Skip main navigation

Pharmacoinvasive Strategy Versus Primary Percutaneous Coronary Intervention in ST-Elevation Myocardial Infarction in Clinical Practice

Insights From the Vital Heart Response Registry
Originally publishedhttps://doi.org/10.1161/CIRCINTERVENTIONS.119.008059Circulation: Cardiovascular Interventions. 2019;12:e008059

    Abstract

    Background:

    Recent clinical trial data support a pharmacoinvasive strategy as an alternative to primary percutaneous coronary intervention (pPCI) in ST-segment elevation myocardial infarction. We evaluated whether this is true in a real-world prehospital ST-segment elevation myocardial infarction network using ECG assessment of reperfusion coupled with clinical outcomes within 1 year.

    Methods:

    Of the 5583 ST-segment elevation myocardial infarction patients in the Alberta Vital Heart Response Program (Cohort 1 [2006–2011]: n=3593; Cohort 2 [2013–2016]: n=1990), we studied 3287 patients who received a pharmacoinvasive strategy with tenecteplase (April 2013: half-dose tenecteplase was employed in prehospital patients ≥75 years) or pPCI. ECGs were analyzed within a core laboratory; sum ST-segment deviation resolution ≥50% was defined as successful reperfusion. The primary composite was all-cause death, congestive heart failure, cardiogenic shock, and recurrent myocardial infarction within 1 year.

    Results:

    The pharmacoinvasive approach was administered in 1805 patients (54.9%), (493 [27.3%] underwent rescue/urgent percutaneous coronary intervention and 1312 [72.7%] had scheduled angiography); pPCI was performed in 1482 patients (45.1%). There was greater ST-segment resolution post-catheterization/percutaneous coronary intervention with a pharmacoinvasive strategy versus pPCI (75.8% versus 64.3%, IP-weighted odds ratio, 1.59; 95% CI, 1.33–1.90; P<0.001). The primary composite was significantly lower with a pharmacoinvasive approach (16.3% versus 23.1%, IP-weighted hazard ratio, 0.84; 95% CI, 0.72–0.99; P=0.033). Major bleeding and intracranial hemorrhage were similar between a pharmacoinvasive strategy and pPCI (7.6% versus 7.5%, P=0.867; 0.6% versus 0.6%; P=0.841, respectively). In the 82 patients ≥75 years with a prehospital pharmacoinvasive strategy, similar ST-segment resolution and rescue rates were observed with full-dose versus half-dose tenecteplase (75.8% versus 88.9%, P=0.259; 31.0% versus 29.2%, P=0.867) with no difference in the primary composite (31.0% versus 25.0%, P=0.585).

    Conclusions:

    In this large Canadian ST-segment elevation myocardial infarction registry, a pharmacoinvasive strategy was associated with improved ST-segment resolution and enhanced outcomes within 1 year compared with pPCI. Our findings support the application of a selective pharmacoinvasive reperfusion strategy when delay to pPCI exists.

    WHAT IS KNOWN

    • The STREAM trial (Strategic Reperfusion Early After Myocardial Infarction) tested a dedicated pharmacoinvasive reperfusion strategy (with half-dose tenecteplase in the elderly) compared with primary percutaneous coronary intervention and found similar 30-day and 1-year clinical outcomes.

    • However, a comparison of these reperfusion strategies in a real-world population of prehospital ST-elevation myocardial infarction patients with adoption of half-dose tenecteplase in the elderly has never been performed.

    WHAT THE STUDY ADDS

    • For the first time, in a large comprehensive Canadian ST-segment elevation myocardial infarction registry comparing 2 contemporary guideline recommended reperfusion strategies, namely pharmacoinvasive treatment (adopting half-dose tenecteplase in the elderly) versus primary percutaneous coronary intervention, we have demonstrated improved reperfusion, as measured by the electrocardiogram (core-laboratory), accompanied by enhanced clinical outcome within 1-year follow-up for those receiving pharmacoinvasive therapy.

    • Reassuringly, no difference in major bleeding or intracranial hemorrhage was observed within 1 year.

    • Our findings support a selective pharmacoinvasive reperfusion strategy in ST-segment elevation myocardial infarction, particularly when delays to primary percutaneous coronary intervention exists.

    Introduction

    According to current ST-segment elevation myocardial infarction (STEMI) guidelines, primary percutaneous coronary intervention (pPCI) remains the preferred reperfusion strategy for STEMI as long as it can be performed in a timely and expert fashion.1–3 However, timely access to pPCI remains a significant challenge given logistic constraints and system delays which mitigate prompt access to such care. Globally (including both North America and Europe), many STEMI patients primarily present to a non-PCI capable hospital requiring transfer to a percutaneous coronary intervention (PCI) site. In this context, between 40% and 75% do not achieve guideline recommended metrics of 120 minutes with pPCI.4–6 These patients portend a worse outcome as compared with those meeting guideline targets.6,7 In such cases, a pharmacoinvasive approach that is defined by early fibrinolysis coupled with timely PCI provides an acceptable alternative reperfusion strategy.s8

    The STREAM trial (Strategic Reperfusion Early After Myocardial Infarction) tested a dedicated pharmacoinvasive reperfusion strategy (with half-dose tenecteplase [TNK] in the elderly) compared with pPCI and found similar 30-day and 1-year clinical outcomes.9,10 Although prior work in a community hospital-based system using half-dose TNK for all patients with transfer for immediate (facilitated) angiography has established the utility of this approach,11 a comparison of a pharmacoinvasive strategy with rescue/urgent or scheduled angiography (adopting half-dose TNK in the elderly) versus pPCI in a real-world population of prehospital STEMI patients has not been undertaken. Accordingly, we compared ECG core-lab assessments of reperfusion and clinical outcomes of patients receiving pharmacoinvasive reperfusion versus pPCI in patients with STEMI followed in a large comprehensive Canadian registry.

    Methods

    The data that support the findings of this study are available from the corresponding author on reasonable request.

    Study Population and Data Collection

    A regional reperfusion protocol in northern Alberta (Canada), the Vital Heart Response (VHR) Program was established in 2006 to implement evidence-based guidelines and to deliver expeditious reperfusion therapy for STEMI patients.12 A linked quality assurance registry tracks all STEMI patients admitted to any of the 5 Edmonton (Alberta, Canada) zone hospitals including 2 tertiary care cardiac catheterization centers and 3 community hospitals (within 12 miles from a PCI-capable hospital). Patients transferred from other community hospitals to the Edmonton referral region were also included. The VHR program includes a prehospital dual reperfusion strategy where patients in the ambulance can receive fibrinolysis (absolute contraindications reviewed) or be directed for pPCI at a PCI hospital at the discretion of the physician who interprets ambulance ECGs on a mobile device. In patients receiving fibrinolysis in the ambulance, most are diverted to a PCI-capable hospital for assessment of reperfusion success. Although contemporary guidelines for STEMI patients recommend pPCI as the preferred reperfusion strategy, a large portion of northern Alberta patients receive fibrinolytic treatment at non-PCI capable hospitals because of recognized delays in achieving timely metrics. Primary PCI is performed when first medical contact (FMC) to device time of 90 to 120 minutes can be achieved. With a pharmacoinvasive approach, the need for rescue intervention is based on <50% ST-segment elevation resolution on the worst lead 90 minutes after TNK. Urgent angiography is performed when hemodynamic instability, refractory ventricular arrhythmias, ongoing ischemic chest pain, or recurrent ST-segment elevation occurs. Otherwise, those with successful reperfusion are scheduled routinely for cardiac catheterization nonurgently. Detailed data on patient demographics, treatment intervals, mode of reperfusion therapy, hospital management, and in-hospital clinical events are collected prospectively on consecutive patients by trained analysts within the VHR program.12 Postdischarge clinical events within 1 year were obtained via the Alberta Strategy for Patient Oriented Research Support Unit, a jointly funded program by Alberta Innovates and the Canadian Institutes of Health Research to support patient-oriented research, using the International Statistical Classification of Diseases and Related Health Problems (ICD)-10 codes in provincial health databases and follow-up status from the Alberta Health Care Insurance Plan registry. Those events included all-cause death, congestive heart failure (CHF), cardiogenic shock, recurrent myocardial infarction (MI), major bleeding, and intracranial hemorrhage (ICH) (Table S1 in the Data Supplement). The study cohort time frame was from 2006 to 2011 (Cohort 1) and from 2013 to 2016 (Cohort 2).12

