The aim of this study was to compare clinical outcome at 5 years in patients with complete and incomplete revascularization treated with coronary artery bypass grafting (CABG) and percutaneous coronary intervention (PCI) with drug-eluting stents. Baseline and procedural angiograms and surgical case-record forms were centrally assessed for completeness of revascularization. Patients treated with PCI for incomplete revascularization were stratified according to Synergy between Percutaneous Coronary Intervention with Taxus and Cardiac Surgery (SYNTAX) score tertiles. Complete revascularization was achieved in 360 of 588 patients (61.2%) in the PCI with sirolimus-eluting stent group and 477 of 567 patients (84.1%) in the CABG group (p <0.05). There was no significant difference in 5-year survival without major adverse cardiac and cerebrovascular events (MACCEs; death, cerebrovascular accident, myocardial infarction, and any revascularization) between patients with complete and incomplete revascularization treated with PCI or CABG. Survival free from MACCEs in patients with incomplete revascularization treated with PCI was significantly lower than those with complete revascularization treated with CABG (hazard ratio 1.66, 0.96 to 1.80, log-rank p = 0.001). The 5-year MACCE-free survival in patients with incomplete revascularization treated with PCI stratified according to SYNTAX score tertiles showed a significantly lower MACCE survival in the higher SYNTAX tertile compared to the low (hazard ratio 0.56, 0.32 to 0.96, log-rank p = 0.04) and intermediate (hazard ratio 0.50, 0.28 to 0.91, log-rank p = 0.02) tertiles, whereas survival between the low and intermediate SYNTAX tertiles was not significantly different (hazard ratio 1.13, 0.60 to 2.13, log-rank p = 0.71). In conclusion, this study suggests that patients with complex coronary disease, in whom complete revascularization cannot be achieved with PCI, should be offered surgical revascularization. However, in those patients with less complex disease, PCI is a valid alternative even if complete revascularization cannot be achieved.
The aim of this study was to compare differences in clinical outcome at 5-year follow-up in patients with complete and incomplete revascularization treated with coronary artery bypass grafting (CABG) and percutaneous coronary intervention (PCI) with drug-eluting stents (DESs).
Methods
The method of the part-II Arterial Revascularisation Therapies (ARTS-II) study has been published previously. In brief, the study was a multicenter, nonrandomized, open-label trial designed to compare the safety and efficacy of the sirolimus-eluting stent in patients with de novo multivessel coronary artery disease, using the surgical group of the ARTS-I study as historical controls. The ARTS-I and ARTS-II studies used the same inclusion criteria.
Patients with stable angina, unstable angina, or silent ischemia who had ≥2 coronary lesions in different major epicardial vessels and/or their side branches (excluding the left main coronary artery) that were potentially amenable to stent implantation were eligible for inclusion. All patients were required to have a lesion with a diameter stenosis >50% in the left anterior descending coronary artery and ≥1 other major epicardial coronary artery.
The goal was to achieve complete anatomic revascularization. One totally occluded major epicardial vessel or side branch could be included. Coronary lesions were required to be amenable to stenting using a sirolimus-eluting stent with diameter of 2.5 to 3.5 mm and length of 13 to 33 mm; there was no restriction on total implanted stent length.
The major exclusion criteria were patients with previous coronary intervention, left main coronary disease, overt congestive heart failure, left ventricular ejection fraction <30%, history of a cerebrovascular accident, transmural myocardial infarction in the preceding week, severe hepatic or renal disease, neutropenia or thrombocytopenia, an intolerance or contraindication to acetylsalicylic acid or thienopyridines, need for concomitant major surgery, and life-limiting major concomitant noncardiac diseases.
Surgical techniques for patients randomized to surgery were also standardized. The left anterior descending coronary artery and/or diagonal branches were revascularized using the left internal mammary artery. Other vessels were bypassed with venous bypass grafts.
This study analyzed clinical outcomes from the 567 patients with CABG from ARTS-I and 588 patients from ARTS-II treated with DESs who had completeness of revascularization assessed.
