Frederick A. Masoudi, MD, MSPH, FACC, Chair; Robert O. Bonow, MD, MACC, FAHA#; Elizabeth DeLong, PhD; N.A. Mark Estes III, MD, FACC, FAHA; David C. Goff, Jr, MD, PhD, FAHA, FACP; Kathleen Grady, PhD, RN, FAHA, FAAN; Lee A. Green, MD, MPH; Ann Loth, RN, MS, CNS; Eric D. Peterson, MD, MPH, FACC, FAHA; Martha J. Radford, MD, FACC, FAHA; John S. Rumsfeld, MD, PhD, FACC, FAHA; David M. Shahian, MD, FACC

Over the past decade, there has been an increasing awareness that the quality of medical care delivered in the United States is inadequate. In its seminal document dedicated to characterizing deficiencies in delivering effective, timely, safe, equitable, efficient, and patient-centered medical care, the Institute of Medicine described a quality “chasm” (1). The recognition of the magnitude of the gap between the care that is delivered and the care that ought to be provided has stimulated interest in the development of measures of quality of care and the use of such measures for the purposes of quality improvement and accountability.

Consistent with this national focus on healthcare quality, the American College of Cardiology (ACC) and the American Heart Association (AHA) have taken a leadership role in developing measures of the quality of care for cardiovascular disease in several clinical areas (Table 1). The ACC/AHA Task Force on Performance Measures was formed in February 2000 and was charged with identifying the clinical topics appropriate for the development of performance measures and assembling writing committees comprising clinical and methodological experts. When appropriate, these committees have included representation from other organizations involved in the care of patients with the condition of focus. The committees are informed about the methodology of performance measure development and are instructed to construct measures for use both prospectively and retrospectively that rely on easily documented clinical criteria and, when appropriate, incorporate administrative data. The data elements required for the performance measures are linked to existing ACC/AHA clinical data standards to encourage uniform measurements of cardiovascular care. The writing committees also are instructed to evaluate the extent to which existing nationally recognized performance measures conform to the attributes of performance measures described by the ACC/AHA and to strive to create measures aligned with acceptable existing measures when this is feasible.

The initial measure sets published by the ACC/AHA focused primarily on processes of medical care, or actions taken by healthcare providers, such as the prescription of a medication for a condition. These process measures are founded on the strongest recommendations contained in the ACC/AHA clinical practice guidelines, delineating actions taken by clinicians in the care of patients. Specifically, the writing committees consider as candidates for measures those processes of care that are recommended by the guidelines as either Class I, which identify procedures/treatments that should be administered, or Class III, which identify procedures/treatments that should not be administered (Table 2). Class II recommendations are not considered candidates for performance measures. The methodology guiding the translation of guideline recommendations into process measures has been delineated explicitly by the ACC/AHA, providing guidance to the writing committees (8).

Although possessing several strengths, processes of care are limited as the sole measures of quality. Thus, current ACC/AHA performance measures writing committees are instructed to consider measures of structures of care, outcomes, and efficiency as complements to process measures. In developing such measures, the committees are guided by methodology established by the ACC/AHA (9). Although the implementation of measures of outcomes and efficiency is currently not as well established as that of process measures, it is expected that such measures will become more pervasive over time.

Although the focus of the performance measures writing committees is on measures intended for quality improvement efforts, other organizations may use these measures for external review or public reporting of provider performance. Therefore, it is within the scope of the writing committee's task to comment, when appropriate, on the strengths and limitations of such external reporting for a particular cardiovascular disease state or patient population. Thus, the metrics contained within this document are categorized as either performance measures or test measures. Performance measures are those metrics that the committee designates appropriate for use for both quality improvement and external reporting. In contrast, test measures are those appropriate for the purposes of quality improvement but not for external reporting until further validation and testing are performed.

All measures have limitations and pose challenges to implementation that could result in unintended consequences when used for accountability. The implementation of measures for purposes other than quality improvement requires field testing to address issues related but not limited to sample size, frequency of use of an intervention, comparability, and audit requirements. The manner in which these issues are addressed is dependent on several factors, including the method of data collection, performance attribution, baseline performance rates, incentives, and public reporting methods. The ACC/AHA encourages those interested in implementing these measures for purposes beyond quality improvement to work with the ACC/AHA to consider these complex issues in pilot implementation projects, to assess limitations and confounding factors, and to guide refinements of the measures to enhance their utility for these additional purposes.

By facilitating measurements of cardiovascular healthcare quality, ACC/AHA performance measurement sets may serve as vehicles to accelerate appropriate translation of scientific evidence into clinical practice. These documents are intended to provide practitioners and institutions that deliver care with tools to measure the quality of their care and to identify opportunities for improvement. It is our hope that application of these performance measures will provide a mechanism through which the quality of medical care can be measured and improved.

Frederick A. Masoudi, MD, MSPH, FACC

Chair, ACC/AHA Task Force on Performance Measures

The ACC/AHA ST-Elevation and Non–ST-Elevation Myocardial Infarction (STEMI/NSTEMI) Performance Measures Writing Committee (the writing committee) was charged with the development of performance measures concerning the diagnosis, treatment, and outcomes of both STEMI and NSTEMI. The purpose of the effort is to develop measures that can be used to improve care for patients with an acute myocardial infarction (AMI). Recognizing that each measure may impose a burden, the writing committee sought to focus on those areas with the most potential for impact, where there was the strongest consensus about the best practice, and where the likelihood for unintended harm was lowest. Moreover, the group sought to keep the measures as straightforward as possible, as aligned with existing measures as possible (when appropriate), and as clinically sensible as possible, giving the clinician the opportunity for judgment about the appropriateness of an intervention to the extent possible. The focus is on in-hospital care, with attention to outpatient care being deferred at this time (even as the importance of the episode of care is acknowledged by the writing committee). Many processes recommended by the guidelines were not translated into measures. The decisions were based on many factors, and common considerations were the complexity of the recommendations (making translation difficult) and the timing of the decision relative to other processes (eg, whether the process was better considered as an outpatient measure). This document is intended to supersede the prior publication of AMI performance measures (5). We present a refinement in 9 measures, the deletion of a measure (early beta-blocker therapy), 4 new performance measures, and 9 test measures (Table 3). The test measures are understood as areas worthy of measurement, but, for reasons related to the strength of evidence, the feasibility of the measure, or other considerations, are not considered to be suitable for accountability or public reporting.

The estimated annual incidence of MI in the United States (including both STEMI and NSTEMI) is 600 000 new and 320 000 recurrent attacks. In 2004, AMI resulted in 695 000 hospital stays and $31 billion in hospital charges (10). The risk of further cardiovascular complications, including recurrent MI, sudden cardiac death, heart failure, stroke, and angina pectoris, for those who survive AMI is substantial (11).

