Author + information
- Received November 19, 2008
- Revision received May 20, 2009
- Accepted May 29, 2009
- Published online August 1, 2009.
- Mariann Gyöngyösi, MD, PhD⁎,⁎ (, )
- Günter Christ, MD⁎,
- Irene Lang, MD⁎,
- Gerhard Kreiner, MD⁎,
- Heinz Sochor, MD⁎,
- Peter Probst, MD⁎,
- Thomas Neunteufl, MD⁎,
- Rosa Badr-Eslam, MD⁎,
- Susanne Winkler, MD⁎,
- Noemi Nyolczas, MD⁎,
- Aniko Posa, PhD⁎,
- Franz Leisch, MD†,
- Ronald Karnik, MD‡,
- Peter Siostrzonek, MD§,
- Stefan Harb, MD∥,
- Matthias Heigert, MD¶,
- Gerald Zenker, MD#,
- Werner Benzer, MD⁎⁎,
- Gerhard Bonner, MD††,
- Alexandra Kaider, MSc‡‡,
- Dietmar Glogar, MD⁎,
- AUTAX Investigators
- ↵⁎Reprint requests and correspondence:
Dr. Mariann Gyöngyösi, Department of Cardiology, Medical University of Vienna, Austria, Waehringer Guertel 18-20, A-1090 Vienna, Austria
Objectives The multicenter AUTAX (Austrian Multivessel TAXUS-Stent) registry investigated the 2-year clinical/angiographic outcomes of patients with multivessel coronary artery disease after implantation of TAXUS Express stents (Boston Scientific, Natick, Massachusetts), in a “real-world” setting.
Background The AUTAX registry included patients with 2- or 3-vessel disease, with/without previous percutaneous coronary intervention (PCI) and concomitant surgery.
Methods Patients (n = 441, 64 ± 12 years, 78% men) (n = 1,080 lesions) with possible complete revascularization by PCI were prospectively included. Median clinical follow-up was 753 (quartiles 728 to 775) days after PCI in 95.7%, with control angiography of 78% at 6 months. The primary end point was the composite of major adverse cardiac (nonfatal acute myocardial infarction [AMI], all-cause mortality, target lesion revascularization [TLR]) and cerebrovascular events (MACCE). Potential risk factor effects on 2-year MACCE were evaluated using Cox regression.
Results Complete revascularization was successful in 90.5%, with left main PCI of 6.8%. Rates of acute, subacute, and late stent thrombosis were 0.7%, 0.5%, and 0.5%. Two-year follow-up identified AMI (1.4%), death (3.6%), stroke (0.2%), and TLR (13.1%), for a composite MACCE of 18.3%. The binary restenosis rate was 10.8%. The median of cumulative SYNTAX score was 23.0 (range 12.0 to 56.5). The SYNTAX score did not predict TLR or MACCE, due to lack of scoring of restenotic or bypass stenoses (29.8%). Age (hazard ratio [HR]: 1.03, p = 0.019) and acute coronary syndrome (HR: 2.1, p = 0.001) were significant predictors of 2-year MACCE. Incomplete revascularization predicted death or AMI (HR: 3.84, p = 0.002).
Conclusions With the aim of complete revascularization, TAXUS stent implantations can be safe for patients with multivessel disease. The AUTAX registry including patients with post-PCI lesions provides additional information to the SYNTAX (Synergy Between Percutaneous Coronary Intervention With TAXUS and Cardiac Surgery) study. (Austrian Multivessel TAXUS-Stent Registry; NCT00738686)
The introduction of drug-eluting stents (DES) in recent years demonstrating superiority over bare-metal stents in reducing the restenosis incidence has narrowed the reintervention gap between coronary artery bypass graft surgery (CABG) and percutaneous coronary intervention (PCI) in multivessel coronary artery disease (CAD) (1–7); multivessel stenting with DES has therefore moved into the foreground again as an alternative option for CABG.
The high revascularization rate with DES compared with CABG remains a critical end point in multivessel CAD (8,9). However, given the approximate 7% frequency of in-hospital vein graft occlusion (10,11) and the degeneration of vein grafts over time, DES may still be a more durable means than CABG as a revascularization strategy. Interestingly, even if stenting of multiple lesions is classified as an “off-label” indication for DES, approximately 10% of all PCIs in Europe and the U.S. are performed in patients with multivessel CAD.
