Download this article's Factoid (PDF & PPT for Gold Subscribers

Stable patients undergoing elective catheterization in Ontario, Canada, were 50% more likely to be diagnosed with obstructive coronary artery disease (CAD) than their counterparts in New York State due to more careful patient selection in the Canadian system, according to a registry study published in the July 10, 2013, issue of the Journal of the American Medical Association. Importantly, the higher diagnostic yield in Ontario did not come at the cost of underdetection of serious disease or increased mortality.

A team led by Dennis T. Ko, MD, MSc, of the Institute for Clinical Evaluative Sciences (Toronto, Canada), compared data on stable patients with no history of cardiac disease who underwent elective cardiac catheterization in Ontario (n = 54,933) with those of similar patients from the New York State Cardiac Catheterization Database (n = 18,114) between October 1, 2008, and September 30, 2011.

The observed rate of obstructive CAD was lower in New York State patients than Ontario patients (30.4% vs. 44.8%; P < 0.001). The same pattern of reduced rates for New York State vs. Canadian patients was seen regardless of anatomic location (P < 0.001 for all comparisons):

• Left main: 2.5% vs. 5.0%
• Proximal LAD:  7.2% vs. 13.4%
• Three-vessel disease: 5.2% vs. 9.8%
• Left main or 3-vessel disease: 7.0% vs. 13.0%

The researchers developed models for predicting the presence of obstructive CAD that were determined to be applicable to both New York State and Ontario patients. They reported “strong concordance” between observed and predicted CAD rates.

The proportion of patients with a low predicted probability of obstructive CAD who underwent catheterization was larger in New York State than Ontario, while the reverse was true for those with intermediate or high predicted risk (table 1).

Table 1. Likelihood of Catheterization According to Predicted Probability of CAD

P < 0.001 for all comparisons.

In a separate analysis limited to patients enrolled before 2011, New York State patients with obstructive CAD (n = 13,824) were more likely than Canadian patients with stenosis (n = 40,794) to undergo PCI (54.9% vs. 34.6%) or bypass surgery (20.4% vs. 14.1%; both P < 0.001) within 30 days of catheterization. However, no difference was seen in 30-day mortality between New York State and Ontario for either patients with CAD (0.73% vs. 0.47%; P = 0.08) or those who received PCI (0.29% vs. 0.25%; P = 0.75) or bypass surgery (0.67% vs. 0.90%; P = 0.52).

Better Patient Selection Would Mean Big Savings

The authors conclude that the lower diagnostic yield of catheterization in New York State was “primarily the result of selecting more patients at low predicted probability of obstructive CAD.”

They highlight 2 important practice differences between the regions. The majority of New York State patients undergoing catheterization did not have typical chest pain as categorized by the Canadian Cardiovascular Society classification. Also, among those who underwent noninvasive ischemic evaluation before the procedure, only 5% of New York State patients vs. 50% of Ontario patients had high-risk findings, although regional differences in physician interpretation may account for some of the gap.

Previous research by the same investigators showed that New York State clinicians perform angiography at about twice the per capita rate as Ontario physicians (Ko DT, et al. Circulation. 2010;121:2635-2644). Assuming the Canadian cost of $3,000 per outpatient procedure, Dr. Ko and colleagues calculated that if New York State, following a more selective catheterization strategy, matched Ontario’s per capita rate, it could save approximately $75 million per year.

High-Risk CAD Not Missed

“One of the primary reasons to perform cardiac catheterization is to detect patients with severe CAD, for which coronary revascularization may improve clinical outcomes,” the investigators acknowledge. But since the per capita detection rate of left-main stenosis and triple-vessel disease was similar between the 2 regions, a more restrictive approach to cardiac catheterization in Ontario did not jeopardize detection of patients with high-risk disease, they conclude.

The study lends support to the notion of using rates of obstructive CAD as a tool to promote the efficiency and improve the quality of catheterization, the authors say. But, they add, “we do not believe [it] can be used to determine the optimal rate of obstructive CAD or optimal selection criteria for cardiac catheterization because decisions for procedure use are based on complex interactions between patients, physicians, and the local environment.”

Another reason an ‘ideal’ rate is elusive, study coauthor Edward L. Hannan, PhD, of the University at Albany (Albany, NY), told TCTMD in a telephone interview, is that for any given rate “there will be some patients who are cathed who don’t have disease and some who aren’t cathed who do have disease. As you move the rate in one direction or the other, one type of error increases as the other decreases, so people would disagree about where to set that line. But to even try to set it intelligently, you’d like to know the [proportion of] patients who have disease who were not cathed, and that is missing.”

However, Dr. Hannan noted that since the study data were collected, the American College of Cardiology and associated professional societies have produced appropriate use criteria for diagnostic catheterization that provide guidance about when the procedure should and should not be performed. “I’ll be curious to see whether there’s a change as a result of those criteria,” he commented.

From Description to Action

In an accompanying editorial, Harlan M. Krumholz, MD, SM, of the Yale University School of Medicine (New Haven, CT), argues that while delineation of practice variations is useful, “perhaps even more emphasis should now be placed on producing innovations that optimize decision making and ensure that any variation is based on differences among the patients and not on the tendencies of the health care practitioners, organizations, or payment systems.”

To that end, he offers some potential next steps:

• Set standards for high-quality decisions, and develop metrics for assessing the quality of decisions
• Teach the science of clinical decision making, and establish it as a competency for those in the medical profession
• Develop tools to facilitate high-quality, patient-oriented decisions

The goal “is not to eliminate variation but to guarantee that its presence throughout health care systems derives from the needs and preferences of patients,” Dr. Krumholz concludes.

Study Details

Obstructive CAD was defined as stenosis of 50% or more in the left main coronary artery and 70% or more in a major epicardial or branch vessel. Triple-vessel disease was defined as stenosis of 70% or more in the LAD, left circumflex, and right coronary arteries.

2. Krumholz HM. Variations in health care, patient preferences, and high-quality decision making. JAMA. 2013;310:151-152.

Related Stories:

• Consensus Document Outlines Guidelines for Cardiac Cath Use
• PCI More Common in NY State than Ontario—for Better and Worse
• CathPCI Registry Report Details Contemporary US Practice Patterns