Percutaneous coronary intervention with drug-eluting stents (DES) may achieve midterm outcomes comparable to coronary artery bypass grafting (CABG) for unprotected left main coronary artery disease, but few real-world, long-term studies have been reported. In this study, 376 patients with unprotected left main coronary artery disease who underwent DES implantation (n = 131) or CABG (n = 245) were evaluated, and outcomes were compared using propensity analyses to adjust for baseline differences. Overall, 367 patients (98%) had complete clinical follow-up for a median of 4.0 years (interquartile range 3.2 to 4.7). Although the overall sample size was limited, there was a trend toward lower mortality with DES versus CABG in unadjusted (hazard ratio [HR] 0.50, 95% confidence interval [CI] 0.20 to 1.22, p = 0.13), multivariate-adjusted (HR 0.37, 95% CI 0.13 to 1.09, p = 0.07), and propensity score–adjusted (HR 0.34, 95% CI 0.12 to 1.03, p = 0.06) analyses. Treatment with DES was associated with a higher rate of target-vessel revascularization (TVR; 18% vs 9%, p = 0.02). However, ischemic TVR was not significantly different between the 2 groups (25% vs 39%, p = 0.15) in patients who received angiographic follow-up. No differences were detected in the occurrence of composite major adverse cardiac and cerebrovascular events between DES and CABG (27% vs 22%, p = 0.42). In conclusion, during 4-year follow-up, overall composite major adverse cardiac and cerebrovascular events were similar after DES and CABG treatment of unprotected left main coronary artery disease, with a trend toward lower mortality after percutaneous coronary intervention with DES. DES were associated with a higher rate of TVR compared to CABG, but ischemic TVR was not significantly different between the 2 groups.
Recently, the Intracoronary Stenting and Angiographic Results: Drug-Eluting Stents for Unprotected Coronary Left Main Lesions (ISAR-LEFT-MAIN) randomized trial provided compelling evidence supporting the value of drug-eluting stents (DES) in patients with unprotected left main coronary artery (ULMCA) disease. In addition, some studies have suggested similar results or even reductions in cardiac events with percutaneous coronary intervention (PCI) compared with coronary artery bypass grafting (CABG) in patients with ULMCA disease, mainly because of a lower rate of cardiac death and myocardial infarction during the peri-interventional period. Favorable recent studies of LMCA DES have been performed in single-center registries, propensity-matched cohorts compared to CABG, meta-analyses of multicenter registries, and randomized trials. However, in-stent restenosis and the risk for late thrombosis have limited the widespread application of DES, especially because of the potential risk for sudden death associated with ULMCA restenosis. The aim of the present study was to evaluate the long-term results of DES versus CABG in patients with ULMCA stenoses.
Methods
From February 2003 to December 2006, 438 consecutive patients with de novo ≥50% ULMCA stenoses who underwent either CABG or PCI with DES were enrolled in this single-center study. We excluded 36 patients who presented with ST-segment elevation myocardial infarctions or with cardiogenic shock, 12 patients who had undergone previous CABG, 8 patients who underwent concomitant valvular or aortic surgery, and 6 patients with severe, life-limiting co-morbid medical illness. The remaining 376 patients constituted the present study population: 131 patients underwent PCI with DES, and 245 underwent CABG. The local ethics committee approved the use of clinical data for this study, and all patients provided written informed consent.
The decision to perform CABG or PCI was dependent on either physician or patient preference. Many patients treated with PCI were considered at high risk for CABG. Patient risk for CABG was evaluated for operative mortality with the European System for Cardiac Operative Risk Evaluation (EuroSCORE) and for in-hospital mortality with the Parsonnet score. Subjects with EuroSCOREs >6 or Parsonnet scores >15 were defined as at high risk.