    Reperfusion Strategies, Time to Treatment, Half-Dose TNK in the Elderly and Clinical End Points

    STEMI patients were classified into 2 groups according to the reperfusion strategy they received based on the VHR care path12: TNK-treated pharmacoinvasive or pPCI approach. The pharmacoinvasive patients were further categorized according to whether they required rescue/urgent cardiac catheterization or had scheduled angiography (aim for within 24 hours). Time to FMC was defined from symptom onset to ambulance arrival or hospital admission if the patient self-presented. Time to treatment was defined from symptom onset to the first medical device in the pPCI group and from symptom onset to time of TNK initiation in the pharmacoinvasive group. Based on the results from the STREAM trial, half-dose TNK was administered to elderly patients in the ambulance (prehospital only) after April 2013. The primary clinical composite end point included in this study was time to first occurrence of all-cause death, CHF, cardiogenic shock, or recurrent MI within 1 year. Safety end points were major bleeding and ICH within 1 year. Follow-up was complete at index hospital discharge in all patients; between discharge and 1 year, 7.2% (n=239) of patients either could not be linked with follow-up data or they left the province during the fiscal year. These patients were censored at either discharge or end of the last fiscal year in the Alberta Health Care Insurance Plan registry. The composite of all-cause death, CHF, cardiogenic shock, and recurrent MI occurring during the index hospitalization as well as individual and combined clinical end points of the composite were considered as secondary end points.

    ECG Analysis

    ECGs were collected at baseline, after reperfusion (≈90 minutes after fibrinolysis, ≈30 minutes post-pPCI for pharmacoinvasive or pPCI). They were then analyzed by experienced ECG readers at the Canadian VIGOUR Centre ECG Core Laboratory (located at the University of Alberta in Edmonton, AB, Canada) without knowledge of the clinical outcomes using standard protocols.12 ST-segment elevation and depression were measured at the J point with magnified calipers to the nearest 0.05 mV. The sum total across all leads except aVR was used to calculate ST-elevation sums and ST-depression sums. Then, the total ST-deviation was calculated by adding the sums of ST-elevation to ST-depression. The percent resolution of sum ST-deviation from baseline to post-PCI/angiogram was dichotomized in accordance with guidelines issued by the European Society of Cardiology and the American College of Cardiology/American Heart Association as either ≥50% or <50% as our primary metric for reperfusion (primary ECG end point).2,3 We intentionally chose the sum of ST-deviation resolution as our primary metric for evaluation given prior work from our ECG laboratory showing incremental advantage over ST-elevation resolution alone.13 We also report worst lead residual ST-elevation as well as worst lead ST-elevation resolution dichotomized, as either ≥50% or <50%, based on guideline recommendations and ease of assessment from our prior work.14

    Statistical Analysis

    Categorical variables are reported as percentages, whereas continuous variables are presented as medians with 25th and 75th percentiles. Differences between groups (pharmacoinvasive versus pPCI, rescue/urgent angiography versus pPCI, scheduled angiography versus pPCI) were tested using the χ2 test for categorical variables and using Wilcoxon rank-sum test for continuous variables. In a prespecified analysis, we compared pPCI to pharmacoinvasive scheduled angiography and pharmacoinvasive rescue/urgent angiography as was done in a post hoc analysis of the STREAM trial.15

    Kaplan-Meier curves were plotted to display the unadjusted relationship between reperfusion strategy and the primary clinical composite end point (ie, the time to the first occurrence of death, CHF, cardiogenic shock, or recurrent MI within 1 year) with comparison between groups using the log-rank test. Kaplan-Meier estimated rates and 95% CI within 1 year were also reported. Landmark analysis was conducted at day 30 to examine the later-term association, and the log-rank test was performed across the treatment groups. The relative association between reperfusion strategy and primary clinical composite end point was examined using Cox proportional hazard regression. Unadjusted and adjusted hazard ratios (HRs) and 95% CI were reported. To account for selection bias and confounders, this association was adjusted using an inverse probability weighting approach.16 For this inverse probability weighting analysis, a propensity model for the administration of either reperfusion strategy (pharmacoinvasive or pPCI) was first developed using logistic regression, and then the inverse of these predicted probabilities was used as weights in the regression model for the primary clinical composite end point. Patient characteristics available at FMC (when a decision for therapeutic approach would be made), such as age, sex, history of diabetes mellitus, history of hypertension, history of angina, prior MI, family history of coronary artery disease, history of hypercholesterolemia, systolic blood pressure, heart rate, body mass index, current smoker, inferior MI and time from symptom onset to FMC, were forced into the propensity model. A restricted cubic spline function was used to test the linearity assumption for continuous variables in the model. Spline transformations were applied when linearity assumption was violated. To examine homogeneity between the 2 study cohorts, interactions between reperfusion strategy and cohorts on the primary clinical composite end point were tested.

    The relative associations between (1), reperfusion strategy and reperfusion success (ie, sum ST-segment deviation resolution ≥50%), and (2), between reperfusion strategy and in-hospital composite end point of death, CHF, cardiogenic shock, and recurrent MI were estimated using logistic regression with (adjusted) and without (unadjusted) the inverse probability weighting, as described above. Odds ratio (OR) and 95% CI were reported.

    A descriptive analysis was performed in elderly patients (≥75 years) according to dosage of TNK administered. Observed primary and secondary end points as well as major bleeding and ICH were reported.

    All statistical tests were 2-sided with P<0.05 considered as statistically significant. Statistical analyses were performed using SAS (version 9.4; Cary, NC). No correction was made for multiple comparison. The study was approved by the University of Alberta Ethics Review Board with individual consent waived because of privacy rules related to this quality assurance registry.

