Clinical characteristics and early mortality of patients undergoing coronary artery bypass grafting compared to percutaneous coronary intervention: Insights from the Australasian Society of Cardiac and Thoracic Surgeons (ASCTS) and the Melbourne Interventional Group (MIG) Registries

Results CABG patients were older ( p < 0.01). The CABG group had a higher incidence of diabetes, heart failure, left ventricular ejection fraction <45%, multi-vessel coronary artery, peripheral vascular and cerebro-vascular disease (all p < 0.01). Patients undergoing PCI had a higher incidence of recent myocardial infarction (MI) as the indication for revascularisation ( p < 0.01). In-hospital and 30-day mortality was 1.8% and 1.7% in the CABG group, and 1.4% and 1.8% in the PCI group, respectively. Independent predictors of 30-day mortality after CABG were age (odds ratio 1.1 per year, 95% confidence interval 1.0–1.1), cardiogenic shock (4.10, 1.7–10.5) and previous CABG (6.6, 2.4–17.7). Predictors after PCI were diabetes (2.7, 1.4–5.1), female gender (3.0, 1.6–5.5), renal failure (3.2, 1.2–8.0), MI < 24 h (4.0, 2.2–7.6), left main intervention (5.4, 1.0–27.7), heart failure (6.0, 2.6–14.0) and cardiogenic shock (11.7, 5.4–25.2). Conclusions In contemporary clinical practice, CABG is preferred in patients with multi-vessel coronary and associated non-coronary vascular disease, while PCI is the dominant strategy for acute MI. Despite this, in-hospital and 30-day mortality rates were similar. Predictors of early mortality after CABG differ to those of PCI. Keywords Coronary artery bypass surgery Percutaneous coronary intervention Introduction Coronary artery bypass grafting (CABG) and percutaneous coronary intervention (PCI) with or without stent implantation are alternative strategies for the treatment of ischaemic obstructive coronary artery disease. Generally, CABG is the preferred strategy in patients with left main or triple-vessel coronary disease with reduced left ventricular function [1] . PCI is generally preferred in patients with single or two-vessel disease unless adverse lesion characteristics are present [2] . The major limitation of PCI with bare-metal stents (BMS) compared to CABG has been a greater need for repeat revascularisation [3,4] . Recently, the introduction of drug-eluting stents (DES) has been shown to significantly reduce the incidence of restenosis and hence the need for repeat revascularisation in certain lesion and patient subsets [5,6] . This has led to widespread uptake of DES in more complex lesions compared with those used in randomised trials [7] . As a result, PCI is increasingly performed for indications in patients who have traditionally been referred for CABG. Despite this, there are presently limited registry and randomised data regarding the clinical characteristics, risk profile, and mortality of patients undergoing CABG compared to PCI in the era of DES. We aimed to compare the clinical characteristics of patients undergoing CABG vs. PCI and determine the risk factors for early mortality utilising two large multi-centre Australian registries. Methods The study population comprised 8258 operations and procedures performed from 1 April 2004 to 30 June 2006. This consisted of 3841 consecutive patients undergoing isolated CABG in the Australasian Society of Cardiac and Thoracic Surgeons (ASCTS) registry (CABG group) and 4417 consecutive patients undergoing PCI in the Melbourne Interventional Group (MIG) registry (PCI group). The ASCTS registry collects data on patients undergoing CABG at all public cardiac surgical centres in Victoria, Australia [8] . The MIG registry is a collaborative venture of interventional cardiologists practicing at seven Victorian public tertiary referral hospitals [9] . Both registries are independently coordinated through an academic outcomes research centre at Monash University, Melbourne, Australia. Demographic, clinical and procedural characteristics were prospectively recorded using standardised definitions for all fields [8,9] . Ethical approval and “Opt-out” informed consent was obtained in all patients, as previously described [8,9] . The ASCTS database is subject to regular random audits where 5% of cases are audited for 38 out of >200 fields. The quality of data collection has been consistently excellent. An independent random MIG audit examined 10 verifiable fields from 3% of all patients enrolled from each site. Overall data accuracy in the MIG registry was 97%. Multi-vessel disease is defined as >50% stenosis in two coronary