Over the past 30 years, advances in cardiovascular care have resulted in a dramatic decline in mortality and morbidity associated with STEMI and NSTEMI (12). However, there remain gaps in the application of the best treatments and strategies for these patients (13- 14). As a result, the outcomes of STEMI and NSTEMI patients are not as good as they could be with more effective and widespread application of the best scientific knowledge to their care.

The members of the ACC/AHA STEMI/NSTEMI Performance Measures Writing Committee included clinicians specializing in cardiology, internal medicine, family medicine, and emergency medicine and individuals with expertise in performance measurement. Moreover, the writing committee included representatives of the American College of Physicians (ACP), American Academy of Family Physicians (AAFP), and the American College of Emergency Physicians (ACEP).

The work of the writing committee was supported exclusively by the ACC and AHA without direct commercial support. Writing committee members volunteered their time to this effort. Meetings of the writing committee were confidential and attended only by committee members, invited observers from the Centers for Medicare and Medicaid Services (CMS) to promote alignment as described further below, and staff from the ACC and AHA. Writing committee members declared all relationships with industry relevant to this topic in writing and at each meeting according to standard reporting requirements of the ACC and AHA. Committee members with relevant relationships to a specific measure did not participate in the voting regarding that measure but were allowed to participate in the discussion after disclosing the relationship. Please see Appendix A for relevant writing committee relationships with industry. In addition, Appendix B includes relevant relationships with industry information for all peer reviewers of this document.

Between June 23 and July 22, 2008, the ACC/AHA STEMI/NSTEMI Performance Measures document underwent a 30-day public comment period during which time ACC and AHA members, as well as other health professionals and members of the general public, had an opportunity to review and comment on the draft document in advance of its final approval and publication. A number of medical specialty societies with an interest in this topic, including the AAFP, ACEP, Society of Hospital Medicine (SHM), and other organizations that develop or implement performance measures, participated in the public comment period.

Official peer and content review of the document was conducted simultaneously with the 30-day public comment period, with 2 peer reviewers nominated by the ACC and 2 reviewers nominated by the AHA. Additional comments were sought from numerous clinical content experts and performance measurement experts.

ACC/AHA Clinical Performance Measures for Adults With ST-Elevation and Non–ST-Elevation Myocardial Infarction was adopted by the respective Boards of the ACC in September 2008 and AHA in October 2008 and are endorsed by the AAFP, ACEP, American Association of Cardiovascular and Pulmonary Rehabilitation, SHM, and Society for Cardiovascular Angiography and Interventions. These measures will be reviewed for the need for update or revision annually or as needed by modifications in relevant practice guidelines. They will be considered valid until they are updated or rescinded by the ACC/AHA Task Force on Performance Measures.

The development of performance systems involves identification of a set of measures targeted toward a particular patient population observed over a particular time period. To achieve this goal, the ACC/AHA Task Force on Performance Measures has outlined and published a methodology of sequential tasks required for the development of process of care measures and for outcomes measures suitable for public reporting (8- 9). The following sections outline how these steps were applied by this writing committee.

The writing committee has incorporated the terms STEMI and NSTEMI throughout this document, along with the all-inclusive term AMI. The writing committee has used the term AMI when the measure refers to both STEMI and NSTEMI patients, whereas the term STEMI was used in cases when the measure is specific to STEMI patients only. In all cases, the measures pertain to patients with an AMI, as defined by the recent statements (15- 16). Unstable angina (UA) is not considered in this document, in part because of the difficulty in defining the population with certainty and concern about the accuracy of the administrative codes for use in the retrospective ascertainment of patients. The measures also are intended for patients admitted to the hospital with an AMI as opposed to patients who have an AMI during the hospitalization as part of another illness. This choice was made because patients with a secondary diagnosis of AMI tend to be complex and are not addressed well by the literature or the guidelines.

The writing committee recognizes that in some cases there will be interest in prospective assessment of performance on quality measures for AMI, but these measures are constructed to permit the retrospective assessment of performance, consistent with contemporary performance measure implementation. For possible use in retrospective analysis of performance, it was thought useful to identify administrative codes that could be used to screen for eligible patients, providing guidance in standardizing case ascertainment. This approach should not preclude modifications of assessments in real time for the purpose of quality improvement, although it should be recognized that differences in case ascertainment may affect the results of the measurements.

For retrospective identification of patients, specific diagnosis codes, based on International Classification of Diseases, 9th revision, clinical modification (Table 4), are recommended in the screening and selection of an inpatient target patient population. These codes correspond to those used by CMS and The Joint Commission (TJC) for the identification of patients with AMI (17). These measures are constructed to include only those patients with a principal discharge diagnosis that identifies the condition for which, in retrospect, the patient was admitted to the hospital. The writing committee also recognizes that in some cases the principal discharge diagnosis code may identify patients who may not be appropriate for these measures. In part because of this, all measures are written with exclusions that permit clinicians to document reasons for not applying particular measures to individual patients.

Given the multiple domains of providing care that can be measured, the writing committee identified and explicitly articulated the relevant dimensions of care for evaluation. As part of the methodology, each potential performance measure was categorized into its relevant dimension of care. Classification into dimensions of care facilitated identification of areas in which evidence was lacking and prevented duplication of measures within the set. The relevant dimensions of care included diagnostics, patient education, and treatment. Self-management and monitoring of disease status may be best evaluated in the outpatient setting (see Table 5). The writing committee focused exclusively on hospitalized patients with AMI. Other ACC/AHA performance measure sets apply to patients with AMI who have made the transition to the outpatient setting. Although focusing primarily on processes of care, the writing committee also considered measures of structures of care (eg, the implementation of dosing protocols for antithrombotic agents) and outcomes (eg, risk-adjusted mortality).

As the primary sources for updating the 2006 STEMI/NSTEMI measure set (5) and for deriving new measures, as specified in the ACC/AHA methodology for developing process measures (8), the writing committee reviewed the 2004 ACC/AHA Guidelines for the Management of Patients With ST-Elevation Myocardial Infarction (STEMI guideline) (19), the 2007 Focused Update of the ACC/AHA Guidelines for the Management of Patients With ST-Elevation Myocardial Infarction (STEMI guideline focused update) (20), and the 2007 ACC/AHA Guidelines for the Management of Patients With Unstable Angina/Non–ST-Elevation Myocardial Infarction (UA/NSTEMI guideline) (21). The chair of this writing committee also participated on the writing committee of the STEMI guideline and the STEMI guideline focused update. In addition, the chair of the 2007 UA/NSTEMI guideline writing committee was a member of this writing committee. As participants on the guideline writing committees, they were able to offer insights into measurement issues and to provide suggestions for clarity and specificity of guideline recommendations.

In addition, existing measure sets, including those developed by TJC, CMS, and the American Association of Cardiovascular and Pulmonary Rehabilitation (AACVPR)/ACC/AHA, were reviewed by the writing committee. See the Discussion section below for details on our efforts to align the ACC/AHA measures with the measure sets of other organizations.

