Download PDF

Liviu-Nicolae Ghilencea1,2, Andreea–Catarina Popescu1,2, Doina Dimulescu1,2, Sean Maher1,2, Ali Al Hassan2, Adrian Linte2, Ionut Stanca2, Mihai Melnic2, Sergiu Barsan2, Cristian Bejan1, Laura Arama2, Brandusa Zamfirescu2, Stefanita Petrea2, Cristian Gîrgel2, Maria-Lorena Roamba2, Simona Huidu2, Luminita Ionescu2, Raluca Aflorii2, Andreea Rachieru2, Maria Marinescu2, Elena Lechea2, Nicoleta Tascan2, Constantin Stefani1, Alexandrina Nas-tasa2, Livia Trasca1,2, Lavinia Matei1,2, Adrian Mereuta1,3, Serban Balanescu1,2

1 “Carol Davila” University of Medicine and Pharmacy, Bucharest, Romania
2 Elias University Hospital, Bucharest, Romania
3 „C.C. Iliescu” Institute of Cardiovascular Disease, Bucharest, Romania

Abstract: Aims – The objective of our study is to compare pharmaco-invasive strategy (PhIS) in terms of safety and effi-cacy, with primary percutaneous coronary intervention (pPCI) strategy as standard therapy for STEMI patients from remote areas. Background – At present, primary percutaneous coronary intervention (pPCI) is the preferred reperfusion strategy for STEMI patients. However, despite pPCI being the gold standard, it is not always achievable due to variables such as lack of cardiac catheterization services and delays in the first medical contact-to-balloon time. Methods – This observational study is based on a prospective analysis of a cohort of 157 patients with ST-elevation myocardial infarction, over nine months, at Elias University Hospital (EUH). The study assessed the safety and efficacy of a sequence of fi brinolytic therapy in the first contact hospital before being referred to our center for PCI (PhIS, 35 patients, 22.29%) versus angioplasty alone (pPCI, 122 patients, 77.70%) during hospitalization. The primary outcomes studied were in-hospital death, and major adverse cardiac events (MACE), while secondary outcomes were the length of in-hospital stay, and the safety of the procedure. Results – The median reperfusion time was lower for the PhIS group compared to the primary PCI group (4 hours, IQR:6.25 vs 7 ho-urs, IQR:12.38, p<0.05). The left ventricular systolic function (%) on arrival at EUH was higher in the PhIS group compared with the PCI alone group (46.55; 95%CI 42.42-49.15 vs 41.73; 95%CI 39.91-43.34; p=0.04). The mean number of diseased vessels, including the culprit-lesion, were similar in the two groups (2.84 vs 2.82, p=0.09). The in-hospital mortality rate was lower in the PhIS group than in the primary PCI group (1 patient; 2.9% vs 18 patients; 14.80%; p=0.05), while the number of in-hospital major adverse cardiac events (MACE) was not signifi cantly different (17.10% vs 26.20%; p=0.27). There was no difference of the median length of hospitalization (6 days, IQR:2 vs 5 days, IQR:3; p=0.67) for the PhIS, and primary PCI groups, respectively. The safety endpoints of the procedures were similar in the two groups. Conclusions – Pharmaco-in-vasive strategy (PhIS) had clinical and procedural outcomes (in-hospital MACE, length of in-hospital stay) similar to primary percutaneous coronary intervention (pPCI), in case of long distances to catheterization laboratories. Efficient thrombolysis makes PhIS a viable alternative in saving lives with a lower rate of in-hospital death than pPCI. Performed thrombolysis in a local non-PCI center and referral of the STEMI patients to a 24/7 catheterization laboratory may be a good option for areas where the infrastructure of such facilities is weak.
Keywords: ST-segment elevated myocardial infarction (STEMI); Pharmaco-invasive strategy (PhIS); Primary percutaneous coronary intervention (pPCI).