systems. Coronary systems are defined as (i) left anterior descending and diagonal arteries, (ii) left circumflex and obtuse marginal arteries and (iii) right coronary artery. Left main coronary artery is considered multi-vessel disease as it gives rise to the left anterior descending and left circumflex artery systems. Current heart failure was defined in the MIG registry as clinical heart failure within 14 days of index procedure whereas in the ASCTS registry, it is defined as clinical heart failure during the same admission. The most recent assessment of left ventricular ejection fraction (LVEF) by SPECT imaging, echocardiography or left ventricular angiography, prior to intervention or surgery was used. In the ASCTS registry, LVEF was expressed as normal (>60%), mildly (45–60%), moderately (30–44%) or severely (<30%) reduced. In the MIG registry, LVEF was estimate and expressed as a percentage. Otherwise, all other field definitions between the two groups were identical. In-hospital complications were recorded at time of hospital discharge. Thirty-day mortality was obtained by telephone contact with patient, family member or treating medical practitioner. Cause of death outside hospital was confirmed with the patient's primary care physician. CABG Procedures Bypass grafting strategy, peri-operative management of antiplatelet therapy and the choice of using cardiopulmonary bypass was at the discretion of the individual surgeon. PCI Procedures The interventional strategy and stent selection was at the discretion of the operator. In 2003, PCI guidelines were developed for use of DES in public hospitals restricting their use for patients at high-risk of restenosis who will theoretically derive the greatest benefit, though this would be considered as largely “off-label” use [7] . The resulting criteria for use of DES included one or more of the following: diabetes mellitus, renal failure, small target vessels (≤2.5 mm), long lesions (≥20 mm) and complex lesions such as chronic total occlusions, in-stent restenosis, bifurcation and ostial lesions. Total stent length was used as a surrogate for target lesion length, and stent diameter for target vessel diameter. Peri-procedural glycoprotein IIb/IIIa inhibitors were used at the operator's discretion. Oral anti-platelet therapy followed guidelines which recommend combination of aspirin and clopidogrel for a minimum of 4 weeks for BMS and between 3 and 12 months for DES [2] . Statistics Datasets were merged after common definitions of all fields were identified. Unadjusted comparison of clinical characteristics, in-hospital and 30-day mortality between registries were performed. Continuous variables were expressed as mean ± S.D., and categorical data expressed as percentages. Continuous variables were compared using Student's t -tests. Categorical variables were compared using Fisher exact tests. Univariate and multivariate predictors of in-hospital and 30-day mortality for PCI and CABG were identified by logistic regression. All significant univariate variables at p < 0.10 were then included in multivariate models to identify independent predictors ( p < 0.05). All statistical analysis was performed using SPSS Version 15.0 for Windows (SPSS Inc., Chicago, IL, USA). Results Baseline Characteristics Patients who underwent CABG compared to PCI were older (66.0 ± 10.2 vs. 64.9 ± 12.0, p < 0.01) ( Table 1 ). However, octogenarians were more likely to undergo PCI than CABG (11.2 vs. 5.7%, p < 0.01) ( Fig. 1 ). Patients undergoing CABG had a higher incidence of diabetes (32.8 vs. 23.0%, p < 0.01), history of heart failure (current, 6.6 vs. 3.7%, p < 0.01 and previous 13.5 vs. 3.5%, p < 0.01), MI >7 days (40.0 vs. 28.9%, p < 0.01), more diffuse vascular disease including multi-vessel coronary artery disease (96.0 vs. 58.8%, p < 0.01), peripheral vascular disease (13.5 vs. 6.9%, p < 0.01) and cerebrovascular disease (12.0 vs. 5.5%, p < 0.01). Conversely, patients undergoing PCI had a higher incidence of recent MI (<24 h, 21.0 vs. 2.6%, p < 0.01 and 1–7 days, 20.8 vs. 12.8%, p < 0.01, respectively). Somewhat surprisingly, there was no difference in the incidence of cardiogenic shock between the two groups (2.1 vs. 2.2%, p = 0.65). Procedural Characteristics CABG Coronary artery bypass grafting is still utilised primarily for three vessels disease (74%) with a mean total of 3.3 ± 1.0 bypass grafts per procedure ( Table 2 ). Arterial bypass grafts were predominantly used; mean of 2.4 ± 1.1 arterial grafts compared to a mean of 0.9 ± 1.0 saphenous venous