Author + information
- Received April 11, 2005
- Revision received June 10, 2005
- Accepted June 20, 2005
- Published online December 6, 2005.
- Eric H. Yang, MD⁎,
- Richard J. Gumina, MD, PhD⁎,
- Ryan J. Lennon, MS†,
- David R. Holmes Jr, MD⁎,
- Charanjit S. Rihal, MD⁎ and
- Mandeep Singh, MD⁎,⁎ ()
- ↵⁎Reprint requests and correspondence:
Dr. Mandeep Singh, Division of Cardiovascular Disease and Internal Medicine, Mayo College of Medicine, 200 First Street SW, Rochester, Minnesota 55905.
Objectives The purpose of the current study was to evaluate the changes in incidence, clinical characteristics, and indications for emergency coronary artery bypass grafting (CABG) in patients undergoing percutaneous coronary intervention (PCI) from 1979 to 2003.
Background Emergency CABG after PCI is associated with significant morbidity and mortality.
Methods Data from 23,087 patients who underwent PCI at Mayo Clinic from 1979 to 2003 were analyzed. Patients were divided into three groups: the “pre-stent” era, 1979 to 1994 (n = 8,905); the “initial stent era,” 1995 to 1999 (n = 7,605); and the “current stent era,” 2000 to 2003 (n = 6,577).
Results Although patients undergoing PCI in the recent time periods had more high-risk features, there was a significant decrease in the incidence of emergency CABG from 2.9% to 0.7% to 0.3% across the groups (p < 0.001). Patients requiring emergency surgery in the recent time periods had a higher prevalence of hypertension, prior revascularization, and left ventricular dysfunction (ejection fraction <40%), as well as more complex coronary lesions. Fewer patients in the current stent era had coronary artery dissections and abrupt vessel closure requiring emergency CABG. The in-hospital mortality rate for emergency CABG patients remains unchanged and ranges from 10% to 14%.
Conclusions The current study demonstrates that despite the increase in high-risk patients undergoing PCI, there has been a marked decrease in the incidence of patients requiring emergency CABG. However, the in-hospital mortality rate for those requiring emergency CABG remains high and unchanged.
Many technologic and pharmacologic advances in percutaneous coronary intervention (PCI) have been made over the past 25 years. Steerable guide wires, coronary artery stents, and new anti-platelet therapies have resulted in an increase in the success rate of PCI procedures (1–3). In addition, high-risk patients who in the past would have undergone surgical revascularization are now undergoing PCI (4,5). Despite the increase in the number of PCI procedures performed, the need for emergency coronary bypass grafting (CABG) has been decreasing (6). Emergency CABG, however, still occurs and is associated with significant morbidity and mortality (7–9). The impact of improvements in technology and the inclusion of high-risk patients on the need and outcome of emergency CABG following PCI are not well known.
We hypothesized that improvements in PCI technology and the inclusion of high-risk patients have resulted in a change in the incidence, clinical characteristics, and indications for emergency CABG in patients undergoing PCI from 1979 to 2003.
Patients undergoing PCI at the Mayo Clinic have been followed in a prospective registry since 1979. Patients in this registry have undergone follow up at 6 months, 1 year, and then annually since their PCI. A total of 23,087 patients undergoing PCI from 1979 to 2003 were included in the study. Patients were divided into three groups on the basis of the time period of their PCI. Group 1 consisted of 8,905 patients who underwent PCI from 1979 to 1994 and represents the “pre-stent era.” During this period, the technology for percutaneous transluminal angioplasty was being developed and operator experience was being gained. Group 2 consisted of 7,605 patients undergoing PCI from 1995 to 1999 and is considered the “early-stent era.” This was a transition period where stents and new parentral anti-platelet agents were introduced. Group 3 included 6,577 patients who underwent PCI from 2000 to 2003 and represents the “current-stent era.” In this recent time period, there was more frequent use of bare-metal stents, parentral glycoprotein IIb/IIIa inhibitors, and dual oral anti-platelet therapy.