Explicit criteria exist for the development of process performance measures so that they accurately reflect the quality of care, including a strong evidence base, quantification of the numerator and denominators of potential measures, and evaluation of the interpretability, applicability, and feasibility of the proposed measure (8). The writing committee sought to identify measures for which there was strong evidence and clear consensus about their importance in the care of AMI patients. To determine the processes of care with adequate evidence support to be considered for inclusion in the performance measurement set, the writing committee reviewed and prioritized the Class I and Class III recommendations from the 2004 STEMI guideline, the STEMI guideline focused update, and the 2007 UA/NSTEMI guideline (19- 21), with particular attention to changes in any guideline recommendations on which the 2006 ACC/AHA STEMI/NSTEMI performance measures (2006 measures) (5) were based.

From the analysis of these recommendations, the writing committee identified potential new measures relevant to the treatment of STEMI and NSTEMI patients and potential revisions of the 2006 measures. After extensive writing committee discussion and additional literature review, consensus was reached on revisions to be made to the 11 measures included in the 2006 document. Ten potential new measures also were considered for full specification. All measures were written to assess high-quality care in appropriate patients, allowing for the exclusion of patients with contraindications to the process of care.

Using the ACC/AHA Performance Measure Rating Form and Guide (Appendix D), writing committee members independently evaluated each of the substantially revised 2006 measures and all of the potential new measures against the ACC/AHA Attributes of Performance Measures (Table 6) using a 5-point Likert scale (1 = lowest rating, 5 = highest rating). Member ratings were collated and discussed by the full writing committee to reach consensus on which measures should advance for inclusion in the final measure set. After additional writing committee discussion and further refinement of the measure specifications, the writing committee conducted a confidential vote on whether to include each measure and whether to designate any of the measures as test measures in the final set. Writing committee members were required to recuse themselves from voting on any measures for which they had significant relevant relationships with industry. The writing committee met again for further discussion to reach consensus on those measures for which the vote was not unanimous. After the comment period, further deliberation occurred, and refinements were made to the measures.

Although measures focusing on processes of care have substantial appeal as a means of reflecting quality, such measures assess only a small proportion of all of the care delivered and apply to only subsets of the population with a particular condition. Furthermore, while determining whether a particular process of care was delivered, such measures do not convey information on the effectiveness of the process. Finally, although patients presumably care about the processes of care that they receive, this interest reflects an assumption that better processes of care ultimately result in better outcomes. For these reasons, outcomes measures have been proposed as a means of complementing process measurement as a reflection of quality (22).

The writing committee considered the development of outcomes measures beyond its scope, but it discussed standards for outcomes measures for AMI. A multidisciplinary AHA Scientific Statement, which is endorsed by the ACC, identified 7 central attributes for the statistical models used for publicly reported outcome measures (9). These attributes include: 1) a clear and explicit definition of an appropriate patient sample, 2) clinical coherence of model adjustment variables, 3) sufficiently high-quality and timely data, 4) designation of an appropriate reference time before which covariates are derived and after which outcomes are measured, 5) use of an appropriate outcome and a standardized period of outcome assessment, 6) application of an analytical approach that takes into account the multilevel organization of data, and 7) disclosure of the methods used to compare outcomes, including disclosure of performance of risk-adjustment methodology in derivation and validation samples. The writing committee recognizes the importance of outcomes measures and their alignment with the published standards but did not endorse a particular measure because that was not its charge.

The target population for these measures consists of hospitalized patients 18 years of age or older with a principal discharge diagnosis of AMI (STEMI and NSTEMI), meaning a focus on patients admitted with this condition. Inclusion and exclusion criteria specific to each inpatient measure were developed. The general period of assessment is the inpatient hospitalization or related emergency department visit, and the specific time period of interest for each measure is further defined in the full measure specifications (see Appendix C).

(Table 7) shows the ACC/AHA STEMI/NSTEMI Performance Measurement Set: those measures with the highest level of evidence and guideline support that met the additional criteria for performance measures and that generated consensus support among the writing committee members. Appendix C provides the detailed specifications for each inpatient performance measure, including numerator, denominator, period of assessment, method of reporting, sources of data, rationale, clinical recommendations, and challenges to implementation. The interest in providing these specifications was for consistency in efforts across institutions. It is understood that the spirit of the measure could be maintained with some modification in the exact specifications to facilitate implementation.

To aid in the collection of hospital data for performance measurement, use of a data collection tool or flow sheet is recommended. The flow sheet may be developed at individual institutions to conform to local workflow issues and data collection practices. To further the use of standardized terminology and data definitions in the field of cardiology, those collecting data on patients with STEMI or NSTEMI are referred to the ACC Key Data Elements and Definitions for Measuring the Clinical Management and Outcomes of Patients with Acute Coronary Syndromes (23).

The ACC/AHA Clinical Performance Measures for Adults With ST-Elevation and Non–ST-Elevation Myocardial Infarction address many of the same processes of care in earlier measurement sets published by other organizations but have been developed through the use of the ACC/AHA methodology for developing performance measure sets (8). The writing committee is cognizant of previous efforts of other organizations and sought to enhance and clarify measures in ways that reflect the advancement of the underlying science, the complexity of care, and the challenges of accurate and complete data collection. As such, the writing committee has made every attempt to align these measures with those promulgated by the CMS and TJC.

In the development of these measures, the writing committee thus considered the specifications of performance measures that have been developed and implemented by the CMS and TJC. In addition, the writing committee reviewed areas of nonalignment between the 2006 measures and corresponding AMI measures currently in use by the CMS and TJC to determine whether to revise the 2006 measures to harmonize the 2 measure sets. Wherever possible, the writing committee incorporated changes to achieve this alignment. For most of the 2006 measures, changes made are limited to changes to the excluded population lists in the denominators to better align the measures with the current CMS/TJC measures. In general, it was considered appropriate to use identical specification for those measures used by the CMS/TJC. In some cases, although the definition of a specific measure inclusion or exclusion criterion used may not be completely identical, the measures shared by the ACC/AHA and the CMS/TJC are conceptually aligned. The writing committee acknowledges that differences in the description of some components of measures specifications might be modified to facilitate implementation.

The current flow of the CMS/TJC measures requires that all patients be assessed for potential contraindications and that all such patients are excluded regardless of whether the treatment was provided at discharge. Because many of the possible contraindications are relative or may resolve between the time of documentation and the time of the provision of the therapy, this approach may result in false exclusions of patients who were appropriately treated from the measure. Thus, despite the provision of care that is aligned with the guidelines, clinicians caring for patients who are falsely excluded are not appropriately rewarded for their actions. In addition to the elimination of false exclusions, this approach also decreases the burden of data abstraction. Furthermore, it is concordant with the approach used with other measure sets for inpatient and outpatient care both within and outside the cardiovascular arena. The ACC/AHA Performance Measures Task Force has supported a change in approach whereby all patients who receive the treatment would be included in the numerator and denominator of the measures and the assessment of potential documented contraindications to therapy would be assessed only among the remaining patients who did not receive the therapy; those without contraindications would join the denominator. The measures in this set have been modified to reflect this approach.