grafts. Single and bilateral internal thoracic arterial grafts were used in 96.1% and 12.6% of cases, respectively. Most CABG procedures were performed with cardiopulmonary bypass; only 4.6% were conducted off-pump. Reoperation was performed in 2.8% of cases. More patients received intra-aortic balloon pump compared to patients who underwent PCI (5.0 vs. 1.8%, p < 0.01). PCI Left main coronary, left anterior descending artery and bypass graft interventions were performed in 0.7%, 32.0% and 3.1% of patients, respectively ( Table 3 ). A mean of 1.22 ± 0.5 lesions were treated per procedure. Stents were implanted in 94.6% of case, of which drug-eluting stents were used in 51.7% of patients. Clinical Outcomes In-hospital and 30-days follow up was complete in 100% and 95%, respectively. Unadjusted in-hospital (1.8 vs. 1.4%, p = 0.14) and 30-day mortality rates (1.7% vs. 1.8%, p = 0.87) in patients who underwent CABG were similar to patients who underwent PCI ( Fig. 2 ). There was no difference in 30-days survival on Kaplan–Meier analysis ( Fig. 3 ). In the PCI group, the rate of stent thrombosis was 0.7% ( n = 32). Predictors of 30-day Mortality Using the predetermined variables listed above, independent predictors of 30-day mortality in patients who underwent CABG were age (odds ratio (OR) 1.1 per year, 95% confidence interval (CI) 1.04–1.13), cardiogenic shock (OR 4.10, 95%CI 1.67–10.48) and previous CABG (OR 6.56, 95%CI 2.43–17.71) ( Table 4 ). Independent predictors of 30-day mortality in patients who underwent PCI were different to those who underwent CABG with the exception of cardiogenic shock. They included diabetes mellitus (OR 2.72, 95%CI 1.24–8.06), female gender (OR 2.99, 95%CI 1.63–5.48), MI < 24 h (OR 4.06, 95%CI 2.17–7.60), current heart failure (OR 5.98, 95%CI 2.55–14.04), left main coronary artery disease (OR 5.37, 95%CI 1.04–27.64) and cardiogenic shock (OR 11.70, 95%CI 5.42–25.22). Discussion There are several important findings from this observational study of two databases (CABG and PCI) that demonstrate current utilisation of these revascularisation strategies. There were significant differences in clinical characteristics and risk profile between the two groups. Octogenarians were more likely to undergo PCI. Diabetics and patients with multi-vessel coronary, peripheral vascular and cerebral vascular disease were more likely to undergo CABG. PCI was overwhelmingly used in patients who presented with recent MI < 7 days. There were no differences in early (30-day) mortality rates between the two revascularisation strategies. Independent predictors for early mortality were different between treatment strategies. Females, diabetics and patients who presented with MI < 24 h who underwent PCI were at increased risk of short-term mortality whereas these variables were not associated with early mortality in the surgical group. For patients who underwent CABG, advancing age and previous CABG were independent predictors of mortality. Cardiogenic shock was the only clinical variable associated with increased mortality for both CABG and PCI. Percutaneous coronary intervention and CABG are complementary methods of coronary revascularisation. The decision to offer patients CABG continues to be largely determined by the extent of coronary artery disease and LVEF. In patients with multi-vessel disease, CABG is still associated with higher rates of complete revascularisation and a greater durability than PCI with BMS, resulting in lower rates of repeat revascularisation. However, in patients who present with recent MI (STEMI or NSTEMI), the speed of reperfusion and the relatively low morbidity of PCI are distinct advantages over CABG. In our study, 21% of PCI compared to 2.6% of CABG were performed in patients who presented with MI < 24 h (mostly acute STEMI). The ACC/AHA STEMI guidelines recommend PCI as the initial reperfusion strategy for acute STEMI contingent upon rapid initiation [10] . This is based on multiple randomised trials demonstrating the superiority of rapid primary PCI over fibrinolysis in STEMI [11] . Moreover, CABG in the setting of acute MI is associated with increased mortality risk [12] . Clinical decision-making in coronary artery disease relies heavily on evidence-based medicine. Although randomised controlled trials (RCT) constitute the highest order of evidence and remain the standard for comparisons between therapies, RCT study populations are often highly selected and extrapolation of trial results to a more heterogeneous general population may be