Unprotected Left Main Coronary Artery Disease in Patients With Low Predictive Risk of Mortality

Unprotected Left Main Coronary Artery Disease in Patients With Low Predictive Risk of Mortality Shun Watanabe, MD, Tatsuhiko Komiya, MD, Genichi Sakaguchi, MD, PhD, and Takeshi Shimamoto, MD, PhD Department of Cardiovascular Surgery, Kurashiki Central Hospital, Okayama, Japan Background. There has been an increasing use of percutaneous coronary intervention with drug-eluting stent (DES) in patients with unprotected left main coronary artery disease. We assessed whether coronary artery bypass grafting (CABG) would be superior to percutaneous coronary intervention with first-generation DES in patients with unprotected left main coronary artery disease. Methods. Between January 2003 and December 2007, a total of 295 patients with unprotected left main coronary artery disease were treated with Kurashiki Central Hospital, Okayama, Japan. Among these patients, 169 and 126 underwent CABG and percutaneous coronary intervention with DES, respectively. The average Society of Thoracic Surgeons predictive risk of mortality score was 2.3 3.8%. We defined higher-risk and lower-risk patients as those with Society of Thoracic Surgeons predictive risk of mortality scores higher than 2.3% (CABG, n 48; percutaneous coronary intervention, n 28) and lower than 2.3% (CABG, n 121; percutaneous coronary intervention, n 98), respectively. Survival, major adverse cardiac-related events, and target lesion revascularization were analyzed by the Kaplan-Meier method. Results. Between-group differences in favor of CABG were seen with respect to the rate of major adverse cardiac-related events and target lesion revascularization in both strata. In patients with lower scores according to the Society of Thoracic Surgeons predictive risk of mortality, the cardiac death-free survival rate was significantly higher in CABG. Conclusions. In this 5-year single-center experience, CABG is recommended for unprotected left main coronary artery disease, especially in patients with low scores in the Society of Thoracic Surgeons predictive risk of mortality. (Ann Thorac Surg 2012;94:1927 33) 2012 by The Society of Thoracic Surgeons Coronary artery bypass grafting (CABG) is considered the optimum revascularization treatment for patients with de novo left main disease, three-vessel disease, or both [1]. Furthermore, of those undergoing coronary angiography, 4% are found to have left main coronary artery disease [2]. Published cohort studies have found that mortality rates are similar at 1, 2, and 5 years of follow-up; however, the risk of needing targetvessel revascularization is significantly higher with stenting than with CABG. The new American College of Cardiology Foundation/American Heart Association guidelines for CABG were released in 2011 and recommend CABG for unprotected left main coronary artery [ULMCA] disease in class I patients regardless of their risk and anatomy. Meanwhile, percutaneous coronary intervention (PCI) for ULMCA is recommended in class IIa patients if the SYNTAX (Synergy Between Percutaneous Coronary Intervention with TAXUS and Cardiac Surgery) score is under 22 and the Society of Thoracic Surgeons predictive risk of mortality score (STS PROM) Accepted for publication June 26, 2012. Presented at the Forty-eighth Annual Meeting of The Society of Thoracic Surgeons, Fort Lauderdale, FL, Jan 28 Feb 1, 2012. Address correspondence to Dr Komiya, Department of Cardiovascular Surgery, Kurashiki Central Hospital, 1-1-1 Miwa, Kurashiki, Okayama 710-8602, Japan; e-mail: komiya@kchnet.or.jp. is over 5%, and in class IIb patients if the SYNTAX is under 33 and the STS PROM is over 2%. The guidelines recommend estimating the operative risk for all CABG candidates by using a standard instrument such as STS PROM [3]. The usability of the STS PROM score as a long-term predictor was revealed in a recent report [4]. The purpose of this study was to compare the long-term outcomes between CABG and PCI with drug-eluting stents (DES) for ULMCA disease among patients stratified according to the STS PROM scores. Patients and Methods The