Primary percutaneous coronary intervention (PCI) is the preferred reperfusion strategy in patients with ST-segment elevation myocardial infarction (STEMI). We evaluated whether presentation of patients with STEMI to a noninterventional facility requiring transfer for primary PCI compared to direct admission to a PCI center has an impact on clinical outcomes. Of 3,602 patients enrolled in the multicenter, prospective HORIZONS-AMI trial, 988 (24.7%) were transferred for primary PCI and 2,614 were directly admitted to an interventional hospital. Clinical outcomes at 30 days and 1 year were evaluated. Median time to reperfusion in patients with transfer was 67 minutes longer compared to patients without transfer (272 vs 205 minutes, p <0.001), and first door-to-balloon time was 47 minutes longer (134 vs 87 minutes, p <0.001). At 30 days and 1 year there were no significant differences between patients with and without transfer in the rates of major adverse cardiac events (30 days 5.8% vs 5.4%, p = 0.68; 1 year 11.6% vs 12.0%, p = 0.74), major bleeding (30 days 7.3% vs 6.9%, p = 0.66; 1 year 7.9% vs 7.4%, p = 0.63), or mortality (30 days 2.6% vs 2.6%, p = 0.92; 1 year 4.0% vs 4.2%, p = 0.81). In transfer and nontransfer patients use of bivalirudin compared to unfractionated heparin plus glycoprotein IIb/IIIa inhibitor was associated with lower rates of bleeding, cardiac death, and net adverse clinical events. In conclusion, in the HORIZONS-AMI trial, 30-day and 1-year survival rates and clinical outcomes were comparable in patients with STEMI requiring and not requiring transfer for primary PCI.
Data are sparse comparing patients with need for transfer to patients without need for transfer for primary percutaneous coronary intervention (PCI). In the prospective, randomized, multicenter Harmonizing Outcomes with Revascularization and Stents in Acute Myocardial Infarction (HORIZONS-AMI) trial, 3,602 patients with ST-elevation myocardial infarction (STEMI) with and without transfer for a primary PCI strategy were enrolled. From this detailed database we examined whether the need for transfer for primary PCI had an impact on 30-day and 1-year clinical outcomes compared to patients without transfer for primary PCI.
Methods
Patients were randomized in a 1:1 ratio to treatment with bivalirudin or unfractionated heparin plus a glycoprotein IIb/IIIa inhibitor. Bivalirudin was administered as an intravenous bolus of 0.75 mg/kg, followed by an infusion of 1.75 mg/kg/hour. Heparin was administered as an intravenous bolus of 60 IU/kg of body weight, with subsequent boluses targeted to an activated clotting time of 200 to 250 seconds. Patients were stratified for administration of heparin before randomization; administration of clopidogrel 300 or 600 mg or ticlopidine 500 mg; selection of abciximab or eptifibatide; start of study drug in emergency room or catheterization laboratory; and location of study site in or outside the United States. After coronary angiography patients were triaged to treatment with PCI, coronary-artery bypass grafting (CABG), or medical management, as described previously. Patients with stent implantation were randomized again in a 3:1 ratio to paclitaxel-eluting stents (TAXUS Express, Boston Scientific, Natick, Massachusetts) or uncoated bare metal stents (Express, Boston Scientific, Natick, Massachusetts).
Patients ≥18 years of age were eligible for inclusion if they presented within 12 hours after onset of symptoms and had ST-segment elevation ≥1 mm in ≥2 contiguous leads, new left bundle-branch block, or true posterior MI. Principal exclusion criteria were contraindications to study medications and previous administration of thrombolytic agents, bivalirudin, glycoprotein IIb/IIIa inhibitors, low-molecular-weight heparin, or fondaparinux for the present event as described in detail elsewhere. The study was approved by the institutional review board or ethics committee at each participating center, and all patients gave written informed consent. All 3,602 patients were randomized in a center with “around-the-clock” primary PCI service. Patients presenting first to a hospital without PCI capabilities were also considered for enrollment if the described inclusion and exclusion criteria after transfer were fulfilled at arrival at the study center. In total, 988 patients (27.4%) were transferred from a non-PCI center to the study hospital and subsequently randomized, whereas 2,614 patients (72.6%) were enrolled after presenting directly at the PCI study hospital.
In HORIZONS-AMI there were 2 primary end points: major bleeding (not related to CABG) and net adverse clinical events (NACEs), defined as the combination of major bleeding or a composite of major adverse cardiovascular events. Major adverse cardiovascular events included death, reinfarction, target vessel revascularization for ischemia, or stroke. The end points definitions have been detailed elsewhere. Bleeding was adjudicated according to the protocol definition and based on the Thrombolysis in Myocardial Infarction (TIMI) and Global Utilization of Streptokinase and Tissue Plasminogen Activator for Occluded Coronary Arteries scales. Stent thrombosis was defined according to the Academic Research Consortium classification.
