BACKGROUND: Current practice guidelines for performance of percutaneous coronary intervention (PCI) in the United States mandate availability of on-site surgical backup. With the decreasing frequency of urgent coronary bypass surgery (UCABG) with newer technologies, it is unclear whether such backup continues to be necessary. METHODS: A database of 5655 consecutive patients undergoing PCI at a single center between August 1, 1992, and December 31, 1997, was analyzed. Outcomes were determined as well as clinical, lesion, and procedural characteristics of patients during 4 time periods preceding and during use of coronary stenting. RESULTS: Frequency of UCABG for failed PCI decreased from 2.2% to 0.6% in the most recent time period (P <.01) with no change in incidence of in-hospital death or myocardial infarction. Incidence of stenting progressively increased to 72% in the latest period. Patients requiring UCABG had a higher prevalence of acute coronary syndromes (95%) and type B lesions (79%), but these characteristics were also common in patients who did not undergo UCABG. Although coronary stents were available during the last 3 periods studied, only 30% of UCABG patients had lesions or complications amenable to stenting, and stenting attempts in these patients were all unsuccessful. Despite stenting and use of perfusion balloons and intra-aortic balloon pumps, only 40% of patients having UCABG were stable and pain free on transfer to the operating room. CONCLUSIONS: Although use of UCABG for a failed PCI is currently very low, there are no satisfactory predictors, patients requiring UCABG are frequently clinically unstable, and availability of stenting does not reliably eliminate the need for UCABG or result in a decrease in mortality. This small group of patients continues to require readily available surgical standby.
BACKGROUND: Current practice guidelines for performance of percutaneous coronary intervention (PCI) in the United States mandate availability of on-site surgical backup. With the decreasing frequency of urgent coronary bypass surgery (UCABG) with newer technologies, it is unclear whether such backup continues to be necessary. METHODS: A database of 5655 consecutive patients undergoing PCI at a single center between August 1, 1992, and December 31, 1997, was analyzed. Outcomes were determined as well as clinical, lesion, and procedural characteristics of patients during 4 time periods preceding and during use of coronary stenting. RESULTS: Frequency of UCABG for failed PCI decreased from 2.2% to 0.6% in the most recent time period (P <.01) with no change in incidence of in-hospital death or myocardial infarction. Incidence of stenting progressively increased to 72% in the latest period. Patients requiring UCABG had a higher prevalence of acute coronary syndromes (95%) and type B lesions (79%), but these characteristics were also common in patients who did not undergo UCABG. Although coronary stents were available during the last 3 periods studied, only 30% of UCABGpatients had lesions or complications amenable to stenting, and stenting attempts in these patients were all unsuccessful. Despite stenting and use of perfusion balloons and intra-aortic balloon pumps, only 40% of patients having UCABG were stable and pain free on transfer to the operating room. CONCLUSIONS: Although use of UCABG for a failed PCI is currently very low, there are no satisfactory predictors, patients requiring UCABG are frequently clinically unstable, and availability of stenting does not reliably eliminate the need for UCABG or result in a decrease in mortality. This small group of patients continues to require readily available surgical standby.