Only 4 out of 10 patients with suspected coronary artery disease (CAD) who undergo elective diagnostic angiography actually turn out to have obstructive disease. The finding highlights the need for better risk stratification to increase the usefulness of cardiac catheterization, according to registry data published in the March 11, 2010, issue of the New England Journal of Medicine.

To assess the effectiveness of current practices, Manesh R. Patel, MD, of the Duke Clinical Research Institute (Durham, NC), and colleagues identified nearly 2 million patients in the American College of Cardiology National Cardiovascular Data Registry who underwent cardiac catherization at 663 US hospitals from January 2004 to April 2008. The researchers then excluded patients who had prior MI; had previously undergone PCI, CABG, a heart transplant, or valve surgery; had emergency indications such as AMI, ACS, or cardiac shock; or had indications for diagnostic catherization apart from suspected CAD.

Winnowing Down to the Lowest Risk Subgroup

Applying these exclusion criteria left 397,954 patients without known CAD (amounting to 20% of all patients who underwent catheterization). Among this cohort, diagnostic angiography identified obstructive disease—defined as ≥ 50% stenosis of the left main coronary artery or ≥ 70% stenosis of a major epicardial vessel—in 37.6% of patients, of whom a little more than half (53%) had multivessel disease. Catheterization found no CAD (< 20% stenosis in all vessels) in 39.2% of patients.

Expanding the definition of obstructive disease to include a ≥ 50% stenosis of any coronary vessel had little effect, raising the prevalence to 41.0%.

Noninvasive testing, including resting ECG, echocardiography, CT, or stress test, was performed in 83.9% of patients before invasive angiography. Patients with a positive result on noninvasive testing (68.6%) had a slightly higher rate of obstructive disease than did those who did not undergo such testing (41.0% vs. 35.0%; P < 0.001). The diagnostic ability of noninvasive tests increased among patients with higher Framingham risk scores and those with anginal symptoms.

Several factors independently predicted obstructive disease (table 1).

Table 1. Independent Predictors of Obstructive CAD

^a Per 5-year increase.
^b Compared with no testing.

Using 4 separate models for predicting obstructive disease, the researchers found that considering noninvasive testing results added little value beyond information provided by the Framingham risk score and by other clinical factors and symptom characteristics.

Based on the poor diagnostic ability of cardiac catheterization in this population, “[c]urrent strategies that are used to inform decisions regarding invasive angiography, including clinical assessment of risk and noninvasive testing, need to be improved substantially,” the authors conclude.

In a telephone interview with TCTMD, Dr. Patel stressed that the findings must be kept in the proper context. “What I think is important to note is that in the entire 2 million patients that went to the cath lab over 4 years, 60% had a significant occlusion,” he said, adding that low diagnostic yield was found not in this entire group but in the 400,000-person cohort without known disease. “People with clear need to go to the lab should go. This [study looks at] an elective population.”

‘Bang for the Rad’

An editorial accompanying the study by David J. Brenner, PhD, DSc, of Columbia University Medical Center (New York, NY), asserts that given the contribution of cardiac imaging to total radiation exposure in the United States, it is important to ask whether current techniques are being used optimally. He asks, “Specifically, are they providing us with the maximum information relative to the population exposure that they involve—the best possible ‘bang for the rad’?”

The current study suggests “that we still have some way to go,” Dr. Brenner continues, characterizing the 38% prevalence of obstructive disease as a “decidedly low proportion considering both the adverse events and radiation exposure associated with invasive coronary angiography.”

News Not All Bad

Gregory Dehmer, MD, of Texas A&M University Health Science Center College of Medicine (Temple, TX) and spokesperson for the Society for Cardiovascular Angiography and Interventions, viewed the situation differently, saying 38% was a “substantial number” of patients found to have CAD.

And for the more than 60% of patients who did not have this diagnosis, the angiogram was not necessarily inappropriate, he asserted. “Even if it turns out that these patients do not need a revascularization procedure, just knowing this has very important implications,” Dr. Dehmer told TCTMD in a telephone interview. “It helps get patients on the right medicines, encourages them to do lifestyle risk factor modification, maybe to exercise more.”