Institutional Review Board of Kurashiki Central Hospital approved this study and waived individual consent because this study was retrospective. A total of 295 consecutive patients with ULMCA disease were treated in our institution between January 2003 and December 2007. Among these patients, 169 underwent CABG (the CABG group), and 126 underwent PCI with DES. The choice of treatment was made by the cardiologists. The decision for patients to undergo PCI instead of CABG was based either on the patient s or the cardiologist s preference or on the high risk associated with CABG. All patients were divided into two strata according to STS PROM. The mean and standard deviation of 2012 by The Society of Thoracic Surgeons 0003-4975/$36.00 Published by Elsevier Inc http://dx.doi.org/10.1016/j.athoracsur.2012.06.060

1928 WATANABE ET AL Ann Thorac Surg TREATMENT FOR ULMCA IN LOWER STS PROM 2012;94:1927 33 the STS PROM was 2.3 3.8% (the 75 value was also 2.3%). According to the STS database, the average score of CABG patients was 2.24% [4], which was almost the same as our average. Therefore, we defined the cutoff value of STS PROM as 2.3%. Patients were divided into two strata: higher-risk patients with a mortality rate higher than 2.3% (CABG, n 48; PCI, n 28) and lower-risk patients with a mortality rate lower than 2.3% (CABG, n 121; PCI, n 98). Patients who had prior PCI or CABG, a concomitant surgical procedure, or acute myocardial infarction were excluded. Propensity-Matched Analysis We performed propensity matching for the patients and analyzed limited patients outcomes as a subgroup analysis. The factors used for propensity matching were age, sex, hypertension, hyperlipidemia, SYNTAX score, diabetes mellitus, hemodialysis, ejection fraction, estimated glomerular filtration rate (egfr), cerebrovascular disease, and peripheral artery disease. After propensity matching, the number of CABG and PCI patients was 82 in each group. Percutaneous Revascularization Patients in the DES group were implanted with either a sirolimus-eluting stent (Cypher; Cordis, Miami Lakes, FL), a paclitaxel-eluting stent (TAXUS; Boston Scientific Corp, Natick, MA), or a biolimus-eluting stent (Bio Matrix; Biosensors Interventional Technologies Pte, Ltd.; Singapore). Aspirin in combination with clopidogrel sulfate or ticlopidine was administered throughout the study period to patients if they experienced no side effects from antiplatelet medications. Angiographic follow-up was scheduled at 3, 8, and 20 months after PCI. Surgical Revascularization CABG was performed through a median sternotomy in all cases. The left internal mammary artery was used to bypass the left anterior descending artery in all possible cases. For patients younger than 60 years of age, we attempted to use bilateral internal thoracic artery and radial artery grafts. For those under 70 years of age, we tried to use no more than two arterial grafts. The off-pump technique was used in elective cases and in patients with good cardiac function. In emergent cases, patients with low ejection fraction, or both, cardiopulmonary bypass was used. Postoperative antiplatelet medication involved only aspirin in patients receiving coumadin or both aspirin and ticlopidine for 3 months in patients not receiving warfarin, after which ticlopidine was discontinued. Study Endpoints The primary clinical endpoint was freedom from major adverse cardiac-related events (MACE) and death. We defined MACE as death, myocardial infarction, and target lesion revascularization (TLR). The TLR was defined as any revascularization performed on the treated lesion. Statistical Analyses Comparisons of categoric variables were performed by use of the 2 test or Fisher s exact test. Continuous variables were analyzed by Student s t test. Survival and event-free survival were analyzed by the Kaplan-Meier method; the log-rank test was used for comparisons between groups. Risk of mortality was analyzed by Cox proportional hazard analysis. All analyses were conducted with SAS software, version 5.0 (SAS Institute, Inc, Cary, NC). Follow-Up Follow-up was obtained by means of a direct telephone