After the index revascularization procedure an independent core laboratory (Cardialysis BV, Rotterdam, The Netherlands) reviewed all diagnostic coronary angiograms to assess completeness of revascularization. The coronary arterial tree was subdivided into 15 segments according to American Heart Association/American College of Cardiology criteria. All lesions with diameter stenosis >50% in a vessel with a reference diameter ≥1.50 mm were scored as potentially amenable to treatment. If all such defined segments had been treated according to the surgical report on the case-record form, the surgical procedure was scored as a complete revascularization. If ≥1 segment was left unbypassed, the patient was considered to have incomplete revascularization. Any patient with grafts bypassing ≥1 nonsignificant lesion and all significant lesions was included in the completely revascularized subgroup. Patients treated with grafts bypassing nonsignificant lesions who also had significant lesions that were left untreated were included in the incompletely revascularized subgroup.
For patients treated with PCI, diagnostic and procedural angiograms were reviewed. Patients were considered to have complete revascularization if all lesions with >50% diameter stenosis had been successfully treated. Those patients in whom attempt was made to treat ≥1 significant lesion or whose treatment resulted in a final diameter stenosis >50% were considered to have incomplete revascularization.
Degree of incompleteness of revascularization with either technique was further specified by dividing coronary artery segments into main and side branches. The proximal left anterior descending coronary artery (segments 6 and 7), proximal left circumflex artery (segment 11 and, in case of left dominance, segment 13), and proximal right coronary artery (segments 1, 2, and 3) were scored as main branches. All other segments were scored as side branches (12). The completeness of revascularization was then scored for the main branches, the side branches, or a combination of such defined vessels.
In addition, a detailed coronary risk score that has been previously published and tested in a subgroup of ARTS-II patients with 3-vessel disease (Synergy between Percutaneous Coronary Intervention with Taxus and Cardiac Surgery [SYNTAX] score) was used to characterize the complexity of the coronary anatomy. In brief, each coronary lesion producing ≥50% luminal obstruction in vessels ≥1.5 mm was separately scored and added to provide the overall SYNTAX score. The SYNTAX score was calculated using a dedicated software that integrates the number of lesions with their specific weighting factors based on the amount of myocardium distal to the lesion according to the score of Leaman et al and the morphologic features of each single lesion, as previously reported. Baseline SYNTAX scores in the ARTS-I study were not calculated because the baseline cine angiograms are no longer available.
Deaths included death from any cause. Cerebrovascular accidents included transient ischemic attacks, reversible neurologic deficits, intracranial hemorrhage, and ischemic stroke.
Myocardial infarction was defined in the first 7 days after the intervention, if there was documentation of new abnormal Q waves and a ratio of serum creatinine kinase-MB isoenzyme to total creatinine kinase that >0.1 or a creatinine kinase-MB value that was 5 times the upper limit of normal. Serum creatinine kinase and creatinine kinase-MB isoenzyme concentrations were measured 6, 12, and 18 hours after the intervention. Commencing 8 days after the intervention (length of hospital stay after surgery), abnormal Q waves or enzymatic changes, as described earlier, were sufficient for a diagnosis of myocardial infarction. Myocardial infarction was confirmed only after the relevant electrocardiograms had been analyzed by the core laboratory and adjudicated by the clinical events committee. Incidence of stent thrombosis was determined according to Academic Research Consortium definitions.
Continuous variables are expressed as mean ± SD and were compared using analysis of variance and Tukey post hoc test for multiple comparisons of all pairs. Categorical data are presented as frequency (percentage) and were compared using chi-square test or Fischer’s exact test. Survival curves were constructed for time-to-event variables using Kaplan-Meier estimates and compared by log-rank test.
Results
Angiograms from 1,155 patients (97.4%) were available for central analysis. In the PCI DES group, 588 (96.9%) of 607 angiograms were reviewed, whereas for bypass surgery, 567 (97.9%) of 579 angiograms were available. Complete revascularization was achieved in 360 of 588 patients (61.2%) in the PCI DES group and 477 of 567 patients (84.1%) in the CABG group (p <0.05).