problematic. While comprehensive observational databases can be subject to numerous biases, they have significant value in validating real-world use of technologies and represent a more accurate accounting of everyday clinical care. We found that despite the introduction of DES in 2003, CABG remained primarily used for patients with multi-vessel, especially triple-vessel coronary artery disease and reduced left ventricular function. Nevertheless, 58.5% of PCI were performed in patients with multi-vessel disease. It is not known whether multi-vessel PCI or complete revascularisation was performed in these patients. This may affect long-term clinical outcome as failure of stenting to achieve complete revascularisation in patients with multi-vessel coronary artery disease has been associated with reduced survival [13] . Although DES reduce the risk of restenosis and the need for repeat revascularisation, there is no convincing evidence of DES reducing the risk of mortality or subsequent MI compared to BMS [14,15] . Furthermore, late (>30 days) and very-late (>12 months) stent thrombosis appears to be a potentially important limitation of DES. Stent thrombosis is associated with a high-risk of MI of 65–70% and mortality of 25–45% [16] . The annual risk of DES thrombosis is estimated at between 1% and 3% depending on the complexity of the lesion, patient comorbidities and use in off-label situations [7] . It is unclear how contemporary PCI with DES compare with CABG in patients with a risk profile that is more advanced than in published RCTs. There are three large ongoing randomised trials (FREEDOM, SYNTAX and CARDIA) comparing DES against CABG in patients with diabetes, multi-vessel and left main coronary disease. Given the recent concern regarding late stent thrombosis with DES, the long-term outcomes of these three studies will be critical in determining the safety and effectiveness of stenting compared to CABG in these high-risk patients population. Limitations Our study has the inherent limitation of being a retrospective observational study. There were minor differences in definitions between the two registries. Discrepancy in the assessment of LVEF and incomplete data in this field may account for the wide variation in LVEF estimates. Conclusions The risk profile of patients undergoing CABG differs to those of patients having PCI. While there was no difference in short-term mortality, the predictors of poor outcome differed according to the revascularisation strategy. Not surprisingly, PCI was the dominant strategy for acute MI while CABG is preferred in patients with multi-vessel coronary and associated vascular disease. Long-term follow-up will aid randomised studies in determining the best revascularisation strategy for specific patient cohorts. Ultimately, instead of being competitive, CABG and PCI will be seen to be complementary. Acknowledgements The Melbourne Interventional Group acknowledges funding from Astra-Zeneca, Biotronik, Boston-Scientific, Johnson & Johnson, Medtronic, Pfizer, Schering-Plough, Sanofi-Aventis, Servier, St. Jude and Terumo. These companies do not have access to the data, and do not have the right to review articles before publication. The Victorian Cardiac Surgery Database funded by the Victorian Department of Health and Human services. Dr. Duffy's work is supported by a NHMRC Centre of Clinical Research Excellence grant to the Alfred and Baker Medical Unit. Appendix A Melbourne Interventional Group Investigators: Alfred Hospital: S.J. Duffy, J.A. Shaw, A. Walton, C. Farrington, A. Dart, A. Broughton, J. Federman, C. Keighley, M.J. Butler. Austin Hospital: D.J. Clark, O. Farouque, M. Horrigan, J. Johns, L. Oliver, J. Brennan, R. Chan, G. Proimos, T. Dortimer, A. Tonkin, L Brown, A Sahar, M. Freeman, H.S. Lim, A. Al-Fiadh, K. Charter. Box Hill Hospital: G. New, L. Roberts, M. Rowe, G. Proimos, N. Cheong, C. Goods, A. Teh, C.C.S. Lim. Frankston Hospital: R. Lew, G. Szto, R. Teperman, R. Templin. Geelong Hospital: A. Black, M. Sebastian, T. Yip, J. Aithal, J. Dyson, T. Du Plessis. Monash University: H. Krum, C. Reid, N. Andrianopoulos, A. Brennan, P. Loane, L. Curran, F. Groen. Royal Melbourne Hospital: A.E. Ajani, R. Warren, D. Eccleston, J. Lefkovits, B.P. Yan, R. Gurvitch, M. Sallaberger. Western Hospital: Y.