Specialized Centers and Systems for Heart Attack Care


Dean J. Kereiakes, MD, The Christ Hospital Heart and Vascular Center/The Lindner Research Center, 2123 Auburn Avenue, Suite 424, Cincinnati, OH 45219 E-mail: lindner@fuse.net


Evidence from randomized controlled clinical trials and registries suggest that primary percutaneous coronary intervention provides superior clinical outcomes when compared with fibrinolytic therapy for the treatment of ST segment elevation myocardial infarction (STEMI). However, the delivery of expert and timely primary percutaneous coronary intervention to the majority of patients with STEMI is extremely challenging. This objective has fueled the concept of regional centers of excellence for the care of patients with STEMI as well as regional STEMI networks similar to those currently available for trauma or stroke victims. This article reviews the rationale behind, as well as the issues inherent to, the development of systems of care for STEMI patients.

Spontaneous atherosclerotic plaque rupture is followed by platelet adhesion, activation, and aggregation with subsequent thrombus formation.1 The severity of the resultant clinical syndrome is manifest in direct proportion to the degree of restriction in coronary blood flow and ranges from asymptomatic (insignificant restriction) to non-ST segment elevation acute coronary syndrome (ACS), including both unstable angina and non-ST segment elevation myocardial infarction, which are associated with severe coronary flow restriction, and finally, ST segment elevation myocardial infarction (STEMI), which is usually secondary to complete coronary occlusion.2

The number of deaths from ACS in the United States is 7-fold higher than that for all-cause trauma in the general population, and for persons aged 65 years or older, this number is 20-fold higher (Figure 1).3 Nevertheless, when it comes to stroke and trauma, the medical community agrees that “all hospitals are not created equal.”4 Indeed, the approach of creating specialized centers of care for treating victims of trauma has recently been shown to improve clinical outcomes. Trauma victims treated in a trauma center had significantly lower mortality compared with patients treated in a nontrauma center.5 More recently, specialized centers for care of stroke patients have been implemented with a standard of care established by the American Heart Association (AHA) and with a formal process provided through the Joint Commission on Accreditation of Healthcare Organizations for the certification of primary stroke centers.6 In this context, the lack of specialized regional centers of care for patients with ischemic heart disease is not commensurate with the magnitude of this public health problem.

Figure 1.

Deaths in the United States stratified by cause for all individuals and for those individuals aged 65 years or older. COPD indicates chronic obstructive pulmonary disease. Adapted form Anderson.3

Although percutaneous coronary intervention (PCI) with coronary stent deployment has been demonstrated to significantly improve the clinical outcomes in patients who present with either non-ST segment elevation ACS7 or STEMI8 (compared with medical therapy alone/no PCI), both where and how patients with ACS are treated has been the subject of recent debate.9,10 In this context, it is appropriate to review our current strategy and system for providing care to patients who present with STEMI.

Anatomy of the Problem

STEMI results when thrombotic occlusion of the coronary artery blocks the nutrient blood flow to heart muscle. Heart muscle death occurs rapidly (within 20 minutes), with damage extending outwardly over time from the innermost border of the heart's muscle wall (Figure 2).11 Both survival and recovery of heart muscle function following STEMI depend on how quickly coronary blood flow can be restored. Despite that early treatment (within 6 hours following STEMI onset) with either clot-dissolving medications (fibrinolytic therapy) or PCI has been demonstrated to salvage viable heart muscle and to improve patient outcomes, up to 30% of patients with STEMI do not receive any reperfusion therapy.12 Based on randomized controlled clinical trials comparing PCI with fibrinolytic therapy for the treatment of STEMI, PCI has demonstrated a significant reduction in death and disability and is the preferred (when available) therapy (Figure 3).8 Although use of PCI in the treatment of STEMI has increased, approximately 75% of US hospitals lack the facilities required to perform the procedure. In addition, PCI is logistically challenging and requires greater resource allocation (specially trained physicians, nurses, and technicians as well as both radiographic and angiographic equipment) than does fibrinolytic therapy. Furthermore, recent studies have demonstrated that the benefits of PCI for STEMI are time-dependent, based on the duration from STEMI onset to balloon catheter inflation.12 The patient's risk of dying increases significantly with each 30-minute increment from chest pain onset until the artery is reopened. Because it usually takes considerably longer to summon the cardiac catheterization laboratory personnel and cardiologist to perform PCI than it does to simply inject an intravenous fibrinolytic medicine in the emergency department, the relative time delay to treatment may negate any benefit of PCI vs fibrinolysis. Furthermore, the time of STEMI onset may be difficult to determine (symptoms may mimic “indigestion” or may wax and wane in a “stuttering” fashion), and another time interval that can be determined more reliably and used as a measure for time delay to treatment is the “door-to-balloon” (DTB) time. This interval begins when the patient registers at the emergency department and ends with the first PCI balloon catheter inflation. Longer DTB times are associated with higher rates of death and major complications (Figure 4).13