Baseline demographics of patients with complete and incomplete revascularization from the CABG and PCI DES groups are presented in Table 1 . Patients in the PCI group whose revascularization was incomplete were significantly older than those treated with CABG whose revascularization was complete.
| Variable | CABG (n = 567) | PCI With DES ⁎ (n = 588) | ||
|---|---|---|---|---|
| Complete (n = 477) | Incomplete (n = 90) | Complete (n = 360) | Incomplete (n = 228) | |
| Men | 367 (76.9%) | 63 (68.9%) | 277 (76.1%) | 176 (77.2%) |
| Age (years), mean ± SD | 61 ± 9 | 62 ± 10 | 62 ± 10 | 63 ± 10 † |
| Stable angina pectoris | 280 (58.7%) | 53 (58.9%) | 181 (50.3%) | 134 (58.8%) |
| Unstable angina pectoris | 197 (41.3%) | 37 (41.1%) | 179 (49.7%) | 94 (41.2%) |
| Previous myocardial infarct | 33 (36.7%) | 204 (46.9%) | 116 (32.2%) | 82 (36.0%) |
| Previous percutaneous coronary intervention | 10 (2.1%) | 3 (3.3%) ‡ | 2 (0.6%) | 0 (0.0%) |
| Previous smoker | 221 (46.3%) | 33 (48.9%) | 151 (41.9%) | 91 (39.9%) |
| Current smoker | 127 (26.6%) | 17 (18.9%) | 70 (19.4%) | 46 (20.2%) |
| Diabetes mellitus | 72 (15.0%) | 21 (23.3%) | 91 (25.3%) | 61 (26.8%) |
| Hypercholesterolemia (>190 mg/dl) | 286 (60.0%) | 42 (46.7%) ‡ § | 276 (69.7%) | 159 (77.1%) |
| Hypertension (>165/95 mm Hg) | 208 (43.6%) | 49 (54.4%) | 232 (64.4%) | 162 (64.4%) |
| Logistic EuroSCORE (%), mean ± SD | 2.02 ± 1.63 | 2.22 ± 1.64 | 2.13 ± 1.48 | 2.21 ± 1.63 |
| SYNTAX score, mean ± SD | — | — | 18.8 ± 8.9 | 23.5 ± 9.6 ⁎ |
| Only main branch untreated/patient | — | 24/90 (26.7%) | — | 3/228 (1.3%) ‡ |
| Only side branch untreated/patient | — | 63/90 (70.0%) | — | 143/228 (62.8%) |
| Main and side branches untreated/patient | — | 3/90 (3.3%) | — | 82/228 (35.9%) ‡ |
| Number of diseased vessels/patient, mean ± SD | 2.30 ± 0.50 | 2.70 ± 0.50 § | 2.47 ± 0.49 | 2.62 ± 0.48 ⁎ |
| Number of lesions/patient, mean ± SD | 2.60 ± 0.80 | 3.70 ± 1.10 § | 3.22 ± 1.11 | 4.05 ± 1.30 ⁎ |
| Number of stents implanted/patient, mean ± SD | — | — | 3.62 ± 1.49 | 3.72 ± 1.52 |
| Total stent length (mm), mean ± SD | — | — | 71.20 ± 30.91 | 73.62 ± 32.72 |
⁎ Not significant for PCI DES complete versus PCI DES incomplete.
† Post hoc multiple comparison analysis: p <0.05 for PCI DES incomplete versus CABG complete revascularization group.
‡ p <0.05 for CABG incomplete versus PCI DES incomplete.