    Results

    Patient Population

    Of the 5583 patients consecutively admitted or transferred to hospitals in the Alberta VHR STEMI Network Registry (Cohort 1 [2006–2011]: n=3593; Cohort 2 [2013–2016]: n=1990), 3287 patients were included in our analysis (Figure 1). Two thousand two hundred ninety-six patients were excluded for the following reasons: (1) late presentation (>12 hours from symptom onset to treatment) or unknown ischemic time (n=986), (2) baseline ST-segment elevation could not be obtained or interpreted (n=562), (3) acute reperfusion was not administered (n=508), (4) baseline ECG was a left bundle branch block or paced rhythm (n=100), (5) STEMI developed while admitted to hospital for another reason (n=78), or (6) fibrinolysis was administered without receiving subsequent cardiac catheterization (n=62). Comparison of study cohort and excluded patients can be seen in Table S2 in the Data Supplement). Among pharmacoinvasive patients, only TNK was used as a fibrin-specific agent. Of these, all patients underwent subsequent coronary angiography and 84.2% had a PCI (4.2% had coronary artery bypass graft). The median time from successful fibrinolysis to nonurgent cardiac catheterization (pharmacoinvasive scheduled) was 23.4 hours. In patients where pPCI was intended as the reperfusion strategy, 91.6% received PCI (1.5% had coronary artery bypass graft).

    Figure 1.

    Figure 1. Study cohort. LBBB indicates left bundle branch block; PCI, percutaneous coronary intervention; and STEMI, ST-elevation myocardial infarction.

    Baseline Characteristics According to Reperfusion Therapy Administered

    Selected patient baseline characteristics are presented in Table 1 according to acute reperfusion strategy employed. Among patients undergoing reperfusion therapy for STEMI (Table 1, left), 45.1% (n=1482) received pPCI and 54.9% (n=1805) received a pharmacoinvasive approach. Compared with pPCI, pharmacoinvasive patients were younger, less often female, and heavier (greater body mass index) but had a lower baseline heart rate. Pharmacoinvasive patients were less likely to have a history of hypertension and atrial fibrillation but were more commonly active smokers and had a premature history of coronary artery disease. A high use of evidence-based medications in the acute setting were noted with both groups; however, pharmacoinvasive patients had a slightly higher use of a P2Y12 receptor antagonist and anticoagulants (unfractionated heparin or enoxaparin) but a substantially lower use of a glycoprotein IIb/IIIa inhibitors compared with pPCI. Slightly higher use of β-blockers and renin-angiotensin system blockers were noted in pharmacoinvasive patients. In-hospital medication use in the 2 VHR cohorts is outlined in Table S3 in the Data Supplement according to assigned reperfusion therapy. A greater use of more potent P2Y12 inhibitors were used in VHR cohort 2 in patients undergoing pPCI consistent with contemporary evidence. Infarct location was more commonly inferior (as opposed to anterior) in pharmacoinvasive patients, and these patients had a shorter symptom onset to FMC and shorter symptom onset to treatment (administration of any reperfusion therapy). PCI-related delay (defined as the difference between symptom onset to treatment between reperfusion strategies) was 84 minutes. Symptom onset to treatment >3 hours was substantially less in the pharmacoinvasive patients. Similar length of hospital stay was noted between both reperfusion strategies.

    Table 1. Selected Patient Characteristics According to Reperfusion Therapy

    NpPCIPIP*PIP
    RescueScheduled
    148218054931312
    Age, y60 (51–70)58 (51–65)<0.00157 (51–65)58 (50–65)0.946
    Female, n (%)362 (24.4)344 (19.1)<0.00177 (15.6)267 (20.4)0.023
    Weight, kg82 (72–98)86 (75–98)<0.00187 (77–100)85 (75–97)0.018
    BMI, kg/m228 (25–32)29 (26–32)<0.00129 (26–32)29 (26–32)0.274
    Heart rate, beats per minute76 (62–89)72 (61–86)<0.00171 (61–86)73 (61–86)0.422
    Systolic blood pressure, mm Hg142 (120–162)139 (120–159)0.183137 (118–155)140 (120–160)0.044
    Diastolic blood pressure, mm Hg87 (73–101)87 (74–99)0.81685 (72–98)88 (74–100)0.326
    Comorbidities, n(%)
     Hypercholesterolemia578 (39.0)723 (40.1)0.539191 (38.7)532 (40.5)0.485
     Hypertension709 (47.8)793 (43.9)0.025205 (41.6)588 (44.8)0.217
     Diabetes mellitus274 (18.5)293 (16.2)0.08983 (16.8)210 (16.0)0.670
     Current smoker599 (40.4)931 (51.6)<0.001237 (48.1)694 (52.9)0.068
     Family history of CAD267 (18.0)465 (25.8)<0.001108 (21.9)357 (27.2)0.022
    Medical history, n (%)
     CAD90 (16.1)102 (15.7)0.83634 (16.0)68 (15.6)0.895
     MI225 (15.2)254 (14.1)0.36959 (12.0)195 (14.9)0.115
     Coronary revascularization185 (13.3)200 (11.5)0.12644 (9.3)156 (12.3)0.077
     Atrial fibrillation36 (2.4)19 (1.1)0.0026 (1.2)13 (1.0)0.675
     Stroke/TIA29/558 (5.2)16/650 (2.5)0.0121/213 (0.5)15/437 (3.4)0.022
    In-hospital medications n (%)
     Aspirin1470 (99.2)1796 (99.5)0.265488 (99.0)1308 (99.7)0.057
     Clopidogrel/ticagrelor1446 (97.6)1780 (98.6)0.027491 (99.6)1289 (98.2)0.029
     Glycoprotein IIb/IIIa inhibitors937 (63.2)459 (25.4)<0.001181 (36.7)278 (21.2)<0.001
     Anticoagulant1394 (94.1)1785 (98.9)<0.001481 (97.6)1304 (99.4)0.001
     Cholesterol-lowering medication1408 (95.0)1730 (95.8)0.250471 (95.5)1259 (96.0)0.688
     β-Blocker1408 (95.0)1754 (97.2)0.001478 (97.0)1276 (97.3)0.733
    ACEi/ARB1358 (91.6)1698 (94.1)0.007467 (94.7)1231 (93.8)0.471
    Presentation
     Inferior MI (ECG), n (%)609 (41.1)948 (52.5)<0.001254 (51.5)694 (52.9)0.602
     Symptom onset to first medical contact, min79 (37–158)70 (33–151)0.04665 (30–128)72 (35–162)0.008
     Symptom onset to treatment, min212 (138–349)128 (80–210)<0.001111 (71–180)133 (83–221)<0.001
     Symptom onset to treatment >3 h, n (%)881 (59.4)581 (32.2)<0.001123 (24.9)458 (34.9)<0.001
     Length of hospital stay, d5 (4–7)5 (4–7)0.0695 (4–6)5 (4–7)<0.001

    Continuous variables presented as median (25th–75th percentiles). ACEi/ARB indicates angiotensin-converting enzyme inhibitor/angiotensin receptor blocker; BMI, body mass index; CAD, coronary artery disease; MI, myocardial infarction; PI, pharmacoinvasive; pPCI, primary percutaneous coronary intervention; and TIA, transient ischemic attack.

    *P comparison between pPCI and PI.

    P comparison between rescue and scheduled.