With this document, we present a current set of AMI performance measures, renewing and refining some old measures, dropping a measure, introducing some new ones, and providing some as test measures. (Table 3) summarizes the changes in this updated measure set. The set remains parsimonious, and we continue to lack measures in self-monitoring and assessment of disease status. We also lack many measures in diagnostics and patient education. In addition, there are no measures that address overuse of tests and procedures. These types of measures are needed.

The assessment of care remains challenging, and this document provides modest changes in the current efforts. Continuing research on which to base future measurement remains necessary, not only to produce new knowledge about interventions to promote better patient outcomes but also to inform the measurement of quality and the promotion of safe and effective care. Nevertheless, this document should be useful to those who want an updated, consensus list of measures that can be used to assess clinical performance in the care of patients with AMI.

The writing committee considered many approaches to modifying the structure of the measures but generally opted to implement the approach used in the first version of these measures. As such, consistent with the prior ACC/AHA performance measures, this writing committee agreed that it was important to maintain exclusion criteria to the measures to recognize the justifiable reasons for not meeting the process performance measures. These reasons are included in the “reasons documented by physician, advanced practice nurse, physician assistant, or PharmD for not. . . .” Documentation of such factors should be encouraged and will provide valuable data for future research and conducting in-depth quality improvement for situations in which there seem to be outliers with respect to the number of patients with medical or patient-centered exclusions for the performance measures.

Challenges to implementation of measures are discussed when applicable. In general, the requirements for documentation are an important challenge of any measurement effort. The acknowledgment of these challenges is not intended as an argument against measurement. Rather, the challenges should be considered cautionary notes that draw attention to areas where additional focus on research and improvement of the measures should be considered.

The ACC/AHA STEMI/NSTEMI performance measurement set should contribute to the evolution of reporting systems that allow physicians to improve care for a critical patient population. Quality improvement is a continuous process, and this document reflects the lessons the practicing community has learned to date in using existing measures and knowledge gained about how they might be improved. The clinical care team should collect data and review adherence to these measures on a routine basis, look for changes, and adjust practice patterns as necessary to improve performance.

The writing committee examined the 11 performance measures included in the 2006 STEMI/NSTEMI Performance Measure set and considered whether any of the measures should be retired or updated based on revised guideline recommendations or experience from implementation of the measures such as very high rates. The writing committee also considered whether measures with very high rates could be retained but changed to an “emeritus” status to designate their clinical importance while recognizing that performance is already high.

Compelling scientific evidence indicates that HMG Co-A reductase inhibitors (statins) reduce the risk of recurrent coronary events and improve survival in patients after MI (24- 28). The benefits of this therapy apply to both men and women, to patients older and younger than 65 years of age, and to diabetics (29- 32). The magnitude of benefit with statins matches or exceeds benefits with other secondary prevention medications such as aspirin, beta-blockers, and angiotensin-converting enzyme inhibitors (ACEIs) in the patient after MI (26,33). On the basis of available data, the majority of individuals are candidates for statins at the time of discharge for AMI.

Despite the effectiveness of statins in altering subsequent cardiovascular mortality, several prior studies have documented low treatment rates in patients with established coronary artery disease (34- 39). Current gaps in care are less well characterized, however, because many of these studies involved patients from a single or a limited number of centers, enrolled in randomized clinical trials, or treated before dissemination of the most convincing clinical trial evidence.

After careful consideration of the guideline recommendations and the data supporting these recommendations, the writing committee voted to adopt statin therapy at hospital discharge as a performance measure. The writing committee discussed whether to include all forms of lipid-lowering therapy in the numerator of this measure. The ACC/AHA STEMI and UA/NSTEMI guidelines recommend somewhat different approaches to lipid-lowering therapy. Although the 2007 STEMI guideline focused update provides a Class I indication for lipid-lowering therapy with relatively little guidance regarding the specific agent used, the 2007 UA/NSTEMI guidelines specifically provide a Class I indication for statin drugs. Both guidelines, however, acknowledge that the preponderance of evidence with respect to post-MI outcomes and low-density lipoprotein (LDL) lowering has been demonstrated with statins. On the basis of this information, the measure was restricted to statin therapy only.

The writing committee decided to exclude patients with a known LDL less than 100 mg/dL. This decision was made to focus the measure on those who are most likely to benefit and because there was a lack of consensus about whether patients with an LDL less than 100 mg/dL should be placed on statins. This exclusion was felt to assist in the acceptance of the measure.

Accumulating data for lipid-lowering therapy, particularly for statin drugs, have substantially increased the proportion of patients with AMI who are potential candidates for lipid-lowering therapy. Indeed, the ACC/AHA UA/NSTEMI guideline considers statin drugs in the absence of contraindications, regardless of baseline LDL cholesterol and diet modification, a Class I recommendation. Both the UA/NSTEMI and STEMI guidelines also consider LDL targets of less than 70 mg/dL reasonable.

Both the STEMI and UA/NSTEMI guidelines support fasting lipid profiles within 24 hours of admission in hospitalized patients to help guide lipid-lowering therapy. The recommendation that such testing be performed earlier is motivated by evidence that lipid values obtained more than 24 hours after an acute coronary event may be misleading (40). On the basis of guideline recommendations, the previous ACC/AHA STEMI/NSTEMI Performance Measures included a measure for lipid testing. However, such a measure generates substantial data collection burden, may be difficult to ascertain from chart review, and may not necessarily improve quality regarding the ultimate goal of ensuring that patients appropriate for lipid-lowering therapy receive a discharge prescription. The current writing committee agreed that the modified construction of the measure of statin therapy at discharge largely renders moot a specific performance measure for LDL testing. Nevertheless, there were varying opinions in the group, because of this the measure was retained as a test measure.

Older clinical trial data show that beta-blockers administered early during AMI hospitalization significantly reduce postinfarction angina and reinfarction (41- 42). Whether early beta-blocker use reduces mortality in AMI patients remains controversial, however. Although some individual clinical trials did show a modest, statistically significant mortality benefit associated with early beta-blocker therapy (41), a large meta-analysis, published in 1999, of 29 260 patients enrolled in 51 clinical trials of early beta-blocker therapy showed no mortality benefit associated with this approach (odds ratio, 0.96; 95% CI, 0.85 to 1.08) (43).

