Yes, We Can! (Should We?)
Historically, percutaneous coronary intervention (PCI) was relegated to hospitals with colocated cardiac surgery because of the potential need for emergent surgical treatment of PCI-related complications. In the current issue of Circulation, Lee and colleagues1 compare outcomes of PCI at hospitals with and without on-site cardiac surgery and show that emergency cardiac surgery is, in fact, rarely needed (<1%). This meta-analysis summarizes 23 studies that include >1 million patients and demonstrates that the incidence of other PCI-related complications, including myocardial infarction, stroke, cardiogenic shock, aortic dissection, and tamponade, as well as early (within 30 days) and late (after 30 days) all-cause mortality, is not different at hospitals with and without colocated cardiac surgery.1
Article see p 388
Because of inconsistencies among the studies, other outcomes indicators of quality, including the need for target vessel revascularization, could not be evaluated in detail. However, when results from the 2 randomized, controlled trials reporting target vessel revascularization (Cardiovascular Patient Outcomes Research Team Trial [CPORT-E]2 and Percutaneous Coronary Intervention [PCI] Outcomes in Community Versus Tertiary Settings [MASS COMM]3) were combined, rates of target vessel revascularization were similar at hospitals with and without on-site cardiac surgery.
This important contribution confirms and puts on a more solid evidence-based foundation current guideline recommendations that allow the performance of primary and nonprimary PCI at hospitals without on-site surgery.4
Like all important studies, this report raises a number of questions. Although PCI at hospitals without on-site cardiac surgery is safe and effective, what is the motivation for extending nonprimary PCI at these facilities? What are the consequences of extending PCI to more hospitals? How should the extensive research knowledge base summarized by Lee and colleagues be applied to the development of PCI programs at hospitals without on-site cardiac surgery?
Why Extend PCI to Hospitals Without On-Site Cardiac Surgery?
The cynical view is that the performance of PCI at hospitals without on-site cardiac surgery is all about money and market share. Economic motivations certainly exist and are understandable, given that those entrusted with the well-being of a hospital, be it a for-profit or nonprofit hospital, are stewards of its financial health. However, this kind of motivation is a given, is reasonable, and is not particularly interesting or illuminating.
The motivation for providing primary PCI at hospitals without on-site cardiac surgery is the survival advantage with which it is associated; nonprimary PCI is associated with no such advantage. Lee and colleagues1 suggest that the reason for extending nonprimary PCI to hospitals without on-site cardiac surgery is patient convenience and continuity of care. Although these are reasonable goals, the strongest motivation for extending nonprimary PCI to hospitals without on-site cardiac surgery is to sustain primary PCI programs. Because of the financial and human resource commitment required, it is at best difficult to sustain stand-alone primary PCI programs that have a low procedure volume and yet must have staff available 24 hours a day, 7 days a week, 365 days a year. If nonprimary PCI also can be performed at these hospitals, then case volume increases, which, in turn, allows additional staff to be hired, could lower per-case cost, and might reduce the on-call burden for physicians, nurses, and technicians. A second important motivation is evidence that, in the absence of a common treatment (PCI) for a ubiquitous disease (coronary artery disease), outcomes for patients with acute coronary syndromes treated at hospitals without revascularization capability may be suboptimal.5,6 Therefore, another reason to allow nonprimary PCI at hospitals without on-site surgery is to improve the outcomes of patients with coronary artery disease who present to those institutions. Finally, if PCI is allowed at hospitals without on-site cardiac surgery, the pressure to develop new cardiac surgery programs at a time when cardiac surgical volume is flat or declining could be minimized.
How Are Safe and Effective PCI Services Extended to Hospitals Without On-Site Cardiac Surgery?
Avoiding application of study results to an inappropriate (ie, unstudied) population requires careful attention to methodological detail. In a meta-analysis, this kind of detail can be obscured. In the report by Lee and colleagues,1 the “population” studied involves not only patients but also institutions, nursing and physician staff in every care area, including the emergency department, cardiac catheterization laboratory, and postprocedure care area, and even PCI equipment. For example, in the 2 largest randomized, controlled trials, high-risk patients (those with an unprotected left main lesion or who those with very poor left ventricular function [ejection fraction <20%]) were excluded. All hospitals participating in CPORT-E and some of the MASS COMM hospitals participated in a 3- to 6-month PCI development program that included training of staff and development of logistics, order sets, and critical pathways in the emergency department, cardiac catheterization laboratory, and postprocedure care areas. Interventional cardiologists had to meet American College of Cardiology/American Heart Association competency criteria to participate, and institutions had to achieve certain volume requirements. The use of devices that might be associated with higher complication rates such as rotational and directional atherectomy was not allowed. Finally, outcomes of sites participating in these studies were reported and monitored, an activity that encourages prudent patient selection and temperate procedures, which can importantly contribute to good outcomes. To what extent these specific patient, institution, device, and provider criteria and training programs influence outcome is not clear. However, failure to incorporate them into a PCI program at a hospital without on-site cardiac surgery makes outcomes comparable to those reported by Lee et al less assured.
What Are the Consequences of Extending PCI to More Hospitals?
Perhaps the single, most problematic consequence of extending PCI to hospitals without on-site cardiac surgery is the effect on institutional procedure volume. Thus, although nonprimary PCI programs may help sustain primary PCI programs, the potential reduction in human and financial resource use has probably not diminished and could increase if more programs develop.