Emergency CABG was defined as cardiac surgery performed within hours of PCI to avoid unnecessary morbidity or death. This surgery takes precedence over elective cases. Indications for emergency CABG included abrupt vessel closure, extensive coronary artery dissection, incomplete revascularization, coronary perforation, unsuccessful dilation, and other situations resulting in hemodynamic instability and requiring surgical intervention. Unstable angina was defined as new onset of chest pain at rest or progression of stable angina to an increased Canadian Cardiovascular Society (CCS) score. Patients presenting with chest pain within two months of coronary revascularization were also considered to have unstable angina if the last episode of pain occurred within one week of the PCI.
The indication for the index PCI was classified as elective, urgent, or emergent. Emergent cases were those that required PCI within hours of presentation in order to avoid significant morbidity and mortality. Urgent cases required PCI before discharge from the hospital and usually occurred within one to three days after presentation with chest pain. All other cases were considered elective. Coronary lesions were classified according to the American College of Cardiology/American Heart Association (ACC/AHA) scoring system (10).
Data are presented as mean ± SD for continuous variables and frequency for discrete variables. Kaplan-Meier methods were used to estimate long-term survival. Group comparisons were made using one-way analysis of variance for continuous data and Pearson’s chi-square test for nominal data. Discrete ordinal data were compared with the Wilcoxon rank-sum test. For pairwise comparisons, a Bonferroni-adjusted significance level of 0.0167 was used so that the total Type 1 error rate from the pairwise comparisons was no more than 0.05. Predictors of emergency CABG were determined by using a backwards selection method with multiple logistic regression. Significant predictors (p < 0.05) were reported as odds ratios and 95% confidence intervals (CI).
Characteristics of patients undergoing PCI
The baseline characteristics of the 23,087 patients who underwent PCI from 1979 to 2003 are shown in Table 1.Compared to Group 1, patients in Groups 2 and 3 were older (63.8 ± 11.4 years vs. 65.5 ± 11.9 years and 66.9 ± 12.1 years for Group 1 versus Groups 2 and 3, respectively, p < 0.001), and more likely to have diabetes (17% vs. 22% and 26%, respectively, p < 0.001), hypertension (49% vs. 61% and 74%, respectively, p < 0.001), and a greater body mass index (27.8 ± 4.6 kg/m2vs. 29.0 ± 5.2 kg/m2and 29.7 ± 5.7 kg/m2, respectively, p < 0.001). Patients in the recent time periods were also more likely to have undergone prior PCI (23% vs. 32% and 36% for Group 1 vs. Groups 2 and 3, respectively, p < 0.001) or CABG (19% vs. 21% and 22%, respectively, p < 0.001) and were more likely to have an ejection fraction <40% (5% vs. 10% and 11%, respectively, p < 0.001). Patients in Groups 2 and 3 were also more likely to have calcified (24% vs. 41% and 33% for Group 1 vs. Groups 2 and 3, respectively, p < 0.001) and type C coronary lesions (32% vs. 46% and 42%, respectively, p < 0.001). Despite a greater percentage of patients presenting with unstable angina (57% vs. 73% and 67%, respectively, p < 0.001), patients in Group 1 were less likely than those in Groups 2 and 3 to undergo urgent (8% vs. 44% and 48%, respectively, p < 0.001) or emergent procedures (10% vs. 17% and 19%, respectively, p < 0.001). There were a significantly greater proportion of stents used (0.1 ± 0.3 stents/patient vs. 1.2 ± 1.1 and 1.4 ± 1 stents/patient, respectively, p < 0.001) and glycoprotein IIb/IIIa inhibitors (0% vs. 34% and 59%, respectively, p < 0.001) in Groups 2 and 3.
The incidences of in-lab death were 35 (0.39%) in Group 1, 22 (0.29%) in Group 2, and 15 (0.23%) in Group 3 (p = 0.06 for trend).