Although current randomized trials involving the comparison of DES and CABG in multivessel CAD have encouraged inclusion of a wide variety of lesion sets, data suggest a significant variation in anatomical complexity between PCI and CABG patients (12,13). In the study by Hoffman et al. (1), only 2% to 12% of the patients screened could be randomized to either PCI or CABG (1). The goal of the SYNTAX (Synergy Between Percutaneous Coronary Intervention With TAXUS and Cardiac Surgery) trial, with randomized and registry cohorts, was to provide guidance on optimal revascularization strategies for patients with high-risk stenosis (12,14). However, the trial included only patients with de novo lesions, whereas the rapid increase in PCI in the last 10 years has led to turn the anatomy of the coronary lesions to present more complex morphology, involving restenosis, aneurysms, vessel remodeling, and phenotypic modulation of intima post-stenting (15). Accordingly, registries of nonrandomized patients with suboptimal lesions (e.g., restenotic stented lesions; or occluded, calcified, bypass; or small vessels) for either revascularization strategy may provide revascularization outcome data as robust as those from randomized analyses (13).
The aim of the AUTAX (Austrian Multivessel TAXUS-Stent) registry was to conduct a multicenter prospective registry including patients with multivessel CAD with/without previous PCI or concomitant cardiac surgery with possible complete revascularization by PCI, and treated solely with multiple TAXUS Express stent (Boston Scientific, Natick, Massachusetts) implantations in a “real-world” setting, and to report the short (30 days), medium (6 months), and long-term (1 and 2 years) angiographic and clinical outcomes.
The prospective academic nonrandomized AUTAX registry included consecutive patients with multivessel CAD at 9 Austrian medium- and high-volume PCI centers.
Between June 2004 and September 2005, 441 consecutive (all incoming) patients (age 64.4 ± 11.8 years, 78% men) with symptomatic multivessel CAD and possible percutaneous complete anatomical revascularization were prospectively included in the registry. Assignment to multivessel PCI with DES, rather than to CABG, was made on the basis of the individual clinical assessments by cardiologists and cardiac surgeons, involving a spectrum of factors such as extent and anatomy of CAD, the feasibility of achieving complete revascularization, age, prior surgery, current cardiac function, comorbidities, general health, pulmonary function, risk of anticoagulation, likelihood of compliance with continuing clopidogrel therapy, urgency of revascularization, assumption of risk of percutaneous procedure, and the waiting period for CABG in nonurgent cases. The multidisciplinary team, which also included noninvasive cardiologists, ensured that the most balanced and appropriate advice was consistently offered regarding the choice of revascularization strategy.
The study protocol conformed to the Declaration of Helsinki and was approved by the Interventional Working Group of Austrian Society of Cardiology and the institutional review committees of all centers. All patients gave their written informed consent and received optimal medical therapy according to current guidelines.
The inclusion criteria were symptomatic 2- or 3-vessel disease with/without left main disease, age >18 years, stable angina, or acute coronary syndrome (unstable angina and non–ST-segment elevation myocardial infarction [UA/NSTEMI]). Exclusion criteria were ST-segment elevation myocardial infarction within 48 h, cardiogenic shock, noncompliance with the study protocol, expected survival less than 2 years, hemorrhagic diathesis, and/or platelet count <100,000/ml3.
Study end points
The primary end point of the registry was the incidence of major adverse cardiac and cerebrovascular events (MACCE), defined as all-cause death, nonfatal acute myocardial infarction (AMI), target lesion revascularization (TLR), and cerebrovascular event during the 2-year clinical follow-up.
The secondary end points of the study included the: 1) break down primary end points at the 2-year (±2 months) follow-up, such as the rate of TLR, nonfatal AMI, all-cause death, stroke, and the composite of death or AMI; 2) the 30-day, 6-month, and 1-year MACCE; 3) the acute, subacute, and late thrombosis rates; and 4) the angiographic end points such as the binary restenosis rate per lesion (defined as diameters stenosis ≥50%), in-stent, and proximal and distal in-lesion late lumen loss.
Stenting procedure and follow-up
Pre- and post-stent dilations were performed at the interventionist's discretion. Loading doses of 300 to 600 mg clopidogrel and 250 mg aspirin were administered either before the procedure (at least 6 h before stenting in 91% of patients) or immediately after diagnostic coronary angiography before the start of PCI. During the 2-year follow-up, patients were treated with a 75-mg daily dose of clopidogrel and 100 mg aspirin.
