Clinical trial data have supported the safety and efficacy of drug-eluting stents (DES) in the treatment of patients with ST-segment elevation myocardial infarctions (STEMIs), but contemporary “real-world” registry data regarding the late safety profiles of DES are limited. This prospective registry-based study included 1,569 consecutive unselected patients with STEMIs who underwent emergency primary percutaneous coronary intervention from January 2001 to December 2009. Of the study cohort, 200 patients (12.7%) received DES, while 1,369 patients (87.3%) underwent bare-metal stent (BMS) placement. The primary end points of the study were all-cause mortality and target vessel revascularization at 1, 2, and 3 years. Survival status was assessed by municipal civil registries. Repeat revascularization procedures were prospectively collected in the hospital database. All-cause mortality was significantly lower in the DES group at 3 years (4.2% vs 13.5%, p = 0.007) compared to BMS-treated patients, but DES use was not an independent predictor of all-cause mortality (adjusted odds ratio 0.5, 95% confidence interval 0.2 to 1.2, p = 0.10). Target vessel revascularization was significantly lower in the DES group compared to the BMS group at 3 years (10.5% vs 21%, p = 0.001). DES use was an independent predictor of reduced target vessel revascularization (adjusted odds ratio 0.44, 95% confidence interval 0.25 to 0.77, p = 0.004). Late definite stent thrombosis occurring after 1 year occurred in 4 (2.5%) patients in the DES group compared to 6 (0.7%) in the BMS group (p = 0.05). DES use was an independent predictor of late stent thrombosis (adjusted odds ratio 8.6, 95% confidence interval 1.9 to 38, p = 0.004). In conclusion, this contemporary registry-based study of patients who underwent primary percutaneous coronary intervention for STEMI demonstrated improved revascularization rates without increased 3-year hazard of adverse clinical outcomes in DES-treated patients.
Although there is relative consensus of the benefit of drug-eluting stents (DES) over bare-metal stents (BMS) in the treatment of patients with ST-segment elevation myocardial infarction (STEMI) on the basis of several clinical trials, more “real-world” registry data regarding the late safety profiles of DES are scarce. To date, these studies have been largely limited by length of follow-up (most <24 months). A series presented by Pierre-Louis et al showed an absolute risk reduction in mortality of 7% at 43 months but evaluated only 376 patients. Data from the large Registro Angioplastiche Dell’Emilia-Romagna (REAL) multicenter registry revealed comparable rates of mortality and stent thrombosis between the 2 stenting groups but significant reductions in reintervention in the DES group at 3-year follow-up. The most recent update of the American College of Cardiology and American Heart Association guidelines consider the use of DES as a “reasonable” alternative to BMS in primary percutaneous coronary intervention (PCI) for STEMI (class IIA recommendation). Because of ongoing concern about the late cardiovascular events associated with the use of these stents in clinical practice, we sought to explore the comparative 3-year outcomes in a large, single-center registry.
Methods
This prospective registry included 1,569 consecutive unselected patients with STEMI who underwent emergency primary PCI at Rabin Medical Centre. This study was conducted from January 2001 to December 2009. The only exclusion criterion to enrollment was cardiogenic shock at the time of admission. The study was approved by the Investigational Review Board (Helsinki Committee) of the Rabin Medical Centre, and all subjects provided written informed consent before cardiac catheterization.
All patients were pretreated with aspirin, clopidogrel, and unfractionated heparin (70 U/kg loading dose). A 300- to 600-mg loading dose of clopidogrel was administered before the index procedure or immediately afterward. Platelet glycoprotein IIb/IIIa receptor inhibitors were used at the discretion of the operator. Coronary angiography was performed through the femoral route. Selection of DES or BMS and predilatation with undersized balloons were left to the operator’s discretion. In the first year, DES were used in proximal coronary segments or in anterior wall acute myocardial infarction. The unavailability of DES >3.5 mm in diameter was a common reason for selecting BMS, even after DES became available. The diameter of a stent could be enlarged by postdilating the stent with a larger angioplasty balloon to leave a minimal residual stenosis and optimize stent apposition to the vessel wall. All stents were implanted with moderate to high deployment pressure (10 to 16 atm). Baseline characteristics, procedural details, and quantitative coronary angiographic parameters were collected and available for all patients. All patients were prescribed lifelong aspirin, and clopidogrel was prescribed for 1 to 3 months after BMS implantation and 6 to 12 months after DES implantation.