Characteristics of patients requiring emergency CABG
Temporal trends in the proportion of patients requiring emergency CABG are shown in Figure 1.A significant decrease in the incidence of emergency CABG from 2.9% to 0.7% to 0.3% (p < 0.001 Armitage test for trend) was observed across the three groups. As shown in Table 2,patients requiring emergency CABG in the recent time periods had a higher prevalence of hypertension (39% vs. 56% and 65% for Group 1 vs. Groups 2 and 3, respectively, p = 0.010), prior PCI (19% vs. 3% and 24%, p = 0.039), and left ventricular dysfunction (5% vs. 13% and 24%, respectively, p = 0.004). These patients were also more likely than those in Group 1 to have undergone urgent (4% vs. 30% and 38% for Group 1 vs. Groups 2 and 3, respectively, p < 0.001) and emergent (15% vs. 45% and 38%, respectively, p < 0.001) procedures and have more complex coronary lesions.
During the study period, there was a significant change in reasons for emergency CABG (Table 2). Compared with the patients in Group 1, those in Groups 2 and 3 had a significantly lower incidence of abrupt vessel closure (21% vs. 4% and 14% for Group 1 vs. Groups 2 and 3, respectively, p < 0.001) and coronary artery dissection (34% vs. 22% and 14%, respectively, p < 0.001) requiring emergency CABG. Patients requiring emergency CABG in Groups 2 and 3 were also more likely to have coronary artery stents placed (0.1 ± 0.4 stents/patient vs. 0.9 ± 1.1 and 1.1 ± 1.4 stents/patient, respectively, p < 0.001) and receive glycoprotein IIb/IIIa inhibitors (0% vs. 29% and 52%, respectively, p < 0.001).
Predictors of emergency CABG
During the pre-stent era (1979 to 1994) the strongest predictor for emergency CABG was pre-procedure shock (OR 2.35, 95% confidence interval 1.33 to 4.13). Other significant predictors (Table 3)included acute myocardial infarction, CCS score ≥3, lesion in an angulated segment (>45°), and multi-vessel coronary disease.
During the stent-era (1995 to 2003) the strongest predictor for emergency CABG was an emergent indication for PCI (odds ratio 3.77, 95% confidence interval 2.02 to 7.02). Other significant predictors (Table 4)were multi-vessel coronary artery disease, the presence of peripheral vascular disease, lesion in an angulated segment (>45°), and history of tobacco abuse.
Mortality rate of emergency CABG
Forty-one patients (31 in Group 1, 8 in Group 2, and 2 in Group 3) who had emergency CABG died during hospitalization. The in-hospital mortality rates of patients undergoing emergency bypass surgery are shown in Figure 2and range from 10% to 14%. There was no significant difference in the mortality rates between the three groups. The most common causes of death (Table 5)were postoperative myocardial infarction (56%) and respiratory failure from pneumonia, acute lung injury, acute respiratory distress syndrome, or pulmonary embolism (19.5%). An unsuccessful PCI attempt was the most common indication for emergency CABG in these patients (53.7%).
In Groups 1, 2, and 3, respectively, 227, 48, and 19 patients who had emergency CABG survived to hospital discharge. The 1-year survival rates were similar in these patients (94.3%, 97.9%, and 87.7%).
The current study demonstrates that despite the increase in high-risk patients undergoing PCI, there has been a significant and sustained decrease in the incidence of emergency CABG. There has also been a change in the indication for emergency CABG, with fewer abrupt vessel closures and coronary artery dissections. These changes are likely due to improvements in interventional technology and medical therapy. The in-hospital mortality rate for those requiring emergency CABG, however, is unchanged and remains high.
Characteristics of patients undergoing PCI
Our results show that older and sicker patients are undergoing PCI. Patients in the more recent time periods were more likely to have hypertension, diabetes, and left ventricular dysfunction. Angiographically, these patients also had higher risk coronary disease with greater calcification and more severe ACC/AHA lesion scores. Similar findings were noted in a prior study involving patients from the National, Heart, Lung, and Blood Institute Registry (4). A possible explanation for the increase in high-risk patients is that the widespread use of coronary artery stenting has resulted in an improvement in PCI outcomes. These improvements may have allowed more high-risk patients to undergo PCI.