questionnaire or at the outpatient clinic. The median duration of follow-up was 56.4 21.7 months; the follow-up rate was 98.6%. Results Patient Characteristics The patients baseline clinical characteristics are summarized in Table 1. Few differences were observed between the two groups. Cerebrovascular disease was more prevalent in the CABG group. The mean STS PROM score was higher in the CABG group, but the difference was Table 1. Baseline Characteristics of Patients Undergoing CABG and PCI With DES Characteristic CABG (n 169) DES (n 126) p Value Age, years 68.5 10.4 69.6 10.4 0.3563 Male sex, 123 (73.7) 95 (75.4) 0.7348 Hypertension, 120 (71.9) 96 (76.2) 0.4040 Hypercholesterolemia, 88 (52.7) 49 (38.9) 0.0190 Diabetes mellitus, 61 (36.5) 44 (34.9) 0.7765 egfr 59.4 26.8 59.0 23.5 0.8779 Hemodialysis, 11 (6.6) 9 (7.1) 0.8518 Cerebrovascular 37 (22.2) 14 (11.1) 0.0136 disease, Peripheral artery 19 (11.4) 11 (8.7) 0.4593 disease, Ejection fraction, n 52.8 16.0 57.6 9.9 0.0023 (%) Left main only, 8 (4.7) 18 (14.3) 0.0079 Left main 1 vessel 14 (28.6) 35 (71.4) 0.0001 disease, 2 vessel disease, 47 (27.8) 49 (38.9) 0.0446 3 vessel disease, 100 (59.2) 24 (19.0) 0.0001 EuroSCORE 5.4 3.4 4.4 2.6 0.0055 STS PROM 2.6 4.6 1.9 2.4 0.1383 SYNTAX score 26.3 10.4 26.7 9.4 0.7348 CABG coronary artery bypass grafting; DES drug eluting stent; egfr estimated glomerular filtration rate; EuroSCORE European system for cardiac operative risk evaluation; PCI percutaneous coronary intervention; STS PROM The Society of Thoracic Surgeons predictive risk of mortality; SYNTAX synergy between percutaneous coronary intervention with TAXUS and cardiac surgery.

Ann Thorac Surg WATANABE ET AL 2012;94:1927 33 TREATMENT FOR ULMCA IN LOWER STS PROM 1929 not statistically significant. However, the distribution of the stenosis sites was significantly different between the two groups. The number of patients who had stenosis only on the left main coronary artery was higher in the DES group. Meanwhile, more patients with multivessel disease were in the CABG group. However, no significant difference was observed in the SYNTAX scores. CABG and PCI Procedures In the CABG group, the off-pump technique was used in 70% of the patients; the on-pump beating technique, in 12%; and the on-pump arrested heart technique, in 17%. A left or right internal mammary artery graft was used for revascularization of the left anterior descending artery in all cases. All arterial grafting was performed in 49% of the cases. In the DES group, 31 patients had stenosis in an ostial or body lesion of the left main trunk, and 95 patients had stenosis on a distal lesion of the left main trunk. All ostial and body stenoses were treated with lone stent. For distal lesions, a two-stent technique was used in 42% of the cases. Kaplan-Meier Analysis The 5-year survival, the cardiac death-free survival, the MACE-free survival, and the TLR-free survival rates were 82.6%, 93.8%, 80.8%, and 91.7%, respectively, in the CABG group and 77.2%, 91.9%, 54.5%, and 70.5%, respectively, in the DES group (Fig 1). No significant difference in mortality was observed. However, the MACE and TLR rates were significantly higher in the DES group (p 0.0001). Stratification of Patients According to STS PROM The 5-year survival and the cardiac death-free survival rates in the patients with the lower STS PROM scores were 92.4% and 97.9%, respectively, in the CABG group and 85.3% and 94.9%, respectively, in the DES group. There was no significant difference in survival (p 0.065) (Fig 2A, B). In terms of cardiac death-free survival, CABG was favorable for patients with a lower STS PROM score (p 0.025) (Fig 2C, D). The MACE and TLR rates were significantly higher in the DES group, irrespective of the patient s risk (Fig 3). Fig 1. (A) Freedom from all-cause death in the entire cohort. (B) Freedom from cardiac death in the entire cohort. (C) Freedom from MACE in the entire cohort. (D) Freedom from target lesion revascularization in the entire cohort. (CABG coronary artery bypass grafting; DES drug-eluting stent; MACE major adverse cardiac-related events; TLR target lesion revascularization.)

