Screening for Coronary Disease

The newspaper recently reported on new research meant to help doctors better identify patients at risk for heart attacks.  Dr. Eric Topol, the director of California's Scripps Translational Science Institute, announced the discovery of a clue found in the blood of people who are apparently on the verge of suffering the big one.

"On Wednesday, Scripps researchers reported a new lead—by searching people's blood for deformed cells that appear to flake off the lining of seriously diseased arteries.  Topol's team measured high levels of those cells floating in the blood of 50 people who had just had a heart attack. Next, Topol said, his team will begin studies to learn how early those cells might appear before a heart attack and if spotting them could allow the use of clot-preventing drugs to ward off damage."

Is this big news or what!?

Actually, I think the "or what" part is more accurate.  As the article correctly points out, we have no testing that can tell us a heart attack is around the corner.  Stress testing, nuclear imaging, and even coronary angiography can identify the presence of atherosclerosis and coronary blockage, but none of these exams predict when a 70% narrowing will destabilize and cork off.  (For a thorough exegesis on this subject please refer to a previous post.  For a thorough definition of exegesis please refer to this link.  If you wonder why I like hyperlinks so much please click here and read the final paragraph.)

The problem with this research is that it doesn't get us much closer to what we really need.

But before I get into what we really need, I have to add that this research is not the first scientific discovery that allows us to identify from blood samples persons either in the throes of a heart attack or whose karma is leading them in that direction.  The most promising predictor for someone in need of a trip to the cath lab is the high sensity C-reactive protein (hs-CRP), a serum assay that has been around for several years.

The hs-CRP is a marker of inflammation—it rises in response to illnesses that stir up your immune system.  Infections, autoimmune disorders like rheumatoid arthritis and lupus, and certain types of cancer can lead to hs-CRP levels that are off the charts.  Coronary atherosclerosis is, at least in part, an inflammatory condition with participation from monocytes, macrophages, interleukins, and lots of other big words.  These substances get agitated in the run-up to plaque rupture and clot formation, and tend to be reflected in the hs-CRP level.

When hs-CRP became more widely available, we doctors initially believed we could use it as a clinical tool to tease out the subset of patients at highest risk of cardiac death.  We simply needed to screen everyone with this relatively simple blood test, identify those with the highest level of hs-CRP, and . . . and . . .

Well, that was the problem.  What, exactly, do we do with the results?  Rush the patients to the cath lab for a prophylactic stent?  Stick them on mega-doses of Lipitor?  Tell them to always stay close to an external defibrillator?  Or, better yet, carry a car battery around with them wherever they go and quickly stick their tongue on it at the first sign of trouble?

The problem with hs-CRP is that a lot of people with elevated levels have something unrelated to coronary disease as the cause of the lab abnormality.  Acting on the results would condemn half of all rheumatology patients to a life lugging around a DieHard battery and jumper cables.  It's highly likely that the same weakness is inherent in the new blood test cited above—there are probably a dozen other medical conditions that can cause deformed blood cells.  What happens when we prophylactically treat all these people with clot-preventing drugs, as suggested by the study's author?

By now you're probably mumbling to yourself "Dr. Van De Graaff is just jealous that he doesn't do cool research that merits national press coverage.  He's turned into another naysayer."

Nay, I say—such is not the case (okay, maybe I'm a little envious—the best research I ever did took place in the 3rd grade and involved a homemade volcano).  I'm all for using screening tools to determine who needs the most aggressive treatment.  My only issue is that I've been in this business long enough to see test after test come and go without any significant improvement on a simple screening method that has worked well for years.

Our current guidelines recommend that we plug some easily identifiable variables into a simple algorithm to determine who (among those with no known coronary disease) is at highest risk for a cardiac illness.  This equation, known as the Framingham Risk Calculator, uses age, gender, lipid levels, smoking status, and blood pressure to produce an estimate of a person's risk of having a heart attack in the next 10 years.  Try it for yourself at this website (you have to know your HDL and total cholesterol values).

Once you know whether you fall into the 2% or the 20% category you have to ask yourself "OK, what next?"  The National Institutes of Health publishes a nice "at-a-glance" summary of their recommendations, but it basically comes down to helping us decide who should receive a cholesterol pill ("lifestyle modification" is recommended for everyone).  It doesn't help us decide who needs to make a trip to the auto parts store.

Still, for all its shortcomings, the Framingham approach has worked remarkably well and is one of the key reasons why the per capita rate of death from cardiovascular disease has dropped in this country.  Having cool blood tests like hs-CRP and "deformed cells" has unfortunately done little to improve on this strategy.

Now, to answer my original question: What do we really need?  We need 2 things:

1. A test that accurately predicts who will have a heart attack in the near future.
2. A way to intervene in those individuals that actually prevents the heart attack.

Neither of these exists yet.  I have to give credit to the people at the Translational Science Institute for at least working on the issue.  Maybe they'll be the people who crack the code in the future.  Until then, excitement about deformed cells is probably a little premature.  At this point you're still better off with the car battery.