I. BACKGROUND
ST-segment Elevation Myocardial infarction (STEMI) is one of the leading causes of death worldwide with a signifi cant impact on healthcare resources and ex-penditure1. All major international guidelines (ESC 2017, ACC/AHA 2013, NICE 2013) clearly state that primary Percutaneous Coronary Intervention (pPCI) is the preferred reperfusion strategy for STEMI pati-ents2-4. Furthermore, its superiority over fibrinolysis alone has been repeatedly demonstrated in numerous clinical trials5. However, despite pPCI being the gold standard, it is not always achievable due to the lack of cardiac catheterisation services and delays in the first medical contact-to-needle time5. Consequently, pharmaco-invasive strategy (PhIS), which consists of a sequence of fi brinolytic therapy performed in the local hospital followed by patient transfer to a PCI-capable centre for coronary angioplasty, is common practice in many countries6.
Should the maximum expected delay from STEMI diagnosis to PCI be greater than 120 minutes for patients presenting within 12 hours of symptoms onset, fibrinolytic therapy has to be performed as an inter-mediate step before referring the patient to a 24/7 PCI-capable facility2. Nevertheless, many patients do not meet these treatment criteria due to multiple lo-gistical, geographical and resource-related issues. As a result, many patients from remote areas are trans-ferred for PCI outside the recommended 120 minu-tes of transportation time without having received thrombolytic therapy (delayed PCI). Furthermore, the effectiveness of pPCI performed outside the recommended transfer time from diagnosis is not apparent, and recent studies suggest that early intervention with PhIS may be superior to delayed PCI7.
Moreover, many clinical studies have investigated delays in the door-to-balloon time and current guide-lines state that a 30 minutes door-to-balloon time is optimal for effective reperfusion2. However, it is unclear how pre-hospital delays such, as the time from chest pain (CP) onset to contacting emergency ser-vices infl uences the outcome of reperfusion therapy.
Due to the lack of interventional facilities and skil-led interventionists, many patients with STEMI cannot immediately benefit from mechanical reperfusion via angioplasty with stents. Similar situations in other Eu-ropean countries led to the reassessment and reintro-duction of fibrin-specifi c thrombolytic agents in STEMI management protocols, when the transfer time from the FMC (local hospital) to the catherization laborato-ry facility exceeds two hours2.

II. OBJECTIVES
Our prospective, observational study sought to in-vestigate the efficacy and safety of pharmaco-invasi-ve strategy (PhIS) versus primary PCI (pPCI) in the context of the Romanian Ministry of Health’s Acute Myocardial Infarction-Priority Action (AMI-PA) Pro-gramme for STEMI. We also considered whether PhIS should be a model of care for STEMI patients who are outside the 120 minutes transfer time from diagnosis to PCI in the context of a real-world system.

III. METHODS
Study design. We prospectively recruited 157 con-secutive patients with STEMI, over nine months, from January 1st to October 1st, 2018, as our institution provides 24/7 catheterization laboratory services on rotation, for patients coming from within an 58,584km2 area of Bucharest with the longest distance of 340 km.
The protocol was designed by a team of clinical and interventional cardiologists at Elias University hospital (EUH) and approved by the hospital’s research and ethics committee. Data collection and data analysis were carried out by the authors at the Heart Depart-ment. The study was not sponsored by any third par-ties; no other party had any involvement in the study design, data analysis, or manuscript preparation. The authors confirm the accuracy of the data and analysis. Patient inclusion criteria. Patients were eligible for inclusion in the study if they had evidence of on-going CP and dynamic ECG changes consistent with STEMI2.
A total number of 157 patients with ST-segment elevation myocardial infarction (STEMI) were ad-mitted to EUH for PCI via ambulance as follows: 96 patients (61.14%) directly from the scene of CP to EUH for primary PCI; 26 patients (16.56%) from the first medical hospital (FMH) without fibrinolytic thera-py (delayed PCI subgroup); 35 patients (22.29%) from community hospitals (FMH) after fi brinolytic therapy (PhIS group). The patients referred from the FMH (26 plus 35 patients) were screened for eligibility by local doctors and communicated via mobile to the Cardio-logist on-duty at EUH (Figure 1).
Patients in remote areas, outside the transport time of 120 minutes (61 patients, 38.86%), were ini-tially assessed at the FMH where they received a loa-ding doses of Aspirin and Clopidogrel (or Ticagrelor). Some of these STEMI patients (35 patients) received fibrinolytic therapy and unfractionated heparin, be-fore being transferred to EUH for percutaneous co-ronary intervention (PCI) of the infarct-related co-ronary artery (IRCA). The choice of fi brinolytic and the dose administered were at the discretion of the FMH cardiologist/acute medical unit (AMU) physician. The patients from an area within a two hours reach of an interventional center (96 patients, 61.14%) were transferred directly by ambulance to our hospital and assessed in the emergency department (ED) before undergoing angiography. The patients were consented accordingly.