With regard to clinical considerations arising from multivessel stenting, all patients were requested to return to angiographic follow-up at 6 ± 1 months after PCI. The patients were controlled clinically 30 ± 7 days, 1 ± 0.2 years, and 2 ± 0.3 years after PCI.
Quantitative coronary angiography
Pre-, post-stenting, and follow-up minimal lumen diameters (MLDs), reference segment diameters, and percentage diameter stenosis were determined by using a quantitative angiography program (ACOMPC, Siemens, Munich, Germany). Acute lumen gain (post- minus pre-angioplastic MLD) and late lumen loss (post-stent MLD minus follow-up MLD) were calculated. Quantitative analysis was performed in the “in-stent” area (“in-stent” analysis) and in the proximal and distal “in-lesion” area, including the stented segment and also the 5-mm margins both proximal and distal to the DES. Binary angiographic restenosis was defined as stenosis of at least 50% of the MLD in the target lesion at follow-up angiography. SYNTAX score was calculated (16) for all lesions intended to be treated with PCI.
Two-vessel disease was defined as disease in 2 of the 3 major epicardial vessels, and 3-vessel disease as disease in all 3 vessels. Complete revascularization was defined as restoration of Thrombosis In Myocardial Infarction (TIMI) flow grade 3 with residual stenosis <30% to all myocardial territories. Complex lesions were defined as type B2 or C according to a modified American College of Cardiology/American Heart Association classification.
The off-label use of stents was defined if the indication of DES implantation did not meet the standard criteria (e.g., restenotic lesion; lesion in a bypass graft; a lesion length >30 mm or reference vessel diameter <2.5 mm or >3.75 mm; or ostial, bifurcated, or totally occluded lesions).
The procedural success of PCI was stated as TIMI flow grade 3, with a final residual stenosis of <30% without death, myocardial infarction, or emergency CABG before hospital discharge.
TLR was recorded as repeated revascularization for stenosis of at least 50% of the luminal diameter anywhere within the stent or within the 5-mm borders proximal or distal to the stent. Myocardial infarction was defined as either Q-wave (significant new Q waves in at least 2 electrocardiographic leads as compared with the pre-treatment electrocardiogram, and chest discomfort associated with creatine kinase [CK] elevation >3× the upper limit of the normal value) or non–Q-wave myocardial infarction (elevation of CK levels >2× the upper limit of normal with positive CK-MB or elevation of CK levels to >2× upper limit of normal without new Q waves). The same myocardial infarction definition was used for the entire follow-up. Death was defined as any post-procedural death and was considered of cardiac origin unless there was documentation of another cause.
Cerebrovascular events were defined as stroke, transient ischemic attacks, and reversible ischemic neurologic deficits lasting more than 24 h, diagnosed by a neurologist, and confirmed by imaging modalities. In cases of multiple events occurring in 1 patient, the first event was entered into the analysis.
Acute, subacute, and late stent thrombosis were defined within 24 h, within 1 month after stent implantation, or after 1 month, respectively, in accordance with the Academic Research Consortium definitions (17).
Major bleeding was determined as clinically significant hemorrhage of any location, or with hemoglobin decrease >4 g/dl, overt bleeding with hemoglobin decrease >3 g/dl, or transfusion of ≥2 U of blood.
Angiographic and clinical outcomes were analyzed according to intention to treat. Descriptive statistical analysis was performed by using continuous variables expressed as mean value ± standard deviation (for parameters with normal distribution) or median with interquartile ranges (for parameters with skewed distribution) and by using categorical variables presented as percent frequency. For comparison between groups of the pre-specified subanalysis, 2-tailed t tests were used for the continuous variables.
Univariate and multiple Cox regression models were used to evaluate the effects of potential risk factors on event-free (MACCE-free) survival. The following variables were considered in the univariate and multiple regression models: age, sex, diabetes, hypercholesterolemia, 3-vessel disease, previous PCI/CABG, incomplete revascularization, UA/NSTEMI, total stent length, and SYNTAX score (both log-transformed values due to skewed distribution). Additionally, a stepwise Cox regression analysis (forward selection method) was performed, including only statistically significant prognostic factors in the multiple regression model. The estimated relative risks with 95% confidence intervals (CIs) were calculated.
