Primary percutaneous coronary intervention (PPCI) is preferred reperfusion therapy for acute ST-elevation myocardial infarction (STEMI). However, the PCI-capable center is still limited in several countries including Thailand. Therefore, fibrinolytic therapy is the main reperfusion therapy for most STEMI patients in our country. Current practice guidelines recommended routine coronary angiogram (CAG) after successful fibrinolysis, the so called pharmacoinvasive strategy [1‐5]. However, early pharmacoinvasive (within 3–24 h after successful fibrinolysis) cannot be performed in a timely fashion due to limitation of PCI-capable hospitals. Previous acute coronary syndrome (ACS) registries, Thailand Registry in Acute Coronary Syndrome (TRACS) showed a low rate of coronary angiography and intervention during index admission and referral centers for early pharmacoinvasive strategy are still limited [6]. Therefore, risk stratification and identify risk of the patients are important in non PCI-capable hospital. Patients with intermediate to high risk for adverse cardiovascular event should be transferred for coronary angiogram as soon as possible. Although several risk scores for acute coronary syndrome (ACS) have been developed for stratified risk of ACS patients, GRACE (Global Registry of Acute Coronary Events) score is developed to focus on clinical risk assessment and to improve the selection of patients for clinical and interventional procedures following an ACS episode [7]. Many studies and meta-analysis demonstrated the accuracy and the usefulness of the GRACE score on the mortality of ACS patients in hospital and follow-up after hospital discharged [8‐13]. This study aimed to evaluate the prognostic utility of the GRACE score in patients receiving delayed intervention after successful fibrinolysis.

Although early pharmacoinvasive strategy (within 3–24 h) after successful reperfusion are recommended by several guidelines [1‐5], timely fashion CAG is not widely available in countries with limited PCI capable hospitals including Thailand. Several randomized trials and meta-analysis have shown that early routine post-thrombolysis angiography with subsequent PCI reduced the rates of re-infarction and recurrent ischemia compared with a watchful waiting strategy, in which angiography and revascularization were indicated only in the patients with spontaneous or induced severe ischemia or left ventricular (LV) dysfunction [15, 16]. The benefits of early routine PCI after thrombolysis were seen in the absence of increased risk of adverse events in many studies [15, 16]. The data from TRACS showed only half (50 %) of STEMI patients performed CAG on index admission. Fibrinolysis (especially streptokinase), is the first choice for treatment in low risk STEMI patients (42.6 % of STEMI patients received streptokinase and 1 % received Tenecteptase) [6]. Because of only one cardiac catheterization (during the period 2007–2012) in Northern Thailand (Maharaj Nakorn Chiang Mai Catheterization laboratory), the geographic and long distance of transfer and few of number of interventional cardiologists, primary PCI and early routine PCI after successful fibrinolysis were very difficult for this situation. Rescue PCI or primary PCI were performed in the patients who failed fibrinolytic therapy or cardiogenic shock at presentation. Hence, most of the STEMI patients in Thailand, especially in Northern of Thailand who successfully fibrinolytic therapy received the long delay coronary intervention (more than 24 h after fibrinolysis) and some of them received elective PCI or very long delayed intervention or elective PCI (after 2 weeks from successful fibrinolytic therapy) [6]. Several studies demonstrated the worst cardiovascular outcomes in the patients who received delay coronary intervention after thrombolysis [15‐21]. The Southwest German Interventional Study in Acute Myocardial infarction (SIAM III) evaluated the effects of transfer early PCI (within 6 h after fibrinolysis) compared with delay PCI strategy (elective PCI 2 weeks after fibrinolysis) [17]. The early PCI showed significant reduction of primary end point (death, re-infarction, target lesion revascularization (TLR) and ischemic events) (HR: 0.61; 95 % CI 0.42–0.88, p = 0.008) and higher long term survival than delayed PCI (p = 0.057) [17]. Similarly to The Trial of Routine Angioplasty and Stenting after Fibrinolysis to Enhance Reperfusion in Acute Myocardial Infarction (TRANSFER-AMI) trial, showed that the patients who transfer from non-PCI center within 6 h after thrombolysis had fewer ischemic complications than standard treatment (delayed PCI) without increasing of major bleeding [18]. A meta-analysis showed mortality benefit at 30-day and 1 year of the STEMI patients with early transfer PCI after fibrinolysis as compared with ischemic-guided intervention (delayed PCI) [15, 16]. The NORwegian study on District treatment of ST-Elevation Myocardial infarction (NORDISTEMI) study also demonstrated a significant reduction in the composite cardiovascular outcome (death, re-infarction, stroke, or recurrent ischemia) at 1 year in the patients with immediate transferred to PCI following with thrombolysis as compared with the patients in conservative arm treatment (6 % vs 16 %, p = 0.01) [19]. Similarly to The Combined Abciximab RE-teplase Stent Study in Acute Myocardial Infarction (CARESS-AMI) study, a more conservative strategy (i.e. angiogram only in cases of failed thrombolysis) was associated with a worse clinical outcome than the strategy of angiogram and intervention (if indicated) in all cases following thrombolysis (composite of death, re-infarction and refractory ischemia at 30-day, 11 % vs 4 %, p = 0.004) [20]. From the previous data, no studies demonstrated of the benefit in the cardiovascular outcomes of the early and/or delay pharmacoinvasive strategies in STEMI patients who received streptokinase for treatment similar to our study. On the data from CARESS-AMI [20] and TRANSFER-AMI [18], The American College of Cardiology (ACC) and the American heart association (AHA) give a class IIa recommendations for high risk features (such as Kilip class >2, extensive ST-elevation, left ventricular ejection fraction (LVEF) <35 %, or hypotension) should be immediate transferred to PCI-capable facilities [3, 4]. The transfer of low and moderate risk STEMI patients to PCI-capable center received a class IIb recommendation. No available data showed the benefit outcome of early transferred for PCI in low and moderate risk patients.