    Selected baseline characteristics in pharmacoinvasive patients according to the need for a rescue/urgent procedure compared with a scheduled procedure is also presented in Table 1. Among patients undergoing a pharmacoinvasive approach (Table 1, right), 27.3% (n=493) required a rescue/urgent procedure and 72.7% received a scheduled procedure (n=1312). Compared with a scheduled procedure, rescue/urgent patients were less likely female, had a higher body weight, and had a lower baseline systolic blood pressure. There were fewer patients with a family history of coronary artery disease, a trend towards fewer patients with prior coronary revascularization procedures and fewer active smokers in those requiring a rescue/urgent procedure. A high use of evidence-based medications in-hospital was observed, however rescue/urgent patients had a marginally higher use of a P2Y12 receptor antagonist and a slightly lower use of an anticoagulant (unfractionated heparin or enoxaparin). There was a higher use of a glycoprotein IIb/IIIa inhibitor with rescue/urgent patients compared with those receiving a scheduled procedure but not as high as that employed in the pPCI group. Infarct location was similar between both groups, however; rescue/urgent patients had a shorter symptom onset to FMC and shorter symptom onset to treatment with TNK.

    Core Laboratory-Evaluated ECG Characteristics at Baseline and Following Reperfusion Therapy

    ECG metrics at baseline and following reperfusion therapy administered are shown in Table 2. Compared with pPCI (Table 2, left), pharmacoinvasive patients had similar symptom onset to baseline ECG (ie, total ischemic time before the reperfusion decision) with similar ST-segment elevation but greater sum ST-segment deviation. Compared with pPCI, pharmacoinvasive patients had shorter baseline ECG to treatment time (63 minutes). Following the initiation of acute reperfusion therapy (≈30 minutes after pPCI or ≈90 minutes after TNK alone), residual ST-elevation was higher in patients receiving TNK (P<0.001) with less ST-segment resolution compared with pPCI (P<0.001; Table 2, left), largely due to the higher residual ST-elevation with less ST-segment resolution seen in rescue/scheduled patients (Table 2, right). However, after cardiac catheterization±PCI (≈30 minutes following primary PCI or ≈30 minutes following a pharmacoinvasive approach), residual ST-elevation was higher in pPCI with less ST-segment resolution (Table 2, left).

    Table 2. ECG Characteristics According to Reperfusion Therapy

    NpPCIPIP*PIP
    RescueScheduled
    148218054931312
    Baseline ECG
     Symptom onset to baseline ECG, min92 (48–180)92 (50–176)0.98378 (43–148)97 (54–187)<0.001
     Baseline ECG to treatment, min89 (60–146)26 (15–48)<0.00124 (16–44)27 (15–50)0.248
     Baseline Q waves, %572 (38.9)658 (36.6)0.183189 (38.6)469 (35.9)0.292
     Worst lead ST-E, mm2 (2–4)3 (2–4)0.1703 (2–4)2 (2–4)<0.001
     Sum ST-deviation, mm10 (7–16)11 (7–17)0.00113 (9–19)11 (7–16)<0.001
    Post-fibrinolysis ECG
     Symptom onset to post-fibrinolysis ECG, min213 (160–299)186 (141–244)224 (168–318)<0.001
     Fibrinolysis to post-fibrinolysis ECG, min82 (65–94)79 (65–91)84 (66–94)0.044
     Worst lead residual ST-E post-fibrinolysis, %<0.001
      <1 mm460 (40.4)59 (18.9)401 (48.5)
      1 to <2 mm350 (30.8)104 (33.3)246 (29.8)
      ≥2 mm328 (28.8)149 (47.8)179 (21.7)
     Worst lead ST-E resolution (baseline to post-fibrinolysis) ≥50%, n/N (%)704/1069 (65.9)137/296 (46.3)567/773 (73.4)<0.001
     Sum ST-deviation resolution (baseline to post-fibrinolysis) ≥50%, n/N (%)628/1137 (55.2)103/312 (33.0)525/825 (63.6)<0.001
    Postcardiac catheterization/PCI ECG
     No. of patients receiving PCI, n (%)1357 (91.6)1520 (84.2)<0.001490 (99.4)1030 (78.5)<0.001
     No. of patients receiving CABG, n (%)22 (1.5)76 (4.2)<0.0013 (0.6)73 (5.6)<0.001
     Symptom onset to post-PCI ECG, min326 (227–521)953 (483–2254)<0.001436 (341–556)1765 (989–3460)<0.001
     Worst lead residual ST-E post-PCI, %<0.0010.002
      <1 mm576 (47.6)651 (57.6)223 (51.3)428 (61.5)
      1 to <2 mm412 (34.1)342 (30.2)145 (33.3)197 (28.3)
      ≥2 mm221 (18.3)138 (12.2)67 (15.4)71 (10.2)
     Worst lead ST-E resolution (baseline to post-PCI ECG) ≥50%, %849/1085 (78.2)921/1065 (86.5)<0.001356/417 (85.4)565/648 (87.2)0.400
     Sum ST-deviation resolution (baseline to post-PCI ECG) ≥50%, %776/1207 (64.3)857/1130 (75.8)<0.001328/435 (75.4)529/695 (76.1)0.785

    Continuous variables presented as median (25th–75th percentiles). CABG indicates coronary artery bypass graft; PI, pharmacoinvasive; PCI, percutaneous coronary intervention; and pPCI, primary PCI.

    *P comparison between pPCI and pharmacoinvasive approach.

    P comparison between rescue and scheduled.

    The pharmacoinvasive strategy was associated with improved ST-segment resolution after cardiac catheterization as seen in Figure 2 (IP-weighted OR, 1.59; 95% CI, 1.33–1.90; P<0.001). When compared with pPCI, both scheduled and rescue/urgent strategies were associated with improved sum ST-segment deviation resolution ≥50% after cardiac catheterization (Figure 2).

    Figure 2.

    Figure 2. Association between treatment strategy (pharmacoinvasive [PI] vs primary percutaneous coronary intervention [pPCI] upper, rescue/scheduled vs pPCI lower) and reperfusion success measured as sum ST-segment deviation resolution ≥50% post-PCI/cath. IP indicates inverse probability; and OR, odds ratio.

    In-Hospital Clinical Outcomes According to Reperfusion Therapy

    In-hospital outcomes, including the composite end point of death, CHF, cardiogenic shock, recurrent MI and its components are summarized in Table 3. A pharmacoinvasive strategy, as compared with pPCI, was associated with a reduction in the in-hospital composite outcome (IP-weighted OR, 0.80; 95% CI, 0.66–0.98; P=0.029). Compared with pPCI, a scheduled pharmacoinvasive strategy was associated with lower likelihood of these in-hospital events (IP-weighted OR, 0.72; 95% CI, 0.58–0.90; P=0.004). This was not observed with a rescue/urgent strategy (IP-weighted OR, 1.03; 95% CI, 0.77–1.36; P=0.857).