1930 WATANABE ET AL Ann Thorac Surg TREATMENT FOR ULMCA IN LOWER STS PROM 2012;94:1927 33 Fig 2. (A) Freedom from all-cause death in patients with lower STS PROM scores. (B) Freedom from all-cause death in patients with higher STS PROM scores. (C) Freedom from cardiac death in patients with lower STS PROM scores. (D) Freedom from cardiac death in patients with higher STS PROM scores. (CABG coronary artery bypass grafting; DES drug-eluting stent; STS PROM The Society of Thoracic Surgeons predictive risk of mortality.) Propensity-Matched Patients There were no significant differences in survival and cardiac death-free survival (p 0.304 and p 0.655, respectively) (Table 2, Fig 4A, B). The MACE and TLR rates were significantly higher in the DES group, irrespective of the patient s risk (Fig 4C, D). Cox Proportional Hazard Analysis The result of the Cox proportional hazard analysis is shown in Figure 5. The superiority of CABG was evident in the patients with a lower STS PROM score. Comment Although CABG has been considered the gold standard for unprotected left main cardiac disease revascularization, more recently PCI has emerged as a possible alternative mode of revascularization in carefully selected patients [3]. Several studies have suggested that coronary stenting is feasible for patients with ULMCA stenosis [5]. Moreover, the results of recent randomized control trials suggest that major clinical outcomes in selected patients with ULMCA disease are similar between CABG and PCI at 1- to 2-year follow-up but that repeat revascularization rates are higher after PCI than after CABG [1, 3, 6, 7]. However, randomized control trials with an extended follow-up of 5 years are required to provide definitive conclusions about the optimal treatment of ULMCA, but none have yet been conducted [3]. In this retrospective study involving patients with ULMCA stenosis, no significant differences in long-term all-cause and cardiac mortality were observed between the CABG and DES groups. The MACE and TLR rates were higher in the DES group. However, long-term cardiac mortality was significantly higher in the DES patients with a lower STS PROM score. Survival analysis of the entire cohort showed no survival benefit for CABG as a treatment of ULMCA disease, similar to previous randomized control trials [1, 6, 7]. The all-cause mortality rate for 5 years was 17.4%, comparable with the value of a previous report [4]. Moreover, the MACE and TLR rates

Ann Thorac Surg WATANABE ET AL 2012;94:1927 33 TREATMENT FOR ULMCA IN LOWER STS PROM 1931 Fig 3. (A) Freedom from MACE in patients with lower STS PROM scores. (B) Freedom from MACE in patients with higher STS PROM scores. (C) Freedom from target lesion revascularization in patients with lower STS PROM scores. (D) Freedom from target lesion revascularization in patients with higher STS PROM scores. (CABG coronary artery bypass grafting; DES drug-eluting stent; MACE major adverse cardiac-related events; TLR target lesion revascularization; STS PROM The Society of Thoracic Surgeons predictive risk of mortality.) were significantly higher in the DES group, which are also similar to the results of previous reports [1, 6 8]. To the best of our knowledge, this is first report to stratify patients by STS PROM scores to compare CABG and DES in the treatment of ULMCA. STS PROM is a very useful tool as a predictor not only of 30-day mortality but also of long-term mortality [4]. In this study, we defined the cutoff value at 2.3% because the average STS PROM score for our entire cohort was 2.3%, with the 75 value at 2.3%. The average of the STS CABG cohort was 2.24%, and the current guidelines recommend PCI for class IIb patients in whom the STS PROM is over 2% and the SYNTAX score is low to intermediate. Hence, our definition of the cutoff value was relatively feasible. The Kaplan-Meier analysis after stratification based on the STS PROM scores showed that CABG was better for cardiac