The 2-year cumulative incidence rates of individual clinical outcomes and composite outcomes were compared in subgroups of patients at high risk of having at least 1 of the independent predictors. Kaplan-Meier estimates (supplemented by log-rank statistics) were used to plot the rates of composite MACCE and death or AMI. Mortality assessment included in-hospital deaths.
Univariate logistic regression analysis was performed to evaluate the influence of chronic total occlusion on incomplete revascularization. The strength of this potential influence is described by the odds ratio (with 95% CI).
Receiver-operator characteristic analysis was performed to determine a potential cutoff point of SYNTAX score for prediction of TLR and MACCE at 2-year follow-up, with a calculation of predictive accuracy. Patients lost to follow-up (n = 19) or with a <2 ± 0.3-year follow-up without events were excluded from this analysis.
All clinical data were recorded in the angiographic core laboratory of the Medical University of Vienna. The quantitative angiography was analyzed by an independent expert, not involved in the trial. The statistical analysis was performed by an independent statistician. Statistical analysis was carried out using the SPSS (version 17.0, SPSS Inc., Chicago, Illinois) and SAS (version 9.2, SAS Inc., Cary, North Carolina) programs in accordance with the previously set plan of the protocol. A p value of ≤0.05 was considered to be statistically significant.
A total of 1,012 patients with multivessel CAD were screened; 473 patients underwent CABG and 542 PCI; 101 patients with PCI met exclusion criteria, and were therefore not included in the registry (Table 1).
Revascularization and baseline clinical characteristics
A total of 1,080 lesions (2.45 ± 0.31 lesions/patient) were treated solely with TAXUS stents (n = 1,401, 3.18 ± 0.24 stents/patient). The procedural success rates were 100%, 98.9%, and 97.1% for the first, second, and third vessels, respectively. Table 2 presents the prevalence rates of risk factors. The lesion and procedural characteristics are listed in Table 3 and Figure 1.Table 3 presents the off-label indications (beyond the multivessel CAD) for lesions.
Pre-dilation was performed in 24.8%, 21.0%, and 37.9% of the first, second, and third lesions, respectively, and post-dilation in 26.8%, 25.3%, and 14.5% of the first, second, and third lesions, respectively.
Complete revascularization could be achieved in 90.5% of patients. The reason for the lack of complete revascularization was procedural complication (3.6% due to side-branch occlusion, vessel dissection, and/or closure), a nonsatisfactory result of PCI (0.4% of patients), or nontreatment of a significant lesion for any reason (5.5% due to the discovery of poor anatomical situation of the lesion during PCI). The acute lumen gain was 1.54 ± 0.48 mm in all lesions. Pre- and post-stenting and follow-up angiographic results are shown in Table 4.
Primary end point
Tables 5 and 6⇓⇓ report the adverse event rates during the follow-ups. The cumulative MACCE probabilities were estimated by the Kaplan-Meier method. Nineteen patients (4.3%) were controlled at 6 months, but then lost to follow-up. The national mortality registry did not record death in any of the lost-to-follow-up patients. Control angiography (routine or clinically driven) was performed in 344 patients (78%) (832 of 1,080 lesions, 77%) 8.5 ± 2.8 months after the index procedure.
During the overall study period, 78 patients suffered from MACCE, which was mainly because of TLR (n = 60). Sixteen of 60 patients underwent TLR of >1 lesion, and 11 patients repeated TLR during the 2-year follow-up.
No differences between 2- or 3-vessel disease with/without left main lesion were recorded regarding the 2-year MACCE, or the breakdown primary end point events. A trend towards higher MACCE rate was documented for the patients with left main coronary artery PCI (n = 30) as compared with MACCE rates seen in patients without left main coronary artery disease (30% vs. 16.5%) at 1 year, with no further events in the second year. Due to the small number of patients with left main coronary artery disease, no additional subanalysis was done.
Secondary end points
The breakdown of MACCE events are listed in Table 6. During the 2-year follow-up, 7 of 15 patients died because of cardiac-related disease; the others died of pneumonia or malignancies. Late thrombosis occurred in only 2 patients, leading to AMI followed by acute TLR.
The in-stent, proximal, and distal in-lesion late lumen loss data are reported in Table 4. Isolated (without in-stent) edge peri-stent restenosis was seen in 1.6% of lesions, with similar proportions in the proximal and distal stent edges. De novo stenosis was documented in 21 patients (4.8%). Patients with/without off-label use for ≥1 lesion had a 2-year MACCE of 18.7% versus 16.5% (p = 0.748) and TLR of 13.5% versus 11.5% (p = 0.858), respectively.