    Table 3. In-Hospital Clinical Outcomes According to Reperfusion Therapy

    npPCIPIP*PIP
    RescueScheduled
    148218054931312
    Efficacy
     Composite of death, CHF, cardiogenic shock and recurrent MI, n (%)247 (16.7)204 (11.3)<0.00169 (14.0)135 (10.3)0.027
     Death/recurrent MI, n (%)85 (5.7)45 (2.5)<0.00115 (3.0)30 (2.3)0.359
     Death/cardiogenic shock/CHF, n (%)243 (16.4)196 (10.9)<0.00166 (13.4)130 (9.9)0.034
     Cardiogenic shock/CHF, n (%)224 (15.1)188 (10.4)<0.00163 (12.8)125 (9.5)0.044
     Death, n (%)77 (5.2)34 (1.9)<0.00112 (2.4)22 (1.7)0.292
     CHF, n (%)118 (8.0)92 (5.1)0.00129 (5.9)63 (4.8)0.352
     Cardiogenic shock, n (%)156 (10.5)123 (6.8)<0.00144 (8.9)79 (6.0)0.029
     Recurrent MI, n (%)11 (0.7)14 (0.8)0.9133 (0.6)11 (0.8)0.620
    Safety
     Major bleeding, n (%)98 (6.6)117 (6.5)0.88033 (6.7)84 (6.4)0.823
     ICH, n (%)5 (0.3)5 (0.3)0.7552 (0.4)3 (0.2)0.524

    CHF indicates congestive heart failure; ICH, intracranial hemorrhage; MI, myocardial infarction; PI, pharmacoinvasive; and pPCI, primary percutaneous coronary intervention.

    *P comparison between pPCI and PI.

    P comparison between rescue and scheduled.

    Long-Term Clinical Outcomes According to Reperfusion Therapy

    Clinical outcomes occurring within 1 year are summarized in Table 4. As seen in the central illustration, the longer-term primary clinical composite end point of death, CHF, cardiogenic shock, and recurrent MI was lower with a pharmacoinvasive strategy compared with pPCI. Mortality, cardiogenic shock, and CHF were similarly lower (Table 4). In a landmark analysis conducted at 30 days, this difference in the primary composite favoring a pharmacoinvasive strategy appeared to widen over the ensuing year (P log-rank <0.001; central illustration right lower). As seen in Figure 3, a pharmacoinvasive approach compared with pPCI was associated with a nearly 20% lower hazard of the primary composite of death, CHF, cardiogenic shock, and recurrent MI within 1 year (IP-weighted HR, 0.84; 95% CI 0.72–0.99, P=0.033). Compared with pPCI, a scheduled pharmacoinvasive strategy was associated with a lower primary composite event rate within 1 year (IP-weighted HR, 0.76; 95% CI, 0.64–0.91; P=0.002; Figure 3). This was not observed with a rescue/urgent strategy (IP-weighted HR, 1.08; 95% CI, 0.87–1.35; P=0.502; Figure 3). As was the case in hospital, the lesser hazard with pharmacoinvasive therapy also applied to both cardiogenic shock and heart failure but not re-MI (Table 4).

    Table 4. Clinical Outcomes Within 1 Year According to Reperfusion Therapy

    NpPCIPIP*PIP
    RescueScheduled
    148218054931312
    Observed Rate, n (%)KM Estimated Rate % (95% CI)Observed Rate, n (%)KM Estimated Rate % (95% CI)Observed Rate, n (%)KM Estimated Rate % (95% CI)Observed Rate, n (%)KM Estimated Rate % (95% CI)
    Efficacy
     Composite of death, CHF, cardiogenic shock, and recurrent MI343 (23.1)23.9 (21.8–26.2)294 (16.3)17.0 (15.2–18.8)<0.00197 (19.7)20.5 (17.2–24.5)197 (15.0)15.7 (13.8–17.8)0.017
     Death/recurrent MI171 (11.5)12.8 (11.1–14.7)106 (5.9)6.6 (5.4–7.9)<0.00133 (6.7)7.5 (5.4–10.4)73 (5.6)6.2 (5.0–7.8)0.363
     Death/cardiogenic shock/CHF301 (20.3)21.1 (19.1–23.3)251 (13.9)14.6 (13.1–16.4)<0.00183 (16.8)17.7 (14.5–21.5)168 (12.8)13.5 (11.7–15.5)0.027
     Cardiogenic shock/CHF262 (17.7)18.6 (16.7–20.7)227 (12.6)13.3 (11.8–15.0)<0.00178 (15.8)16.8 (13.7–20.5)149 (11.4)12.0 (10.3–13.9)0.011
     Death124 (8.4)9.1 (7.7–10.8)58 (3.2)3.5 (2.7–4.6)<0.00114 (2.8)3.0 (1.8–5.0)44 (3.4)3.7 (2.8–5.0)0.581
     CHF162 (10.9)12.4 (10.7–14.3)138 (7.6)8.4 (7.2–9.9)0.00146 (9.3)10.4 (7.9–13.6)92 (7.0)7.7 (6.3–9.4)0.099
     Cardiogenic shock156 (10.5)11.2 (9.7–13.0)123 (6.8)7.4 (6.2–8.8)<0.00144 (8.9)9.7 (7.3–12.9)79 (6.0)6.5 (5.3–8.1)0.029
     Recurrent MI68 (4.6)5.7 (4.5–7.2)62 (3.4)4.0 (3.1–5.1)0.09119 (3.9)4.7 (3.0–7.2)43 (3.3)3.7 (2.8–5.0)0.549
    Safety
     Major bleeding, n (%)111 (7.5)7.9 (6.2–10.8)138 (7.6)8.1 (6.3–11.1)0.86733 (6.7)7.4 (5.5–9.5)84 (6.4)7.1 (5.2–9.3)0.823
     ICH, n (%)9 (0.6)0.7 (0.4–1.4)10 (0.5)0.7 (0.4–1.3)0.8414 (0.8)1.0 (0.4–2.8)6 (0.5)0.6 (0.3–1.3)0.367

    CHF indicates congestive heart failure; ICH, intracranial hemorrhage; KM, Kaplan-Meier; MI, myocardial infarction; PI, pharmacoinvasive; and pPCI, primary percutaneous coronary intervention.

    *P comparison of observed rates between pPCI and PI.

    P comparison of observed rates between rescue and scheduled.

    Figure 3.

    Figure 3. Association between treatment strategy (pharmacoinvasive [PI] vs primary percutaneous coronary intervention [pPCI] upper, rescue/scheduled vs pPCI lower) and time to first occurrence of all-cause death, congestive heart failure (CHF), cardiogenic shock, or recurrent myocardial infarction (MI) within 1 year. IP indicates inverse probability; and HR, hazard ratio.

    No heterogeneity in outcome between a pharmacoinvasive strategy and pPCI was observed between chronological cohorts (P-interaction=0.740). No heterogeneity was observed in the primary end point when stratified by time from symptom onset to FMC (P-interaction=0.898).

    Finally, we considered 2 alternative methods for accounting for treatment bias. First, we reassessed the association between treatment and the primary composite end point within 1 year by including the individual covariates in the model and found similar results (adjusted HR, 0.84; 95% CI, 0.71–0.99; P=0.04). Second, we included the propensity of the treatment (ie, estimated probability of receiving pharmacoinvasive treatment versus pPCI) as a covariate in the model, and again, the estimated association was supportive of our original findings for the primary composite end point within 1 year (adjusted HR, 0.83; 95% CI, 0.71–0.98; P=0.03).

    Major bleeding events were not different in those receiving a pharmacoinvasive approach compared with pPCI (7.6% versus 7.5%, P=0.867) within 1 year. Similarly, no difference in ICH was observed recognizing the implementation of half-dose TNK in the elderly prehospital patients in April 2013 (0.6% versus 0.6%, P=0.841).