Coronary angiograms were recorded at baseline and after PCI. Experienced cardiologists at the institute’s angiographic core laboratory, using the MDView QA System (Medcon Telemedicine Technology McKesson, Tel-Aviv, Israel), analyzed all angiographic images using automated edge detection techniques independent of clinical outcomes. A contrast-filled guiding catheter (6Fr or 7Fr) was used as calibration standard. Reference and minimal luminal diameters were determined before and after PCI. Standard morphologic criteria were used to identify lesion location, luminal diameter, stent length, and thrombus size. On the basis of these measurements, percentage diameter stenosis was determined before and after intervention. Thrombolysis In Myocardial Infarction (TIMI) flow grade (0 to 3) was measured before and at the completion of the procedure.
Immediate and in-hospital events were recorded, and each patient completed a standardized questionnaire either by telephone or in the outpatient clinic at set time intervals. Follow-up rates at 1-, 2-, and 3-year evaluations were 100%, 81%, and 72%, respectively. Survival status at follow-up was assessed by municipal civil registries. Repeat revascularization procedures and episodes of acute myocardial infarction were prospectively collected in the hospital database. Records from peripheral hospitals were obtained to confirm outcome diagnoses of patients in the follow-up period. All events were further adjudicated by a research coordinator and reviewed by an experienced cardiologist from our research team.
The primary end points of the study were all-cause mortality and target vessel revascularization (TVR). Secondary end points included the occurrence of major adverse cardiac events, defined as composite cardiac death, lesion-related acute myocardial infarction, repeat target lesion revascularization, or TVR at 1, 2, and 3 years. TVR was defined as any revascularization that involved the target vessel. Target lesion revascularization was defined as a successful revascularization procedure (PCI or surgical bypass) driven by a stenosis reappearing at the treated site (including 5 mm proximal and distal to the lesion borders). Reinfarction was defined as a clinical episode of prolonged chest pain suggestive of acute myocardial infarction and an increase in serum cardiac enzyme levels to ≥2 times the upper normal limit or the appearance of ≥1 new pathologic Q wave after hospital discharge. Stent thrombosis was defined according to the Academic Research Consortium as “definite” in the context of acute coronary syndromes and/or reinfarction in the culprit coronary territory with angiographically proved thrombosis of the previously implanted stent.
All clinical parameters assessed during the study period were found to be normally distributed by the Kolmogorov-Smirnov test. Continuous variables are presented as mean ± SD and categorical variables as frequencies and percentages. Continuous variables were compared using Student’s t test. Comparisons between categorical variables were performed using chi-square tests or Fisher’s exact tests. Time-dependent primary clinical end points were further analyzed using Kaplan-Meier survival analysis. Group comparisons were made using the log-rank test. Because of the nonrandomized study design, we used propensity score analytic methods. Propensity scores were calculated for each patient on the basis of a number of prespecified clinical, angiographic, and procedural variables: age, year of PCI, gender, diabetes mellitus, previous myocardial infarction, previous cerebrovascular accident, aspirin and clopidogrel pretreatment, renal failure, stent diameter, and stent length. Because of the limited number of patients in the DES group, propensity score matching was not possible. However, multivariate models adjusting for propensity score and any additional variable with a p value <0.10 in univariate selection models were created for each of the primary end points and for late stent thrombosis. Analysis was performed using Statisca software (StatSoft, Inc., Tulsa, Oklahoma), and p values <0.05 were considered significant.