Comparison of patients with transfer to patients without transfer was performed according to an intention to treat. Categorical outcomes were compared by chi-square test or Fisher’s exact test. Continuous variables were compared by Wilcoxon rank-sum test. Time-to-event outcomes for unadjusted events, determined with Kaplan-Meier methods, were compared by log-rank test. Independent correlates of 30-day and 1-year mortality in the entire randomized population and in the cohort undergoing primary PCI were determined by Cox proportional hazards regression. Variables were selected by stepwise selection with transfer status forced into the model. The following list of potential baseline covariates was considered for inclusion in the model: randomization to bivalirudin (vs unfractionated heparin plus glycoprotein IIb/IIIa inhibitor); age; gender; race; study site in or outside the United States; body mass index; Killip class; anterior MI; anemia; platelet counts; creatinine clearance; history of hypertension, hyperlipidemia, smoking, diabetes, MI, PCI, CABG, coronary artery disease, angina, congestive heart failure, or peripheral vascular disease; and medications 5 days before enrollment including acetylsalicylic acid, β blocker, thienopyridine, calcium channel blocker, angiotensin-converting enzyme inhibitor/angiotension receptor blocker, and diuretics. In addition, time from symptom onset was included for the PCI cohort as indicated.
Results
Management strategy after angiography did not differ in patients with versus patients without transfer. Treatment included primary PCI in 92.7% versus 92.9%, deferred PCI in 0.0% versus 0.1%, primary CABG in 1.5% versus 1.9% and medical treatment in 5.8% versus 5.0%, respectively.
Patients without versus with transfer had higher rates of diabetes mellitus, hyperlipidemia, and previous MI and PCI; other baseline features were balanced between groups ( Table 1 ). In patients with versus those without transfer, the interval from symptom onset to arrival at the first hospital was significantly shorter; the interval from symptom onset to arrival at the PCI hospital was significantly longer; and the interval from arrival at the PCI hospital to the catheter laboratory was significantly shorter ( Figure 1 ). As a result, in the cohort undergoing PCI, median first door-to-balloon time was ∼47 minutes longer in patients with compared to without transfer. There was no difference in intervals between patients treated with bivalirudin versus unfractionated heparin plus a glycoprotein IIb/IIIa stratified by transfer status (data not shown).
| Variable | With Transfer (n = 988) | Without Transfer (n = 2,614) | p Value |
|---|---|---|---|
| Age (years), median (range) | 60.0 (52.3–70.2) | 60.2 (52.4–69.8) | 0.93 |
| Men | 747 (76%) | 2,013 (77%) | 0.38 |
| Diabetes mellitus | 141 (14%) | 452 (17%) | 0.03 |
| Hypertension | 515 (52%) | 1,409 (54%) | 0.34 |
| Hyperlipidemia | 381 (39%) | 1,169 (45%) | 0.0009 |
| Smoking | 626 (64%) | 1,652 (64%) | 0.96 |
| Previous myocardial infarction | 90 (9%) | 302 (12%) | 0.04 |
| Previous percutaneous coronary intervention | 68 (7%) | 318 (12%) | <0.0001 |
| Previous coronary bypass | 21 (2%) | 84 (3%) | 0.08 |
| Body mass index (kg/m 2 ), median (range) | 27.0 (24.5–30.0) | 27.1 (24.6–30.3) | 0.41 |
| Killip class I | 904 (92%) | 2,383 (91%) | 0.72 |
Target vessel, treatment with aspirin or clopidogrel, and compliance with protocol-specified study medication did not differ ( Table 2 ). Use of preprocedure heparin before randomization and selection of abciximab for glycoprotein IIb/IIIa inhibition at randomization were significantly more frequent in patients with transfer compared to patients without transfer. Sum of ST-segment elevation did not differ with medians 7.0 mm (interquartile range 4.1 to 12.3) in patients with transfer and 7.4 mm (4.4 to 12.2) in patients without transfer (p = 0.40). TIMI grade 0/1 flow before PCI was present less often in transferred compared to nontransferred patients (62.3% vs 66.5%, p = 0.02). TIMI flow rates after PCI did not differ according to transfer status, with TIMI grade 3 flow achieved in 91.6% of patients with transfer versus 91.5% of patients without transfer.