Major bleeding occurred in 3 patients during the in-hospital phase (retroperitoneal hematoma and local femoral access bleeding), and in 2 patients (intracranial bleeding leading to death) within 1 and 6 months post-procedure.
Predictors of MACCE
In the multivariate Cox regression model (Fig. 2), the significant predictors of 2-year MACCE were age (hazard ratio [HR]: 1.03, p = 0.019) and UA/NSTEMI (HR: 2.10, p = 0.001). Diabetes mellitus (HR: 1.50, p = 0.081) was close to significant in prediction of MACCE. Previous PCI/CABG and incomplete revascularization exhibited a time dependence for prediction of MACCE. The different clinical outcomes of subgroups of patients (UA/NSTEMI at clinical presentation, age ≥70 years, diabetes mellitus, and previous PCI/CABG) are illustrated by Kaplan-Meier estimates of MACCE-free survival (Figs. 3 and 4).⇓⇓
When all-cause mortality was chosen as a dependent variable, the Cox model identified UA/NSTEMI (HR: 3.28, 95% CI: 1.18 to 9.78, p = 0.024) and incomplete revascularization (HR: 3.71, 95% CI: 1.30 to 10.34, p = 0.011) as significant variables. Mortality and/or AMI was predicted significantly by incomplete revascularization (HR: 3.84, 95% CI: 1.60 to 9.11, p = 0.002) and UA/NSTEMI (HR: 2.40, 95% CI: 1.01 to 5.75, p = 0.046) (Fig. 5). Patients with incomplete revascularization already had a higher composite of death or AMI (9.1% vs. 1.42%, p = 0.001) in the in-hospital phase (Fig. 5). Additional analysis revealed that chronic total occlusion did not predict incomplete revascularization (odds ratio: 0.572, 95% CI: 0.249 to 1.317, p = 0.184).
The median of the cumulative SYNTAX score was 23.0 (range 12 to 56.5, with quartiles of 19.0 and 29.0). No scores were applicable for bypass vessels or restenotic lesions or for vessels with previously successful coronary procedure. Furthermore, the SYNTAX score of a right coronary artery lesion seemed to be underscored (basic score of 1 as compared with a basic score of 2.5 of the proximal left circumflex artery if left dominance and of 3.5 of the proximal left anterior descending coronary artery) even though a right dominance of coronary circulation was found in 358 patients (81.2%).
Patients with/without MACCE or TLR during the 2-year follow-up had similar initial SYNTAX scores of 22.0 (19.0 to 30.0) versus 23.0 (19.0 to 28.0) (p = 0.26) for MACCE and 23.0 (19.0 to 28) versus 22.0 (19.0 to 31.5) (p = 0.206) for TLR. Receiver-operator characteristic analysis revealed low predictive accuracies of 0.518 (p = 0.206) and of 0.510 (p = 0.205), with no clear cutoff value being predictive for occurrence of TLR or MACCE, respectively. Patients with SYNTAX score values within the second quartiles (values between 19 and 23) had similarly poor outcomes as did patients with the highest scores of >29.0 (MACCE rates of 18.9%, 24.7%, 9.4%, and 22.2%, p = 0.04 for SYNTAX score <19, 19 to 23, >23 to 28, and ≥29, respectively). Further subanalysis revealed that 18 of the 19 patients with right coronary artery TLR had a cumulative score in the second quartile range (between 19 and 23).
The major findings of our evaluation of the outcome of patients after multivessel TAXUS implantation are the low MACCE rates at 30 days, 6 months, and 1 and 2 years, which are comparable to, or even better than, the results of other studies in which the outcome of multivessel PCI was assessed. As expected, the MACCE rate was mainly driven by a 13.1% need for TLR. The binary restenosis rate of 10.8% reflected a positive selection bias (all patients with clinical symptoms were controlled angiographically), but was paralleled by an excellent in-stent and in-lesion late lumen loss.