Intervention. We performed diagnostic angiogra-phy on all patients diagnosed with STEMI and implanted stents when technically possible. The stents implanted were drug-eluting, covered under the Romanian Mi-nistry of Health’s national programme (AP-IMA). The protocol allowed for the use of unfractionated hepa-rin, glycoprotein IIb/IIIa (Gp IIb/IIIa) antagonists at the discretion of the interventional Cardiologist on-duty. A TIMI 3 flow, grade III myocardial blush and less than 10% residual coronary stenosis were the aims of PCI. Endpoints and definitions. The primary en-dpoints of the study were in-hospital major adverse cardiac events (MACE), in-hospital death of any cause, and a combination of in-hospital MACE, death and re-current chest pain. Secondary endpoints included the length of hospitalization and the safety of interventi-onal procedures: the amount of contrast agent used (ml), time of X-ray exposure (minutes), radiation dose air kerma (AK), dose-area product (DAP). In-hospital major adverse cardiac events (MACE) were a com-bination of cardiogenic shock, recurrent ischemia/MI, clinically driven target revascularization, stroke, ven-tricular tachycardia /fibrillation (VT/VF), pulmonary oedema and heart failure.

Statistical analysis. Categorical data are reported as numbers (percentages %) for its variables such as gender, in-hospital deaths, MACE, risk factors, recur-rent chest pain, Killip class on admission and infarct-re-lated coronary artery (IRCA). We used the Pearson’s chi-square test and Fischer’s exact test for group com-parisons. Data for continuous variables are presented as mean±SE (%) when the distribution is uniform for age, hs-cTnI on admission (ng/l) and as medians and interquartile range (IQR) when the distribution is not uniform for treatment times.
We performed comparisons of the central tenden-cy of the baseline characteristics and endpoints of the two treatment groups using the t-test for normally distributed continuous variables, and nonparametric tests (Wilcoxon rank-sum test or Mann-Whitney rank sum test) to compare the numerical variables and the abnormally distributed continuous variables.
We used the Kaplan-Meier method to assess the time to primary endpoints and to create survival esti-mates. Chi-square test was also used to compare the rates of death or recurrent chest pain between the treatment groups. All p-values were two-sided, and a p-value <0.05 was considered statistically significant. We assessed the odds ratio for in-hospital MACE and death. The statistical analysis was performed with an SPSS program, version 21 (Statistical Package for the Social Sciences) software (IBM SPSS Statistics, USA).

Figure 1. Flow chart with the randomization of patients with STEMI.

FT – fibrin therapy; A&E – accident and emergency department; dPCI – delayed PCI; FMH – first medical hospital.

