Off-Pump Bypass Surgery—Ready for the Big Dance?

Every spring, college basketball teams from around the United States gear up for a competition to determine the best team of the year. This tournament is often referred to as the "big dance," and it represents the ultimate challenge of peer against peer. In this contest, dominance in prior years is no guarantee of continued success. In medicine, new technologies are continuously being developed to challenge the champions of prior years, but there is no calendar marking when a new therapy is ready for its "big dance"—the large multicenter randomized trial that can determine whether it should displace the reigning champion of conventional care in clinical practice. If the assessment comes too early as the therapy is still evolving, a promising treatment may be inappropriately discarded. If the assessment comes too late, practitioners may have already decided to adopt the therapy based on less rigorous evidence and may refuse to enroll patients into a trial based on "ethical grounds."

Over the last few years, much interest has focused on the evolving contest between coronary artery bypass graft (CABG) surgery with cardiopulmonary bypass and off-pump coronary artery bypass (OPCAB) surgery. Observational data suggest significant outcome advantages for the newcomer, OPCAB, yet it remains controversial. Many senior surgeons have resisted abandoning the CABG procedure they have performed comfortably for decades in favor of an unproven new procedure that is substantially more difficult for the surgeon and the anesthesiologist. Thus, the contest between conventional CABG with cardiopulmonary bypass and the newer OPCAB has all the makings of a classic matchup. But is OPCAB ready for its "big dance"?

Before answering this question, consideration of some additional background is appropriate. The development of cardiopulmonary bypass using the heart-lung machine is a watershed accomplishment of medicine in the 20th century. With varying details, CABG with cardiopulmonary bypass has been used for more than 30 years and has evolved into a safe procedure that is performed in more than 300 000 patients each year in the United States.¹ However, research has shown that routing the patient's circulation through the bypass machine activates a number of undesirable pathogenic processes that have been implicated in major complications following surgery.

Looking to improve patient outcomes, pioneering cardiac surgeons began attempting bypass grafting on a "beating" heart. Early OPCAB procedures identified some important challenges to this approach, including intraoperative hemodynamic instability during manipulation of the heart, difficulty providing complete revascularization (particularly of the lateral and posterior walls), and higher risks for anastomotic failures. Since these early days, OPCAB has rapidly evolved with better devices for stabilizing the heart, anesthetic techniques, and improved surgeon experience.

As the technique has been refined, its proponents have ventured to compare it directly with CABG with cardiopulmonary bypass in randomized trials.^{2 - 10} However, these initial trials have a number of important limitations. They generally have been small, single-center (and often single-surgeon) trials. Additionally, these early trials have tended to exclude severely ill patients as demonstrated by mortality rates that are 2-fold to 3-fold lower than national norms in both the CABG with cardiopulmonary bypass and OPCAB patients. Nonetheless, some trials observed benefits in OPCAB including reduced perioperative myocardial damage (elevations in creatinine kinase levels) and lower rates of renal dysfunction, atrial fibrillation, stroke, sternal wound infection, and need for transfusion. The OPCAB procedure has also been associated with reduced postoperative ventilation time, shorter hospital stays, and lower costs.

However, the available trials have not been uniformly supportive of the OPCAB approach. One area in which the newer procedure may be less satisfactory is in completeness and durability of revascularization. On these points, recent trials have reached different conclusions. At one end are the results from the Prague-4 trial, a 400-patient, 4-surgeon trial conducted at a single center in the Czech Republic.^{6 ,11} While these investigators found favorable 30-day clinical outcomes in the OPCAB group compared with conventional CABG with cardiopulmonary bypass,⁶ patency of saphenous vein grafts at 1 year was 49% vs 59%, respectively.¹¹ A decrease in graft patency for OPCAB was also reported in the Royal Brompton Hospital trial, which is a 104-patient, 2-surgeon randomized study.⁷ Overall, Khan et al⁷ found lower 3-month graft patency rates in those receiving OPCAB (88%) compared with standard CABG with cardiopulmonary bypass (98%) (P = .002). Unlike the Prague-4 trial, the lower patency in this study was largely accounted for by radial artery and internal mammary grafts rather than saphenous vein grafts.

