|Home | About | Journals | Submit | Contact Us | Français|
Background/Objectives. We aimed to investigate the incidence and clinical outcome of coronary artery bypass grafting (CABG) performed in contemporary patients with ST-elevation myocardial infarction (STEMI) within 30 days after presentation.
Methods. All 1071 patients enrolled in the Thrombus Aspiration during Percutaneous coronary intervention in Acute myocardial infarction Study (TAPAS) were included in this analysis. CABG was indicated for both ischaemic and anatomical reasons according to the current treatment guidelines for STEMI. For all surgical as well as non-surgical patients, clinical outcome was assessed at both 30 days and one year.
Results. CABG was performed within 30 days of presentation in 59/1071 (5.5%) patients, in 13 (22%) within 24 hours, in eight (14%) between one and three days, and in 38 (64%) between four and 30 days. Compared with non-surgical patients, surgical patients required more initial intra-aortic balloon pump support (33 vs. 5%, p<0.001) and more often had multi-vessel disease (p<0.001). Overall, rethoracotomy was performed in 9/59 (15%) patients. In patients operated within three days, the rethoracotomy rate was markedly higher than after three days (33 vs. 5%, p=0.004). Cardiac mortality at 30 days and one year was 1.7% in the surgical group and 3.2 and 5.3%, respectively, in the non-surgical group.
Conclusion. STEMI patients treated with CABG within three days after presentation are at increased risk of rethoracotomy. However, despite this higher incidence of surgical complications and multiple high-risk features at presentation, surgical management during the acute and subacute phase is associated with excellent 30-day and one-year survival. (Neth Heart J 2010;18:348–54.)
To optimise the treatment of patients with ST-elevation myocardial infarction (STEMI), timely reperfusion of the occluded coronary artery is essential.1,2 A reperfusion strategy consisting of primary percutaneous coronary intervention (PCI) has improved short- and long-term outcome in patients with STEMI and is now the preferred treatment modality.3,4 However, coronary angiography also identifies patients eligible for coronary artery bypass grafting (CABG) during the acute and subacute phase of STEMI.5,6 In some patients with a coronary anatomy unsuitable for PCI, CABG is used as the primary reperfusion modality either in the acute phase or after initial stabilisation. In most patients, CABG is performed after PCI as definitive or adjunctive revascularisation. A decade ago, Stone et al.7 reported that 11% of STEMI patients require CABG during hospital admission, and that these patients have a similar outcome to STEMI patients not requiring surgery. In the meantime, various advancements have been made in the interventional and surgical management of STEMI patients. In the contemporary interventional approach of STEMI patients, the majority of patients are treated with stenting and modern adjunctive antiplatelet therapies including glycoprotein (GP) IIb/IIIa inhibitors and clopidogrel. These developments may have an impact on the clinical management and outcome of contemporary STEMI patients. To place these recent advancements in perspective, we investigated the incidence and clinical outcome of CABG performed within 30 days in a contemporary cohort of patients presenting with STEMI.
This study included all 1071 patients who were enrolled in the Thrombus Aspiration during Percutaneous coronary intervention in Acute myocardial infarction Study (TAPAS), a randomised clinical trial investigating the effect of thrombus aspiration on myocardial perfusion in patients presenting with STEMI.8,9 Inclusion criteria were symptoms suggesting acute myocardial ischaemia of more than 30 minutes, time from onset of symptoms of less than 12 hours, and ST elevation of more than 0.1 mV in two or more ECG leads. Exclusion criteria were rescue PCI after thrombolysis and an expected life expectancy of less than six months. All patients received aspirin (500 mg), heparin (5000 IU), and clopidogrel (loading dose of 300 mg) after electrocardiographic confirmation of STEMI, usually in the ambulance on the way to the catheterisation laboratory. Weight-adjusted abciximab (0.25 mg/kg) was administered during PCI, combined with additional heparin guided by the activated clotting time.
The decision to perform CABG was made by a joint heart team consisting of a cardiothoracic surgeon, an interventional cardiologist, and a clinical cardiologist according to current guidelines.5,6 We considered CABG as a treatment option in the presence of ischaemia if one of the following criteria was met: persistent or recurrent ischaemia refractory to medical therapy, cardiogenic shock, CABG performed as a concomitant procedure to a postinfarction ventricular septal rupture or mitral valve insufficiency, or as primary reperfusion therapy within 24 hours if the coronary anatomy was unsuitable for PCI. In addition, we considered CABG a treatment option in the absence of ischaemia when an anatomical indication was present. Anatomical indications are defined as described in the guidelines:5,6 in summary, class I indications for CABG including significant left main disease, three-vessel disease and two-vessel disease with significant involvement of the proximal left anterior descending coronary artery and either depressed left ventricular function or noninvasive evidence of ischaemia.