Figure 2.

The extent of necrosis (heart muscle cell death) extends gradually over time following blockage of a coronary artery. Restoring blood flow within the first hour after a heart attack gives the best chance of salvaging viable heart muscle. Adapted from Reimer et al.11

Figure 3.

Pooled analysis of 23 randomized clinical trials (N=7739) of patients presenting with ST-segment elevation myocardial infarction who were randomly assigned to treatment with either thrombolysis or angioplasty. At 4 to 6 weeks' follow-up posttreatment, adverse clinical events were reduced by angioplasty. SHOCK indicates “Should We Emergently Revascularize Occluded Coronaries for Cardiogenic Shock?” Adapted from Keeley et al.8

Figure 4.

Clinical outcomes to 30 days in cumulative analysis of 1186 patients from 4 contemporary trials of primary percutaneous coronary intervention (PCI) for ST segment elevation myocardial infarction (trials are, from left to right, Enhanced Myocardial Efficacy and Removal by Aspiration of Liberated Debris [EMERALD], Cooling as an Adjunctive Therapy to Percutaneous Intervention in Patients With Acute Myocardial Infarction [COOL-MI], Acute Myocardial Infarction Hyperbaric Oxygen Treatment [AMIHOT], and Intravascular Cooling Adjunctive to Primary Coronary Intervention [ICE-IT]). Outcomes of death and major adverse cardiovascular events, which is the composite occurrence of death, recurrent myocardial infarction, and requirement for additional revascularization (PCI or bypass surgery) are stratified by 30-minute increments in door-to-balloon times. Adapted from O'Neill et al.13

Despite that the American College of Cardiology (ACC)/AHA clinical practice guidelines (CPGs) recommend that PCI be performed with DTB times of <90 minutes,2 only 30% to 40% of STEMI patients receive treatment within that time period.14 According to the National Registry of Myocardial Infarction, average DTB times fall below 2.2 hours only between the hours of 8 am and 5 pm.15 Furthermore, patients initially admitted to community hospitals without PCI availability who require transport to a tertiary facility for PCI have an average DTB time of ≈3 hours.16 This prolonged time delay to PCI treatment has been observed despite the fact that ≈80% of the US adult population resides within 60 minutes' ground transport from an existing PCI center.17 Such a discrepancy in time to treatment likely reflects the lack of an integrated system for care of STEMI in the United States. However, several US metropolitan areas have developed integrated systems for rapidly diagnosing, transporting, and treating PCI in patients with STEMI (these are similar to systems previously established in Czechoslovakia and Denmark18). For example, despite being 90 to 120 minutes from the tertiary PCI center by ground transportation, participating community hospitals in the Minneapolis, Minnesota, region have been able to rapidly diagnose and transport STEMI patients to the regional center (Abbott-Northwestern Hospital, Minneapolis, MN) with total DTB (community hospital to tertiary hospital) times of <120 minutes.19 Other US metropolitan areas such as Rochester, Minnesota, and Boston, Massachusetts, in addition to the state of North Carolina have initiated regionalized systems for STEMI care.20–22 The establishment of regional centers of excellence for STEMI care is intuitively attractive on several levels.

First, a direct relationship between both operator (physician) and facility cardiovascular procedural volumes and optimal outcomes for both elective and primary (STEMI) PCI as well as for coronary bypass surgery has been established.23,24 Doctors and hospitals performing the highest volumes of procedures demonstrate the best outcomes, including survival. Indeed, the relative benefit of primary PCI vs fibrinolysis for the treatment of STEMI may be completely lost when primary PCI is performed in a low-volume institution.25 The establishment of regional centers of excellence for treatment of STEMI should provide procedural volumes that are both commensurate with optimal outcomes as well as adequate to allow meaningful measurements and comparisons between centers. Such quality and process measure comparisons are difficult among low-volume institutions.