survival in patients with lower STS PROM scores. In a recent review, low-risk patients with good left ventricular function, nondistal and non-calcified left main stenosis, ostial left main lesions, midshaft left main lesions, and very few additional lesions on the other coronary vessel were defined as favorable for stenting. However, in our study, the superiority of CABG was seen only in the lower-risk patients. The analysis of the higher-risk patients showed no showed significant differences, although this might have resulted from the small sample size. To date, limited data exist regarding the long-term outcomes of coronary stenting with DES. Furthermore, the long-term safety of DES has been questioned by recent reports that suggest an increased risk of late stent thrombosis, mortality, and myocardial infarction. [9, 10] The LMCA differs from the other coronary arteries by its relatively higher elastic tissue content, which can explain the elastic recoil and the high restenosis rate after balloon angioplasty [11]. Therefore, stenting in the ULMCA should be done only in carefully selected patients. Most of our patients underwent implantation with first-generation DESs. However, the new generation of DESs, known as dedicated bifurcation stents, and an improvement in procedural techniques will probably improve the clinical outcome [11]. However, the long-

1932 WATANABE ET AL Ann Thorac Surg TREATMENT FOR ULMCA IN LOWER STS PROM 2012;94:1927 33 Table 2. Baseline Characteristics of Propensity-Score Matched Patients Undergoing CABG and PCI With DES Characteristic CABG (n 82) DES (n 82) p Value Age, years 68.2 10.5 67.7 10.1 0.7440 Male sex, 66 (80.5) 66 (80.5) 1.0000 Hypertension, 63 (76.8) 58 (70.7) 0.4779 Hypercholesterolemia, 41 (50.0) 38 (46.3) 0.7547 Diabetes mellitus, 29 (35.4) 32 (39.0) 0.7468 egfr 60.0 22.7 61.1 22.8 0.7715 Hemodialysis, 4 (4.9) 4 (4.9) 1.0000 Cerebrovascular disease, n 14 (17.1) 12 (14.6) 0.8311 (%) Peripheral artery disease, n 6 (7.32) 9 (11.0) 0.5894 (%) Ejection fraction 56.4 14.8 55.7 10.5 0.7434 STS PROM 1.9 2.5 1.7 2.3 0.4767 Syntax score 26.4 10.7 26.2 9.4 0.9228 CABG coronary artery bypass grafting; DES drug eluting stent; egfr estimated glomerular filtration rate; PCI percutaneous coronary intervention; STS PROM The Society of Thoracic Surgeons predictive risk of mortality. term results of such devices in the treatment of ULMCA disease remain unknown. The long-term results from the EXCEL (Evaluation of Xience Prime vs Coronary Artery Bypass Surgery for Effectiveness of Left Main Revascularization) trials may provide further insights into the optimal treatment for patients with left main disease [1]. The present study had several limitations. First, this study was not a prospective, randomized study, and the sample size was small. Hence, the evaluation of a large number of patients will be necessary. A heterogenous cohort with a small number of subjects, three types of DESs, differences in baseline criteria (cardiovascular disease, ejection fraction, type of coronary heart disease, European system for cardiac operative risk evaluation), the use of different grafts, on/off pump technique, two types of antiplatelet therapy (given for only 3 months) are factors known to interfere with the chosen endpoint; hence, a larger dataset is needed to overcome this major methodologic issue in a retrospective study. Second, the study had patient selection bias. The median follow-up time was shorter in the DES group. This is because more patients received CABG in 2003 and Fig 4. (A) Freedom from all-cause death in propensity-matched patients. (B) Freedom from cardiac death in propensity-matched patients. (C) Freedom from MACE in propensity-matched patients. (D) Freedom from target lesion revascularization in propensity-matched patients. (CABG coronary artery bypass grafting; DES drug-eluting stent; MACE major adverse cardiac-related events; TLR target lesion revascularization.)