More recent data from the Clopidogrel and Metoprolol in Myocardial Infarction Trial (COMMIT) also raised questions about the early use of beta-blockers in patients with AMI (44). In the COMMIT study, 45 852 AMI patients (93% with STEMI and 50% receiving fibrinolytic therapy) were randomized to 15 mg metoprolol intravenously over 10 minutes immediately after presentation and then 50 mg metoprolol orally every 6 hours afterward or placebo. Importantly, patients with cardiogenic shock were excluded, but those with heart failure on presentation (Killip class 2 or 3) were not explicitly excluded.

The primary outcome (composite outcome of death, reinfarction, or cardiac arrest) and all-cause mortality at 30 days were similar between groups. Although beta-blockers significantly reduced the risk of arrhythmic death and reinfarction, they significantly increased the risk of cardiogenic shock within the first 24 hours of hospitalization. The most potent patient risk factors associated with the increased risk of developing cardiogenic shock with early beta-blockers included heart failure (Killip class 3) and hemodynamic instability on presentation.

Balancing the evidence from COMMIT and the earlier studies, the ACC/AHA STEMI and UA/NSTEMI guidelines currently give Class I (Level of Evidence: B) recommendation for early oral beta-blockers, a Class IIa recommendation for early intravenous beta-blockers in hypertensive patients without specific contraindications (including signs of heart failure, evidence of a low output state, increased risk for cardiogenic shock [defined as age more than 70 years, systolic blood pressure less than 120 mm Hg, heart rate of 110 bpm or higher, and increased time since onset of symptoms]), and Class III (Level of Evidence: A) recommendation for intravenous beta-blockers in patients with specific contraindications to early beta-blocker therapy.

The writing committee carefully considered these guideline recommendations. Because of the complexity of integrating these recommendations, which would require the distinction between intravenous and oral administration and the ascertainment of a large number of patient factors that constituted contraindications, the writing committee chose to omit early beta-blocker use from this performance measure set.

Left ventricular systolic function (LVSF) is important from a therapeutic and prognostic standpoint for patients with AMI. Patients with left ventricular systolic dysfunction (LVSD) may be candidates for specific drug therapies (eg, ACEI and angiotensin receptor blocker [ARB]) or may warrant prompt invasive management during acute coronary syndrome (ACS) hospitalization (eg, coronary angiography). In addition, systolic dysfunction after AMI predicts long-term survival. Accordingly, clinical practice guidelines have incorporated the assessment of LVSF, by any method, as a Class I recommendation in patients with AMI (NSTEMI or STEMI).

The writing committee discussed modeling the LVSF assessment for AMI measure on the corresponding measure for patients with heart failure. However, given that AMI patients have experienced an event that may acutely affect LVSF, the writing committee felt that LVSF assessments performed before the AMI hospitalization should not be considered as meeting the performance measure.

The writing committee voted to adopt LVSF assessment for AMI patients as a performance measure. As with the existing performance measure for heart failure, credit would also be given in cases in which there is a documented plan for LVSF testing after discharge because there may be instances when it is difficult to obtain the test during the stay (eg, very short stays or weekend admissions).

Acute reperfusion remains an important focus of quality assessment because of both the positive impact of timely reperfusion on clinical outcomes and the understanding of persistent gaps in the delivery of this effective therapy. The measurement of the quality of reperfusion therapy, however, involves greater complexity than many other process measures and has raised questions regarding the scope of the existing reperfusion performance measures and the possible need for additional measures to better characterize quality in this domain.

In response to these questions, the ACC/AHA Performance Measures Task Force convened a workgroup to evaluate the existing reperfusion measures and to suggest additional measures for consideration. A complete discussion of the proceedings of this workgroup is reported elsewhere (45); however, the reperfusion measures reported in this document reflect a consideration of all of the workgroup's recommendations.

In brief, specific issues addressed in detail in the document deserve mention. First, the reperfusion measures contain exclusions for those situations when a patient-centered factor results in a delay in providing therapy. An example of a patient-centered factor is the initial refusal of a patient. Systemic reasons for delay do not result in exclusions. With respect to the measures of primary percutaneous coronary intervention (PCI), the time at which measurement stops is the time of the first use of a device intended to restore flow (eg, balloon, stent, or thrombectomy device). Although this does not account for the relatively small number of cases when flow is present before device deployment, it also does not create penalties for the failure to achieve procedural success.

A particular recommendation of the workgroup was to include a measure for the timeliness of primary PCI in patients who are transferred from the facility to which they present to another facility for the procedure. In the current era, total door-to-balloon time for these transferred patients is less than 2 hours in a little more than 25% of patients, between 2 and 4 hours in a little more than 50% of patients, and 4 hours or greater in about 20% of patients (46). The previous measures explicitly excluded such patients, rendering invisible the performance of those institutions that routinely use transfer for PCI as their principal approach to reperfusion. This measure does not have a set benchmark, acknowledging the controversy about a time that represents an unacceptable delay. It is intended to make clear the time involved in obtaining reperfusion therapy for these patients. For patients who can receive fibrinolytic therapy, referring clinicians should have a sense of the time that will be required to provide primary PCI. This knowledge can inform the decision about which form of reperfusion therapy is in the patient's best interest. Moreover, such knowledge may stimulate efforts for referral and receiving hospitals, along with transportation companies and agencies, to sit together to review and improve their joint performance. The writing committee understands that in rural areas there may be long distances that are required for transfer. The opinion of the group, however, is that if reasonable primary PCI times could not be achieved then fibrinolytic therapy should be administered, which is consistent with recommendations of the STEMI guidelines. Because patients with contraindications to fibrinolytic therapy may have different considerations regarding the time to primary PCI, the committee recommends that that group be reported separately. The committee also recommends that times be collected on all patients, even those with patient reasons for delay, for the purpose of internal quality improvement and review. Two additional measures are included to reflect the timeliness of primary reperfusion: (1) measuring the time from arrival to and discharge from an emergency department in patients transferred for primary PCI (“door-in-door-out” time) and (2) a comprehensive measurement of the time from presentation at the first facility to the time of PCI at the receiving institution.

A consideration in the measurement of time to transfer for primary PCI is the subgroup of patients for whom fibrinolytic therapy is contraindicated. Although the time to transfer is undoubtedly important in this population, because the option of providing fibrinolytic therapy is not available, clinicians may opt for transfer even if the capacity to do so in a timely manner is not available. In contrast, among patients for whom fibrinolytic therapy is a therapeutic option, fibrinolysis should be provided if transfer will be delayed. Thus, the workgroup concluded that these transfer measures should be reported separately for patients with and without documented contraindications to fibrinolytic therapy.

Currently, evidence-based recommendations or accepted national performance benchmarks for measures of the time of transfer for primary PCI do not exist. Thus, although the writing committee believed that targets of 30 minutes for time from presentation to transfer and 90 minutes for time from presentation at 1 facility to PCI at another were reasonable targets given current guideline recommendations for reperfusion timeliness, no specific performance target is prescribed by the measures.

