Debate on rationality of coronary angiography in low risk patients continues

By Dr. Ananya Mandal, MDAug 9 2011

A new study reveals that American hospitals vary widely in how they decide who's eligible for an invasive type of heart scan.

This brings to front the debate on the rationality of coronary angiography, which carries a price tag of several thousand dollars and a risk of side effects. Researchers found that at some U.S. hospitals, fewer than a quarter of the scanned patients turned out to have heart disease. At others, all of them did. This could mean that some hospitals use the procedure liberally, while others reserve it for high-risk patients, said Dr. Pamela S. Douglas of Duke University Medical Center in North Carolina, who led the study. She said, “Clearly we have no standards… There is differential use, which tells you there may be a quality issue.”

Coronary angiography involves guiding in of a thin tube, called a catheter, through a blood vessel into the heart, where a special dye is injected. Using high-dose x-rays, they then look for cholesterol blockages that restrict blood flow to the heart and in some cases might eventually lead to a heart attack. If there is a big blockage, doctors may choose to open up the artery and put in a stent -- a fine metal tube that props the artery open.

This procedure remains a useful and recommended one for diagnosing coronary artery disease but there is no agreement on who will benefit from it outside of patients with ongoing heart attacks or previous heart disease. Douglas' team looked at data from a national registry and found more than 565,000 patients who'd undergone non-emergency coronary angiography and had never had heart disease before. Most, but not all, had first completed a non-invasive stress test to give the doctor a preview of their heart health.

The total of 691 hospitals had very different overall test results, ranging from 23 percent of patients with confirmed heart disease to 100 percent. Hospitals with a lower rate of positive tests tended to perform angiography on younger patients at a lower risk of heart disease and often without symptoms.

Some patients can get kidney damage from the dyes used in coronary angiography, and the high-dose x-rays may lead to a small increase in cancer risk explained Douglas. There is also a small risk of bleeding and blood clots due to the procedure, although less than one in 10,000 healthy patients experiences serious complications, according to Douglas.

She pointed out that not all negative tests are wasted efforts, because they could provide reassurance to patients, families and doctors. “Negative doesn't mean unnecessary,” said Douglas, whose findings appear in the Journal of the American College of Cardiology. She added that there was also no way of knowing which approach would lead to better outcomes for patients.

Dr. William Boden, a cardiologist at the State University of New York at Buffalo who wasn't involved in the study, said the results suggested that some doctors and hospitals may use the procedure too liberally. “Performing angiography in that setting [low risk] is likely not to lead to much benefit. You will basically confirm the obvious…We need to do a better job in terms of getting all physicians to adhere to a more evidence-based approach.”

So far, there aren't any clear criteria for when to use diagnostic coronary angiography. But Douglas said the American College of Cardiology expects to publish such criteria this winter. “Based on these results we need some standards,” she said.

This study is a follow-up to a 2010 paper by the same group of researchers that found that “the rate at which obstructive coronary artery disease was found nationally was much lower than everybody expected,” Douglas said.

In a statement about the study, the Society for Cardiovascular Angiography and Interventions points out that because “these data were collected between 2005 and 2008, standards have advanced, and new appropriate-use criteria that will soon be released will help better define issues such as these related to diagnostic angiography.”

Another study showed that for people without known coronary artery disease (CAD), both non-obstructive and obstructive disease on coronary CT angiography are associated with greater mortality risk.

The absence of CAD on imaging indicated a very low risk of death from any cause through an average follow-up of 2.3 years (0.28% per year), according to James Min, MD, of Cedars-Sinai Medical Center in Los Angeles, and colleagues. But mortality risk was greater for non-obstructive disease and for obstructive disease, with the risk rising in a dose-response fashion as the number of diseased vessels increased, the researchers reported in the Aug. 16 issue of the Journal of the American College of Cardiology.

“That coronary CT angiography can effectively risk stratify individuals without known CAD should be invaluable for guiding the development of clinical practice guidelines and appropriate use criteria,” they wrote. Although coronary CT angiography has been shown to be accurate for the detection of obstructive CAD, “real-world” effectiveness data - particularly regarding age- and sex-specific outcomes - are lacking, according to Min and colleagues.

The team data from CONFIRM (Coronary CT Angiography Evaluation for Clinical Outcomes: An International Multicenter Registry), which included patients undergoing non-invasive imaging at one of 12 centers from 2005 to 2009. The current analysis included 24,775 patients (mean age 57) who did not have known CAD before the scans. Through an average follow-up of 2.3 years, there were 404 deaths from any cause.

