PCI without on-site cardiac surgery not business as usual

The rapid proliferation of such centers should slow until more data are available.

In an effort to provide urgent coronary revascularization for acute transmural myocardial infarction, hospitals have increasingly begun to provide primary PCI without on-site cardiac surgery. Data have been supportive of this approach in several trials. Most specifically, the Atlantic Cardiovascular Patient Outcomes Research Team (C-PORT) trial reported [JAMA, 2002] improved outcomes for patients undergoing primary PCI without cardiac surgery backup compared to patients receiving thrombolytics in comparative community hospitals.

Likewise, Wharton and colleagues reported comparative primary PCI outcomes in select hospitals — again, without on-site cardiac surgery — compared to the PAMI-AIR trial [JACC, 2004].

These trials specifically assessed the feasibility of primary PCI in hospitals without on-site cardiac surgery, notable in that the involved hospitals, staff and operators were appropriately trained and qualified for such procedures.

Data for elective coronary intervention in hospitals with on-site surgical backup are less well studied. The Mayo Clinic reported satisfactory outcomes for both acute and elective intervention without surgery backup in a satellite hospital using consultative telemedicine (Low-risk Percutaneous Coronary Interventions Without On-site Cardiac Surgery).

PCI without on-site surgery

Despite limited data there appears to be a significant proliferation of elective coronary intervention in hospitals without surgical backup. Reported reasons include expansion of existing AMI programs and improved ease of patient access in more remote areas. However, in reality, other reasons may be operative. Diagnostic cath labs in hospitals without surgical intervention have generally seen volumes drop because patients, and to some degree insurance carriers, demand performance of diagnostic procedures in laboratories where ad hoc coronary intervention can be performed.

George W. Vetrovec

In addition, hospital competition and income undoubtedly play a role. There are institutions literally within blocks of a major heart center who because of a perceived competitive need and/or for economic considerations now offer elective PCI without on-site surgical backup. Given this reality, it is important to understand the impact of elective PCI performed in hospitals without on-site surgery on patient outcomes.

In the current era of stenting, the need for urgent surgical backup has markedly diminished. Although interventionalists can prospectively identify certain high-risk patients, serious complications following PCI are frequently not predictable. Furthermore, the lack of an available cardiac surgeon limits the potential joint review of a case to determine the optimal revascularization approach. Lastly, distant hospitals offering surgical backup are assumed to lessen the risk of needed bailout surgery, though the latter is largely unproven.

National Medicare study

Recently, Wennberg and colleagues reported the comparative outcomes from the 1999 through 2001 national Medicare database for urgent and elective PCI performed in hospitals with and without on-site surgical backup. In this retrospective analysis, mortality for patients undergoing PCI in institutions without cardiac surgery was significantly worse in the elective population while there was no difference in the outcome for primary or rescue PCI in either hospital setting. The higher mortality was predominantly seen in hospitals performing less than 50 procedures a year. Of note, the absence of cardiac surgery did not predict outcomes.

While these data only represent Medicare patients, it is likely this is a representative sample, since about half of PCIs performed in the United States encompass the Medicare population, often representing the sickest population because of age-related comorbidities. The reality is that if these data are true, the performance of elective PCIs in hospitals without in-house cardiac surgery —particularly with low volumes — can be associated with a 38% increase in mortality. Given an increased risk of mortality, one wonders how many patients or families would accept these adverse odds for purported “convenience?”

More data needed

Does this mean that the concept is wrong? My answer at present is we do not have the data to know. Certainly the rapid proliferation of such centers should slow until more data are available. Outcomes issues appear to be more complex than the lack of on-site surgery. In-hospital surgery is likely only a marker of a hospital’s resources for the PCI program.

Thus, more data should be captured in a prospective, scientific fashion comparing outcomes between hospitals with and without on-site cardiac surgery. Examples include appropriate training of a dedicated and trained procedure staff, a well-stocked laboratory including special catheters and emergency equipment that may not be anticipated prior to the procedure.

Finally, qualified physicians, in terms of training, certification and procedure volume, should only perform procedures in such laboratories. Given data that higher volume laboratories have better outcomes, more experienced operators might be most appropriate for low-volume laboratories without surgical backup. Regardless of qualifications, the best interventionalist must have adequate staff and equipment.

With well-designed studies or perhaps data from the ACC NCDR Registry, guidelines for staff, physicians, equipment and minimum procedure numbers can be developed. For now, physicians, hospital administrators, regulators and the public must demand that PCI programs performing elective procedures without on-site surgery demonstrate outcomes comparable to established high-volume PCI centers with on-site surgery.

In summary, to date the concept of urgent PCI in hospitals without in-house surgery has not been shown to be associated with adverse outcomes. Conversely, we currently lack data to support widespread application of elective coronary intervention without appropriate surgical backup. Until further data are available, caution must be exercised to avoid reducing the major successes that have been achieved in PCI over the last 25 years under the guise of enhanced patient convenience.

George W. Vetrovec, MD, from the Virginia Commonwealthe University Medical Center, Richmond, is an editorial board member for Cardiology Today's Interventional Cardiology section.

For more information:

• Aversano T, Aversano P, Paffanni E, et al: Thrombolytic therapy versus primary percutaneous intervention for myocardial infarction in patients presenting to hospitals without on-sight cardiac surgery; a randomized controlled trial [Erratum appears in JAMA. 2002;287:3212] JAMA.2002;287:1943-1951.
• Wharton PP Jr, Grimes LL, Turto MA, et al: Primary angioplasty in acute myocardial infarction in hospitals with no surgery on-sight. The PAMI-NO SOS Study versus transfer to surgical centers for primary angioplasty. J AM Coll Cardiol. 2004;43:1943-1950.
• Ting HH, Garratt KN, Fingh M, et al. Low-risk percutaneous coronary interventions without on-sight cardiac surgery; Two-year's observational experience and follow-up. Am Heart J. 2003;145;278-284.
• Wennberg DE, Lucas FE, Siewers AE, et al. Outcomes of percutaneous coronary interventions performed in centers without and with on-sight coronary artery bypass graft surgery. JAMA. 2004;292:1962-1968.