RESULTS
Baseline characteristics. Baseline characteristics of the patients were well balanced across the two groups, with no differences among age, gender and risk factors, except for a higher non-signifi cant pre-valence of diabetes mellitus (p=0.08) and active smo-king (p=0.08) in the PhIS group compared with the standard treatment (pPCI) group (Table 1). Of the 61 patients from rural areas, outside a 2 hours reach of Cath lab facilities, only 35 patients of them (57.37%) received fibrinolytic therapy in the local hospital, be-fore being transferred to our hospital (EUH) (Table 1).
The median chest pain duration time (hours) from chest pain onset (CP) to local hospital (FMH) (CP-to-FMH time) assessment is similar in the PhIS group and in the primary PCI group (4h, IQR:6.25 vs 5h, IQR:11.50, p=0.43), but the median time from CP onset (hours) to EUH is non-significantly longer in the PhIS group compared with the pPCI group (9h, IQR:7.25 vs 7h, IQR:12.38, p=0.07). This may be explained by the time necessary for fibrinolytic therapy at the local hospital, for the PhIS group. We have to mention that from 35 patients of the PhIS group, only 23 patients (65.71%) arrived within 12 hours, while from the 122 patients of the pPCI group only 82 patients (67.21%) arrived within 12 hours to EUH (Figure 2).
The median revascularization time, from CP onset to fibrinolytic therapy at the FMH for the PhIS group was significantly lower compared with the time from CP to primary PCI at EUH (CP-to-EUH time) for primary PCI group (4h, IQR:6.25 vs 7h, IQR:12.38; p<0.05).
The third treatment time of our research, the medi-an door-to-needle time was similar in the two groups: (59 minutes, IQR: 49 vs 60 minutes, IQR: 49; p=0.93) for the PhIS group and the pPCI group, respectively.
The need for cardio-pulmonary resuscitation (CPR) was slightly less frequent in the pharmaco-invasive strategy group compared to primary PCI (1 patient, 2.90% vs 14 patients, 11.50%; p=0.12).
We assessed the patients on admission and noted the Killip class. Overall, the pharmaco-invasive stra-tegy (PhIS) group had a better Killip class on admission than the pPCI group, but it was not statistically signi-ficant (p=0.07).
We found a significantly higher level of the high-sen-sitivity cardiac troponin I mean (hs-cTnI) on admission (expressed in ng/l) in the PhIS group compared with the pPCI group (112.62±73.27, 95%CI=86.20-139.04 vs 76.03±73.15, 95%CI=61.66-90.40; p=0.01) (Figure 3).
The trimmed mean LVEF (%) on admission was sig-nificantly higher for patients who underwent fibrinoly-tic therapy before PCI (PhIS group) compared with patients in the primary PCI group (46.55±9.79; 95%CI: 42.42-49.15 vs 41.74%±12.27; 95%CI: 38.91-43.34; p=0.04) (Figure 3).

