Note to our followers: Our nearly 13-year run of daily publication of new content on came to a close at the end of 2018. Publisher Gary Schwitzer and other contributors may post new articles periodically. But all of the 6,000+ articles we have published contain lessons to help you improve your critical thinking about health care interventions. And those will be still be alive on the site for a couple of years.
Read Original Story

Optimal balance of competing views on value of CT heart scans

Rena H. Barnett, 65, has high cholesterol, but a scan of her heart found zero calcium plaque. A study found people like her have very little chance of having a heart attack in the next decade. CreditMark Makela for The New York Times

Treatment guidelines suggest that nearly half of those over age 40 — nearly 50 million people in the United States — at least consider acholesterol-lowering statin to reduce heart attack risk.

But a new large study of people who had an inexpensive heart scan found that half of those who were statin candidates actually had no signs of plaque in their heart and very little chance of having a heart attack in the next decade.

Some cardiologists say the results could go a long way toward helping patients make a more informed choice about whether to begin taking the drugs.

The test is a CT scan that looks for calcium in coronary arteries, a signal that plaque is present. It used to be expensive — about $500 — but now typically costs between $75 and $100. Still, it is generally not covered by insurance and so is not often used to assess risk. The X-ray dose is about that of a mammogram.

Advocates for the scan say it should be used to “de-risk” people. It can let those who do not want to take statins know whether their chance of a heart attack is actually extremely low.

“Maybe this is a tool to actually do less,” said Dr. Harlan M. Krumholz, a Yale cardiologist and senior author of the paper, published on Monday in the Journal of the American College of Cardiology.

For those who have no objections to taking a statin, there is no need for a heart scan, Dr. Krumholz said. But for those who are reluctant to take them, he said, “I am willing to use this to refine their risk estimate.”

Others say that the test can lead to an array of other medical problems, some of which are gravely serious.

“The only reason to do things is to feel better or to live longer,” said Dr. Peter Libby, a cardiovascular disease specialist at Harvard’s Brigham and Women’s Hospital. With the scans, he said, that has yet to be established.

Heart researchers have long known that plaques in coronary arteries start out as pimple-like bumps but get waxy and hard and filled with calcium as time passes. Calcium shows up as white flecks in CT scans. The hard plaques are not the dangerous ones — it is the softer ones that rupture and cause a heart attack. But the amount of calcium in arteries can give a good idea of the presence or extent of coronary artery disease.

Dr. Krumholz and Dr. Khurram Nasir, a preventive cardiologist at Baptist Health South Florida, who conceived the new study, reasoned that research on heart scans had not been designed to help doctors make treatment decisions they face today.

Current guidelines have vastly increased the number of people who are eligible to take statins. Many people, though, are reluctant to take them. So, the investigators asked, could a scan identify those whose actual risk is lower than what was calculated?

The study subjects were nearly 4,000 men and women aged 45 to 84 and included blacks, whites, Hispanics and Asians who were recruited in 2000 through 2002. According to today’s guidelines, half had risk scores high enough that a statin would be recommended or should be considered. But guidelines then were more conservative, and in accordance with them, the subjects did not take the drugs. All had heart scans and half had no calcium visible on the scans. The subjects were followed for 10 years.

It turned out that the actual incidence of heart attacks or disabling chest pain in those with zero calcium was half or less than what the risk calculator predicted.

For example, a person who, according to the current risk calculator, has a 12 percent risk of a heart attack in the next decade should take a statin, the guidelines say. But if that person has a calcium score of zero, the actual risk turned out to be 4 percent, below the 7.5 percent threshold for recommending a statin according to the guidelines and below the 5 percent risk for considering a statin.

Dr. Nasir said he has been using the study’s findings in his clinic. First, he asks patients for whom statins are recommended according to the current guidelines if they would want to avoid taking the drugs if they turn out to have a calcium score of zero and an actual risk of less than 5 percent. Most tell him they would. He then sends them for a scan.

Continue reading the main story


The heart has four chambers, which are separated by valves and surrounded by muscle.

The right side pumps blood back to the lungs for more oxygen.

The left side pumps oxygen-rich blood through the body.