Second, basic resources such as subspecialized nurses and trained cardiologists are in critically limited supply.26,27 One strategy for dealing with the mismatch between evidence in favor of an interventional approach and the ability to deliver such care routinely would be to establish centers of excellence for STEMI. Such centers would provide state-of-the-art digital radiographic equipment, a broad inventory of PCI supplies, intra-aortic balloon pumps, and ancillary staff well versed in their use and maintenance. Conversely, a strategy of proliferation of multiple small programs with reduplication of services will further tax already critically limited resource pools.

Third, the prehospital care phase for STEMI is critically important, and shortening the time delay to definitive PCI treatment requires establishing the diagnosis sooner. “Systems” for STEMI care must involve integration of the emergency medical system (EMS) with the hospital as well as the performance and transmission of the 12-lead electrocardiogram (ECG) from the point of first EMS patient contact (home or workplace) to the emergency department physician, who then “activates” the hospital response team so that in-hospital treatment can be provided more quickly.28 In recent National Registry of Myocardial Infarction analyses, hospitals with the shortest DTB times incorporate prehospital diagnosis (transmitted 12-lead ECG) with a multidisciplinary “team” approach in which the emergency physician activates the catheterization laboratory before cardiology consultation.29–31

Fourth, centers of excellence for STEMI care should facilitate CPG adherence. Indeed, recent data suggest that our current “nonregionalized” system of care is suboptimal in dispensing guideline-adherent care, even to high-risk ACS patients.32 In a manner similar to the direct relationship between procedural cardiovascular volume and quality outcomes previously discussed, a direct relationship has also been demonstrated between hospital ACS patient volume and both quality clinical outcomes and adherence to ACC/AHA CPG-recommended care.33 In this regard, the process of care as measured by ACC/AHA CPG adherence has been linked to both in-hospital as well as late (6–12 month) survival following presentation with ACS.34,35 It has been suggested that for every 10% increase in guideline adherence, a 10% relative reduction in in-hospital mortality is observed (Figure 5).34 While others have found a less clear correlation between process and outcomes,36 recent data from the Global Registry of Acute Coronary Events provide a compelling argument in favor of guideline adherence to improve clinical outcomes.37 Both compliance with CPGs as well as the ability to monitor or audit CPG adherence appears to be enhanced in higher-volume regional programs.33 Lower-volume small community hospitals are unlikely to allocate the resources and personnel required to track, collate, analyze, and report clinical outcomes or process measures. Indeed, in a recent survey commissioned by the AHA, only slightly more than half of the hospitals queried were systematically tracking times to STEMI treatment, infection rates, readmission or stroke rates (to 30 days postprocedure), recurrent myocardial infarction, or mortality following either PCI or coronary bypass surgery.38

Figure 5.

Hospital composite quarterly quartiles for American College of Cardiology/American Heart Association guideline adherence correlated with in-hospital mortality both adjusted (for age, sex, body mass index, race, insurance status, family history of coronary disease, hypertension, diabetes, smoking, hypercholesterolemia, prior myocardial infarction/angioplasty/coronary bypass surgery/congestive heart failure/stroke, renal insufficiency, blood pressure, heart rate, ST segment shift, and positive cardiac biomarkers) and unadjusted. Increments in clinical practice guideline adherence are associated with a reduction in mortality. ↑ indicates increase; →, results in; ↓, decrease. Adapted from Peterson et al.34

Finally, the current trend for proliferation of centers without the capacity for on-site cardiac surgical “backup” for the performance of primary PCI in STEMI as well as for elective “low-risk” PCI procedures may be associated with suboptimal clinical outcomes. For example, in an analysis of 625,854 Medicare patients who underwent PCI, in-hospital and 30-day mortality was significantly increased in those centers without on-site cardiac surgery.39 Even in the context of a completely integrated community-tertiary hospital system, the performance of primary PCI without on-site cardiac surgery may be associated with a trend toward increased hospital mortality when compared with primary PCI performed at the tertiary center (Table I).40 Although single-center studies reported excellent outcomes in patients undergoing PCI in hospitals without on-site cardiac surgery,41 the 1 randomized trial that compared fibrinolysis to primary PCI in hospitals without on-site surgery was flawed by an inadequate sample size and by a majority of patients enrolled at a single site.42