Procedural and safety outcomes. There were five leading operators, who performed interventional procedures for both groups, according to the cathe-terization laboratory on-duty Rota. The number of vessels affected was similar in the PhIS group compa-red with the primary PCI group (2.82 vs 2.84; p=0.09). The median total length of stents (mm) implanted is similar in the two groups, PhIS and primary PCI (25, IQR=18 vs 28, IQR=23, p=0.09). The post-interventional need for Gp IIb/IIIa is significantly lower in the PhIS strategy group than in the primary PCI group (8.60% vs 25.40%, p=0.03).
Safety. We assessed the safety of procedures by assessing time of exposure to X-Ray, kinetic energy released per unit mass (radiation dose air kerma), do-se-area product (DAP), and the amount of contrast used during Cath lab interventions for both groups. All the measured safety outcomes were similar for the two groups, (PhIS vs pPCI) (Figure 4).
Efficacy outcome. The procedural and clinical outcomes results are displayed below (Table 2, Figu-res 5-7). The median length of hospital stay was similar in the two groups (6 days, IQR: 3 vs 5 days, IQR: 3, p=0.67), in the PhIS group and the pPCI group, re-spectively.
In-hospital major adverse cardiac events (MACE). The primary endpoint, in-hospital major adverse cardiac events (MACE) is similar in the PhIS group compared with the pPCI group (17.10% vs 26.20%, p=0.27). The odds ratio (OR) for developing MACE is with no statistical signifi cance lower in the PhIS group (OR: 0.582, 95%CI: 0.221-1.531; p=0.27). The relative risk for developing MACE is with 34.6% less in the PhIS group than the primary PCI group (RR:0.654; 95% CI:0.298-14.35, p=0.37), but was found to not be statistically significant (Table 2, Figures 5-6). The risk estimate for the presence of MACE was 18.9% higher in pPCI group (OR:1.189, 95%CI=0.570-2.478).
In-hospital death. The other primary endpoint, in-hospital mortality, was more frequent in the primary angioplasty (pPCI) group compared to patients from remote areas who underwent fibrinolytic therapy at the FMH initially (PhIS), approaching the statistical significance (18 patients, 14.80%, vs 1 patient, 2.90%, p=0.05). The major causes for in-hospital death are reported in table 6. The relative risk (RR) of death for patients with pPCI is 5.12 times higher than the PhIS group, but not statistically signifi cant (95% CI: 0.71-36.94; p=0.07). The odds ratio for developing in-ho-spital death was lower for the PhIS group compared with the pPCI group but was not statistically significant (OR:0.17; 95%CI: 0.022-1.321; p=0.09) (Table 2, Figu-re 7).
Recurrent chest pain. There is no significant difference in the prevalence of recurrent chest pain between the two groups, PhIS vs pPCI (6.3% vs 3.3%, p=0.61).
In-hospital MACE distribution, according to the infarct-related coronary artery (IRCA). Since the primary en-dpoint (in-hospital MACE) proved similar among the two groups (PhIS, pPCI), we queried if the in-hospital MACE is dependent upon the infarct-related coro-nary artery. When we tested the null hypothesis with Independent-Samples Kruskal-Wallis test, we found that the distribution of major adverse cardiac events (MACE) is not similar across the four vessels (LMCA, LAD, LCx, RCA; p=0.02).
In-hospital mortality distribution, according to the IRCA. When we compared the rate of in-hospital death according to the culprit lesion, irrespective to the treatment strategy group, we found significant di-fferences between the LMCA and each of the other culprit vessels: LAD (p=0.02), RCA (p=0.005), LCx (p=0.03), There is no significant difference of the in
hospital mortality between the 2 groups (PhIS and pPCI) when the cause of STEMI is one of the other three vessels: LAD, RCA or LCx.
Composite endpoint of in-hospital MACE, recurrent CP, and in-hospital death, according to the IRCA. When we analyzed the primary composite endpoint (MACE, re-current CP, death) we found significant differences in respect to the location of the culprit lesion. The num-ber of composite endpoint events was signifi cantly hi-gher when LMCA was the IRCA compared with LAD (p=0.002) and with RCA (p=0.01), irrespective of the treatment time or fibrinolytic therapy at FMH.

Correlations. We assessed the correlations among the variables of the two groups, to find any relationship, before using regression tests to identify the quantitative relationship between the dependent variables (in-hospital MACE, deaths, duration of hospi-talization) and independent variables (age, gender, risk factors, treatment times, hs-cTnI, out-of-the hospital resuscitation, cardiac arrest), and Killip class or LVEF on admission (Table 3).
There is a strong correlation of the number of in-hospital death, with in-hospital MACE (p=0.0001), with the Killip class on admission (p=0.0001), and a medium correlation with LVEF on admission (p=0.0001). Inte-restingly, there is no correlation between the number of in-hospital deaths and time from CP onset to FMH (p=0.77) or to EUH (p=0.88).
There is a moderate correlation of in-hospital MACE with the Killip class on admission (p=0.0001), with active smoking (r= -0.302, p=0.0001). There is no correlation between in-hospital MACE and the time from CP onset to FMH (r= 0.007, p=0.92) or to EUH (r= -0.045, p=0.58), and no correlation with fibrinoly-tic therapy at FMH (r= 0.088, p=0.27).
There is a moderate correlation of the length of hospitalization with LVEF on admission (r= 0.320, p=0.0001), and a weak correlation with hs-cTnI (r= 0.231, p=0.009), fluoroscopy time (r= 0.122, p=0.17). We remark that there is no correlation between du-ration of hospitalization and time from CP onset to FMH (r = 0.108, p = 0.20) or EUH (r= 0.096, p = 0.25), or with Killip class on admission (r = 0.051, p = 0.54).
There is a weak correlation of fibrinolytic therapy at local hospital with the in-hospital death (r= -0.152, p=0.05), and with in-hospital MACE (r= 0.088, p=0.27). There is a strong correlation between the number of stents and the contrast volume (r= 0.502, p=0.0001), and a moderate correlation with air kerma (r= 0.353, p=0.0001), or fl uoroscopy time (r= 0.304, p=0.0001).
Interestingly, there is no correlation between re-suscitation and any of the following characteristics: in-hospital death, length of in-hospital stay, recurrent chest pain, number of additional diseased vessels, hs-cTnI on admission, or age.

