Reducing the risk of emergency bypass surgery for failed percutaneous coronary interventions.

In this issue of the Journal, Yang et al. (1) from the Mayo Clinic report the changing incidence and outcomes of emergency coronary artery bypass surgery for failed percu-taneous coronary interventions (PCIs). The good news is that the rate of emergency bypass surgery declined dramatically , falling from 2.9% in the pre-stent era (1979 to 1994) to 0.7% in the early-stent era (1995 to 1998) and showing a further decrease to 0.3% in the current-stent era (1999 to 2003). The bad news is that the in-hospital mortality for emergency bypass surgery remained constant and high at 10% to 14%. A description of what went wrong in each case could identify areas for improvement. The Mayo group analyzed their database and found that patients sent for emergency bypass surgery had a higher incidence of prior revascular-ization, reduced ejection fractions, and more complex coronary anatomy than those with uneventful PCIs. After the introduction of stents, fewer patients with abrupt closure and dissections required emergency bypass surgery than in the pre-stent era. The database analysis generated helpful broad clinical observations but could not define the root cause of PCI failures that continued to occur after the availability of stents. A dutiful fellow-in-training could have reviewed charts, interviewed the laboratory staff, and pored through narratives to identify whether catastrophes such as vessel perforation, balloon non-deflation, or other types of device failure played a role; but the additional information might have been too heterogeneous for a database analysis. Additional insights into reasons for emergency bypass surgery in the current era have come from investigators at the Cleveland Clinic (2). Like the Mayo Clinic group, Seshadri et al. (2) reported a dramatic 10-fold reduction in referrals for emergency bypass surgery between 1992 and 2002 and also decried a persistent high mortality rate of 15%. These investigators found that vessel perforation or cardiac tamponade accounted for emergency bypass surgery in 20% of cases. Unfortunately, such events have been difficult to predict (3). Additional improvements have been needed to enhance the safety of PCI. Platelet glycoprotein (GP) IIb/IIIa inhibitors have been commonly theorized as reducing the need for emergency bypass surgery, but there is little direct proof for this hypothesis. The Cleveland Clinic group (2) identified stent-ing as independently associated with the decreased need for emergency bypass surgery, whereas the use of GP IIb/IIIa inhibitors was not. Several randomized trials of GP IIb/IIIa inhibitors (4 –7) have failed …