Table I.  Primary PCI for Patients Presenting with STEMI to either SMH or ISJ
 SMH (n=285)ISJ (n=285)P Value
Angiographic success279 (98)280 (98).76
Procedural success274 (96)266 (93).085
In-hospital death4 (1)10 (4).050
Any recurrent in-hospital MI1 (0.4)4 (1).17
Recurrent in-hospital Q-wave MI1 (0.4)2 (1).56
In-hospital emergency CABG0 (0)0 (0)-
Abbreviations: CABG, coronary artery bypass grafting; ISJ, Saint Joseph's Hospital, a community hospital 80 miles away from SMH with no on-site cardiovascular surgery support; MI, myocardial infarction; PCI, percutaneous coronary intervention; SMH, Saint Mary's Hospital of the Mayo Clinic, a tertiary cardiovascular service provider; STEMI, ST segment elevation myocardial infarction. Values are No. (%). SMH and ISJ are completely integrated with respect to imaging and information technology and share both protocol-driven algorithms for patient care and physician and ancillary personnel. Despite that patients in cardiogenic shock requiring pressor therapy or refractory ventricular arrhythmias were excluded from PCI at ISJ, a trend toward increased mortality following PCI at ISJ compared with SMH was observed. Modified from Ting et al.40

The Real Issues

Most concerns regarding strategies to regionalize care for patients with ACS have focused on the lack of a clear consensus on the specific nature of regionalization as well as the economic and market impact for such a strategy. Cardiac services supplement low profit margins in other services for many US hospitals, and operating margins for PCI as well as cardiac surgical procedures (3.6–9.5% and 8.7–14.1%, respectively) compared very favorably to the overall operating margin for all US hospitals in 2002 (4.3%).38 In addition, the loss of PCI or cardiac surgical volumes could lower a hospital's case mix index, which in turn could lower overall Medicare reimbursement. Certainly, financial incentives for the local hospital to transfer patients to regional centers and the assurance of the return of the patient to the local practice would be needed as part of any system for care. Finally, the potential financial losses to individual hospitals must be weighed against potential societal benefits in terms of reduced morbidity and mortality that would likely accompany a regionalized STEMI care strategy. If one assumes that even half of the ≈30% of patients who present with STEMI and do not receive reperfusion therapy would be able to undergo primary PCI as part of an integrated, regionalized system and that the absolute risk reduction for mortality would be ≈4% (compared with no reperfusion), approximately 2600 lives per year could be saved.38 In addition, when primary PCI is performed in a timely fashion in experienced high-volume centers, approximately 20 lives are saved per 1000 patients treated when compared with fibrinolytic therapy. Indeed, patients with contraindications to receiving fibrinolytic therapy are often not referred for primary PCI. Furthermore, if only half of the 31% of STEMI patients who are currently receiving fibrinolytic therapy could undergo primary PCI, an additional 1240 lives could be saved. Thus, the implementation of strategies that increase the number of patients with timely access to PCI could save ≈4000 lives yearly.38

Although a regionalized approach to care has been endorsed and validated to improve clinical outcomes in trauma patients and a similar approach for developing and credentialing specialized centers for care of stroke has been proposed by the AHA, the collaborative spirit encountered among community hospitals participating in integrated regional systems for trauma or stroke may be facilitated by the stark reality that it is frequently not profitable to treat trauma and stroke patients.43 Thus, the current scheme for relative disproportionate reimbursement for cardiovascular procedures may constrain the evolution of an optimal care process for STEMI.