Comparison with the SYNTAX study
In contrast with the AUTAX registry (included patients with possibly complete revascularization with multivessel PCI), the aim of the SYNTAX trial was to investigate the noninferiority of multivessel PCI with DES to CABG. Besides the exclusion of patients with post-PCI lesions in the SYNTAX study, there are other important differences between the SYNTAX and AUTAX trials: the patients included in the AUTAX registry less frequently exhibited left main, bifurcation, or trifurcation lesion and 3-vessel disease (33.5%, 72.4%, 10.7%, and 65.4%, respectively, in the SYNTAX study PCI randomized group). The patients in the SYNTAX study PCI arm had more treated lesions (mean 3.6), longer stented segments (mean 86.1 mm), and more frequent total occlusion (24.2%) than the AUTAX study patients, with a consequently higher SYNTAX score (28.4 ± 11.5) (14). Therefore, the AUTAX registry population represents a lower risk profile as compared with the SYNTAX study patients, which could also explain the lower incidence of events. In contrast, the AUTAX study patients more frequently demonstrated small vessel disease and UA/NSTEMI. Additionally, 78% of the AUTAX study patients were controlled angiographically, even if the high frequency of angiographic follow-up might contrarily contribute to the high incidence of nonclinically driven revascularization, especially in some patient groups (e.g., left main disease). The 2-year follow-up period (in contrast with the 1-year follow-up in the SYNTAX study) yielded additional important information on the occurrence of MACCE, as concerns have recently also been raised about the higher late stent thrombosis rate in patients with multivessel disease treated with DES (18).
In spite of the differences between the patients in the SYNTAX and AUTAX trials, the 1-year cumulative MACCE rates and the rates of TLR, death, and stroke were similar. However, less stent thrombosis (1.2%) and less myocardial infarction (0.9%) were documented in the AUTAX registry (3.3% and 4.8% in the SYNTAX study PCI randomized arm), suggesting the possible preventive role of control angiography in multivessel CAD.
The SYNTAX PCI only registry arm included inoperable patients (with de novo lesions), and this patient cohort is therefore not comparable with the cohort in the AUTAX registry.
Cypher stent in multivessel CAD
After the treatment of multivessel CAD with sirolimus-eluting stents (Cordis, a Johnson&Johnson, Bridgewater, New Jersey), Orlic et al. (6) reported a MACE rate of 22.3% at 6.5 months in unselected lesions, which is higher than our 1-year and 2-year MACCE rates. The earlier recruiting period of the Orlic et al. (6) study may suggest an improved procedural technique and optimized use of dual antiplatelet therapy at the time of the AUTAX registry.
Outcomes of PCI and CABG in multivessel CAD
Recent studies have reported an absolute benefit of CABG in multivessel CAD, with significantly lower rates of death, AMI, and composite MACE as compared with multivessel treatment with DES (9,19), due to off-pump surgery or use of arterial grafts (20). However, the ARTS-II (Arterial Revascularization Therapies Study Part II: Sirolimus-Eluting Stents for the Treatment of Patients With Multivessel De Novo Coronary Artery Lesions) (21) and SYNTAX trials reported similar rates of death and AMI for both PCI and CABG, with a significantly higher rate of TLR but less stroke in the PCI arm.
Even if CABG is traditionally favored over PCI, the repeat revascularization rate is poorly reported at the time frames (5 to 10 years) at which CABG graft degeneration occurs. Relative benefits of PCI are the improved recovery and reduced neurologic complications (e.g., stroke and the less often reported subtle neurocognitive dysfunction). In addition, the virtually higher morbidity of CABG (e.g., sternal wound infection or leg infection from harvesting, ventilator pneumonia, early graft closure, pericarditis, or atrial fibrillation) also favors PCI. The hospital stay and the neurologic complications were similarly low in the AUTAX and SYNTAX trials, and the number of patients with prolonged hospital stay for noncardiac reasons was also very low.
Although less invasive (and thus preferred by patients), PCI may not provide complete revascularization. In our study, incomplete revascularization predicted mortality and the combined end point of death or myocardial infarction. Similar to our findings, Hannan et al. (22) reported that stent patients with incompletely revascularized multivessel CAD were significantly more likely to die at any time than completely revascularized patients. Interestingly, Alamanni et al. (23) investigated the impact of completeness of coronary revascularization on long-term outcomes after on- or off-pump CABG, and reported that both incomplete revascularization and increasing numbers of distal anastomoses (even when controlling for completeness of revascularization) were significant predictors of perioperative myocardial infarction. Complete revascularization could be achieved in 63.2% to 74.7% of the SYNTAX CABG randomized arm and registry and 56.7% to 36.5% of the PCI randomized arm and registry, which was not reported to be in association with increased event rate. In our study, incomplete revascularization was a result of diverse causes not predictable before procedure, and was likewise not predicted by chronic total occlusion, and the unfavorable procedural outcome did not merit the indication for emergency or elective bypass surgery. In contrast with the reported studies (e.g., 69% by Hannan et al.  or the SYNTAX study), the rate of complete revascularization in the AUTAX trial was much higher (90.5%), which mirrors the striving for achievement of the complete restoration of myocardial perfusion in a percutaneous way, if possible.