    Half-Dose Tenecteplase as a Pharmacoinvasive Approach in the Elderly

    In the 428 elderly patients (≥75 years), the observed primary clinical composite end point was higher when compared with younger patients, irrespective of reperfusion strategy (pharmacoinvasive group [n=158]: 30.4% versus 13.5%, P<0.001; pPCI group [n=270] 30.7% versus 16.8%, P<0.001). Similar findings were observed for major bleeding and ICH within 1 year (pharmacoinvasive group: 15.2% versus 6.9%, P<0.001; pPCI group: 13.7% versus 6.1%, P<0.001).

    In the 82 elderly patients with a prehospital pharmacoinvasive strategy, 24 (29.3%) received half-dose TNK. Similar sum ST-segment deviation resolution ≥50% and rescue intervention rates were observed irrespective of TNK dose (full-dose TNK: 75.8% versus half-dose TNK: 88.9%, P=0.259; full-dose TNK: 31.0% versus half-dose TNK: 29.2%, P=0.867, respectively). The primary clinical composite end point was 31.0% with full-dose TNK and 25.0% with half-dose TNK (P=0.585). Major bleeding and ICH was 17.2% with full-dose TNK and 16.7% with half-dose TNK (P=0.950).

    In the overall elderly patient population (n=428), ST-segment resolution was greater with pharmacoinvasive compared with pPCI (79.3% versus 59.3%; P=0.006) with similar primary composite within 1 year (29.3% versus 31.6%; P=0.698).

    Discussion

    In this large comprehensive STEMI registry, utilizing 2 contemporary guideline recommended reperfusion strategies, several novel findings were observed. Patients selected to receive pharmacoinvasive treatment, compared with pPCI, showed improved reperfusion as measured by the ECG parameters, which were accompanied by enhanced clinical outcome both at hospital discharge and at follow-up within 1 year. Reassuringly and congruent with prior findings, roughly one-quarter of the pharmacoinvasive patients required rescue/urgent angiography and experienced acceptable clinical outcomes. Patients with successful fibrinolysis triaged to scheduled angiography had superior ST-segment resolution with best clinical outcomes.

    The STREAM trial provided justification for a pharmacoinvasive strategy in STEMI patients who could not undergo pPCI within 1 hour of FMC. In the 1892 patients who presented within 3 hours after symptom onset, similar rates of death, cardiogenic shock, CHF, or recurrent MI occurred at 30 days and 1 year with pharmacoinvasive compared with pPCI.9,10 However, in a subanalysis, superior results were noted with a pharmacoinvasive strategy when delays to pPCI were prolonged.17 Our real-world evidence from the VHR STEMI Registry demonstrates improved clinical outcomes for pharmacoinvasive treated patients at 1 year. This benefit may be related in part to the differences in the ischemic time intervals between reperfusion therapies (84 minutes) in the current study as compared with STREAM (78 minutes). Furthermore, 59.4% of pPCI patients within the registry had ischemic times >3 hours as compared with 32.2% with a pharmacoinvasive strategy.

    To our knowledge, our study is the first to show superior outcome with a dedicated pharmacoinvasive strategy compared with pPCI in a clinical registry. The French registry on Acute ST-elevation Myocardial Infarction have reported similar 1- and 5-year outcomes of STEMI patients who received fibrinolysis (roughly two-thirds prehospital) followed by coronary angiography versus pPCI within 12 hours of symptoms.18,19 The Ottawa Heart Institute reported no difference in the composite of in-hospital death, recurrent MI or stroke in STEMI patients receiving pharmacoinvasive versus pPCI strategy as part of a regional STEMI system registry (without prehospital activation) where geographic proximity to a PCI center dictates reperfusion strategy.20 The Mayo Clinic STEMI network reported no difference in long-term mortality with pharmacoinvasive compared with pPCI patients presenting to non-PCI capable hospitals (even in patients with short door-to-balloon times).21 The Korea Acute Myocardial Infarction Registry investigators reported similar rates of death and major adverse cardiac events (composite of death, recurrent MI, target-vessel revascularization, and coronary artery bypass graft surgery) from their 12-month propensity-matched STEMI patients receiving either in-hospital fibrinolysis with coronary angiography or pPCI.22

    To help explain our findings, we think that the implementation of a balanced prehospital reperfusion strategy used in our cohort facilitates best practice for patients based on clinical presentation and total ischemic time. Symptom onset to treatment (reperfusion) was short with fibrinolysis compared with pPCI (median time: 128 minutes versus 212 minutes) and reflects the reduction in total ischemic time that accounts (in part) for improved outcomes observed in the current study. In addition, we observed a significant reduction in the incidence of both in-hospital cardiogenic shock and heart failure with a pharmacoinvasive strategy, which was sustained up to 1 year, and likely accounts for the concomitant reduction in in-hospital mortality and a remarkably low 1-year mortality of 3.2% as compared with the pPCI patients. The low rates of ICH and systemic bleeding, likely related to careful patient selection, are also probable contributors to our results. Overall, this explains the reduction in the composite outcome seen in our study (unique compared with other registry studies) and aligns with a recent meta-analysis.23 It is interesting to note the landmark analyses performed at 30 days demonstrates a continued lower event rate with a pharmacoinvasive strategy providing further support that earlier reperfusion with fibrinolysis likely reduces infarct size leading to reductions in subsequent CHF within the first year after STEMI. However, this speculation deserves confirmation in a randomized prospective study.

    Unique to this registry are the ECG core laboratory measurements of ST-segment resolution after reperfusion. Using the sum of ST-segment deviation resolution as a correlate of myocardial perfusion (and predictor of outcomes),14 we found improved ST-segment recovery following coronary angiography with enhanced clinical outcome within 1 year with a pharmacoinvasive strategy compared with pPCI. This was most notable in pharmacoinvasive patients receiving scheduled angiography affirming timely TNK if successful leads to faster tissue reperfusion of jeopardized myocardium. Supportive findings are noted in the In the EARLY-MYO trial (Early Routine Catheterization After Alteplase Fibrinolysis Versus Primary PCI in Acute ST-Segment–Elevation Myocardial Infarction) where epicardial and myocardial perfusion (defined as thrombolysis in MI flow grade 3, thrombolysis in MI myocardial perfusion grade 3, and ST-segment resolution ≥70%) were improved using a half-dose alteplase pharmacoinvasive approach compared with pPCI in STEMI patients presenting ≤6 hours after symptoms.24

    In the elderly, prehospital half-dose TNK as part of a pharmacoinvasive strategy appeared safe and as effective as a full-dose strategy based on ST resolution data and similar rates of rescue PCI. In this regard, it is relevant to note the results from the Minneapolis Heart Institute where all patients (irrespective of age) whose geographic location precludes pPCI, receive half-dose fibrinolysis (most commonly TNK) with transfer for immediate PCI. When compared with those receiving pPCI, Larson et al found no difference in 30-day clinical outcomes and major bleeding rates were similar supporting the safety of this regimen.11 Given the small number of patients in our elderly subset, these data remain hypothesis generating and this is currently being tested in the ongoing STREAM-2 trial (Strategic Reperfusion in Elderly Patients Early After Myocardial Infarction trial; NCT02777580).