In this issue of THE JOURNAL, Puskas et al¹² report 1-year angiographic follow-up data from the Surgical Management of Arterial Revascularization Therapies (SMART) trial, a single-center, single-surgeon, 197-patient trial conducted at Emory University that represents the other end of the success spectrum. The study is notable for several reasons. First, Puskas et al randomized a majority of the patients screened for the study. Second, they also examined a broad array of inhospital and 1-year outcomes, including functional status and health care expenditures. Third, the completeness of their patient follow-up is commendable.

Regarding their results, Puskas et al were able to place a substantially higher number of grafts per patient (3.2 per patient for OPCAB and 3.4 for CABG with cardiopulmonary bypass) than most of the earlier trials in this area. In assessing durability of results, they found 1-year graft patency rates in both OPCAB (94%) and CABG with cardiopulmonary bypass (96%) cases to be similarly outstanding (P = .32).

What explains the difference between the SMART trial results and the other recently reported studies? Patient selection, while a possibility, is not likely to be the answer. Patients enrolled in the SMART trial were slightly younger and healthier than those treated with CABG with cardiopulmonary bypass, yet they were comparable with those enrolled in other randomized studies. Chance could have played a role. Specifically, given the limited number of observations in any of these trials, a few adverse events one way or another could have had a measurable impact on an individual's score. Moreover, surgeons' skill may have differed. Studies have demonstrated that there is a significant learning curve associated with OPCAB, affecting both completeness of revascularization and patient outcomes.¹³ Puskas is one of OPCAB's pioneers and routinely uses this technique in more than 90% of his patients, in contrast to surgeons in the other studies who had less OPCAB experience.⁷

How are these data being received in the cardiovascular practice community? At present, some surgeons are believers in OPCAB and some are not. According to recent data from the Society for Thoracic Surgeons database, OPCAB accounts for just 23% of all isolated bypass procedures performed in the United States (E.D.P., unpublished Society of Thoracic Surgeons data, 2003). Additionally, the adoption curve for OPCAB has not been uniform among surgeons. While certain hospitals have embraced OPCAB for a majority of their coronary artery bypass cases, others have steadfastly been loyal to CABG with cardiopulmonary bypass.

This state of equipoise and uncertainty creates the ideal stage for a definitive evaluation of OPCAB vs CABG with cardiopulmonary bypass. Early randomized comparisons such as the SMART trial demonstrate the proof of concept for OPCAB. In the right hands and for the right patients, OPCAB offers safe, complete, and durable revascularization that may reduce complications relative to conventional CABG with cardiopulmonary bypass. The question of whether OPCAB should become the new standard for coronary bypass surgery rests on the generalizability of these findings.

A large, multicenter randomized trial of CABG with cardiopulmonary bypass compared with OPCAB surgery would address many important questions. First, and most important, it could clarify whether the SMART trial findings may be extrapolated to the larger community of experienced cardiac surgeons in practice. Second, a larger multicenter trial could be powered to address any potential difference of these procedures on important patient outcomes, something these smaller, single-center studies were not able to do. Finally, such a study could compare results in important patient subgroups. Specifically, the observational literature suggests that the benefits of OPCAB may be greater in those with higher surgical risks (including elderly patients, those with renal impairment, and patients with significant carotid disease). However, existing trials have generally underenrolled these higher-risk subgroups.

To this end, the US Department of Veterans Affairs (VA) has initiated a randomized comparison of CABG with cardiopulmonary bypass compared with OPCAB (Fred Grover, MD, written communication, March 2004). Of the planned enrollment of 2200 patients over 4 years, more than 900 have been enrolled to date. While this VA study is an important start, ideally a large, non-VA, multicenter randomized trial is needed to compare these techniques in women and among surgeons in a wider range of practice settings.

In subjecting this new approach to a rigorous evaluation, the SMART investigators have set the stage for the more definitive large-scale randomized trials. In medicine, as in athletics, the process of competition itself can be a force for improvement. The "big dance" brings out the best in its competitors. Regardless of the actual trial results, it is patients needing revascularization who are the real winners when physicians are driven to perfect their craft.