Patients with non-obstructive disease had an increased risk of death (HR 1.62, 95% CI 1.20 to 2.19), which has important implications, “as patients with non-obstructive [disease] comprise the majority of patients who experience myocardial events and for whom functional stress testing aimed at detecting flow-limiting coronary artery stenoses would be expectedly negative,” Min and colleagues wrote. The risk of death also increased as the number of vessels with obstructive disease (stenosis of 50% or more) increased.

Compared with patients 65 and older, younger patients had higher risks of death for two-vessel obstructive disease (HR 4.00 versus HR 2.46) and three-vessel obstructive disease (HR 6.19 versus 3.10). “Although numerous explanations exist to account for these findings, it may be that younger patients with greater extent and severity of CAD represent a cohort with more aggressive forms of atherosclerosis than their older counterparts, thus resulting in a higher risk than for older patients with more insidious atherosclerosis,” the authors wrote. Women had a greater risk of dying compared with men after imaging revealed three-vessel obstructive disease (HR 4.21 versus HR 3.27).

They acknowledged some limitations of the analysis, including the use of only one outcome; the possibility of referral bias; the lack of information on treatments performed after coronary CT angiography; and the exclusion of patients with known CAD at baseline.

Study suggests improved patient selection and practice patterns may lower number of negative tests

Hospitals vary markedly when it comes to the rate at which diagnostic coronary angiography or catheterization – an invasive procedure that allows doctors to see the vessels and arteries leading to the heart – actually finds obstructive coronary artery disease (CAD) in people without known heart disease.

In fact, while some U.S. hospitals report that 100 percent of patients undergoing this procedure were found to have CAD, others had rates as low as 23 percent, meaning the majority of patients selected for elective catheterization did not have blockages, according to a new study published in the August 16, 2011, issue of the Journal of the American College of Cardiology. Researchers say these findings warrant further efforts to improve the patient selection and decision-making processes used by institutions to limit exposure to costly, invasive procedures when not needed.

"This procedure has associated costs and it's not without risk," said Pamela S. Douglas, MD, Ursula Geller Professor of Cardiovascular Research, Duke Clinical Research Institute, Durham, NC, and lead author of the study. "This study is an important step in assessing quality of care and is integral to efforts to improve it. Our findings indicate that there may be an opportunity to increase the likelihood of finding CAD at catheterization, and perhaps reduce the number of procedures that don't find disease."

The study – a follow up to the team's 2010 study – is the first to investigate the degree to which hospitals differ in the rate of discovering CAD with coronary angiography and the factors that might predict this. Hospital-level variability appears to be predictable based on differing patterns of patient selection and pre-catheterization evaluation, testing and treatment.

"Because hospitals maintained a similar rate of finding disease relative to other hospitals year in and year out, and these rates are also related to patient characteristics, it suggests that decision-making processes and clinical practice patterns are a highly influential factor guiding the use of diagnostic coronary angiography and could be a target for quality improvement efforts," explained Dr. Douglas.

While the study was not able to capture all of the reasons for performing angiography, its findings suggest that some procedures might be avoided if decisions about whether to perform cardiac catheterization in the hospitals with the lowest rates of finding CAD were more similar to those hospitals with the highest rates.

Hospitals with lower rates of finding CAD by catheterization were more likely to perform the procedures on younger patients, women, those with a lower likelihood of disease (as indicated by a low Framingham or Diamond and Forrester risk score), who presented with atypical symptoms and who had negative non-invasive diagnostic testing done prior to angiography. Not surprisingly, patients at higher risk of being diagnosed with obstructive coronary artery disease were more likely to have hypertension, diabetes and other expected risk factors. In general, the size and type of hospital (e.g., teaching hospital, public vs. private) and setting did not change the findings.

Researchers identified 565,504 patients without known heart disease who underwent elective cardiac catheterization at 691 hospitals nationwide over a three-year-period to evaluate the rate of finding obstructive CAD. Authors defined CAD as any major epicardial vessel stenosis at 50 percent or higher, but similar patterns remained even when alternate definitions were applied.

"The decision to perform catheterization should be selective and ideally limited to patients with moderate to high pre-test probability for CAD," said Dr. Douglas, who also cautions that there are other, valid reasons to do a catheterization besides the expectation of finding obstructive CAD, such as patient reassurance.

To provide additional guidance to cardiologists, ACC is currently working on appropriate use criteria for ordering invasive diagnostic coronary angiographies with release expected this winter. In the meantime, Dr. Douglas advocates for closer consideration of those clinical factors known to be most strongly associated with CAD, including advancing age, risk factors and typical symptoms. A careful assessment of patients' risk and presenting symptoms, as well as results of any stress and other non-invasive tests should be considered.