Cardiac arteries supply oxygen-rich blood to the heart muscles.

If an artery becomes clogged or blocked, the downstream muscle is starved of blood.

This is a heart attack.

But the study was observational, not the highest level of evidence. The problem, critics say, is that there has never been a rigorous study randomly assigning people to a change in treatment based on a scan and demonstrating that the change improves outcomes.

Dr. Libby of Harvard, for one, is leery. Although very few heart attacks may have occurred over a decade in people with no calcium, he said non-calcified plaques may be developing that could cause trouble. And the time span for worrying about a heart attack is more than a decade, he said.

One problem with the scans is what doctors call incidentalomas — unexpected incidental findings, like lung nodules. The new study reported such findings in 5 percent of patients, but radiologists have reported incidences in the double digits. All too often these findings start patients on a diagnostic odyssey, getting tests and biopsies, sometimes exploratory surgery, only to find that there was nothing wrong.

Routine heart scans of tens of thousands of people would uncover a “not negligible” number of incidentalomas, Dr. Libby said.

Then there will be the people, with no symptoms of heart disease, who turn out to have a high calcium score. Dr. Libby explains what often happens next: The doctor suggests an angiogram, an X-ray of the arteries. It shows one of the arteries is 70 percent blocked.

The cardiologist inserts a stent, a wire cage to keep the artery open, although many researchers doubt stents will prevent heart attacks in symptomless people on today’s medical therapy. Now the patient, with the newly inserted stent, has to take a powerful anti-clotting drug and aspirin for at least the next year. Because the drugs make bleeding more likely, the patient notices blood in his stool. Before he can have a colonoscopy to check on the blood, he has to stop the anti-clotting drugs for a week. But without them he risks getting a clot at the site of the stent and having what could be a fatal heart attack.

“Now we have taken a healthy person, asymptomatic, and turned him or her into a patient,” Dr. Libby said.


But others, like Dr. Daniel Soffer, a general internist at the University of Pennsylvania, see a real benefit in using heart scans to de-risk patients. “All the other biomarkers get blown away compared to the calcium score,” he said, adding that it is “far and away the best marker of risk.”

The new study justifies, to him, a practice he began years ago of using scans with the belief that they could be better than a risk calculator for some patients.

One of his patients, Rena H. Barnett, 65, had a scan a decade ago. Her mother died of a heart attack and Mrs. Barnett knew her level of LDL cholesterol, the bad kind, was very high at 190. But she said statins and other cholesterol-lowering drugs made her depressed and made her muscles ache so much she could not get out of bed in the morning. She tried lowering her LDL level by becoming a vegetarian, but it did not budge. Relaxation with yoga did not help.

But when Mrs. Barnett had a heart scan, she learned that her calcium score was zero. Five years later she had another scan. Zero again.

For now, Mrs. Barnett is not taking anything to lower her cholesterol levels. “It’s not that I feel good about it. It would be nice to have it lower,” she said. But her anxiety has lifted.

Heart Scan Can Fine-Tune Risk Estimate for Patients Considering Statins

Our Review Summary

Credit NHLBI.

Credit NHLBI.

This story hits almost all the high points we look for in a news report. It gives readers not only the basic summary of an analysis of how CT scan results showing levels of calcium deposits in coronary arteries are associated with 10-year rates of heart attacks and other cardiovascular disease, but it includes more than one skeptical voice and plainly discusses study limitations and potential harms. Still, our criteria are demanding, and the story falls a bit short on two points: disclosures of researcher relationships with industry and then alternatives available to people who might consider getting one of these scans. But the strengths of this story are an example of how to get it right.


Why This Matters

Too often, stories about medical tests fall into the “why not” category. This story lays out the potential benefits reported by researchers, but it also provides readers clear and useful explanations of the nuances and potential pitfalls of this sort of CT scan. The story is important because the number of people placed on statins has progressively increased over the years — and if some people can avoid these drugs without putting themselves at increased risk, then it’s probably good that they avail themselves of the opportunity.


Does the story adequately discuss the costs of the intervention?