Risk stratification of the STEMI patients were very important for the clinicians in non-PCI capable hospital to use to guide for judged and selected the STEMI patients for early invasive strategy. GRACE risk score, one of clinical risk score, has been shown to be a good risk stratification score in population with STEMI and NSTE-ACS. Several studies demonstrated the validation and the usefulness of GRACE score in stratified the STEMI patients for an early invasive management (AUC = 0.81; 95 % CI 0.80–0.82 for STEMI and AUC = 0.80; 95 % CI 0.74–0.89 for NSTE-ACS) [12]. The AuROC of 6-month mortality and the composite cardiovascular outcome of our study were 0.794 (95 % CI 0.75–0.83) and 0.641 (95 % CI 0.52–0.76). From our study, the GRACE score seem to be better performance in the cardiovascular mortality rather than the composite cardiovascular outcome of the patients with long delay pharmacoinvasive as similar as the previous study [12]. But the usefulness of GRACE score for predict the composite cardiovascular outcome is still unclear. A subgroup analysis of TRANSFER-AMI trial revealed the beneficial outcome of early pharmacoinvasive strategy only in patient with a low to intermediate GRACE risk score (<155), while the early invasive strategy was associated with worse outcome in high-risk patients (≥155) [14]. The pharmacoinvasive strategy was associated with a lower risk of death/re-MI in the low-intermediate GRACE risk group (HR = 0.52, 95 % CI 0.32–0.86, p = 0.010), but a higher risk of death/re-MI in the GRACE high-risk group (HR = 1.98, 95 % CI 1.06–3.67, p = 0.031) [14]. From this subgroup analysis from TRANSFER-AMI, risk score may also guide the best strategy to achieve and maintain myocardial reperfusion after administration of fibrinolytic therapy [14]. Similar to our study, the longer delay pharmacoinvasive strategy (24 h to 2 weeks after successful fibrinolysis) in non PCI-capable facilities may associate with the worst of composite cardiovascular outcome (death, re-hospitalized with ACS, re-hospitalized with HF and stroke) at 30-day and 6-month when compared with the patients with low GRACE score (15.6 % vs 2.3 % at 30 days, p = 0.003 and 16.7 % vs 6.8 % at 6 months, p = 0.024). Therefore, the patients with intermediate to high GRACE risk score should be early transferred to PCI-capable center after fibrinolytic therapy.

The in-hospital mortality and 6-month mortality of our study was lower than the previous registry (TRACS) because the difference in baseline patient characteristics, the severity of the patients and the number of the patients received of the percutaneous coronary intervention on admission (in-hospital mortality 5.3 % vs 3.1 % and 6-month mortality 12.1 % vs 3.1 %) [6]. Most of the patients in our study had multivessel disease but underwent culprit vessel PCI only in significant proportion of patients. A small number of the patients underwent multivessel PCI during index hospitalization (10.7 % in low GRACE group vs 7.7 % in intermediate to high GRACE group). The meta-analysis and systematic review of Moretti et al. [22] in management multivessel coronary disease in STEMI patients, 5855 patients from 6 studies (1 RCT) compared between only culprilt vessel PCI vs complete PCI performed during index hospitalization. No difference in major adverse cardiovascular events (MACE) at short-term (90 days) and long term outcome at 1 year but significant reduced the repeat revascularization at 1 year similar to culprit vessel PCI vs complete revascularization during PCI. The rate of CABG was high especially in intermediate to high GRACE group because of high prevalence of multivessel disease and complex coronary artery disease (Type B2 and C) which may suitable for CABG after acute phase of STEMI. Previous ACS registry in Thailand (TRACS) showed the lower rate of CABG but the revascularization data was collected only in hospital-phase of STEMI [6]. The selective biased in enrolled patients who survived during index admission may contribute to low cardiovascular event in our study. We showed the performance of GRACE score for mortality of in-hospital, short term (30 days) and 6-month Therefore, the GRACE risk score is useful for prediction in short- and long-term mortality of the STEMI patients with successful fibrinolysis and delay intervention in non PCI-capable hospital.

There are some limitations in our study that may compromise clinical implication. Our study was a retrospective observational study (non-randomized). The large number of excluded patients reflected the limited accessibility to coronary intervention within 2 weeks. The mortality was lower than the previous study because the small number of patients with high GRACE risk were included in our study.

The long delay pharmacoinvasive strategy in intermediate to high GRACE score after successful fibrinolysis in non PCI-capable facilities were associated with worse cardiovascular outcomes (death, re-hospitalized with ACS, re-hospitalized with HF and stroke) than the patients with low GRACE score at 30 days and 6 months. GRACE risk score may be helpful and guided the clinicians in non PCI-capable center in early transferred to early intervention in STEMI patients after fibrinolytic therapy.