Because volume is traditionally used as a surrogate for quality, a major concern is that quality will decline as a relatively fixed PCI volume is diluted among more facilities. The contemporary move away from using volume as a quality surrogate and toward using outcomes measures of quality may help address this problem. Questions are frequently raised about how well outcomes measures can be identified and defined and how reliably they can be measured, but rarely addressed is that accurate, meaningful measurements of quality require a certain volume. Particularly with infrequent events (such as death), it may take many years to identify substandard outcomes in hospitals with low procedure volumes because of the wide confidence intervals around that outcome estimate. Although volume may not equal quality, measurement of quality requires volume.
Low procedure volume also makes difficult the maintenance of physician and nurse complication management competency, made still more problematic because complications in the cardiac catheterization laboratory and in the postprocedure care areas are infrequent.
Although we tend to focus on the effect of lower procedure volume in hospitals without on-site cardiac surgery, there can also be important effects in tertiary and academic medical centers. Reducing the volume of “bread-and-butter” PCI procedures at tertiary hospitals may affect their financial viability, reduce their ability to support newer minimally invasive treatments of structural heart, and adversely affect research and training of interventional cardiology fellows.
The ability to treat patients with ST-segment–elevation myocardial infarction with prompt PCI is undoubtedly a positive consequence of extending primary PCI to hospitals without colocated cardiac surgery and has been lifesaving for many. Although extending nonprimary PCI to these hospitals has not directly reduced morbidity and mortality, there are positive consequences.
Although the convenience of patients and continuity of care were not specific motivations for the CPORT primary or nonprimary PCI projects, they are reasonable motivations for performing PCI at hospitals without cardiac surgery on site. Those who work exclusively in large tertiary or academic facilities can underestimate the emotional and financial burden imposed not only on the patient but also on the patient’s support structure, that is, the patient’s family and friends, of moving from a community hospital to a larger, more distant institution. The argument that, if people can drive 20 miles to buy a car or to have a great dinner, they can drive that distance to have an angioplasty is specious: Comparing consumerism to treatment of a disease that for most is frightening, life threatening, and life changing is fatuous. Being able to remain in one’s community hospital for a treatment that is as safe and effective there as it is at a distant, tertiary center is good medicine.
The experience of hospitals participating in CPORT projects was that the development of PCI services raised the standard of care not just for patients with coronary artery disease but for all hospital patients. Perhaps the positive effect of a PCI program is best revealed by examining what happens in hospitals without such a program. In many regions, selective triage transports patients with suspected acute coronary syndromes only to PCI-capable hospitals. In hospitals that cannot perform PCI, cardiac catheterization services are frequently abandoned because most physicians do not want to submit patients to diagnostic catheterization without the option for PCI. Domino-like, once cardiac catheterization services are abandoned, other invasive services such as pericardiocentesis and right heart catheterization are increasingly difficult to obtain. This limited selection of cardiac services makes the facility’s environment less attractive to cardiologists in general. As time goes on, the extent and even the quality of cardiac services in that facility can decline, potentially affecting patient care in all medical and surgical services. In community hospitals with internal medicine residency programs, the absence of patients with acute ischemic heart disease, one of the most common diseases in the United States, is a major deficiency. The hospitals without colocated cardiac surgery that have a PCI program avoid the atrophy of their cardiac services, can better serve their general hospital patient population, and provide a better learning environment for trainees.
Clearly, there are important positive and negative consequences of extending PCI capability to hospitals without on-site cardiac surgery. The report by Lee and colleagues1 in this issue of Circulation says that we can safely perform PCI in this setting. The question is, Even if we can, should we? And if we do, how should we?
To balance the myriad factors that influence the potential benefits and harms of extending PCI capability, state departments of health, in concert with physicians and citizens, can develop evidence-based policies that simultaneously allow rational distribution of high-quality PCI services without excessive restriction. For example, the Maryland Health Care Commission, which regulates cardiovascular services in the State of Maryland, has modified the state’s health plan to accommodate first primary PCI and more recently nonprimary PCI as routine clinical procedures performed at hospitals without on-site cardiac surgery. The plan specifies licensure requirements that include the need for formal program development, institutional and operator volume benchmarks, data reporting, outcomes monitoring, and peer review for performance of PCI at all hospitals, regardless of cardiac surgery. The plan is evidence based and represents a collaborative effort among commission staff, local physician and nursing experts, regional representatives of national cardiovascular organizations, and citizen stakeholders. It can be reviewed online (Code of Maryland Regulations COMAR 10.24.17; http://www.dsd.state.md.us/artwork/10241701.pdf).7
Performing PCI at a hospital without on-site cardiac surgery is not simply a matter of stocking a catheterization laboratory with angioplasty equipment, recruiting interventional cardiologists, and applying national care guidelines. Rather, the development of such programs should replicate, to the extent possible, the research methods on which their safety and efficacy are based. Ideally, all PCI programs, regardless of colocated cardiac surgery, fit into an overall plan for regional cardiovascular services that provides optimal access to the highest-quality care at sustainable cost and is concordant with and supportive of the multifaceted missions of community, tertiary, and academic hospitals.
The opinions expressed in this article are not necessarily those of the editors or of the American Heart Association.
- © 2015 American Heart Association, Inc.
- Lee JM,
- Hwang D,
- Park J,
- Kim KJ,
- Ahn C,
- Koo BK
- Dehmer GJ,
- Blankenship JC,
- Cilingiroglu M,
- Dwyer JG,
- Feldman DN,
- Gardner TJ,
- Grines CL,
- Singh M
- Roe MT,
- Boden WE,
- Chen AY
- 7.↵Maryland Health Care Commission. Code of Maryland Regulations (COMAR 10.24.17). Baltimore, MD: Maryland Health Care Commission; 2015. http://www.dsd.state.md.us/artwork/10241701.pdf. Accessed July 21, 2015.