Another bit of good news is that the majority of patients (83.9%) underwent noninvasive testing before angiography, he added: “I actually think that percentage is reasonably high. It doesn’t mean it can’t be improved on, but I’m comforted by that.”

In a telephone interview, Ajay J. Kirtane, MD, SM, of Columbia University Medical Center (New York, NY), also highlighted that finding. “It shows that people are actually practicing evidence-based medicine and doing noninvasive testing prior to sending somebody to the cath lab. That’s reassuring,” he told TCTMD.

Dr. Dehmer noted that the argument on behalf of diagnostic angiography is furthered by the study’s observation that 31.5% of people who had CAD had no symptoms. “That’s a little bit worrisome,” he said. “It just points out that our screening tests right now have a lot of imperfections. We’re missing some people.”

Another indication of the limitations of noninvasive tests is that among patients with obstructive disease, more than half had negative results yet were considered high risk according to their Framingham scores. “Right now, risk stratification for coronary artery disease is based on a number of factors, and anatomic delineation is still critical,” said Dr. Kirtane. “What has happened to some extent is that there has been over-reliance on the pure results of stress testing. What we need to focus on is putting all the tests we perform . . . in the appropriate clinical context to then determine how best to treat a patient.”

“A common misperception is that the cath lab equals a stent,” Dr. Kirtane continued. But the reality is that the cath lab process “starts with a diagnostic procedure that then helps you appropriately risk-stratify patients and decide what the next best thing for the patient is, whether that’s medical therapy, PCI, or surgery,” he said, adding that the BARI 2D and COURAGE trials both involved upfront diagnostic angiography prior to treatment.

How to Remedy the Situation

Dr. Patel explained that the current findings speak not so much to what happens in the cath lab as to the decisions beforehand. “Our conclusion has been we need to reconsider how we evaluate and care for the patients before we get them to the cath lab in an elective situation,” he commented.

“Like most interventionalists, I don’t like doing a lot of clean coronary catheterizations. I want to find patients who have obstructive disease that I can help,” noted Dr. Patel, who predicted that there will probably be a “national conversation about how we get people to the cath lab and what the rate of clean cath should be.”

Meanwhile, the National Heart, Lung, and Blood Institute is funding a randomized trial of 10,000 patients. Known as PROMISE (Prospective Multicenter Imaging Study for Evaluation of Chest Pain), the trial will compare CT angiography with functional testing in symptomatic patients without known CAD. “That may be a large study that informs us on how we manage patients [and catheterization] in the future,” said Dr. Patel.

In his editorial, Dr. Brenner also notes that CT angiography may be a “better gatekeeper test” than myocardial perfusion imaging. “In summary, it is impossible to imagine the current practice of medicine without modern-day imaging. It is also axiomatic that, in the final analysis, the clinician is in the best position to assess the imaging needs of his or her patient,” he concludes. “But with so many high-tech imaging tools currently available, it is essential to optimize their use. Clinical-decision guidelines represent a proven methodology in this regard. This is not easy to implement on a national level, but it can be done and it should be done.”

2. Brenner DJ. Medical imaging in the 21st century—Getting the best bang for the rad. N Engl J Med. 2010:362:943-945.

Disclosures:

• Dr. Patel reports no relevant conflicts of interest.
• Disclosures for Dr. Brenner are available online at http://media.nejm.org/GetFile.aspx?Type=Disclosure&DOI=NEJMe1000802.
• Dr. Kirtane reports receiving consulting fees from Abbott Vascular, Boston Scientific, and Medtronic.
• Dr. Dehmer reports serving on the National Cardiovascular Data Registry management board.

Related Stories:

• Coronary CTA Bests Exercise Testing as Diagnostic Tool for Suspected CAD
• Zero Calcium Score Cannot Rule Out CAD in Symptomatic Patients
• Imaging Procedures Impart Sizeable Radiation Dose to Patients