Beyond the specification of these measures, the issue of attribution of these times is critical. In the case of the “door-in-door-out” time (Measure 10), attribution is straightforward (ie, the facility at which the patient presents is largely accountable for all aspects of the process). For the measure of time of presentation to PCI among patients who are transferred (Measure 11), the question of accountability is less clear given the participation of the hospital to which the patient presents, the providers of the transfer, and the hospital at which the PCI occurs. Although arguments for several approaches are reasonable, both institutions providing care for a patient who is transferred for primary PCI should be invested in ensuring that the transfer is performed in a timely manner and, if this is not possible, should consider fibrinolytic therapy. Thus, the writing committee recommends that for the measurement of the time from presentation at 1 hospital to the time of PCI in another, the results should be attributed to both institutions. This approach to attribution will stimulate efforts for both types of institutions to collaborate with each other to optimize the care of their patients with STEMI who require acute reperfusion therapy.

The workgroup also considered the issue of the use of the time of first system contact rather than the time of hospital presentation as the start time for the reperfusion measures. The workgroup concluded that measures used for the purposes of accountability should migrate toward including the time before hospital presentation in measurement. However, until several issues regarding this approach are resolved, it was proposed that measures starting with the time of first system contact were more appropriate for the purposes of quality improvement within systems and that systems should be encouraged to measure and improve these times.

Finally, it is possible that attempts to decrease the time to reperfusion for STEMI may result in the delivery of reperfusion strategies to patients who do not meet reperfusion criteria. Identifying a population for whom angiography or fibrinolytic therapy is clearly inappropriate through the use of retrospective criteria is likely to pose substantial challenges if public accountability for such measurement is considered. However, for the purposes of quality improvement, it may be useful to review cases of “false alarm” catheterization laboratory activations or cases when fibrinolysis is administered when it is unclear that reperfusion criteria were met. Such measures are proposed as secondary measures for consideration for quality improvement.

Data on the benefits of dual antiplatelet therapy (aspirin plus clopidogrel) for patients with ACS have accumulated over the past several years. Accordingly, the prescription of clopidogrel for ACS patients has been incorporated into the ACC/AHA clinical practice guideline recommendations. Specifically, clopidogrel at hospital discharge for patients presenting with ACS, including UA, NSTEMI, and STEMI, received a Class I guideline recommendation in the 2007 updates of the STEMI and UA/NSTEMI guidelines. Class I recommendations are relevant to several patient populations, including all patients receiving coronary stents and patients not receiving stents who are managed medically. After careful consideration of the guideline recommendations and the data supporting these recommendations, the writing committee agreed to adopt clopidogrel at hospital discharge for medically treated AMI patients as a test performance measure. The rationale for this recommendation is discussed further below.

Data from the NCDR ACTION Registry-GWTG, a national ACS registry, demonstrate significant variability in the prescription of clopidogrel at hospital discharge for ACS patients depending on in-hospital treatment. Among those undergoing PCI and stenting, clopidogrel is prescribed to a very high percentage of patients. Because rates of clopidogrel prescription are already very high in these patients, the writing committee decided to exclude them from the test measure. This decision was based on balancing considerations of the burden of data abstraction among a population for which evidence suggests that gaps in care are not substantial. The decision does not question the importance of thienopyridine therapy in the population receiving stents.

The writing committee also discussed this therapy among patients undergoing coronary artery bypass graft surgery during AMI hospitalization. Because of the limited data on the benefit of dual antiplatelet therapy in this population, the writing committee concluded that patients undergoing coronary artery bypass graft during AMI hospitalization should also be excluded from the measure.

In contrast, there is evidence of substantially greater variability in rates of clopidogrel prescription at hospital discharge for medically treated patients (46). The population that does not undergo angiography and PCI is likely very heterogeneous, including some of the sickest and frailest patients and those who refuse treatment. However, given the demonstrated benefit of clopidogrel in medically treated ACS patients enrolled in clinical trials and the potential gaps in care identified in contemporary registries, the writing committee considered thienopyridine therapy in medically treated patients as a potential opportunity to improve care. A test performance measure focused on these patients would be important in this regard and would provide a better understanding of AMI patients treated medically in clinical practice. Furthermore, as with all performance measures, the heterogeneity of this patient population is acknowledged with the exclusion of those patients for whom a clinician documents any reason for not prescribing the therapy.

The writing committee also discussed whether to restrict the measure to clopidogrel only or to include the entire class of thienopyridine derivatives. Current clinical practice guidelines specify an explicit preference for clopidogrel, reserving ticlopidine for patients with contraindications to clopidogrel. Because of the approach in the guidelines and no evidence for clinically meaningful occurrence of contraindications specific to clopidogrel, the writing committee limited the measure to clopidogrel only. Although emerging evidence suggests the benefits of other agents, current guidelines do not yet include recommendations for their use.

Recommended doses for anticoagulant therapy and intravenous glycoprotein IIb/IIIa inhibitors are well established. However, excess dosing in patients with UA/NSTEMI is a common occurrence (47- 49), particularly in vulnerable populations (eg, the elderly, those with impaired renal function). Although these patients may stand to benefit the most from anticoagulant therapy, they also are the most likely to receive excess dosing and experience bleeding complications. Importantly, in these observational studies, higher rates of bleeding and in-hospital mortality were associated with excess dosing after accounting for potential confounders.

Given the high frequency of dosing errors that have been reported and their potential negative consequences, the writing committee believed that performance measures focused in this area (and including intravenous glycoprotein IIb/IIIa inhibitors) would have an important impact on quality improvement and patient care despite the lack of definitive randomized clinical trial data and the potential burden of data collection for institutions. The burden of data collection is due primarily to assessments of glomerular filtration rates for many agents. Estimations of glomerular filtration rates are usually performed with either the Cockroft-Gault or the Modification of Diet in Renal Disease formula. Hospitals may vary in their preference for using a specific formula, which could lead to minor differences. It is noteworthy that clinical studies have relied primarily on the Cockroft-Gault formula to generate dosage adjustments. An additional concern is that these agents are frequently administered urgently in the emergency department (particularly for unfractionated heparin) before a patient's weight is obtained. A measure therefore could potentially delay or diminish the use of these agents in this setting. However, this concern needs to be balanced against the significant risk for bleeding associated with excess dosing. The fact that measures for unfractionated heparin and enoxaparin have an added margin of error well above recommended doses also emphasizes true outlier doses.