| Variable | With Transfer (n = 988) | Without Transfer (n = 2,614) | p Value |
|---|---|---|---|
| Infarct-related coronary artery | |||
| Left anterior descending | 389 (40%) | 1,065 (41%) | 0.60 |
| Left circumflex | 163 (17%) | 403 (16%) | 0.36 |
| Right | 412 (42%) | 1,088 (42%) | 0.79 |
| Left main | 2 (0.2%) | 18 (0.7%) | 0.08 |
| Saphenous vein graft | 7 (0.7%) | 28 (1.1%) | 0.33 |
| Heparin before procedure | 757 (77%) | 1,796 (69%) | <0.0001 |
| Randomization | |||
| Heparin plus glycoprotein IIb/IIIa inhibitor | 508 (51%) | 1,294 (49%) | 0.31 |
| Bivalirudin | 480 (49%) | 1,320 (51%) | 0.31 |
| Antithrombin during procedure | |||
| Heparin | 511 (52%) | 1,313 (50%) | 0.43 |
| Bivalirudin | 467 (48%) | 1,282 (49%) | 0.32 |
| Glycoprotein IIb/IIIa inhibitor | |||
| Any | 525 (53%) | 1,428 (55%) | 0.39 |
| Abciximab | 318 (32%) | 708 (27%) | 0.003 |
| Eptifibatide | 204 (21%) | 710 (27%) | <0.0001 |
| Tirofiban | 3 (0.3%) | 11 (0.4%) | 0.77 |
| Aspirin use | |||
| Before admission | 249 (25%) | 719 (28%) | 0.17 |
| During hospitalization | 984 (100%) | 2,602 (100%) | 1.0 |
| At discharge | 942 (98%) | 2,484 (98%) | 0.98 |
| Thienopyridine use | |||
| Before admission | 48 (5%) | 95 (4%) | 0.09 |
| Loading dose, any | 959 (97%) | 2,524 (97%) | 0.42 |
| Clopidogrel 300 mg | 321 (33%) | 892 (34%) | 0.36 |
| Clopidogrel 600 mg | 627 (64%) | 1,589 (61%) | 0.14 |
| At discharge | 891 (92%) | 2,382 (94%) | 0.21 |
| Medication at discharge | |||
| β blockers | 861 (88%) | 2,317 (89%) | 0.21 |
| Angiotensin-converting enzyme inhibitor or angiotensin receptor blocker | 784 (81%) | 2,055 (81%) | 0.77 |
| Statins | 907 (94%) | 2,386 (94%) | 0.85 |
At 30 days and 1 year there were no significant differences in rates of major adverse cardiovascular events, major bleeding, or NACEs between patients with and without transfer ( Figure 2 , Table 3 ). Furthermore, no significant differences were observed with respect to all-cause death, cardiovascular death, reinfarction, stroke, target vessel revascularization due to ischemia, bleeding events according to TIMI or Global Utilization of Streptokinase and Tissue plasminogen Activator for Occluded Coronary Arteries scales, or stent thrombosis according to transfer time. By multivariable analysis, after accounting for differences in baseline features and medication use between groups (but not treatment times), transfer status was not an independent predictor of mortality at 30 days or 1 year ( Table 4 ). Similarly, in patients with PCI, after adjusting for baseline covariates, medication use, transfer status, and total symptom to balloon time, transfer status was not an independent predictor of mortality at 30 days or 1 year ( Table 4 ). There was no significant relation among 30-day mortality, 1-year mortality, and presence of TIMI grade 0/1 flow and reperfusion time for patients with and without transfer.
| Variable | 30-Day Follow-up | 1-Year Follow-up | ||||
|---|---|---|---|---|---|---|
| With Transfer (n = 988) | Without Transfer (n = 2,614) | p Value | With Transfer (n = 988) | Without Transfer (n = 2,614) | p Value | |
| Net adverse clinical events | 113 (11.5%) | 283 (10.9%) | 0.63 | 159 (16.3%) | 441 (17.2%) | 0.59 |
| Major adverse cardiac events | 57 (5.8%) | 141 (5.4%) | 0.68 | 112 (11.6%) | 307 (12.0%) | 0.74 |
| Death (all cause) | 26 (2.6%) | 67 (2.6%) | 0.92 | 39 (4.0%) | 108 (4.2%) | 0.81 |
| Cardiac | 22 (2.2%) | 62 (2.4%) | 0.79 | 25 (2.6%) | 80 (3.1%) | 0.40 |
| Noncardiac | 4 (0.4%) | 5 (0.2%) | 0.26 | 14 (1.5%) | 28 (1.1%) | 0.39 |