We collected clinical and procedural data from all patients from the TAPAS database. For patients who were treated with CABG within 30 days after presentation, the perioperative and postoperative data were collected from a registry containing all cardiothoracic surgical procedures at our centre and from digital hospital records. From all patients, follow-up information was obtained at 30 days and one year, as previously described.8,9
Continuous data were expressed as mean ± standard deviation or median (interquartile range), where appropriate. Categorical data were expressed as counts (percentages). Continuous variables were compared using the two-sided Student’s t test or Mann-Whitney U test if samples were not normally distributed. Categorical variables were analysed with the χ2 test or Fisher’s exact test depending on the sample size. An α of 0.05 (two-tailed) was considered statistically significant. For comparison of ordinal variables of more than two groups, the p value for trend was obtained. The Statistical Package for the Social Sciences (SPSS Inc, Chicago, IL) version 14.0 was used for all statistical analyses.
CABG was performed within 30 days of presentation in 59 (5.5%) of the 1071 STEMI patients. All CABG procedures were performed for a class I or IIa indication. In five patients, a concomitant valvular procedure was performed (three mitral valve repairs, one mitral valve replacement, and one aortic valve replacement).
The clinical and angiographic characteristics of CABG and non-CABG patients are shown in table 1. Compared with the 1012 patients who were not treated with CABG within 30 days, CABG patients smoked less often (26 vs. 48%, p=0.002), had multivessel disease more often (p<0.001), underwent PCI and stent implantation less often (70 vs. 95% and 50 vs. 94%, respectively, both p<0.001), and were treated with intra-aortic balloon pump (IABP) support more often (33 vs. 5%, p<0.001).
Of the 59 (5.5%) patients requiring surgery, CABG was performed as the initial reperfusion therapy in seven (0.7%) patients (figure 1). In ten (0.9%) patients, pharmacological stabilisation was followed by CABG after a median of 4.5 days (range 1 to 12). In 42 (3.9%) patients, CABG was performed within 30 days after PCI as an additional or definitive treatment. CABG was required in 42/1005 (4%) patients who were initially treated with primary PCI and in 17/66 (26%) patients who were not treated with PCI.
Of the 59 CABG patients, 13 (22%) underwent surgery within 24 hours, 8 (14%) between one and three days, and 38 (64%) between four and 30 days. Overall, CABG was performed for an ischaemic indication in 18 (31%) patients and for an anatomical indication in 41 (69%) patients. Ischaemia was the indication to perform surgery in 85% of patients operated within 24 hours compared with in 25% of patients operated between one and three days and in 13% of patients operated between four and 30 days (figure 2). Ischaemia was more often the reason to perform CABG in patients who were not treated versus those treated with PCI (59 vs. 19%, p=0.003).
Table 2 shows the in-hospital characteristics of CABG patients in relation to time to surgery. Compared with patients undergoing surgery after three days, those operated within three days were treated with PCI and GP IIb/IIIa inhibitors less frequently and had a higher estimated operative risk as expressed in the European System for Cardiac Operative Risk Evaluation (EuroSCORE) and a longer duration of surgery.
Overall, rethoracotomy was performed in nine (15%) patients. When surgery was performed within 3 days, the rethoracotomy rate was significantly higher than after three days (33 vs. 5%, p=0.004, figure 3). In addition, patients receiving surgery within three days experienced more bleeding complications, required more postoperative use of IABP and inotropic agents and a longer intensive care and total hospital stay (table 2).
Perioperative enzymatic evidence of reinfarction was observed in six (10%) patients, with new Q waves present in one (1.7%) patient. During postoperative hospital stay, mediastinitis developed in two (3.4%) patients and atrial fibrillation or supraventricular tachycardia in 15 (26%) patients. Occlusion of an arterial graft was observed in one (1.7%) patient, for which PCI of the native vessel was successfully performed. One (1.7%) patient died due to mediastinitis after surgery was performed on the fourth day after admission.
Clinical follow-up was obtained in 1060/1071 (99%) patients. In the surgical group, no new deaths, reinfarctions or additional target vessel revascularisations occurred after hospital discharge. A non-infarct-related vessel was revascularised with PCI after seven weeks in one (1.7%) surgical patient as part of a hybrid strategy. While cardiac mortality at 30 days and one year was 3.2 and 5.3%, respectively, in the non-surgical group, mortality in the surgical group was 1.7%.
CABG has played an important role in the management of selected patients presenting with STEMI. In the current study, CABG was performed within 30 days after presentation in 5.5% of STEMI patients. They represent a high-risk group with more haemodynamic instability and more severe coronary artery disease. In these patients, in whom antiplatelet and anticoagulant therapy was administered to all patients and GP IIb/IIIa inhibitors in a large majority, the rate of rethoracotomy is markedly higher when surgery is performed within the first three days after presentation. Nevertheless, surgical management of these selected patients is associated with excellent 30-day and one-year survival. Patients included in the TAPAS trial are representative for a contemporary and unselected population of patients presenting with STEMI.