In recent clinical practice, DES are also implanted for lesions with off-label indications, which may be at higher risk of subsequent cardiac events. However, current data suggest that the use of DES in most lesion subsets is at least as safe as bare-metal stent use and clearly more efficacious than the use of bare-metal stents (24). Almost 60% of our patients had ≥2 lesions, and additionally 21% of patients had 1 lesion with an off-label indication, with a diverse combination of lesion subsets. In comparison with other studies involving “off-label” DES implantation, the rates of TLR and MACCE events were similar in our study (25).
In contrast with the randomized or observational TAXUS studies including de novo native lesions, the AUTAX registry also included patients with small vessel disease (32.7%), with an expected lower late lumen loss. Further, we analyzed the proximal and distal reference segments separately, which again results in some differences between the AUTAX and other trials as regards the quantitative angiographic data.
Even if the number of events (78 events) was low for the number of predictors (10 predictors) in multivariate approach, we investigated the known predictive factors for occurrence of MACCE and death or AMI.
The SYNTAX score
Interestingly, the SYNTAX score in our population did not predict worse outcome, probably for several reasons, including underscoring of right coronary artery lesions in right coronary dominance, lack of scoring of bypass vessels or restenotic lesions, and vessels with previous coronary procedures. In the real-world setting, increasing numbers of patients are repeatedly treated with PCI, with an apparently higher risk of restenosis and consequent revascularization, which should be considered in the SYNTAX score system. Almost half of our patients had previous PCI or CABG; exclusion of these patients from the scoring system would have led to a severe bias.
With the aim of complete revascularization, multiple TAXUS stent implantation is safe for patients with multivessel CAD, with low incidence of late stent thrombosis, even with off-label indications or complex coronary anatomy unfavorable for bypass surgery.
- Abbreviations and Acronyms
- acute myocardial infarction
- coronary artery bypass graft surgery
- coronary artery disease
- confidence interval
- creatine kinase
- drug-eluting stent(s)
- hazard ratio
- major adverse cardiac and cerebrovascular event(s)
- minimal lumen diameter
- non–ST-segment elevation myocardial infarction
- percutaneous coronary intervention
- target lesion revascularization
- unstable angina
- Received November 19, 2008.
- Revision received May 20, 2009.
- Accepted May 29, 2009.
- American College of Cardiology Foundation
- Hoffman S.N.,
- Tenbrook J.A.,
- Wolf M.P.,
- et al.
- Rodriguez A.,
- Bernardi V.,
- Navia J.,
- et al.
- Legrand V.M.G.,
- Serruys P.W.,
- Unger F.,
- et al.,
- Arterial Revascularization Therapy Study (ARTS) Investigators
- Orlic D.,
- Bonizzoni E.,
- Stankovic G.,
- et al.
- Berger P.B.,
- Alderman E.L.,
- Nadel A.,
- Schaff H.V.
- DeCanniere D.,
- Jansens J.L.,
- Goldschmidt-Clermont P.,
- Barvais L.,
- Decroly P.,
- Stoupel E.
- Kappetein A.P.,
- Dawkins K.D.,
- Mohr F.W.,
- et al.
- Hao H.,
- Gabbiani G.,
- Camenzind E.,
- Bacchetta M.,
- Virmani R.,
- Bochaton-Piallat M.L.
- Laskey W.K.,
- Yancy C.W.,
- Maisel W.H.
- Maree A.O.,
- Mieres J.,
- Berrocal D.,
- et al.
- Hannan E.L.,
- Racz M.,
- Holmes D.R.,
- et al.
- Alamanni F.,
- Dainese L.,
- Naliato M.,
- et al.
- Harjai K.J.,
- Orshaw P.,
- Shenoy C.,
- et al.
- Applegate R.J.,
- Sacrinty M.T.,
- Kutcher M.A.,
- et al.