease risk, with higher ratios corresponding to higher risks. Note, though, that no Primal Blueprint adherents were among the groups analyzed, ever.


What they say: Get it as low as humanly possible! I want that low density lipoprotein so low as to be nearly nonexistent. Your body obviously hates you; otherwise, it wouldn't be producing a potently toxic substance and sending it directly into your endothelial cells to form atherosclerotic plaque! Of course, we're not actually measuring the number of low density lipoproteins, just the amount of cholesterol contained in them, but still!

My take: While a high LDL-C may indicate a problem, remember that LDL-C only indicates the total amount of cholesterol in your LDL particles. You could easily have a few large particles (good) or a bunch of smaller, denser ones (bad, might indicate poor LDL receptor activity and an LDL that likes to hang out in the blood), but LDL-C alone isn't enough to know. It's also just a moment in time, whereas what you're interested in is the trend. If the trend indicates a steady rise in LDL-C, however, that could hint at poorer LDL clearance and lower LDL receptor activity (and greater susceptibility to oxidation).


What they say: Lower would be better, sure, but you really gotta do something about that LDL! Anything less than 150 mg/dl is fine.

My take: High triglycerides correlate strongly with low HDL and smaller, denser LDL. High triglycerides, then, could indicate more oxidized (or oxidizable) LDL. The triglycerides of most Primal eaters, especially those on the lower carb side of things, usually hover well below 100 mg/dl. Triglycerides come packaged in VLDL, or very low density lipoproteins (which are calculated by dividing your triglyceride count by 5).

So, what can we learn from a standard lipid test? Not much, actually. We can learn from standard lipid tests, however. If we take a series of regular ol' lipid measurements, preferably one pre- and several peri-Primal, we can get an idea of our metabolic health.
Look for:

  1. Trends - Are your triglycerides going down over time? That's great. Is your HDL trending up? Also good.
  2. Normal fluctuations - Your numbers can jump around 20-30 points in either direction between readings without it necessarily meaning anything.
  3. TC:HDL-C ratio - Lower is better and indicates fewer LDL particles.
  4. Triglyceride:HDL-C ratio - Lower is better and indicates larger LDL (and, usually, fewer) particles. Ideally, this will be close to 1 or lower; one study (PDF) found that 1.33 was the cut off.
If you're going to get your cholesterol tested, and the basic labs just aren't cutting it, you might as well go for one of the premium lab tests: the NMR LipoProfile or maybe the VAP. Rather than rely on indirect estimates and formulas, NMR and VAP directly measure the size of your lipoproteins. I find NMR to be far more useful, because in addition to measuring particle size, it measures particle count (whereas VAP only estimates the count).

But you probably have holiday shopping to do, and I don't want to drone on for too long, so I'll leave it at that for now. Next week, I'll pick up where I left off and get into what you can expect from NMR and VAP testing, including the downsides and the advantages. After that, I'll go into some strategies for improving your numbers - or, rather, improving your health which in turn should improve your numbers.