The writing committee specifically focused on 5 performance measures for the most commonly used agents (unfractionated heparin, enoxaparin, eptifibatide, tirofiban, and abciximab) and focused on initial doses (bolus and infusion), including recommendations for maximum acceptable doses when applicable. We excluded patients who received treatment initially in the catheterization laboratory because doses for these agents may vary in the setting of PCI or may be adjusted directly by monitoring coagulation studies like activated clotting times. A comparable performance measure focused on dosing of fibrinolytic therapy in STEMI also was considered, although data on the impact of overdosing in this population are less conclusive (50). The writing committee believed that because of the smaller number of patients with STEMI and recent declines in the use of fibrinolytic therapy in the United States, the impact of such a measure may be more limited. Future performance measure development efforts may need to reconsider this issue in STEMI. In addition to a process measure assessing the dosing of commonly used anticoagulant and antiplatelet agents, the writing committee has developed 2 structural performance measures that assess formal approaches within a facility to minimize dosing errors for anticoagulant therapy and similar agents. This would be relevant for all patients, including those with NSTEMI, UA, and STEMI.

All measures dedicated to assessing anticoagulation dosing were unanimously considered most appropriate as test measures by the writing committee. Although we recognize that the 5 performance measures related to dosing of specific agents are based primarily on observational studies, are complex, and may add to the potential burden of data collection for institutions, contemporary data suggest that there is a substantial opportunity to improve patient processes of care and outcomes in this area. As test measures, these metrics are considered most appropriate for use for internal quality improvement programs but not other functions (eg, pay for performance, physician ranking, or public reporting) until the validity of these measures and the effort needed to collect the necessary data are better understood.

There is vast scientific evidence that physical activity is beneficial to health in general and for the prevention of ischemic heart disease and its complications specifically. The growing problem of obesity, which in turn has spurred an epidemic of diabetes, is related in part to the low level of physical activity among adults in the United States. Patients with cardiovascular disease are even less likely than the general public to participate in regular physical activity (51). The AHA/ACC and the federal government advocate regular physical activity for all persons, including those with established heart disease. Meta-analyses and systematic reviews indicate that exercise-based cardiac rehabilitation programs improve risk factors among patients with established heart disease. Pooled data from randomized clinical trials of cardiac rehabilitation demonstrate a reduction in total mortality of approximately 20% to 30% and a reduction in cardiac mortality of approximately 30% (52- 57). Trials to date have not demonstrated superiority of comprehensive cardiac rehabilitation programs over those that incorporate exercise only (53,57).

In 2007, the ACC/AHA, in conjunction with the AACVPR, published a performance measurement set related to referral to cardiac rehabilitation programs and more specific measures regarding the structure and process of cardiac rehabilitation for patients with cardiovascular disease (6). It was the expectation of that group that the general measure related to referral for cardiac rehabilitation would be incorporated into the performance measurement sets developed by other ACC/AHA groups. The STEMI/NSTEMI Performance Measures Writing Committee reviewed the recently published AACVPR/ACC/AHA Cardiac Rehabilitation measures. The measure specifically relevant to the inpatient AMI population is that all patients hospitalized with a primary diagnosis of AMI should be referred to an early outpatient cardiac rehabilitation/secondary prevention program.

After extensive discussion and deliberation, the writing committee ultimately concluded that the AACVPR/ACC/AHA process measure should be adopted as published, restricted in this case to the survivors of AMI hospitalization. This will promote consistency across measurement sets, more feasible data collection, and better opportunities for providers to develop a system that addresses their care as it relates to multiple cardiac conditions rather than require different strategies to deal with different performance measure sets for similar conditions.

The writing committee strongly endorses the use of outcomes measures to complement process measures provided that these measures meet the criteria delineated by the AHA for the public reporting of outcomes measures as discussed above. Several outcomes could be the focus of such measures, including mortality, morbidity, health status, and symptom severity. At this point in time, however, few of these outcomes can be practically measured in large populations. Currently, only measures of risk-adjusted mortality have been implemented on a large scale. On the basis of existing knowledge about the feasibility and validity of measures of outcomes, the writing committee endorsed hospital-level 30-day risk-adjusted mortality as an appropriate outcomes performance measure for AMI. Although the writing committee did not consider official endorsement of any particular measure as part of its change, the CMS currently reports a previously validated measure of hospital-level 30-day risk-adjusted mortality after AMI that conforms to the attributes delineated by the AHA and thus would be considered reasonable for use in public reporting. The writing committee acknowledges that other measures of mortality or other patient outcomes that meet the criteria delineated by the AHA may emerge over time and that, after adequate evaluation, further outcomes measures may be adopted. Ideally, any future outcomes measures would be endorsed by the National Quality Forum because this endorsement process provides the necessary scrutiny by multiple stakeholders.

Although the writing committee considered a number of additional potential measures that focus on equally important aspects of care, either the evidence base or more significant challenges to measurement of these components of care across all patients undermined the benefits that might be gained.

The writing committee investigated early clopidogrel therapy as a potential performance measure. Areas discussed were: 1) clopidogrel administration within 24 hours after hospital arrival in patients with aspirin hypersensitivity or intolerance, 2) upstream clopidogrel in patients undergoing early invasive strategy, and 3) clopidogrel administration within 24 hours in patients undergoing conservative strategy. For patients with aspirin hypersensitivity or intolerance, both the STEMI and UA/NSTEMI guidelines recommend administration of clopidogrel in lieu of aspirin therapy. However, the writing committee felt that the number of patients with aspirin hypersensitivity or intolerance would be too small for this potential measure to be useful given the burden of abstraction that would be required.

With regard to the upstream clopidogrel administration, the writing committee felt that the complexity of decision making regarding this therapy precluded translation into a performance measure. The recommendations for clopidogrel in the early stages of AMI are dependent on several factors, including treatment strategy (interventional versus early conservative), and other “upstream” medical therapy with glycoprotein IIb/IIIa inhibitors. Because of the complexity of decision making in the determination of the appropriate antiplatelet therapy in medically managed patients and the difficulty in identifying appropriate populations for the denominator, the writing committee thought that it would be extremely difficult, if not impossible, to develop a meaningful measure in this subgroup of patients.

Clinical trial data support the use of anticoagulant therapy in patients with UA/NSTEMI (21). However, the specific agent recommended depends on the type of initial treatment approach chosen (ie, early invasive versus selective invasive strategy) and patient factors (ie, high bleeding risk or chronic renal insufficiency). Although the level of evidence for each agent varies, the UA/NSTEMI guidelines currently support 4 options as Class I recommendations: unfractionated heparin, enoxaparin, bivalirudin, and fondaparinux. In patients with STEMI, use of anticoagulant therapy is a Class I recommendation after fibrinolytic therapy with options including unfractionated heparin, enoxaparin, and fondaparinux (20). For primary PCI, use of anticoagulant therapy typically is limited to the cardiac catheterization laboratory.

The writing committee strongly considered a performance measure in this area. Ultimately, however, a measure was not developed for 2 reasons. First, the complexity of clinical options and scenarios involving this therapy made the construction of a measure challenging and potentially confusing for clinicians. Second, use of anticoagulant therapy is already high among patients with ACS, approaching 90% for unfractionated heparin or low-molecular-weight heparin (58). This suggests that a performance measure in this area would identify only limited opportunities for quality improvement.

