Read Original Release

Release prematurely recommends replacing stress tests with CT scans for spotting clogged arteries

Heart CT Scans Outperform Stress Tests in Spotting Clogged Arteries

Our Review Summary

atherosclerosisThis release from Johns Hopkins describes results from a head-to-head comparison study that found computed tomography (CT) scans of the heart’s vessels outperformed single-photon emission computed tomography (SPECT) stress tests in detecting clogged arteries. The analysis came from a trial that compared how well the two noninvasive imaging techniques (CT scans and SPECT stress tests) matched up with invasive coronary angiography.

A quick read of this news release could erroneously lead readers to believe the current standard initial imaging test for a blocked heart artery, SPECT stress testing, should be immediately tossed aside in favor of CT angiography, and that the switch would improve heart attack prevention.

While the release had several strengths, it inflated the benefits and soft-pedaled study limitations. Our reviewers suggest the findings are of modest importance because while identifying blocked arteries is important, the bigger question is: What is the prognosis (what will happen to me if I don’t do anything differently?) and how should patients be treated? Identifying blocked vessels is part of that, but it doesn’t provide a complete picture. The release doesn’t mention that elements of the standard stress test, unlike the CT scan, can help predict prognosis. The release presumes that finding blocked arteries is the goal when really what patients and providers need to know is whether or not to undergo additional interventions.


Why This Matters

About 15 million Americans seek medical attention each year for chest pain, shortness of breath and extreme fatigue — symptoms that may signal a clogged artery. A study that definitively showed that one non-invasive imaging test outperformed another in accurately detecting clogged arteries would be of wide interest. However, one study rarely changes clinical practice. And even when results clearly indicate a newer test does a better job of measuring what it measures, that doesn’t necessarily mean it will automatically improve the health of patients. News releases shouldn’t confuse advancing the ball with scoring a touchdown.


Does the news release adequately discuss the costs of the intervention?


The release says the price tags of the two tests are similar, between $750 and $1200. But the release says it is unclear whether there are differences in total costs, after taking into account follow-up testing.

The journal article provides a clearer picture by citing some other studies that found CT angiography may lead to more follow-up testing and higher rates of revascularization and one study that did not see a difference.

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


The reported findings appear impressive: the CT scans correctly identified blockages in nine out of 10 people, while stress tests picked up blockages in six out of 10. The release further quantifies the findings when it says:

Overall, non-invasive CT angiograms accurately detected or ruled out artery blockages in 91 percent of patients, compared with 69 percent for stress testing. When researchers analyzed test performance in a subgroup of 111 patients with very high-risk disease as identified on traditional angiography, the diagnostic accuracy of CT angiograms went up to 96 percent, compared with 80 percent for stress testing.

Does the news release adequately explain/quantify the harms of the intervention?

Not Satisfactory

Here again, the release soft-pedals concerns that are pointed out in the journal article. All the tests looked at in the study expose patients to radiation. The journal article notes that while the latest-generation CT angiography machine that they used in the study exposes patients to less than half as much radiation as the SPECT machine they used, the older CT angiography machines in general used around the country actually expose patients to higher doses of radiation than their SPECT machine, and there are ways of doing SPECT that can reduce radiation exposure.

In contrast, the release gave only a muddled nod to the issue, saying “While many newer CT scans deliver substantially lower doses of radiation than the scans used in nuclear stress tests, the dose and protocols can vary widely from hospital to hospital.” It never says the average doses in clinical practice are higher.

Does the news release seem to grasp the quality of the evidence?

Not Satisfactory

The release says that this is the first head-to-head comparison of CT angiography and SPECT stress testing that also included coronary angiography (inserting a catheter to directly check suspected blockages) and that it included many more participants than earlier studies. The release includes a pretty good description of what happened in the study.

But we had a problem with the lead sentence of the release, which says CT scans of heart arteries “are far better at spotting clogged arteries that can trigger a heart attack” (emphasis added). However, this study did not look at heart attack rates or other real health outcomes. The researchers looked only at how well the two noninvasive imaging techniques matched up with coronary angiography, which includes threading a catheter directly into the heart artery to directly measure a suspected blockage. The release should have been more clear that the study did not look at how patients fared after the tests.

Does the news release commit disease-mongering?


The release notes in the lead sentence and in the body of the release that the study involved only patients who had chest pain or other symptoms that could indicate a blocked heart artery. However, it should have mentioned that the researchers cautioned their results may not apply to low-risk patients.

Does the news release identify funding sources & disclose conflicts of interest?

Not Satisfactory

The release got this one only half right. It mentions that the study was partly funded by Toshiba Medical Systems, the maker of the CT angiography device they tested. However, it fails to include that some of the researchers receive other grant funding, consulting or speaker’s fees from Toshiba and other device makers, even though these financial relationships are clearly noted in the journal article.

Does the news release compare the new approach with existing alternatives?


The release is clear that the point of the study was to compare CT angiography with the current standard imaging test for these patients. It would have been beneficial to also mention what follows the test.

Does the news release establish the availability of the treatment/test/product/procedure?


The release notes that both types of testing are available so we gave it a satisfactory. However, it would have been beneficial if the release had noted somewhere the differences between the machines and methods used by the researchers and those commonly used in clinical practice.

