CT Scans for Statin Eligibility: A Pandora's Box to Avoid

Steven E. Nissen, MD; Peter Libby, MD


December 07, 2015

This feature requires the newest version of Flash. You can download it here.

Steven E. Nissen, MD: Hello. I'm Steve Nissen. I'm a cardiologist with the Cleveland Clinic, and I'm here with Dr Peter Libby of the Brigham and Women's Hospital in Boston. We're going to talk about CT scanning for calcium scoring, and specifically using CT scanning to determine who should get statins.[1]

Peter, you and I agree on this particular topic. Perhaps you can start by giving your perspective.

Peter Libby, MD: Well, let me just tell you my objections to the recent proposition that we should do mass screening with calcium score to allocate statins. My first concern is that any technique that involves radiation has a radiation risk. Then, I'm concerned about the "incidentaloma" effect. I'm concerned about what I call "Pandora's box," and I think that we agree about that. We also have to worry about clinical effectiveness and cost-effectiveness. I'd like to discuss those issues and put them on the table.

Dr Nissen: Let's start with a couple of these. I was incredulous of the idea that you do a test, which—if you pay full price—is $800-$1000 [Editor's Note: This price refers to CT angiography]. Now, a generic statin can be obtained at a big-box store for $10 for a 3-month supply. So we're going to do a $1000 test to figure out whether we should give a drug that costs $40 or $50 a year? It would take 20 years to recoup the investment. How can that possibly make any sense?

Dr Libby: Because there are a lot of patients who are reluctant to take statins. I think, with the marketing campaigns for the proprotein convertase subtilisin/kexin type 9 (PCSK9) inhibitors, that we're going to see an epidemic of statin intolerance—or, I call it, "statin reluctance." So people are eager to find that they don't need statins.

In fact, one finding from the recent studies was consistent with the overprediction of risk in the American College of Cardiology (ACC) and American Heart Association (AHA) Pooled Cohort Equations Risk Calculator, which really was hotly debated from its release in 2013.[1,2] Many people pointed out that there are circumstances under which that Pooled Cohort Risk Calculator may overestimate risk, so that's not a surprise.

Dr Nissen: One of the problems I have is the application of calcium scanning in the United States and what happens to patients—to real patients. It's one thing to publish an erudite analysis of the effectiveness of [calcium scanning]. But in reality, what happens is, people set up a CT scanner. They often advertise to get patients to have scans done. It's not paid for by insurance, for good reason. The calcium scan then reveals a score that is high or very high, for [the patient's] age. The next thing that happens is the physician solemnly tells the patient, "I've got good news for you, and I've got bad news. The bad news is, you've got a high calcium score. The good news is, we can do your cath tomorrow."

Dr Libby: Right. That's what I called the Pandora's box problem. If you do an imaging studying and screening, you're going to find some grief. And then when you do, this is a perfect pathway to take someone who's healthy, an asymptomatic person—they may end up, even without routine fractional flow reserve (FFR) measurements, getting a stent that obligates them to dual-antiplatelet therapy. Let's say they have an undercurrent issue, which requires cessation of their dual-antiplatelet therapy. Then they're at risk for a fatal disease: abrupt stent thrombosis.

So we've taken a perfectly normal, asymptomatic person and given them a disease that can lead to a fatal complication.

Dr Nissen: I have to be frank here. I see about one perfectly healthy patient a week who has been harmed by a calcium score found incidentally. The person ended up getting overly aggressive treatment, including intervention. They got in-stent restenosis. They got stent thrombosis, and something terrible happened. So, we turn people, sometimes people who are very healthy—runners, very active—into cardiac patients simply by doing an inappropriate screening test.

If there's one thing that's clear, it's that we've moved away from mass screening, not toward mass screening as a society, because it's turned out that in many cases it hasn't benefited patients.

Dr Libby: Right. And one of the kinds of harms that a mass screening within imaging technology can generate is the incidentaloma effect. In the radiology literature, it's pretty clear that it's in the double-digits percent. Between 10% and 14% of screening studies will show some findings that require follow-up. This generates a great deal of anxiety on the part of the patient and [his or her] family, scrambling around to find a prior film. Often biopsies are ordered. Those can lead to pneumothoraces, bleeds, and other complications. Again, we've taken a healthy person, created anxiety, and put them at risk for really serious complications because of incidentalomas.