The writing committee discussed a performance measure centering on the provision of an influenza vaccination for patients after an AMI. The 2007 ACC/AHA UA/NSTEMI guidelines (21) have a Class I recommendation: An annual influenza vaccination is recommended for patients with cardiovascular disease (Level of Evidence: B). The 2007 STEMI guideline focused update (20) also has a Class I recommendation: Patients with cardiovascular disease should have an annual influenza vaccination (Level of Evidence: B). Over the past decade, more chronic diseases have been added to the list of indications for this vaccine, and there appears to be little, if any, risk of harm. However, seasonal administration and the potential difficulty of finding vaccine administration documentation if previously given outside the hospital have presented barriers to measurement feasibility in other settings. Given these challenges, the writing committee felt that influenza vaccination should not be considered for a performance measure specifically for AMI at this time.

The writing committee discussed a performance measure on nonsteroidal antiinflammatory drug (especially COX-2 inhibitor) avoidance in AMI patients. The 2007 ACC/AHA UA/NSTEMI guidelines and the 2007 STEMI guideline focused update both recommend that nonsteroidal antiinflammatory drugs with increasing degrees of relative COX-2 selectivity should not be administered to AMI patients with chronic musculoskeletal discomfort when therapy with acetaminophen, small doses of narcotics, nonacetylated salicylates, or nonselective nonsteroidal antiinflammatory drugs provides acceptable levels of pain relief (Level of Evidence: C). However, previous experience with measures implementation reveals the challenges of constructing a “negative” measure (ie, one focused on measuring a therapy that is given inappropriately) because of the need to identify a denominator for which the therapy is clearly inappropriate. Furthermore, given the extensive publicity regarding COX-2 inhibitors, it is not clear whether these agents are still being prescribed acutely in the hospital setting in this patient population. For these reasons, the writing committee concluded that a measure of avoiding nonsteroidal antiinflammatory drugs and COX-2 inhibitors should not be pursued at this time.

The writing committee carefully reviewed the evidence and guideline recommendations in regard to aldosterone blockade in patients hospitalized with AMI. The principal evidence for this therapy derives from the Eplerenone Post-AMI Heart Failure Efficacy and Survival Study (EPHESUS), in which aldosterone blockade with eplerenone initiated within 3 to 14 days improved outcomes in post-AMI patients with either heart failure or diabetes (59). All patients were receiving optimal medical therapy, including ACEIs, beta-blockers, and aspirin when appropriate. Half of the population was treated with statins. Reflecting these findings, current clinical guidelines give Class I recommendations to long-term aldosterone receptor blockade for AMI patients without significant renal dysfunction or hyperkalemia who are already receiving therapeutic doses of an ACEI, have a left ventricular ejection fraction less than or equal to 0.40, and have either symptomatic heart failure or diabetes mellitus.

The writing committee considered the addition of a new performance measure for aldosterone blockade but believed that a measure for this treatment for hospitalized AMI patients should not be developed. Several factors influenced this decision. First, identifying candidates for the denominator of this measure would create significant abstraction burden and likely identify a relatively small proportion of AMI patients (those with estimated creatinine clearance higher than 30 mL/min, patients with potassium of 5 mEq/L or lower, those receiving therapeutic doses of ACEI, those with left ventricular ejection fraction of 40% or lower, and patients with either symptomatic heart failure or diabetes). Second, patients enrolled in EPHESUS were randomized to eplerenone between 3 and 14 days after AMI, which for most patients represents an early postdischarge period. Accordingly, the writing committee felt that initiation of aldosterone blockade as a layered therapy (in patients treated with ACEI and beta-blockers) may be most appropriate in the early postdischarge setting. Finally, the writing committee also had some concerns about recent evidence in regard to the use of aldosterone blockade in patients with contraindications to this therapy (60- 67), which in some cases puts patients at risk for hyperkalemia. The committee believed that, in addition to an outpatient measure for the use of aldosterone antagonists, a parallel measure of inappropriate use may be warranted.

In the 2007 STEMI guideline focused update, facilitated PCI refers to “a strategy of planned immediate PCI after administration of an initial pharmacological regimen intended to improve coronary patency before the procedure.” Pharmacological regimens for facilitated PCI have been variably defined and include high-dose heparin, glycoprotein IIb/IIIa inhibitors, and fibrinolytic therapy. Clinical trial data suggest that the routine use of this approach does not provide any advantages and may result in harm when full-dose fibrinolytic therapy is used as the initial pharmacological regimen. The latter approach was considered a Class III recommendation in the 2007 STEMI guideline focused update. The writing committee considered a performance measure in this area to assess the use of this potentially harmful strategy. In the end, however, the writing committee chose not to pursue this further because of the challenges of constructing a performance measure that could accurately distinguish between facilitated PCI in which full-dose fibrinolytic therapy is used as the initial pharmacological regimen and other forms of facilitated PCI or rescue PCI.

The UA/NSTEMI guidelines recommend an early invasive strategy (ie, coronary angiography with PCI if appropriate) for patients with UA/NSTEMI who have evidence of refractory symptoms and hemodynamic or electric instability (Class I; Level of Evidence: B) or an elevated risk for clinical events based on clinical characteristics, including elevated biomarkers or electrocardiographic abnormalities (Class I; Level of Evidence: A). A conservative (or selectively invasive) strategy also is considered reasonable (Class IIb; Level of Evidence: B) for stable patients, including those with elevated biomarkers. The writing committee considered an AMI performance measure to evaluate the use of an early invasive strategy in patients with NSTEMI. However, a measure was not endorsed at this time because of the complexity of the guideline recommendations and the challenges in translating these recommendations into a measure that can be implemented feasibly. Particular considerations include concerns about identifying high-risk clinical characteristics reliably from abstracted data, particularly with respect to the accurate classification of ECG abnormalities, and the importance of considering overuse given the invasive nature of coronary angiography. Current initiatives through registries (eg, ACTION or the National Cardiovascular Data Registry CathPCI) may be valuable in exploring feasible approaches to identifying the “eligible” population for early invasive strategy and to inform the construction of a quality or performance measure on this topic in the future.

American College of Cardiology Foundation

John C. Lewin, MD, Chief Executive Officer

Charlene May, Senior Director, Science and Clinical Policy

Tilithia McBride, Associate Director, Performance Measurement Policy

Melanie Shahriary, RN, BSN, Associate Director, Performance Measures and Data Standards

Erin A. Barrett, Senior Specialist, Science and Clinical Policy

American Heart Association

M. Cass Wheeler, Chief Executive Officer

Rose Marie Robertson, MD, FACC, FAHA, Chief Science Officer

Gayle R. Whitman, PhD, RN, FAHA, FAAN, Senior Vice President, Office of Science Operations

Judy Bezanson, DSN, CNS, RN, Science and Medicine Advisor