All interventions were performed according to current practice guidelines. DES implantation in ULMCA stenosis was performed with full coverage of the diseased segment. Lesions at the ostium or shaft without involvement of the bifurcation were treated with single stents. Bifurcation lesions were treated using 1 of 4 stenting strategies at the operator’s discretion: crossover, T stent, culotte stent, or the crush technique. Final kissing balloon dilation was performed in most cases and was mandatory in cases with suboptimal results at the branch ostium. The choice of particular DES and antithrombotic agent was made by the operator. Overall, 126 patients received sirolimus-eluting stents, including Cypher (Cordis Corporation, Miami Lakes, Florida), Firebird (Microport Co. Ltd., Shanghai, China), Partner (Lepu Medical Technology Co. Ltd., Beijing, China), and Excel (JW Medical Co. Ltd., Shandong, China) stents, and 5 received zotarolimus-eluting stents (Endeavor; Medtronic Inc., Minneapolis, Minnesota). Cypher stents were available in sizes of 2.5 to 3.5 mm and in lengths of 8 to 33 mm; Firebird, Partner, and Excel stents were available in sizes of 2.5 to 4.0 mm and in lengths of 13 to 33, 12 to 36, and 14 to 28 mm, respectively; endeavor stents were available in sizes of 2.25 to 4.0 mm and in lengths of 14–30 mm.
All patients who underwent PCI were prescribed aspirin plus clopidogrel (loading dose, 300 or 600 mg) or ticlopidine (loading dose, 500 mg) before or during the intervention. After discharge, aspirin was continued indefinitely (100 mg/day); clopidogrel (75 mg/day) or ticlopidine (250 mg/day) was continued for ≥12 months. Since 2004, when available, adjunctive cilostazol (200 mg/day) was administrated for 6 months to patients with maximal platelet aggregation >50% with 5 μmol/L adenosine diphosphate tested 2 days after the procedure, but before discharge.
Standard bypass techniques included a left internal mammary artery for revascularization of the left anterior descending coronary artery whenever possible. For those patients taking aspirin and clopidogrel, surgery was delayed for 5 days after discontinuing clopidogrel.
Clinical follow-up after PCI and after CABG was performed at 1 month, 6 months, 1 year, and then annually thereafter. Angiographic follow-up was scheduled in the DES group from 8 to 12 months after revascularization or whenever clinically indicated. For patients who underwent CABG, angiographic follow-up was recommended if there were ischemic symptoms or signs during follow-up. Clinical, angiographic, procedural, and outcomes data were collected. All outcomes of interest were confirmed by source documentation and were adjudicated by the local events committee at Capital Medical University, Anzhen Hospital, Beijing, China.
Baseline, postprocedural, and follow-up coronary angiograms were digitally recorded and assessed off-line with an automated edge detection system (Cardiovascular Analysis Software, Beijing Crealife Technology Co. Ltd., Beijing, China) by 2 independent experienced physicians unaware of treatment allocation. Quantitative analysis was performed on the left main coronary artery (LMCA) area. The LMCA stenoses were classified as ostial (stenosis located within 3 mm of the LMCA ostium), midshaft (in the medial part of the LMCA, having ≥3 mm of apparently nondiseased artery before bifurcation), and distal (involving the distal part of the LMCA and bifurcation or trifurcation with the proximal left anterior descending coronary artery, left circumflex coronary artery, or ramus intermedius).
Assessed outcomes included death; target-vessel revascularization (TVR); the composite of death, nonfatal myocardial infarction, or stroke; and major adverse cardiac and cerebrovascular events (MACCEs). ULMCA disease was defined as LMCA stenosis ≥50% and no bypass grafts to the left anterior descending or left circumflex coronary artery. Revascularization was considered complete when all vessels >1.5 mm in diameter with diameter stenoses ≥50% were treated. Death was defined as postprocedural death from any cause. Periprocedural myocardial infarction (<7 days after intervention) was defined as elevated serum creatine kinase-MB isoenzyme ≥5 times the upper limit of normal after CABG and ≥3 times the upper limit of normal after PCI. Myocardial infarction after the periprocedural period was defined as any typical increase and decrease of biochemical markers of myocardial necrosis with ≥1 of the following: cardiac symptoms, development of Q waves on electrocardiography, or electrocardiographic changes indicative of ischemia. Cardiac enzymes were not measured routinely unless there was clinical suspicion of myocardial ischemia. Stroke, as indicated by neurologic deficits, was confirmed by a neurologist on the basis of imaging studies. TVR was defined as repeat revascularization of the treated coronary artery, including the corresponding segments of the left anterior descending and left circumflex coronary arteries. MACCEs were defined as the occurrence of death, nonfatal myocardial infarction, TVR, or stroke. In PCI patients, stent thrombosis was classified as definite, probable, or possible according to the Academic Research Consortium criteria. All events were based on clinical diagnoses assigned by the patients’ physicians and were centrally adjudicated by an independent group of clinicians. In the cumulative analysis of end points, events were counted only once, whichever occurred first.