If you screen millions of people, if there's even a 5% incidentaloma effect of findings that require some follow-up, you can see a stream of morbidity that's going to cascade. So it's another issue.

Dr Nissen: We see this in other areas. There's a movement away from getting prostate-specific antigen (PSA) tests on every male because of some of the downstream effects: You get a high value, a series of biopsies. You have a disease—unclear whether it is life-threatening—that ends up getting treated and causing a lot of morbidity. I worry that we're in the same realm with calcium scanning.

Dr Libby: There's another issue with calcium scans that we found in the intravascular imaging ultrasound (IVUS) study that I was able to participate in through the Study of Coronary Atheroma by Intravascular Ultrasound: Effect of Rosuvastatin Versus Atorvastatin (SATURN),[3] confirmed by the IBIS 4 trial.[4] That is, high-intensity statin treatment, which reduces events, actually increases calcification.The concept that statins—in proportion to their low-density lipoprotein (LDL)-lowering potency, roughly—will increase calcification is actually corroborated by a meta-analysis recently published by Paolo Raggi, who's one of the pioneers in this area.[5] So, serial calcium scanning holds no water whatsoever, because we know that there's a directionally opposite effect: increase in calcium with statin therapy, decrease in events.

Dr Nissen: The advocates say, "Well, we should do this because if you have a calcium score of zero, that you're essentially immune from an event. That's why we should do it: to look for people who have zero values." What is your thought about that?

Dr Libby: The Multi-Ethnic Study of Atherosclerosis (MESA) researchers found a very low incidence—over 5 per 1000 patient-years—of an event.[1] But that's over 10 years. I don't know about your perspective, Steve, but I expect to be around a lot longer than 10 years. A lot of my patients have a lifetime perspective, not a 10-year perspective. We know that this disease percolates in the arteries for many decades. I wouldn't be quite so sanguine, if I had a zero calcium score, that I was out of the woods.

Dr Nissen: In fact, it isn't really the calcified plaque that leads to events, is it? It's that lipid-laden noncalcified plaque that in many cases is responsible for myocardial infarction.

Dr Libby: That result came out of the IVUS studies as well, and also the CT angiography (CTA) studies. The morphologic features on CTA that correspond to future events includes spotty calcification—not necessarily the kind of macrocalcification that we see on CT scans—and outward expansion rather than stenosis.

Dr Nissen: Let me raise one more issue. A lot of the advocates have pointed out that electron-beam computed tomography (EBCT) uses a relatively low radiation dose. But one of the things I've noticed is that, in practice, often people have the calcium score obtained during a CTA—that's a full-dose CTA. In that case, the calculated incidence of malignancy is really quite significant.[6] I worry that you do a screening test that isn't needed, and what you end up doing is producing, years later, a malignancy that will then have a life-threatening effect on the patient.

Dr Libby: And then this other side: not using a zero calcium score to absolve people of the need for statin therapy, but also trying to target therapy based on calcium. And what I'd like to challenge the imaging community to do is a clinical trial where therapy is allocated on the basis of calcium score, and we show a clinical benefit. That's what is called clinical efficacy.

With respect to inflammatory biomarkers, that's exactly what we did in the JUPITER (Justification for the Use of Statin in Prevention: An Intervention Trial Evaluating Rosuvastatin) trial: 17,000 people, years of our lives, to actually do a trial to show that if we allocate a therapy on the basis of a biomarker, we could produce a clinical benefit.[7] And we haven't seen that, nor do I think we will with calcium scores.

Dr Nissen: That's a great point, and we're going to leave it there. I'm Steve Nissen, and I'm delighted you could join us. Thank you for listening to this discussion about calcium scoring with CTA.


Comments on Medscape are moderated and should be professional in tone and on topic. You must declare any conflicts of interest related to your comments and responses. Please see our Commenting Guide for further information. We reserve the right to remove posts at our sole discretion.