Results
A total of 1,569 consecutive patients with STEMI were included in the final analysis. Of the study cohort, 200 patients (12.7%) underwent DES implantation, while 1,369 (87.3%) were treated with BMS. In the DES group, 87 patients (43.5%) received sirolimus-eluting stents, 69 (34.5%) paclitaxel-eluting stents, 21 (10.5%) zotarolimus-eluting stents (Endeavor; Medtronic, Inc., Minneapolis, Minnesota), 14 (7%) biolimus-eluting stents, and 9 (4.5%) everolimus-eluting stents.
Baseline clinical characteristics are listed in Table 1 . Patients in the DES group were younger, had higher rates of previous myocardial infarction, and presented more commonly with anterior wall infarcts. BMS were implanted in patients with higher rates of current smoking, peripheral vascular disease, and lower admission hemoglobin levels. Notably, mean Controlled Abciximab and Device Investigation to Lower Late Angioplasty Complications (CADILLAC) scores and rates of higher Killip classes (>1) were similar between the 2 groups.
| Variable | BMS | DES | p Value |
|---|---|---|---|
| (n = 1,369) | (n = 200) | ||
| Age (years) | 60.7 ± 12.6 | 58.8 ± 12.0 | 0.05 |
| Men | 1,108 (81%) | 174 (87%) | 0.40 |
| Diabetes mellitus | 357 (26%) | 49 (24.5%) | 0.60 |
| Renal insufficiency ⁎ | 167 (12.2%) | 22 (11%) | 0.50 |
| Hypertension † | 662 (48.4%) | 88 (44%) | 0.20 |
| Current smokers | 613 (44.8%) | 71 (35.5%) | 0.02 |
| Family history of coronary artery disease | 423 (31%) | 81 (40.5%) | 0.02 |
| Hyperlipidemia ‡ | 660 (48%) | 100 (50%) | 0.70 |
| Peripheral vascular disease | 55 (4%) | 5 (2.5%) | 0.05 |
| Previous myocardial infarction | 115 (8.4%) | 28 (14%) | 0.06 |
| Previous percutaneous coronary interventions | 186 (13.6%) | 30 (15%) | 0.80 |
| Previous coronary artery bypass graft surgery | 31 (2.3%) | 4 (2%) | 0.80 |
| Previous stroke | 73 (5.3%) | 6 (3%) | 0.10 |
| Left ventricular ejection fraction <40% | 547 (40%) | 85 (42.5%) | 0.60 |
| Anterior wall myocardial infarction | 609 (44.5%) | 126 (63%) | 0.002 |
| Hemoglobin (g/dl) | 13.9 ± 1.6 | 14.2 ± 1.5 | 0.006 |
| Peak creatine kinase (×1,000 U/L) | 1.9 ± 1.9 | 2.2 ± 2.1 | 0.30 |
| Killip class >1 | 185 (13.5%) | 19 (9.5%) | 0.40 |
| CADILLAC score | 4.1 ± 3.6 | 3.7 ± 3.1 | 0.10 |
⁎ Creatinine clearance <60 ml/min/m 2 .
† Blood pressure ≥140/90 mm Hg at enrollment or use of antihypertensive medication.
‡ Total cholesterol ≥200 mg/dl at enrollment or use of lipid-lowering medication.
Preprocedural medication utilization and mean PCI-related time intervals are listed in Table 2 . Patients who underwent DES implantation were more likely to receive dual antiplatelet therapy but were less likely to receive β blockers and angiotensin-converting enzyme inhibitors or angiotensin receptor blockers compared to the BMS group. There were no differences between the 2 groups in terms of time to presentation and time to intervention.