Statistical analysis
Continuous variables were compared using Student’s t test or the Mann-Whitney rank-sum test, as appropriate, and are presented as mean ± SD or as median (interquartile range). Categorical variables were compared using chi-square statistics or Fisher’s exact test, as appropriate. A p value <0.05 was considered statistically significant. To reduce the effect of treatment selection bias and potential confounding in this observational study, a propensity score analysis was performed to adjust for differences in baseline characteristics of patients. Briefly, for each patient, a propensity score, indicating the likelihood of undergoing PCI, was calculated by use of a multivariate logistic regression that identified variables independently associated with PCI. Exact Cox regression models were used to perform unadjusted univariate analyses and adjusted multivariate analyses with or without propensity scores in all patients. Results are reported as hazard ratios (HRs) together with associated exact 95% confidence intervals (CIs). Statistical analyses were performed using SPSS version 15.0 (SPSS, Inc., Chicago, Illinois).
Results
Overall, 331 patients (88%) were eligible for either PCI or CABG. Of 131 patients who underwent PCI, 117 (89%) were eligible for either PCI or CABG; the remaining 14 patients had underlying conditions that made them poor candidates for surgery (5 were ≥80 years of age and had poor performance status, 3 had no suitable bypass conduits, 5 had poor distal runoff, and 1 had a porcelain aorta). Of 245 patients who underwent CABG, 214 (87%) were eligible for either PCI or CABG; the remaining 31 patients had underlying conditions that were unsuitable for PCI (25 had ≥2 total occlusions of other major epicardial coronary arteries, 4 had severe calcification of the LMCA, and 2 had contraindication to antiplatelet therapy).
As listed in Table 1 , 80% of patients were men, 28% had diabetes mellitus, and acute coronary syndromes were present in nearly 70% of overall cohort. Renal failure was more often presented in the DES group than in the CABG group (6.9% vs 2.3%, p = 0.06). More than 20% of patients in the 2 groups were considered at high surgical risk on the basis of EuroSCOREs ≥6 and Parsonnet scores ≥15. During follow-up, patients in the CABG group received fewer adjunctive medications. The DES group compared to the CABG group were more frequently treated with antiplatelet medications (92% vs 26% for thienopyridine, p <0.001, and 24% vs 0.4% for cilostazol, p <0.001). The length of time between admission and procedure, the duration of the postprocedural hospital stay, and hospitalization (16.5 ± 7.5 vs 7.5 ± 6.3 days, p <0.001) were significantly longer with CABG than with DES.
| Variable | CABG | DES | p Value |
|---|---|---|---|
| (n = 245) | (n = 131) | ||
| Men | 203 (83%) | 99 (76%) | 0.09 |
| Age (years) | 63.6 ± 9.1 | 61.9 ± 10.8 | 0.13 |
| Diabetes mellitus | 71 (29%) | 35 (27%) | 0.64 |
| Hypertension ⁎ | 153 (62%) | 85 (65%) | 0.64 |
| Hypercholesterolemia † | 75 (31%) | 42 (32%) | 0.77 |
| Previous myocardial infarction | 37 (15%) | 16 (12%) | 0.44 |
| Previous stroke | 27 (11%) | 10 (8%) | 0.29 |
| Peripheral vascular disease | 3 (1.2%) | 4 (3.1%) | 0.21 |
| Current smokers | 96 (39%) | 51 (39%) | 0.96 |
| Renal failure | 17 (6.9%) | 3 (2.3%) | 0.06 |
| LVEF (%) | 59.2 ± 12.0 | 60.0 ± 11.8 | 0.50 |
| LVEF ≤30% | 2 (0.8%) | 2 (1.5%) | 0.61 |
| EuroSCORE | 4.3 ± 2.4 | 4.2 ± 2.7 | 0.63 |
| EuroSCORE >6 | 69 (28%) | 40 (31%) | 0.74 |
| Parsonnet score | 8.8 ± 6.4 | 8.9 ± 6.4 | 0.91 |
| Parsonnet score >15 | 52 (21%) | 32 (24%) | 0.48 |
| Clinical presentation | 0.58 | ||
| Stable angina pectoris | 76 (31%) | 42 (32%) | |
| Unstable angina pectoris | 158 (64%) | 80 (61%) | |
| NSTEMI | 11 (5%) | 9 (7%) | |
| Time from admission to procedure (days) | 6.9 ± 5.5 | 3.5 ± 3.8 | <0.001 |
| Postprocedural hospital stay (days) | 9.6 ± 4.8 | 4.0 ± 4.8 | <0.001 |
| Hospitalization (days) | 16.5 ± 7.5 | 7.5 ± 6.3 | <0.001 |
| Medical therapy at 1 year | |||
| Aspirin | 205 (84%) | 122 (93%) | 0.009 |
| Thienopyridine | 63 (26%) | 120 (92%) | <0.001 |
| β blocker | 194 (79%) | 109 (83%) | 0.35 |
| Cilostazol ‡ | 1 (0.4%) | 32 (24%) | <0.001 |
| Statin | 185 (75%) | 113 (86%) | 0.01 |
| ACE inhibitor or angiotensin II receptor antagonist | 139 (57%) | 90 (69%) | 0.02 |
⁎ Systolic blood pressure ≥140 mm Hg, diastolic blood pressure ≥90 mm Hg, or use of an antihypertensive drug.
