Author + information
- Manesh R. Patel, MD⁎ ()
- ↵⁎Reprint requests and correspondence:
Dr. Manesh R. Patel, Duke Clinical Research Institute, Duke University Medical Center, 2400 Pratt Street, Durham, North Carolina 27705
“We are what we repeatedly do.
Excellence is not an act, but a habit.”
Each physician caring for patients with obstructive coronary artery disease attempts daily to determine the patient's individualized risk for future cardiovascular events, the patient's symptom burden, and then provide recommendations on therapies aimed at improving the symptom/functional status of the patient and reducing the risk of cardiovascular events. Inherent in this complex decision-making process is the physician's evaluation and application of the available evidence to the specific patient with subsequent shared decision making with patients' wishes. For stable ischemic heart disease, these decisions often revolve around revascularization with either percutaneous coronary intervention (PCI) or coronary artery bypass grafting (CABG).
Recently, concerns over utilization, mainly around overuse, have led to significant changes in the review and process for payment for cardiovascular procedures in general, and PCI specifically. In this context, the American College of Cardiology developed and recently updated appropriate use criteria (AUC) for coronary revascularization (2) in an effort to provide structure and inform revascularization decisions and to understand current patterns of care. However, the AUC for coronary revascularization have been met with important critiques, particularly questions about the outcomes of patients by AUC criteria, terminology, and possible underutilization.
In this issue of the Journal, Ko et al. (3) provide an important step in the evaluation and improvement of the AUC in clinical practice. The authors perform a retrospective review of over 1,628 patients with stable coronary artery disease undergoing cardiac catheterization between April 2006 and March 2007, from a representative sample of all angiograms performed in the province of Ontario, Canada, during the study period. The authors had trained nurse abstractors review the charts to determine the appropriateness score of all the patients, both those undergoing revascularization and those treated medically, with 3-year follow-up for death or acute coronary syndrome (ACS). The nurse abstractors focused on the 4 key variables for AUC: 1) clinical presentation and symptom severity; 2) intensity of anti-ischemic medical therapy; 3) ischemic burden as determined by noninvasive testing; and 4) extent of coronary stenosis on coronary angiography. Several of the findings from the study represent important answers to questions around AUC.
The investigators found that 14% of coronary revascularization procedures (18% for PCI) were classified as inappropriate in this population of patients with stable coronary artery disease. This rate from the Canadian population is similar to the 12% rate for elective stable angina PCI observed in the larger observational study from the National Cardiovascular Data Registry in the United States (4). As previously noted by many, the rate of revascularization procedures classified as inappropriate should not be expected to be zero, as there are a myriad of clinical features and exceptions that are not accounted for by the limited scenarios for coronary revascularization. These rates do start to provide benchmarks so the providers and institutions can review practice patterns, and highlight indications and clinical areas that need better clarity in future AUC documents.
However, the most important finding from the analysis by Ko et al. (3) does not revolve around inappropriate coronary revascularization. Rather, the investigators describe important findings around the potential underuse of coronary revascularization. They found that only 69% of patients with an appropriate indication for coronary revascularization underwent revascularization (57% with PCI and 43% with CABG), and the patients who underwent revascularization had a significantly lower risk of death or repeat ACS at 3 years (hazard ratio [HR]: 0.61; 95% confidence interval [CI]: 0.42 to 0.88) compared with similarly classified patients who were treated medically, even after adjustment for baseline variables. This reduction in clinical events was not seen in patients with a revascularization classification of uncertain (HR: 0.57; 95% CI: 0.28 to 1.16) or inappropriate (HR: 0.99; 95% CI: 0.48 to 2.02).
The study is limited by its retrospective nature, and therefore the inability to fully account for and control unmeasured confounders may affect the decision to perform coronary revascularization. Additionally, the investigators do not have any information on quality of life, symptom, or function status, limiting the study's ability to identify other benefits in the patients. Nevertheless, the study provides a large, well-conducted analysis that demonstrates through the use of the published AUC for coronary revascularization that there are likely patients where coronary revascularization is underused, in addition to previously described concerns about overuse. As with the inappropriate classified procedures, it is important to note that the appropriate classification does not mean that revascularization is mandated or that a 100% rate would be expected. Rather this classification highlights patients at the highest risk for coronary artery disease. Additionally, the investigators found a gradient with adverse clinical outcomes across the AUC categories. So what do these findings mean to clinicians, patients, and policy makers in the era of healthcare reform, and how can they be reconciled with prior studies?