    Our study has strengths and limitations. The current study on contemporary STEMI treatment data is acquired from systematic chart review and provincial claims data, which is subject to unmeasured confounders and constitutes an inherent limitation of observational studies. The lack of baseline data about cardiogenic shock, Killip class, and cardiac arrest is a limitation that could account for residual confounding and contribute to the improved clinical outcomes seen with a pharmacoinvasive strategy. However, propensity-score inverse probability weighting was intentionally selected to mitigate bias for the treatment selected in this setting. Selection bias by the clinicians managing STEMI patients within the Alberta VHR program was likely operational as they chose a reperfusion therapy influenced by the STEMI risk, the risk of fibrinolysis, ischemic time, and anticipated PCI-related delay. Notably, these choices were made by experienced interventionalists from the only 2 pPCI centres, serving a population of >1 million residents and they performed pPCI procedures using state of the art guideline-based techniques and concomitant therapies. As commonly occurs in clinical practice, the majority of pharmacoinvasive patients received scheduled coronary angiography, and the post-PCI ECG was performed much later than pPCI patients and may have influenced the improved ST-segment resolution. Finally, late presentation STEMI patients (>12 hours of symptoms) were excluded from our study given the lack of efficacy with fibrinolysis in these patients.

    Conclusions

    In a large comprehensive STEMI network registry based in an integrated prehospital and community dual reperfusion program, a selective pharmacoinvasive strategy was associated with improved ST resolution after PCI with enhanced clinical outcome within 1 year compared with pPCI. Our findings support the application of a selective pharmacoinvasive reperfusion strategy particularly when rapid access to pPCI is challenging to achieve.

    Acknowledgments

    We thank Lisa Soulard for her editorial assistance.

    Footnotes

    The Data Supplement is available at https://www.ahajournals.org/doi/suppl/10.1161/CIRCINTERVENTIONS.119.008059.

    Kevin R. Bainey, MD, MSc, University of Alberta Hospital, 2C2.12 WMC, 8440 112 St, Edmonton, AB T6G 2B7, Canada. Email