Although more high-risk patients are undergoing PCI, the in-lab death rate has remained the same. This finding, along with the decrease in emergency PCI rates, suggests that improvements in PCI have resulted in fewer complications, but if a major complication occurs, the mortality rate remains unchanged by the newer technologies.
Incidence of emergency CABG
Despite the increase in high-risk patients undergoing PCI, we found a 10-fold reduction in the incidence of emergency CABG, from 2.9% to 0.3%. These data are similar to the rates reported by Seshadri et al. (6) from the Cleveland Clinic, who reported a rate of 0.61%. Our results are also consistent with those from six trials comparing coronary artery stenting to angioplasty. This analysis demonstrated that 0.31% of patients in the most recent time period undergoing primary PCI required emergency CABG (11). The decline in the incidence of emergency CABG is most likely due to the development of coronary artery stents and glycoprotein IIb/IIIa inhibitors (12,13).
The patients requiring emergency CABG in the current-stent era were also older and sicker than the patients in the pre-stent era. This may be a reflection of the increase in high-risk patients undergoing PCI.
Indications for emergency CABG
There was a significant change in the indications for emergency CABG after PCI, with a decrease in the incidence of abrupt vessel closure and coronary artery dissection resulting in the need for emergency CABG. Both of these changes are most likely due to the increased use of coronary artery stents and glycoprotein IIb/IIIa inhibitors, which reduce the risk of abrupt vessel closure and can be used to treat coronary artery dissections (14).
There has also been a change in the predictors of emergency CABG. In the pre-stent era, the presence of pre-procedural shock was the strongest predictor. In the stent era, shock is no longer a significant predictor and again may be due to improvements in PCI technology.
Outcomes of emergency CABG
Although there has been a change in the characteristics and indications for emergency CABG, the in-hospital and one-year mortality rate for these patients remains the same, underscoring the need to recognize variables associated with higher incidence of emergent CABG, and tailor strategies to avoid or reduce such complication. We observed a mortality rate of 10% to 14%, which is similar to the rate of 15% reported by Seshadri et al. (6). These rates are much higher than those for elective CABG, and a previous study suggests that the high mortality rate may be due to the hemodynamic instability that is present at the time of emergency surgery (15). This instability can lead to a lower probability of receiving an internal mammary graft and a greater requirement for inotropic support and blood products (15,16). The most common causes of death in these patients were postoperative myocardial infarction and cardiac arrhythmias. Emergency CABG has also been associated with longer hospital stays, increased risk of postoperative myocardial infraction, and a greater prevalence of ventricular arrhythmias (16).
A limitation of the current study is that it was a retrospective review of outcomes at a single high-volume center. The results may therefore not be applicable to other centers with lower volumes.
In conclusion, the results of the current study show that despite the increase in high-risk patients undergoing PCI, there has been a dramatic decrease in the incidence of patients requiring emergency CABG. In addition, there has been a change in the indication for emergency CABG, with a decrease in the incidence of coronary artery dissections and abrupt vessel closure. The mortality rate associated with emergency CABG, however, remains high and unchanged.
- Abbreviations and Acronyms
- American College of Cardiology/American Heart Association
- coronary artery bypass grafting
- Canadian Cardiovascular Society
- percutaneous coronary intervention
- Received April 11, 2005.
- Revision received June 10, 2005.
- Accepted June 20, 2005.
- American College of Cardiology Foundation
- Stone F.,
- Gregg W.,
- Brodie F.,
- et al.
- Williams D.O.,
- Holubkov R.,
- Yeh W.,
- et al.
- Singh M.,
- Rihal C.S.,
- Berger P.B.,
- et al.
- Seshadri N.,
- Whitlow P.L.,
- Acharya N.,
- Houghtaling P.,
- Blackstone E.H.,
- Ellis S.G.
- Ryan T.J.,
- Faxon D.P.,
- Gunnar R.M.,
- et al.
- Singh M.,
- Ting H.H.,
- Berger P.B.,
- et al.
- Blankenship J.C.,
- Sigmon K.N.,
- Pieper K.S.,
- O’Shea C.,
- Tardiff B.E.,
- Tcheng J.E.
- Moliterno D.J.,
- Chan A.W.