† Total cholesterol ≥200 mg/dl or medication treated.
The main angiographic and procedural characteristics of the CABG and PCI procedures are listed in Table 2 . Overall, 69% had bifurcation or trifurcation lesions, and total occlusions of major coronary arteries (other than the LMCA) were identified in 31%. More patients treated with CABG had additional disease in the right coronary artery (82% vs 44%, p <0.001) and concomitant 3-vessel disease (68% vs 35%, p <0.001). In the DES group, 83 patients (63%) received PCI treatment in vessels other than the LMCA (49 in the right coronary artery, 46 in the left anterior descending coronary artery, and 36 in the left circumflex coronary artery). The number of stented vessels was 2.0 ± 0.9, and the number of overall stents was 2.5 ± 1.3, while the number of stents implanted in the LMCA was 1.4 ± 0.5, the length of the stents in the LMCA was 24.0 ± 11.6 mm, and stent diameter was 3.4 ± 0.4 mm. In the CABG group, 189 patients (78%) underwent on-pump surgery, and 191 (78%) received ≥1 arterial conduit. The mean number of grafts used was 2.9 ± 0.8 (0.9 ± 0.6 arterial grafts and 2.0 ± 0.9 venous grafts).
| Variable | CABG | DES | p Value |
|---|---|---|---|
| (n = 245) | (n = 131) | ||
| Involved location | 0.71 | ||
| Ostium, midshaft, or both | 74 (30%) | 42 (32%) | |
| Distal bifurcation/trifurcation | 171 (70%) | 89 (68%) | |
| Number of coronary artery narrowed | <0.001 | ||
| LMCA only | 7 (3%) | 11 (9%) | |
| LMCA plus single-vessel disease | 9 (4%) | 24 (18%) | |
| LMCA plus double-vessel disease | 63 (26%) | 50 (38%) | |
| LMCA plus triple-vessel disease | 166 (67%) | 46 (35%) | |
| RCA dominant | 179 (73%) | 99 (76%) | 0.60 |
| RCA disease | 201 (82%) | 58 (44%) | <0.001 |
| Chronic total occlusion | 97 (40%) | 20 (15%) | <0.001 |
| Complete revascularization | 224 (91%) | 103 (79%) | <0.001 |
| Coronary artery bypass grafting | |||
| On pump | 189 (78%) | ||
| Grafts per patient | 2.9 ± 0.8 | ||
| ≥1 IMA used | 191 (78%) | ||
| Drug-eluting stents | |||
| Stent type | |||
| Cypher | 87 (66%) | ||
| Endeavor | 5 (4%) | ||
| Chinese sirolimus-eluting stent | 39 (30%) | ||
| Final kissing balloon | 57 (64%) | ||
| PCI technique for bifurcation disease | |||
| Crossing over LCX or LAD | 41 (46%) | ||
| T stenting | 27 (30%) | ||
| Cullote stenting | 12 (14%) | ||
| Crush technique | 9 (10%) |
Three hundred sixty-seven patients (98%) received complete clinical follow-up for a median of 4.0 (interquartile range 3.2 to 4.7) (3.8 years [interquartile range 3.1 to 4.4] in the DES group vs 4.2 years [interquartile range 3.6 to 4.8] years in the CABG group). The MACCEs during follow-up are listed in Table 3 . Overall, during the median 4-year follow-up period, 29 patients (7.7%) died; among them, 18 (62%, or 5% of the overall cohort) died from cardiovascular causes. Twenty patients (5.3%) had myocardial infarctions, 12 (3.3%) had strokes, TVR was performed in 47 (13%), and 6 patients (1.6%) had graft occlusions (n = 4 [1.6%]) or stent thrombosis (n = 2 [1.5%]).