It appears that the need for surgical revascularisation in patients presenting with STEMI has decreased compared with a decade ago. While Stone et al.7 have reported that cardiac surgery was performed in 11% of patients with acute myocardial infarction, including 7% of patients treated with PCI initially, and 45% of patients not treated with PCI, our rates were 5.5, 4 and 26% respectively. It seems plausible that the reduced incidence of patients requiring CABG after PCI is related to advances in interventional cardiology such as the frequent use of stents and triple antiplatelet and anticoagulant therapy.7,10,11 Yang et al.11 have reported that during the last decades the incidence of patients requiring emergency CABG after PCI has decreased since the introduction of stents. Furthermore, routine pretreatment of STEMI patients with aspirin, heparin and clopidogrel is associated with higher initial patency of the infarct-related artery.12,13 As in our cohort both stents and GP IIb/IIIa inhibitors were used in 92% of STEMI patients undergoing PCI, and aspirin, heparin and clopidogrel in all patients,8 this may have contributed to the low need for surgical revascularisation, either as an emergent or elective procedure.
Although the overall rethoracotomy rate was only slightly higher than that in clinical practice and, after three days, resembled the rate reported in elective procedures, the incidence of rethoracotomy was strikingly higher in patients undergoing surgery within the first three days compared with that in patients operated after three days. However, this did not translate into a high mortality rate. Both at 30 days and one year after STEMI, mortality was low in the surgical (1.7%) as well as the non-surgical (3.2 and 5.3%) patients, while earlier reports have shown a mortality rate of 7.8 to 12% at six to 12 months of follow-up.7,14 The high occurrence of rethoracotomy in patients operated within the first three days can be attributed to the modern pharmacological management of STEMI including aspirin, heparin, clopidogrel, and abciximab. Consistently, the use of these antiplatelet and anticoagulant medications prior to CABG, either alone or in combination, has been associated with increased rethoracotomy and bleeding complications but not with in-hospital or 30-day mortality in patients undergoing CABG.15-18 In patients requiring CABG, several recently published guidelines recommend to withhold clopidogrel for five to seven days before CABG.5,19 However, this recommendation cannot be followed in most patients presenting with STEMI who require urgent CABG.
The optimal timing of CABG for patients with STEMI remains difficult to establish. In some studies, timing of surgery did not independently predict hospital mortality once corrected for parameters such as poor left ventricular ejection fraction, preoperative shock, age over 70 years, prior CABG, infarct extension, and stroke.14,20 In other reports, however, there is evidence that mortality is increased in patients operated within the first 24 hours to three days after STEMI.21-24 In the current study, the limited incidence of clinical events did not allow multivariate regression analysis to be performed. Nevertheless, the excellent 30-day and one-year survival of the CABG patients observed in this study, including those operated within the first three days, seem to justify the choice for early surgery in these patients. Although a direct comparison between the surgical and non-surgical group is not completely appropriate due to selection bias and natural selection in the surgical group, the favourable clinical outcome of the surgical group suggests a judicious selection of patients who are most likely to benefit from surgery. In fact, the selection of appropriate patients for surgery will have positively contributed to the excellent long-term outcome of the overall patient population enrolled in this trial. The excellent outcome after CABG in these selected patients implicates that complete revascularisation in multivessel disease patients presenting with a STEMI may be an attractive concept. Randomised trials with specific coronary anatomical criteria for revascularisation or with functional testing (fractional flow reserve, magnetic resonance imaging, and/or nuclear perfusion studies) will be needed to clarify these issues.
Our analysis suffers from all the limitations inherent in a cohort study from a high-volume single centre. The number of patients was not sufficient to allow separate analyses on the surgical patients based on their indication for surgery. Ideally, large randomised clinical trials should determine the role of surgical treatment in the management of STEMI patients. However, patients treated with CABG during the acute and subacute phase of STEMI represent a heterogeneous group, for whom surgery is required for a variety of reasons depending on complex combinations of clinical and anatomical characteristics including the patient’s coronary anatomy and comorbidity, and the surgeon’s and interventional cardiologist’s experience. Appropriate adjustments for all these factors are not likely to be achievable, even in large registry studies. Therefore, data from observational studies will remain the mainstay of our knowledge in this field in the foreseeable future.
CABG was performed in 5.5% of STEMI patients within 30 days after presentation in a contemporary patient cohort. These surgical patients are at increased risk of rethoracotomy when operated within the first three days. However, despite a high incidence of surgical complications and multiple high-risk features at presentation, 30-day and one-year clinical outcome is excellent. Therefore, the appropriate use of CABG is an integral component of the invasive approach in STEMI patients.