Hemingway et al. (5) previously described the underuse of coronary revascularization and the associated adverse clinical outcomes. This was an important observation in the percutaneous transluminal coronary angioplasty era, which included both stable and ACS patients. The current analysis goes the next step and provides a real-world contemporary analysis of stable coronary artery disease patients without prior CABG or PCI in the preceding year. The findings are also in line with the evolving consensus toward ischemia-driven revascularization (6).
Review of these features of the real-world patients described helps possibly explain why patients categorized as appropriate for revascularization might have improved clinical outcomes when considered in context of the COURAGE (Clinical Outcomes Utilizing Revascularization and Aggressive Drug Evaluation) trial (7). As noted by the authors, some patients in the current analysis may have been excluded from COURAGE due to significant left main stenosis or 3-vessel disease, high-risk stress test findings, or other high-risk features. The majority of these patients with high-risk features would be categorized as appropriate for revascularization and likely represents the patients with the largest possible benefit from revascularization. In fact, review of the spectrum of patients in each revascularization strata (appropriate, uncertain, and inappropriate) demonstrates that the patients at highest risk (and potentially at highest likelihood of benefit) do not always get coronary revascularization, another example of the “risk-treatment paradox” in the practice of medicine.
Furthermore, as more studies with invasive hemodynamic lesion assessment, such as fractional flow reserve, demonstrate clinical benefit (8), the science behind what improves patient outcomes is becoming clearer. So how do clinicians, patients, and health systems put processes in place that provide the most efficient care of patients with stable coronary artery disease, a process that increases the use of revascularization in patients that may benefit the greatest, and reduce procedures where the benefit may be much less to nonexistent?
The first step requires a framework for thinking about all cardiac procedures, including coronary revascularization (Fig. 1). In this framework, the initial decision to perform a procedure (right patient–right procedure) is made on the basis of clinical guidelines and in large part to AUC. Here the available evidence is discussed in the context of these criteria and guidelines so that patients can make informed and shared decisions. It should be noted that medical therapy alone without revascularization is a decision for which the side effects, risks, and possible benefits have to be discussed. Patients can then make their choices (preferences) known.
For patients undergoing coronary revascularization, a checklist that includes the AUC and is used at the timeout before the procedure should be considered. Research with surgery has shown that a simple safety checklist, which includes ensuring the team is aware of patient confirmation, expected risks, and possible complications, reduces mortality (9). Once the procedure is started, the remaining aspects of quality then include performing the procedure in the right way (effected by ongoing clinical trial evidence), and obtaining the right or expected clinical outcomes (quality metrics and possible public reporting). This framework, if electronically captured in real time with decision support, will hopefully inform clinicians and allow our patients and our society to evaluate the care of the patients with stable coronary artery disease. It is with this type of learning network that physicians will obtain timely feedback and the AUC will obtain real-world patient data, so that definitions and criteria can be continuously improved.
Unfortunately, the current infrastructure and ongoing actions with healthcare reform do not support the robust collection of data for understanding practice patterns, but simply the implementation of mechanisms to control services such as prior authorization. These mechanisms are inherently flawed because patients who are not getting services that might be essential or beneficial are not reviewed or previously authorized. Rather, payers should consider systems that allow shared risk and gains with providers that implement efficient care processes without the current control mechanisms. It should be noted that the decision process for care of patients with ischemic heart disease starts well before the catheterization laboratory, and tools to provide support in electronic health records are not present in the current clinical environment.
Therefore, the American College of Cardiology, along with other stakeholders, should continue to encourage innovation and ensure systems are put into place to inform patients and physicians about coronary revascularization. Ideally, real-time clinical decision support tools with outputs that include AUC and patient-specific risk scores would be embedded in portable electronic health records that eventually become parts of the structured reports at the end of the revascularization procedure. In the interim, simple paper and web checklists should be used. The current study highlights the need for these types of tools to help guide the use of revascularization, reduce the risk-treatment paradox, and improve patient outcomes. It is with these types of tools that cardiovascular physicians can continue to have the discipline and daily habits to ensure efficient and excellence in cardiac care.
Dr. Patel has received research grants from Johnson & Johnson, NHLBI, AHRQ, and AstraZeneca; and is a consultant/advisory board member for Genzyme, Bayer, Jansen, Baxter, and Otsuka.
↵⁎ Editorials published in the Journal of the American College of Cardiology reflect the views of the authors and do not necessarily represent the views of JACC or the American College of Cardiology
- American College of Cardiology Foundation
- Durant W.
- Patel M.R.,
- Dehmer G.J.,
- Hirshfeld J.W.,
- Smith P.K.,
- Spertus J.A.
- Ko D.T.,
- Guo H.,
- Wijeysundera H.C.,
- et al.
- Park S.J.,
- Ahn J.M.,
- Kang S.J.