The story reports that a CT scan looking for calcium in coronary arteries “typically costs between $75 and $100. Still, it is generally not covered by insurance and so is not often used to assess risk.” That’s enough information to clear our bar handily, although we’d note that such scans can also turn up lung nodules that require additional follow-up and related costs.

Does the story adequately quantify the benefits of the treatment/test/product/procedure?


The story reports that among study participants, those with a zero calcium score on a heart CT scan had only half the expected number of heart attacks or disabling chest pain over a 10-year period. That’s a relative comparison, which as we frequently note does not provide the full picture regarding benefit. But the story goes on to give readers a better sense of what these numbers mean in absolute terms. It explains the example of a person with a 12 percent risk of a heart attack over a decade based on the risk calculator recommended by the American college of Cardiology and others. That person actually turned out to have only a 4 percent 10-year risk, below the widely-used 7.5 percent risk threshold for recommending a statin.

The story makes clear that the scan does not provide a direct health benefit, but it provides information to people who are trying to decide about whether to start taking cholesterol-reducing statins. To help with that decision, it would have been nice for the story to include some discussion of the benefits associated with statin use, especially the numbers needed to treat (NNT) with a statin in order to save a life.

The story includes only one patient anecdote, a woman who was feeling depressed and had trouble getting out of bed when she took statins, and then decided to stop taking the drugs after having a CT scan that showed zero calcium. We often find fault with stories that introduce readers to only one patient, who claims to have greatly benefited; however, the placement of the anecdote low in the story and the preceding context that included limitations and expert cautions mean this single patient story doesn’t overwhelm the other information.

Link to online Atherosclerotic Cardiovascular Disease (ASCVD) Risk Estimator:

Does the story adequately explain/quantify the harms of the intervention?


The story clearly lays out some of the ways that testing can go awry. It reports that the CT scans expose people to about the same amount of radiation as a mammogram, implying a level that many people routinely accept, though it does not go into detail about potential radiation effects. The story also explains the risk of “incidentalomas”, things that can pop up on this sort of scan besides coronary artery calcium, such as nodules in lungs, that might lead a person down a path of further testing and treatment. And it spells out the potential consequences of having a scan that reveals high levels of calcium deposits, prompting further testing and treatments, perhaps leading to substantial harms.Overall, we think the harms are admirably covered.

Does the story seem to grasp the quality of the evidence?


The story points out that the study was observational, “not the highest level of evidence”. It goes on to explain that since people in the study were not randomly assigned to testing or no testing, the results do not demonstrate that changing current practice would change real health outcomes.

Does the story commit disease-mongering?


The story reports that proponents are suggesting this sort of scan only for people who need more information to make a decision about statin treatment. It also includes skeptics who doubt the value of the test for most people. The tone is calm, not shrill. We saw no evidence of disease mongering.

Does the story use independent sources and identify conflicts of interest?

Not Satisfactory

This was a split decision that we ultimately ruled Not Satisfactory. There are multiple independent sources in the story, so it clearly addresses that element of the criterion. However, the story does not report the financial relationships some of the study authors have with manufacturers of scanners and other medical devices. One of our reviewers argued that these scanning machines are essentially generic and that therefore a conflict of interest disclosure is not relevant in this case. Two others voted that the potential conflict, minimal though it might be, was still directly relevant to the story and merited a brief mention. This is an instance where a binary Satisfactory/Not Satisfactory rating does not capture the nuance involved in our decision-making, which is why we always encourage readers to pay more attention to the comments than the ratings.

Does the story compare the new approach with existing alternatives?

Not Satisfactory

There is no mention of alternative risk calculators that include lifestyle and other factors, which people on the fence can use to help clarify their thinking about statin treatment. It also does not clearly point out that people who are uncertain about statin treatment can try the drug to see if they experience bothersome side effects. In other words, there are alternative ways for people to gather information that could help them decide about statin treatment.

Does the story establish the availability of the treatment/test/product/procedure?


The story makes clear that this test is readily available, and points out that it is generally not covered by insurance.

Does the story establish the true novelty of the approach?


The story notes that CT scans for coronary artery calcium are not new, and that what is new is the analysis of how test scores are associated with 10-year rates of heart attacks and other cardiovascular disease.