| Reinfarction | 17 (1.7%) | 49 (1.9%) | 0.75 | 40 (4.2%) | 98 (3.9%) | 0.68 |
| Q wave | 12 (1.2%) | 36 (1.4%) | 0.70 | 20 (2.1%) | 54 (2.1%) | 0.93 |
| Non Q wave | 6 (0.6%) | 14 (0.5%) | 0.80 | 22 (2.4%) | 47 (1.9%) | 0.40 |
| Target vessel revascularization | 20 (2.0%) | 65 (2.5%) | 0.41 | 57 (6.1%) | 173 (6.9%) | 0.35 |
| Stroke | 11 (1.1%) | 15 (0.6%) | 0.09 | 13 (1.3%) | 27 (1.1%) | 0.47 |
| Major bleeding | ||||||
| Excluding coronary bypass related | 72 (7.3%) | 179 (6.9%) | 0.66 | 77 (7.9%) | 191 (7.4%) | 0.63 |
| Including coronary bypass related | 97 (9.9%) | 230 (8.9%) | 0.35 | 104 (10.6%) | 243 (9.4%) | 0.28 |
| Blood transfusion | 32 (3.3%) | 74 (2.9%) | 0.53 | 35 (3.6%) | 83 (3.2%) | 0.59 |
| Thrombolysis In Myocardial Infarction classification | ||||||
| Major bleeding | 45 (4.6%) | 107 (4.1%) | 0.54 | 48 (4.9%) | 113 (4.4%) | 0.49 |
| Minor bleeding | 38 (3.9%) | 97 (3.8%) | 0.87 | 41 (4.2%) | 98 (3.8%) | 0.60 |
| Global Utilization of Streptokinase and Tissue plasminogen Activator for Occluded Coronary Arteries classification | ||||||
| Life-threatening/severe | 6 (0.6%) | 14 (0.5%) | 0.80 | 8 (0.8%) | 18 (0.7%) | 0.70 |
| Moderate bleeding | 49 (5.0%) | 101 (3.9%) | 0.15 | 52 (5.3%) | 109 (4.2%) | 0.16 |
| Stent thrombosis | ||||||
| Definite/probable | 15 (1.7%) | 60 (2.6%) | 0.15 | 24 (2.8%) | 79 (3.5%) | 0.36 |
| Definite | 14 (1.6%) | 46 (2.0%) | 0.49 | 22 (2.6%) | 63 (2.8%) | 0.77 |
| Probable | 1 (0.1%) | 14 (0.6%) | 0.07 | 2 (0.2%) | 16 (0.7%) | 0.12 |
| Variable | Hazard Ratio (95% confidence interval) | p Value |
|---|---|---|
| Model 1: 30-day mortality, all patients, including transfer status but not treatment times | ||
| Baseline creatinine clearance | 0.97 (0.96–0.97) | <0.0001 |
| Diabetes mellitus | 1.90 (1.19–3.02) | 0.007 |
| Killip class 2–4 | 2.94 (1.82–4.75) | <0.0001 |
| Transfer vs no transfer | 1.04 (0.64–1.68) | 0.87 |
| Model 2: 1-year mortality, all patients, including transfer status but not treatment times | ||
| Bivalirudin | 0.68 (0.47–0.98) | 0.04 |
| Baseline creatinine clearance | 0.97 (0.96–0.97) | <0.0001 |
| Anterior myocardial infarction | 1.76 (1.21–2.57) | 0.003 |
| United States | 1.78 (1.21–2.62) | 0.004 |
| History of peripheral vascular disease | 2.02 (1.14–3.57) | 0.02 |
| History of congestive heart failure | 2.03 (1.06–3.87) | 0.03 |
| Killip class 2–4 | 2.74 (1.79–4.20) | <0.0001 |
| Transfer vs no transfer | 0.89 (0.59–1.36) | 0.60 |
| Model 3: 30-day mortality, patients with percutaneous coronary intervention, including transfer status and symptom-to-balloon time | ||
| Baseline creatinine clearance | 0.97 (0.96–0.98) | <0.0001 |
| Anterior myocardial infarction | 1.62 (1.00–2.61) | 0.049 |
| Diabetes mellitus | 1.98 (1.20–3.27) | 0.008 |
| Killip class 2–4 | 3.01 (1.80–5.05) | <0.0001 |
| Transfer vs no transfer | 1.01 (0.60–1.70) | 0.98 |
| Model 4: 1-year mortality, patients with percutaneous coronary intervention, including transfer status and symptom-to-balloon time | ||
| Baseline creatinine clearance | 0.97 (0.96–0.98) | <0.0001 |
| Time from symptom onset to first hospital | 1.12 (1.04–1.21) | 0.005 |
| Anterior myocardial infarction | 2.06 (1.34–3.19) | 0.001 |
| United States | 2.10 (1.34–3.27) | 0.001 |
| Killip class 2–4 | 2.62 (1.60–4.29) | 0.0001 |
| History of congestive heart failure | 3.16 (1.57–6.33) | 0.001 |
| Transfer vs no transfer | 0.97 (0.61–1.54) | 0.89 |
Stay updated, free articles. Join our Telegram channel
Full access? Get Clinical Tree