Does the news release establish the true novelty of the approach?


The release notes that for the first time in a trial, patient volunteers underwent all three tests (the two imaging tests) as well as invasive cardiac angiography as part of the study.

Does the news release include unjustifiable, sensational language, including in the quotes of researchers?

Not Satisfactory

While the tone of the release is measured, it suggests that clinicians should go ahead and switch tests now, without acknowledging that other experts may want to see confirmation or other kinds of studies before changing how they practice. Our reviewers saw this as problematic.

Total Score: 6 of 10 Satisfactory

Comments (2)

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

ekaterina pesheva

November 2, 2015 at 11:08 am

Thank you for your thorough and instructive review of our press release.

We take the work of very seriously and value its mission as a watchdog to ensure accurate reporting and proper contextualization of medical research that has the potential to shape health care decisions.

Johns Hopkins Medicine has a long and rich tradition in reporting research findings. Improving the public understanding of biomedical science and enlightening the national discourse on matters of health care has always been central to the mission of the institution. To ensure accurate, responsible and balanced reporting of our research, our principal investigators are asked to vet our releases not only for scientific and factual accuracy but also for proper emphasis.

The PI of the study described in the press release, Dr. Armin Zadeh, and the writer of the news release agree with some of the points raised in your review, but feel that your overall criticism misses the mark.

We are baffled by the charge that our release “prematurely recommends replacing stress testing with CT scans for spotting clogged arteries.” We believe our release makes no such leap. The release uses emphatic language to describe the comparative accuracy of the two tests. The preponderance of evidence from our study, as well as previous research, shows the better diagnostic value of CT over stress testing. And since patient outcomes and prognosis are predicated on accurate diagnosis, our release makes the case that the comparative accuracy of both tests should be taken into consideration when choosing the appropriate test, suggesting that many might benefit from one over the other. We believe such a statement is well justified. If a standard-of-care, first-line test is only 60 percent accurate and an alternative test offers much higher accuracy, patients should be made aware of it. We argue that patients with symptoms suggestive of an underlying cardiac pathology may benefit from the superior diagnostic accuracy of CT scans, but nowhere does our release call for a change to current clinical guidelines, diagnostic and screening protocols.Our release does not suggest change in practice but rather highlights an important difference in diagnostic accuracy that both physicians and patients should be aware of. Such awareness is an essential part in the patient-clinician encounter and should inform clinical decision-making and patient choice.

Furthermore, the release clearly states that any change to current recommendations ought to take into consideration the downstream costs of testing, which remain unclear.

Therefore, we respectfully disagree with your read of the release on that particular point and feel that in leveling this charge, reviewers commit the same leap in judgment they vigorously guard against.

Regarding our funding disclosures: It is a longstanding Johns Hopkins policy to disclose any funding received from private industry and to list relevant relationships that our faculty members have with pharmaceutical companies, device manufacturers, foundations and other private or non-governmental organizations. Such disclosures were made in the release. We do not report on financial relationships that co-authors from other institutions might have as we have no way of verifying such arrangements.

Your review raises some important points that we agree with and we thank you for highlighting them.

• The intent of the lead sentence was to point out that the two tests are currently used to detect blockages that can cause myocardial infarction. We agree that by itself, the sentence could inadvertently imply that the study tracked how many patients suffered heart attacks. We agree that we could have crafted the sentence in a manner that removes any ambiguity. This said, read in its entirety, the release leaves no room for difference in interpretation. It clearly describes the methodology, how the researchers assessed diagnostic accuracy and the variables they used to measure it.

• Potential exposure to higher doses of radiation with CT scans: We agree with the overall point. Yes, we should have highlighted the fact that community providers may be using significantly higher radiation doses in their CT protocols. Nonetheless, it is important to keep in mind that major CT laboratories report radiation doses in the same range reported in the study and even lower. The study was conducted a few years ago and our radiation doses have decreased since. The bottom line is that modern CT technology yields lower radiation doses compared to doses currently used with nuclear stress testing technology.


    Andrew Holtz

    November 5, 2015 at 1:09 pm

    As one of the reviewers of this news release, I and my colleagues welcome and applaud the thoughtful response by Ekaterina Pesheva and colleagues at Johns Hopkins. This sort of engagement and consideration of the content and effect of news releases is a key goal of the endeavor.

    As to the main point of contention in the response, we stand by our assessment that most readers would take the lead sentence stating “CT scans of the heart’s vessels are far better at spotting clogged arteries that can trigger a heart attack than the commonly prescribed exercise stress that most patients with chest pain undergo” as a recommendation to change practice. The body of the release asserted that the study “should help settle lingering doubts” among physicians and patients and repeatedly highlights the “much higher accuracy” of CT scans. The release does note that costs should be considered in crafting new testing recommendations. Nevertheless, if the doubts are settled, and CT scans have much higher accuracy, then why would doctors and patients choose a test that the release paints as clearly inferior? To us and, we believe, to most journalists and others reading this release, these statements come across as a recommendation to change practice.

    Again, thank you for your response. It is clear that researchers and others at Johns Hopkins considered our critique, and that is at the heart of what we hope for.