Ann Thorac Surg WATANABE ET AL 2012;94:1927 33 TREATMENT FOR ULMCA IN LOWER STS PROM 1933 CABG: 48 PCI: 28 Higher STS = 2.3% Lower O utcome Hazard Ra o (95% CI) P value for Interac on All cause death 081(04116) 0.81 (0.41-1.6) 055 0.55 C ardiac death 0.63 (0.19-2.01) 0.44 M ACE 0.53 (0.30-0.97) 0.04 TLR 024(0 0.24 (0.08-0.75) 001 0.01 All cause d eath 0.46 (0.20-1.06) 0.07 Cardiac death 0.16 (0.03-0.94) 0.04 Fig 5. Cox proportional hazard analysis. (CABG coronary artery bypass grafting; CI confidence interval; DES drug-eluting stent; MACE major adverse cardiac-related event; PCI percutaneous coronary intervention; STS The Society of Thoracic Surgeons; TLR target lesion revascularization.) CABG: 121 PCI: 98 M ACE 0.19 (0.10-0.36) < 0.0001 T LR 0.28 (0.13-0.59) 0.0008 0.25 0.50 1.00 2.00 4.00 CABG Be er DES Be er 2004. The number of patients who received CABG in 2003 and 2004 was 72. However, the number of patients who received DES was only 10. Because our hospital has produced excellent results in operating on patients in Japan, there appears to be some value in examining the results of such a single institution. Moreover, this study reflected the real-world results in Japan. Nonetheless, a large, 5-year randomized trial will be necessary to confirm the present results. In this 5-year single-center experience, CABG seemed to be a favorable treatment for ULMCA disease, especially in patients with lower STS PROM scores. Cardiologists should consider the patients surgical risk and selection of PCI carefully. CABG should be used to treat ULMCA in patients with lower STS PROM scores. References 1. Kappetein AP, Mohr FW, Feldman TE, et al. Comparison of coronary bypass surgery with drug-eluting stenting for the treatment of left main and/or three-vessel disease: 3-year follow-up of the SYNTAX trial. Eur Heart J 2011;17:2125 34. 2. Ragosta M, Dee S, Sarembock IJ, et al. Prevalence of unfavorable angiographic characteristics for percutaneous intervention in patients with unprotected left main coronary artery disease. Catheter Cardiovasc Interv 2006;68:357 62. 3. Hillis LD, Smith PK, Anderson JL, et al. ACCF/AHA guideline for coronary artery bypass graft surgery: a report of the American College of Cardiology Foundation/American Heart Association Task Force on practice guidelines. Circulation 2011;124:e652 735. 4. Puskas JD, Kilgo PD, Thourani VH, et al. The Society of Thoracic Surgeons 30-Day Predicted Risk of Mortality Score also predicts long-term survival. Ann Thorac Surg 2012;93: 26 35. 5. Park SJ, Park DW. Percutaneous coronary intervention with stent implantation versus coronary artery bypass surgery for treatment of left main coronary artery disease: is it time to change guidelines? Circ Cardiovasc Interv 2009;2:59 68. 6. Buszman PE, Kiesz SR, Bochenek A, et al. Acute and late outcomes of unprotected left main stenting in comparison with surgical revascularization. J Am Coll Cardiol 2008;51: 538 45. 7. Park SJ, Kim YH, Park DW, et al. Randomized trial of stents versus bypass surgery for left main coronary artery disease. N Engl J Med 2011;364:1718 27. 8. Lee MS, Jamal F, Kedia G, et al. Comparison of bypass surgery with drug-eluting stents for diabetic patients with multivessel disease. Int J Cardiol 2007;123:34 42. 9. Daemen J, Wenaweser P, Tsuchida K, et al. Early and late coronary stent thrombosis of sirolimus-eluting and paclitaxel-eluting stents in routine clinical practice: data from a large two-institutional cohort study. Lancet 2007;369:667 78. 10. Lagerqvist B, James SK, Stenestrand U, Lindback J, Nilsson T, Wallentin L. Long-term outcomes with drug-eluting stents versus bare-metal stents in Sweden. N Engl J Med 2007;356:1009 19. 11. Fajadet J, Chieffo A. Current management of left main coronary artery disease. Eur Heart J 2012;33:36 50.