    References

    • 1. Wong GC, Welsford M, Ainsworth C, Abuzeid W, Fordyce CB, Greene J, Huynh T, Lambert L, Le May M, Lutchmedial S, Mehta SR, Natarajan M, Norris CM, Overgaard CB, Perry Arnesen M, Quraishi A, Tanguay JF, Traboulsi M, van Diepen S, Welsh R, Wood DA, Cantor WJ; members of the Secondary Panel. 2019 Canadian Cardiovascular Society/Canadian Association of Interventional Cardiology guidelines on the acute management of ST-elevation myocardial infarction: focused update on regionalization and reperfusion.Can J Cardiol. 2019; 35:107–132. doi: 10.1016/j.cjca.2018.11.031CrossrefMedlineGoogle Scholar
    • 2. Ibanez B, James S, Agewall S, Antunes MJ, Bucciarelli-Ducci C, Bueno H, Caforio ALP, Crea F, Goudevenos JA, Halvorsen S, Hindricks G, Kastrati A, Lenzen MJ, Prescott E, Roffi M, Valgimigli M, Varenhorst C, Vranckx P, Widimský P; ESC Scientific Document Group. 2017 ESC Guidelines for the management of acute myocardial infarction in patients presenting with ST-segment elevation: The Task Force for the management of acute myocardial infarction in patients presenting with ST-segment elevation of the European Society of Cardiology (ESC).Eur Heart J. 2018; 39:119–177. doi: 10.1093/eurheartj/ehx393CrossrefMedlineGoogle Scholar
    • 3. O’Gara PT, Kushner FG, Ascheim DD, Casey DE, Chung MK, de Lemos JA, Ettinger SM, Fang JC, Fesmire FM, Franklin BA, Granger CB, Krumholz HM, Linderbaum JA, Morrow DA, Newby LK, Ornato JP, Ou N, Radford MJ, Tamis-Holland JE, Tommaso CL, Tracy CM, Woo YJ, Zhao DX, Anderson JL, Jacobs AK, Halperin JL, Albert NM, Brindis RG, Creager MA, DeMets D, Guyton RA, Hochman JS, Kovacs RJ, Kushner FG, Ohman EM, Stevenson WG, Yancy CW; American College of Cardiology Foundation/American Heart Association Task Force on Practice Guidelines. 2013 ACCF/AHA guideline for the management of ST-elevation myocardial infarction: a report of the American College of Cardiology Foundation/American Heart Association Task Force on Practice Guidelines.Circulation. 2013; 127:e362–e425. doi: 10.1161/CIR.0b013e3182742cf6LinkGoogle Scholar
    • 4. Vora AN, Holmes DN, Rokos I, Roe MT, Granger CB, French WJ, Antman E, Henry TD, Thomas L, Bates ER, Wang TY. Fibrinolysis use among patients requiring interhospital transfer for ST-segment elevation myocardial infarction care: a report from the US National Cardiovascular Data Registry.JAMA Intern Med. 2015; 175:207–215. doi: 10.1001/jamainternmed.2014.6573CrossrefMedlineGoogle Scholar
    • 5. Terkelsen CJ, Sørensen JT, Maeng M, Jensen LO, Tilsted HH, Trautner S, Vach W, Johnsen SP, Thuesen L, Lassen JF. System delay and mortality among patients with STEMI treated with primary percutaneous coronary intervention.JAMA. 2010; 304:763–771. doi: 10.1001/jama.2010.1139CrossrefMedlineGoogle Scholar
    • 6. Lambert L, Brown K, Segal E, Brophy J, Rodes-Cabau J, Bogaty P. Association between timeliness of reperfusion therapy and clinical outcomes in ST-elevation myocardial infarction.JAMA. 2010; 303:2148–2155. doi: 10.1001/jama.2010.712CrossrefMedlineGoogle Scholar
    • 7. Scholz KH, Maier SKG, Maier LS, Lengenfelder B, Jacobshagen C, Jung J, Fleischmann C, Werner GS, Olbrich HG, Ott R, Mudra H, Seidl K, Schulze PC, Weiss C, Haimerl J, Friede T, Meyer T. Impact of treatment delay on mortality in ST-segment elevation myocardial infarction (STEMI) patients presenting with and without haemodynamic instability: results from the German prospective, multicentre FITT-STEMI trial.Eur Heart J. 2018; 39:1065–1074. doi: 10.1093/eurheartj/ehy004CrossrefMedlineGoogle Scholar
    • 8. Claeys MJ, de Meester A, Convens C, Dubois P, Boland J, De Raedt H, Vranckx P, Coussement P, Gevaert S, Sinnaeve P, Evrard P, Beauloye C, Renard M, Vrints C. Contemporary mortality differences between primary percutaneous coronary intervention and thrombolysis in ST-segment elevation myocardial infarction.Arch Intern Med. 2011; 171:544–549. doi: 10.1001/archinternmed.2011.57CrossrefMedlineGoogle Scholar
    • 9. Sinnaeve PR, Armstrong PW, Gershlick AH, Goldstein P, Wilcox R, Lambert Y, Danays T, Soulat L, Halvorsen S, Ortiz FR, Vandenberghe K, Regelin A, Bluhmki E, Bogaerts K, Van de Werf F; STREAM investigators. ST-segment-elevation myocardial infarction patients randomized to a pharmaco-invasive strategy or primary percutaneous coronary intervention: Strategic Reperfusion Early After Myocardial Infarction (STREAM) 1-year mortality follow-up.Circulation. 2014; 130:1139–1145. doi: 10.1161/CIRCULATIONAHA.114.009570LinkGoogle Scholar
    • 10. Armstrong PW, Gershlick AH, Van de Werf F; STREAM Study Group. Fibrinolysis or primary PCI in myocardial infarction.N Engl J Med. 2013; 369:280–281; discussion page 281. doi: 10.1056/NEJMc1305999CrossrefMedlineGoogle Scholar
    • 11. Larson DM, Duval S, Sharkey SW, Garberich RF, Madison JD, Stokman PJ, Dirks TG, Westin RK, Harris JL, Henry TD. Safety and efficacy of a pharmaco-invasive reperfusion strategy in rural ST-elevation myocardial infarction patients with expected delays due to long-distance transfers.Eur Heart J. 2012; 33:1232–1240. doi: 10.1093/eurheartj/ehr403CrossrefMedlineGoogle Scholar
    • 12. Zheng Y, Bainey KR, Tyrrell BD, Brass N, Armstrong PW, Welsh RC. Relationships between baseline Q waves, time from symptom onset, and clinical outcomes in ST-segment-elevation myocardial infarction patients: insights from the Vital Heart Response Registry.Circ Cardiovasc Interv. 2017; 10:e005399. doi:10.1161/CIRCINTERVENTIONS.117.005399LinkGoogle Scholar
    • 13. Tjandrawidjaja MC, Fu Y, Westerhout CM, White HD, Todaro TG, Van de Werf F, Mahaffey KW, Wagner GS, Granger CB, Armstrong PW; APEX-AMI Investigators. Resolution of ST-segment depression: a new prognostic marker in ST-segment elevation myocardial infarction.Eur Heart J. 2010; 31:573–581. doi: 10.1093/eurheartj/ehp494CrossrefMedlineGoogle Scholar
    • 14. Buller CE, Fu Y, Mahaffey KW, Todaro TG, Adams P, Westerhout CM, White HD, van ‘t Hof AW, Van de Werf FJ, Wagner GS, Granger CB, Armstrong PW. ST-segment recovery and outcome after primary percutaneous coronary intervention for ST-elevation myocardial infarction: Insights from the Assessment of Pexelizumab in Acute Myocardial Infarction (APEX-AMI) trial.Circulation. 2008; 118:1335–1346. doi: 10.1161/CIRCULATIONAHA.108.767772LinkGoogle Scholar
    • 15. Welsh RC, Van de Werf F, Westerhout CM, Goldstein P, Gershlick AH, Wilcox RG, Danays T, Bluhmki E, Lopes RD, Steg PG, Armstrong PW. Outcomes of a pharmacoinvasive strategy for successful versus failed fibrinolysis and primary percutaneous intervention in acute myocardial infarction (from the STrategic Reperfusion Early After Myocardial Infarction [STREAM] study).Am J Cardiol. 2014; 114:811–819. doi: 10.1016/j.amjcard.2014.06.011CrossrefMedlineGoogle Scholar
    • 16. Robins JM. Marginal Structural Models.In: 1997 Proceedings of the American Statistical Association. Alexandria, VA: 1998; 1–10.Google Scholar
    • 17. Gershlick AH, Westerhout CM, Armstrong PW, Huber K, Halvorsen S, Steg PG, Ostojic M, Goldstein P, Carvalho AC, Van de Werf F, Wilcox RG. Impact of a pharmacoinvasive strategy when delays to primary PCI are prolonged.Heart. 2015; 101:692–698. doi: 10.1136/heartjnl-2014-306686CrossrefMedlineGoogle Scholar
    • 18. Danchin N, Coste P, Ferrières J, Steg PG, Cottin Y, Blanchard D, Belle L, Ritz B, Kirkorian G, Angioi M, Sans P, Charbonnier B, Eltchaninoff H, Guéret P, Khalife K, Asseman P, Puel J, Goldstein P, Cambou JP, Simon T; FAST-MI Investigators. Comparison of thrombolysis followed by broad use of percutaneous coronary intervention with primary percutaneous coronary intervention for ST-segment-elevation acute myocardial infarction: data from the French registry on Acute ST-elevation Myocardial Infarction (FAST-MI).Circulation. 2008; 118:268–276. doi: 10.1161/CIRCULATIONAHA.107.762765LinkGoogle Scholar
    • 19. Danchin N, Puymirat E, Steg PG, Goldstein P, Schiele F, Belle L, Cottin Y, Fajadet J, Khalife K, Coste P, Ferrières J, Simon T; FAST-MI 2005 Investigators. Five-year survival in patients with ST-segment-elevation myocardial infarction according to modalities of reperfusion therapy: the French Registry on Acute ST-Elevation and Non-ST-Elevation Myocardial Infarction (FAST-MI) 2005 Cohort.Circulation. 2014; 129:1629–1636. doi: 10.1161/CIRCULATIONAHA.113.005874LinkGoogle Scholar
    • 20. Rashid MK, Guron N, Bernick J, Wells GA, Blondeau M, Chong AY, Dick A, Froeschl MP, Glover CA, Hibbert B, Labinaz M, Marquis JF, Osborne C, So DY, Le May MR. Safety and efficacy of a pharmacoinvasive strategy in ST-segment elevation myocardial infarction.JACC Cardiovasc Interv. 2016; 9:2014–2020. doi:10.1016/j.jcin.2016.07.004CrossrefMedlineGoogle Scholar
    • 21. Siontis KC, Barsness GW, Lennon RJ, Holmen JL, Wright RS, Bell MR, Gersh BJ. Pharmacoinvasive and Primary Percutaneous Coronary Intervention Strategies in ST-Elevation Myocardial Infarction (from the Mayo Clinic STEMI Network).Am J Cardiol. 2016; 117:1904–1910. doi: 10.1016/j.amjcard.2016.03.036CrossrefMedlineGoogle Scholar
    • 22. Sim DS, Jeong MH, Ahn Y, Kim YJ, Chae SC, Hong TJ, Seong IW, Chae JK, Kim CJ, Cho MC, Rha SW, Bae JH, Seung KB, Park SJ; Korea Acute Myocardial Infarction Registry (KAMIR) Investigators. Pharmacoinvasive strategy versus primary percutaneous coronary intervention in patients with ST-segment–elevation myocardial infarction.Circ Cardiovasc Interv. 2016; 9:e003508. doi:10.1161/CIRCINTERVENTIONS.115.003508LinkGoogle Scholar
    • 23. Vanhaverbeke M, Bogaerts K, Sinnaeve PR, Janssens L, Armstrong PW, Van de Werf F. Prevention of cardiogenic shock after acute myocardial infarction.Circulation. 2019; 139:137–139. doi: 10.1161/CIRCULATIONAHA.118.036536LinkGoogle Scholar
    • 24. Pu J, Ding S, Ge H, Han Y, Guo J, Lin R, Su X, Zhang H, Chen L, He B; EARLY-MYO Investigators. Efficacy and safety of a pharmaco-invasive strategy with half-dose alteplase versus primary angioplasty in ST-segment-elevation myocardial infarction: EARLY-MYO Trial (Early Routine Catheterization After Alteplase Fibrinolysis Versus Primary PCI in Acute ST-Segment-Elevation Myocardial Infarction).Circulation. 2017; 136:1462–1473. doi: 10.1161/CIRCULATIONAHA.117.030582LinkGoogle Scholar