-L. Lim, D. Eccleston, A. Walton. References [1] S. Yusuf D. Zucker P. Peduzzi L.D. Fisher T. Takaro J.W. Kennedy K. Davis T. Killip E. Passamani R. Norris Effect of coronary artery bypass graft surgery on survival: overview of 10-year results from randomised trials by the Coronary Artery Bypass Graft Surgery Trialists Collaboration Lancet 344 8922 1994 563 570 [2] S.C. Smith Jr. T.E. Feldman J.W. Hirshfeld Jr. A.K. Jacobs M.J. Kern S.B. King III D.A. Morrison W.W. O’Neill H.V. Schaff P.L. Whitlow D.O. Williams E.M. Antman C.D. Adams J.L. Anderson D.P. Faxon V. Fuster J.L. Halperin L.F. Hiratzka S.A. Hunt R. Nishimura J.P. Ornato R.L. Page B. Riegel ACC/AHA task force on practice guidelines; ACC/AHA/SCAI 2005 guideline update for percutaneous coronary intervention Circulation 113 1 2006 156 175 [3] P.W. Serruys F. Unger J.E. Sousa A. Jatene H.J. Bonnier J.P. Schonberger N. Buller R. Bonser M.J. van den Brand L.A. van Herwerden M.A. Morel B.A. van Hout Arterial Revascularization Therapies Study Group Comparison of coronary-artery bypass surgery and stenting for the treatment of multivessel disease N Engl J Med 344 15 2001 1117 1124 [4] SoS Investigators Coronary artery bypass surgery versus percutaneous coronary intervention with stent implantation in patients with multivessel coronary artery disease (the Stent or Surgery trial): a randomised controlled trial Lancet 360 Sep (9338) 2002 965 970 [5] J.W. Moses M.B. Leon J.J. Popma P.J. Fitzgerald D.R. Holmes C. O'Shaughnessy R.P. Caputo D.J. Kereiakes D.O. Williams P.S. Teirstein J.L. Jaeger R.E. Kuntz SIRIUS Investigators Sirolimus-eluting stents versus standard stents in patients with stenosis in a native coronary artery N Engl J Med 349 14 2003 1315 1323 [6] G.W. Stone S.G. Ellis D.A. Cox J. Hermiller C. O'Shaughnessy J.T. Mann M. Turco R. Caputo P. Bergin J. Greenberg J.J. Popma M.E. Russell TAXUS-IV Investigators A polymer-based, paclitaxel-eluting stent in patients with coronary artery disease N Engl J Med 350 3 2004 221 231 [7] H.K. Win A.E. Caldera K. Maresh J. Lopez C.S. Rihal M.A. Parikh J.F. Granada S. Marulkar D. Nassif D.J. Cohen N.S. Kleiman EVENT Registry Investigators Clinical outcomes and stent thrombosis following off-label use of drug-eluting stents JAMA 297 18 2007 2001 2009 [8] C.M. Reid M. Rockell P.D. Skillington G.C. Shardey J.A. Smith M. Yii S. Seevanayagam M. Mohajeri M. Rowland Initial twelve months experience and analysis for 2001–2002 from the Australasian Society of Cardiac and Thoracic Surgeons–Victorian database project Heart Lung Circ 13 3 2004 291 297 [9] A.E. Ajani G. Szto S.J. Duffy D. Eccleston D.J. Clark J. Lefkovits D.P. Chew R. Warren A. Black G. New A. Walton R. Lew J. Shaw M. Horrigan M. Sebastian B.P. Yan A. Brennan A. Meehan C. Reid H. Krum Melbourne Interventional Group investigators The foundation and launch of the Melbourne Interventional Group: a collaborative interventional cardiology project Heart Lung Circ 15 1 2006 44 47 [10] E.M. Antman D.T. Anbe P.W. Armstrong E.R. Bates L.A. Green M. Hand J.S. Hochman H.M. Krumholz F.G. Kushner G.A. Lamas C.J. Mullany J.P. Ornato D.L. Pearle M.A. Sloan S.C. Smith Jr American College of Cardiology; American Heart Association; Canadian Cardiovascular Society ACC/AHA guidelines for the management of patients with ST-elevation myocardial infarction—executive summary J Am Coll Cardiol 44 3 2004 671 719 [11] E.C. Keeley J.A. Boura C.L. Grines Primary angioplasty versus intravenous thrombolytic therapy for acute myocardial infarction: a quantitative review of 23 randomised trials Lancet 361 9351 2003 13 20 [12] D.C. Lee M.C. Oz A.D. Weinberg W. Ting Appropriate timing of surgical intervention after transmural acute myocardial infarction J Thorac Cardiovasc Surg 125 1 2003 115 119 [13] E.L. Hannan R. Michael D.R. Holmes S.B. King III W. Gary J.A. Ambrose S. Samin K. Stanley L.T. Clark R.H. Jones Impact of completeness of percutaneous coronary intervention revascularization on long-term outcomes in the stent era Circulation 113 2006 2406 2412 [14] M.N. Babapulle L. Joseph P. Belisle J.M. Brophy M.J. Eisenberg A hierarchical Bayesian meta-analysis of randomised clinical trials of drug-eluting stents Lancet 364 9434 2004 583 591 [15] A. Kastrati J. Mehilli J. Pache C. Kaiser M. Valgimigli H. Kelbaek M. Menichelli M. Sabate M.J. Suttorp D. Baumgart M. Seyfarth M.E. Pfisterer A. Schomig Analysis of 14 trials comparing sirolimus-eluting stents with bare-metal stents N Engl J Med 356 10 2007 1030 1039 [16] I. Iakovou T. Schmidt E. Bonizzoni L. Ge G.M. Sangiorgi G. Stankovic F. Airoldi A. Chieffo M. Montorfano M. Carlino I. Michev N. Corvaja C. Briguori U. Gerckens E. Grube A. Colombo Incidence, predictors, and outcome of thrombosis after successful implantation of drug-eluting stents JAMA 293 17 2005 2126 2130