STEMI is the logical initial objective for a regionalized ACS treatment strategy in that 5 of Medicare's 10 quality indicators focus on STEMI care.38 Thus, several system process/quality measures are already in place to provide a performance incentive to define “regional networks” for STEMI care. As previously noted, regionalized approaches to the provision of primary PCI therapy for STEMI have been successfully implemented in major metropolitan areas of the United States.19–22 Through partnerships with community hospitals and protocol-driven algorithms for care, designated transport systems and enhanced multidisciplinary communication (among the EMS, emergency physician, interventional cardiologist), the ability to promptly access and treat STEMI patients who originate from a broad area (90–120 minutes' ground transit) has been demonstrated. By focusing on collaboration and integration of resources, community hospitals may initiate adjunctive pharmacotherapy to patients who present with STEMI and then emergently transport these patients to the interventional team waiting at the tertiary regional center. Using this approach, remarkably short DTB times and good clinical outcomes have been achieved.19–22

System Limitations

The rapid transport of patients with STEMI to the nearest appropriate facility for their care may be hampered by several factors. First, only a minority (4%–5%) of EMS-transported patients with chest pain actually have STEMI.38 Second, only a minority (≈10%) of EMSs have 12-lead ECG capabilities. Finally, a precedent “mandate” exists for transport of the patient to the nearest facility even when fibrinolysis may be contraindicated and the facility does not provide primary PCI. Evolution in the current process of prehospital care is further complicated by the fact that there are 329 different EMS regions within the United States, with >993 hospital-based EMSs. Hospital-based EMSs represent only 6.5% of all EMS providers, while the remainder is composed of private third-party systems (48.6%) and fire station-based systems (44.9%).38 Although the transport time to a specialized center of excellence may appear long, this point may be more than counterbalanced by an integrated EMS and prenotification.

It has been estimated that a coordinated system for the provision of PCI could prevent an estimated 6 to 8 major cardiovascular and/or cerebrovascular events per 100 patients with STEMI treated and thus might affect 35,000 patients yearly.44 Credentialing and criteria for the development of a “level 1 heart attack center” should include the established ability to provide prehospital diagnosis of STEMI with a transported 12-lead ECG via integration with the local/regional EMS. Suggested criteria for a level 1 heart attack center are listed in Table II.45 Professional societies and organizations should develop the credentialing criteria for these centers. State and/or local government agencies can be charged with oversight to assure the legitimacy of regional STEMI networks. CPG adherence and monitoring could be performed by established “STEMI systems” rather than by individual hospitals/centers.

Table II.  Criteria for a Level 1 Heart Attack Center
24-Hour cardiac catheterization laboratory availability
24-Hour cardiovascular surgery availability
Comprehensive interventional cardiology and cardiovascular surgery services
>200 PCI patients/y (>36 STEMI) per hospital
>75 PCI patients/y per interventional cardiologist
EMS integration with prehospital 12-lead ECG transmission
Standardized protocols at referral and receiving hospitals
Transfer agreements in place
Education and training programs for transport, referral, and receiving hospital personnel
Quality assurance program
Abbreviations: ECG, electrocardiogram; EMS, emergency medical system; PCI, percutaneous coronary intervention; STEMI, ST segment elevation myocardial infarction. Modified from Henry et al44 and Kereiakes and Jacobs.45

The process for the development of systems of care for STEMI patients has already been initiated by the AHA and is the subject of ongoing meetings.46 The AHA will participate in a needs assessment and analysis of the effectiveness of EMSs for STEMI care and will facilitate the development of an implementation plan to build the appropriate infrastructure that can be tailored to the specific geographic region. In addition, the AHA will convene stakeholders at the state and/or local level to identify initiatives to improve the care of STEMI patients and to establish STEMI systems.46 These stakeholders will include patients, medical personnel (physicians, nurses, EMS providers), public and private payors, STEMI referral hospitals, STEMI receiving hospitals, policy makers, state health departments, and quality improvement organizations. The stakeholder evaluation will include assessment of the financial impact on STEMI referral and STEMI receiving hospitals, rural implications and inclusion, overutilization and potential for false positives, disparate population impact, resource allocation in regional areas, and allocation of resources within STEMI receiving hospitals to allow accommodation of additional patient volume. In addition, the AHA will develop recommendations for certification of STEMI referral and STEMI receiving hospitals that will include performance and outcome measurements for use in quality improvement. Given the importance of time to treatment for STEMI patients, the ideal system of care will likely be different for various geographic locales (urban/suburban vs rural) and will consider the risk of the myocardial infarction in an individual patient.