Does the story appear to rely solely or largely on a news release?


The story clearly does not rely on a news release. There is original reporting that includes feedback from the original authors of the paper as well as independent experts.

Total Score: 8 of 10 Satisfactory

Comments (3)

Please note, comments are no longer published through this website. All previously made comments are still archived and available for viewing through select posts.

Jeanne Lenzer

October 8, 2015 at 4:27 pm

It’s hard for me to imagine how HealthNewsReview could award an 8/10 score to the New York Times article on CT scanning. The story was based on the bizarre premise that to limit overprescribing of statins, people should undergo a test that has no proven clinical benefit and does have significant harms.
There are two key problems with the premise of the article; first, overprescribing of statins should be dealt with by not following compromised guidelines recommending widespread prescribing; instead prescribers should follow the independent assessments of groups like Therapeutics Initiative and Second, readers should review independent assessments of the science regarding coronary calcium CT scanning; despite attempts to prove clinical value, none has been found. Promoters of the scan rely on surrogate markers and have ignored the serious downsides of the test, including false positives, false negatives and the devastating health and financial costs of downstream testing caused by incidentalomas and false positives.
Finally, the NYT piece relied on anecdote to persuade (it had to since it had no quality outcomes data to do otherwise). The reaction engendered by promoting this test as a way to reduce overtreatment could be anticipated and comprises a marketing strategy used by promoters of the test, who had encountered opposition when it was promoted simply as a way to pick up more people who they say need more treatment but weren’t getting it. There is no clinical proof to date that the net effect of the scan will be anything other than, as Vikas Saini, President of the Lown Institute put it: a recipe for mayhem. For more on the science (and the moneyed interests) see my article on a “documentary” promoting the scan in The BMJ that appeared on the same day as the NYT article:


    Andrew Holtz

    October 9, 2015 at 8:59 am

    The issues Jeanne Lenzer raises are real, but it is important to distinguish between the debate over proper use of these calcium scans and the separate matter of the quality of reporting on that debate.

    Our review of the story is based on how it rated on our criteria, not whether CT scans for coronary artery calcium are a good thing to do. The story summarized the journal article by Krumholz et. al, but it also clearly pointed out important limitations in the work and prominently featured experts who presented a different view of the evidence. Jeanne Lenzer comments that the test “has no proven clinical benefit”. We noted that the story reported, “there has never been a rigorous study randomly assigning people to a change in treatment based on a scan and demonstrating that the change improves outcomes.” Lenzer also comments that the test “does have significant harms”. We gave the story credit for reporting both that this sort of scan can turn up “incidentelomas” leading to further testing and treatment, along with accompanying risks, and for presenting a detailed scenario of the potential consequences of a scan that identifies calcium deposits in coronary arteries, perhaps even resulting in death from treatment complications. She criticizes the use of an anecdote of a patient who was glad she was scanned. We noted that we often fault stories for including only one positive anecdote, but that in this case that anecdote was placed deep in the story, following the reporting on study limitations, several critiques from independent sources and detailed scenarios about potential harms resulting from this sort of scan.

    We did not see the story as blindly favoring proponents of testing. Rather, we saw that in reporting on a study that suggests a reason for more CT scans, the reporter and editors noted limitations in the study, included skeptical independent voices and explained to readers some of the ways such scanning could harm people.


      Jeanne Lenzer

      October 9, 2015 at 10:54 am

      I appreciate your comments, Andrew and the need to separate a decision about the science from the nature of reporting. However, it’s rather undeniable that the thrust of the story was that the scan can prevent overtreatment by reassuring patients they might not need statins (just look at the Twitter comments of some doctors). Simply noting that no RCT has been performed and then giving emotive anecdote about a woman saved from a life of worry and statins (which may have been false reassurance, yet that’s not noted with the appropriate stats for false negatives) is bad journalism.
      If publishing “maybe it works” stories with powerful anecdote to drive the premise home is acceptable, then we might as well start reporting on coffee enemas, crystals to cure cancer and all other manner of “but maybe it works” claims.