| Event | DES (n = 131) | CABG (n = 245) | HR (95% CI) Unadjusted | Multivariate Adjusted | Propensity Score Adjusted | p Value ⁎ |
|---|---|---|---|---|---|---|
| Death | ||||||
| 30 days | 1 (0.8%) | 8 (3.3%) | 0.23 (0.03–1.86) | 0.11 (0.01–1.33) | 0.06 (0.01–1.04) | 0.05 |
| 1 year | 2 (1.5%) | 14 (5.7%) | 0.26 (0.06–1.16) | 0.14 (0.02–0.78) | 0.11 (0.02–0.72) | 0.02 |
| 2 years | 5 (3.8%) | 17 (6.9%) | 0.54 (0.20–1.46) | 0.18 (0.05–0.72) | 0.17 (0.04–0.71) | 0.02 |
| 3 years | 5 (3.8%) | 20 (8.2%) | 0.46 (0.17–1.21) | 0.22 (0.06–0.80) | 0.22 (0.06–0.81) | 0.02 |
| Overall | 6 (4.6%) | 23 (9.4%) | 0.50 (0.20–1.22) | 0.37 (0.13–1.09) | 0.34 (0.12–1.03) | 0.06 |
| Target-vessel revascularization | ||||||
| 30 days | 0 | 0 | — | — | — | — |
| 1 year | 15 (12%) | 8 (3.4%) | 3.49 (1.48–8.24) | 4.96 (1.86–13.2) | 5.03 (1.89–13.4) | 0.001 |
| 2 years | 19 (15%) | 16 (6.8%) | 2.26 (1.16–4.39) | 2.66 (1.21–5.81) | 2.65 (1.21–5.81) | 0.02 |
| 3 years | 23 (18%) | 18 (7.6%) | 2.44 (1.32–4.52) | 2.68 (1.30–5.50) | 2.69 (1.30–5.57) | 0.008 |
| Overall | 24 (18%) | 23 (9.4%) | 2.14 (1.21–3.80) | 2.27 (1.15–4.51) | 2.27 (1.14–4.51) | 0.02 |
| Death + myocardial infarction + stroke | ||||||
| 30 days | 4 (3.1%) | 14 (5.7%) | 0.53 (0.18–1.62) | 0.58 (0.17–1.98) | 0.43 (0.12–1.57) | 0.20 |
| 1 year | 7 (5.3%) | 24 (9.8%) | 0.54 (0.23–1.25) | 0.58 (0.22–1.48) | 0.45 (0.17–1.19) | 0.11 |
| 2 years | 11 (8.4%) | 31 (13%) | 0.65 (0.33–1.29) | 0.59 (0.27–1.30) | 0.53 (0.24–1.18) | 0.12 |
| 3 years | 11 (8.4%) | 34 (14%) | 0.59 (0.30–1.17) | 0.55 (0.25–1.21) | 0.51 (0.23–1.12) | 0.09 |
| Overall | 11 (8.4%) | 36 (15%) | 0.56 (0.19–1.11) | 0.52 (0.24–1.13) | 0.48 (0.22–1.06) | 0.07 |
| MACCEs | ||||||
| 30 days | 4 (3.1%) | 14 (5.9%) | 0.52 (0.17–1.58) | 0.59 (0.17–2.01) | 0.44 (0.12–1.58) | 0.21 |
| 1 year | 22 (17%) | 31 (13%) | 1.30 (0.76–2.25) | 1.62 (0.86–3.05) | 1.50 (0.79–2.85 | 0.21 |
| 2 years | 30 (23%) | 44 (19%) | 1.27 (0.80–2.02) | 1.39 (0.81–2.39) | 1.31 (0.76–2.27) | 0.33 |
| 3 years | 34 (26%) | 48 (20%) | 1.33 (0.85–2.06) | 1.42 (0.85–2.37) | 1.36 (0.81–2.28) | 0.25 |
| Overall | 35 (27%) | 55 (22%) | 1.26 (0.83–1.93) | 1.29 (0.78–2.11) | 1.23 (0.75–2.04) | 0.42 |
Stay updated, free articles. Join our Telegram channel
Full access? Get Clinical Tree