| Variable | BMS | DES | p Value |
|---|---|---|---|
| (n = 1,369) | (n = 200) | ||
| Medications | |||
| Aspirin | 1,183 (86.4%) | 184 (92%) | 0.08 |
| Clopidogrel | 673 (49.2%) | 130 (65%) | 0.003 |
| β blockers | 130 (9.5%) | 8 (4%) | 0.03 |
| Angiotensin-converting enzyme inhibitors/angiotensin receptor blockers | 110 (8%) | 7 (3.5%) | 0.06 |
| Statins | 436 (31.9%) | 75 (37.5%) | 0.30 |
| Time Intervals | |||
| Symptom onset to emergency department (hours) | 0.50 | ||
| Mean ± SD | 3.2 ± 2.7 | 3.1 ± 24 | |
| Median (range) | 2 (1.2–4) | 2 (1.5–4) | |
| Emergency department to coronary intervention (hours) | 0.20 | ||
| Mean ± SD | 1.7 ± 3.1 | 1.5 ± 2.1 | |
| Median (range) | 1 (1–2) | 1 (0.7–1.5) |
Lesion and procedural information are listed in Table 3 . Consistent with anterior wall predominant clinical presentations, the left anterior descending coronary artery was more often the culprit vessel in DES patients compared to the BMS group. Lesion and stent length were longer in the DES-treated patients. Glycoprotein IIb/IIIa inhibitors were used in most patients in the 2 groups but with higher frequency in the DES group. Procedural success was obtained in 96% of the patients, regardless of stent type.
| Variable | BMS | DES | p Value |
|---|---|---|---|
| (n = 1,369) | (n = 200) | ||
| Culprit coronary artery | |||
| Left anterior descending | 589 (43%) | 122 (61%) | 0.001 |
| Left circumflex | 210 (15%) | 16 (8%) | |
| Right | 522 (38%) | 54 (27%) | |
| 2- or 3-vessel disease | 781 (57%) | 116 (58%) | 0.70 |
| Pre-PCI TIMI flow grade 0 or 1 | 877 (64%) | 118 (59%) | 0.20 |
| Post-PCI TIMI flow grade 3 | 1,316 (96%) | 192 (96%) | 0.90 |
| Pre-PCI culprit stenosis (%) | 96 ± 7 | 95 ± 7.5 | 0.90 |
| Post-PCI culprit stenosis (%) | 2 ± 7 | 3 ± 8 | 0.60 |
| Pre-PCI reference diameter (mm) | 3.1 ± 0.5 | 3.0 ± 0.5 | 0.05 |
| Lesion length (mm) | 15.6 ± 7.4 | 16.1 ± 67 | 0.03 |
| No reflow | 68 (8%) | 12 (6%) | 0.60 |
| Glycoprotein IIb/IIIa inhibitors | 996 (73%) | 161 (80.5%) | 0.02 |
| Visible or suspected thrombus | 141 (83.4%) | 165 (82.5%) | 0.60 |
| Thrombectomy devices | 213 (16%) | 22 (11%) | 0.40 |
| Stent length (mm) | 18.5 ± 5.7 | 22.0 ± 59 | 0.0003 |
| Mean stent width (mm) | 3.1 ± 0.5 | 3.1 ± 0.4 | 0.40 |
| Procedural success | 1,316 (96%) | 192 (96%) | 0.80 |
Data regarding all clinical outcomes at each year of follow-up are listed in Table 4 . The unadjusted rates of all-cause mortality were no different at 30 days and 1 year between the BMS and DES groups. However, mortality was significantly higher in the BMS group at 2-year (9.8% vs 3.6%, p = 0.008) and 3-year (13.5% vs 4.2%, p = 0.007) follow-up compared to the DES-treated patients. Figure 1 illustrates the significant unadjusted survival difference between the DES and BMS groups (log-rank p = 0.02). After accounting for variable baseline risk profiles between the 2 groups using multivariate and propensity score techniques, DES use was not an independent predictor of all-cause mortality at final follow-up (adjusted odds ratio [OR] 0.5, 95% confidence interval [CI] 0.2 to 1.2, p = 0.10). In fact, CADILLAC score was the only independent predictor of mortality (adjusted OR 1.4, 95% CI 1.3 to 1.6) after risk adjustment.