Angiographic characteristics are presented in Table 2 . Number of diseased vessels and number of lesions per patient were significantly higher in the incompletely revascularized groups compared to the completely revascularized groups irrespective of type of revascularization. Patients with incomplete revascularization who were treated with PCI had significantly more lesions that were >20 mm in length and moderately/heavily calcified and totally occluded compared to those with incomplete revascularization treated with CABG. Number of calcified and totally occluded lesions was significantly higher in patients with PCI DESs and incomplete revascularization compared to patients with PCI DES and complete revascularization. Procedural success rate for PCI of calcified and totally occluded lesions was 65.6% (22 of 32).
| Variable | CABG (n = 1,635 lesions) | PCI With DES (n = 2,160 lesions) | ||
|---|---|---|---|---|
| Complete (n = 1,312 lesions) | Incomplete (n = 323 lesions) | Complete (n = 1,175 lesions) | Incomplete (n = 985 lesions) | |
| Narrowed coronary artery | ||||
| Right | 397 (30.3%) | 84 (26.0%) | 353 (30.2%) | 275 (27.9%) |
| Left circumflex | 375 (28.6%) | 103 (32.8%) | 337 (28.7%) | 297 (30.2%) |
| Left anterior descending | 540 (41.1%) | 133 (41.2%) | 483 (41.1%) | 413 (41.9%) |
| Lesion characteristics | ||||
| Lesion length (visual) (percent lesions) | ||||
| Discrete (<10 mm) | 840 (67.2%) | 217 (72.3%) ⁎ | 715 (60.9%) | 543 (55.1%) † |
| Tubular (10–20 mm) | 329 (26.3%) | 62 (20.7%) ⁎ ‡ | 291 (24.8%) | 272 (27.6%) |
| Diffuse (>20 mm) | 81 (6.5%) | 21 (7.0%) ⁎ | 125 (10.6%) | 120 (12.2%) |
| Small vessels (<2.5 mm) | 71 (6%) | 128 (13%) † | ||
| Moderate/heavy calcium | 177 (14.2%) | 52 (17.3%) ⁎ | 321 (27.3%) | 322 (32.7%) † |
| Thrombus containing lesions | 19 (1.5%) | 4 (1.3%) ⁎ | 9 (0.8%) | 2 (0.2%) |
| Long-term total occlusion | 76 (5.8%) | 23 (7.1%) ⁎ | 18 (1.5%) | 32 (3.2%) † |
| Bifurcation with side branch involvement | 384 (36.0%) | 108 (30.7%) | 364 (31.0%) | 337 (34.2%) |
⁎ p <0.05 for CABG incomplete versus PCI incomplete.
† p <0.05 for PCI complete versus PCI incomplete.
Five-year Kaplan-Meier curves for major adverse cardiac and cerebrovascular events (MACCEs; composite of death, cerebrovascular accident, myocardial infarction, and any revascularization) after complete and incomplete revascularization with CABG and PCI are shown in Figure 1 .
Survival free from MACCEs ( Figure 1 ) in patients with incomplete revascularization and PCI was numerically lower than in patients with incomplete revascularization and CABG (hazard ratio 1.52, 0.91 to 0.2.52, log-rank p = 0.10) and significantly lower than the completely revascularized CABG group (hazard ratio 1.66, 0.96 to 1.80, log-rank p = 0.001). The Kaplan-Meier curve for MACCEs of the PCI DES incompletely revascularized group shows a crossing point with the incompletely revascularized CABG group at 6 months, whereas the survival curve of the PCI DES completely revascularized group diverges from the CABG completely revascularized group after 2 years with a decrease of survival rate free from MACCEs in the PCI completely revascularized group from 87% at 2 years up to 75% at 5-year follow-up.
Figure 1 shows 5-year survival free from MACCEs in the incompletely revascularized PCI DES group stratified according to SYNTAX score tertile distribution. Of note, the 5-year event-free survival in the low (SYNTAX score <19) and intermediate (SYNTAX score ≥19 to ≤26.5) tertiles was significantly better than in the higher tertile (SYNTAX score >26.5) and similar to the completely and incompletely CABG groups or the completely revascularized PCI group.
Five-year Kaplan-Meier curves for death, for the composite of death, cerebrovascular accident, and myocardial infarction, for any revascularization, and for the composite of definite/probable stent thrombosis after complete and incomplete revascularization with CABG and PCI are shown in Figure 2 .