Odds ratio. We used a multivariable regression model for predictors of the primary efficacy outco-me. In the multivariate analysis for the STEMI patients, the following factors were associated with major car-diovascular events and in-hospital death, irrespective of fibrinolytic therapy before PCI (Tables 4-5, Figures 6-8).
The odds of developing in-hospital major adverse cardiac adverse (MACE) are mentioned in table 4 and displayed in figure 6. Patients aged over 60 yo (OR: 2.414; p=0.02), resuscitated before EUH, with Killip class IV on admission (OR:4.267; p=0.009), with LVEF less than 40% (OR:2.663; p=0.01), or with LMCA in-volvement (OR:8.864; p=0.001) have higher odds risk for developing in-hospital MACE. Interestingly, acti-ve smoking patients (OR 0.197, p=0.0001) have the lowest risk of developing in-hospital MACE, after PCI for an ST-elevation myocardial infarction.
There is no difference between the risks of deve-loping in-hospital MACE for the patients who recei-ved thrombolytic therapy before PCI (PhIS) compared with patients in the pPCI group (OR:0.582, 95%CI: 0.221-1.531; p=0.27).
Similarly, the odds of developing in-hospital death are displayed in table 5 and fig. 8. Patients older than 70 years of age, the need for resuscitation before EUH, LMCA involvement, Killip class IV on admission, a total length of stents over 33 mm, troponin level over 80 ng/l, more than two diseased vessels (inclu-ding the IRCA) have statistically significant higher odds of death during hospitalization.
Interestingly and intriguing, active smoking (OR: 0.114, 95%CI 0.025-0.514, p=0.005) and high choles-terol patients (OR: 0.096, 95%CI 0.021-0.43, p=0.002) have the lowest risk of in-hospital death with a better prognostic irrespective of the therapy group they be-long. The correlations described earlier in our resear-ch are in line, though, with these fi ndings.
Killip class I have the lowest OR to develop in-hos-pital MACE (OR: 0.183, p=0.001) or death (OR: 0.088, 95%CI 0.03-0.251; p=0.0001). When we compared the Killip classes on admission, for the two groups, the in-hospital survival time (days) to a patient’s death is better for PhIS group compared with the pPCI group (Killip I: 30 vs 9.96; Killip II: 14.50 vs 4; Killip III: 11 vs 7; Killip IV: 14 vs 7).
Also, although the odds of in-hospital death are lower for the PhIS group compared with the pPCI group (OR: 0.17, 95% CI 0.22-1.321; p=0.09) it is not statistically significant.

Survival. The Kaplan-Meier survival curve during hospital stay shows a better result for the PhIS group compared to the pPCI group (Figures 8-9). When we compared the length of hospital stay, there are no sig-nifi cant differences among the groups with different IRCA (p=0.54).

Figure 2. Therapeutical times for the two groups, pPCI (no fibrin therapy) and PhIS (thrombolysed before PCI).

1. Time from chest pain onset to FMH (hours);
2. Chest pain to EUH time (hours);
3. Door-to-needle time (hours).

Figure 3. Clinical characteristics boxplots for the 2 groups, pPCI (no fi brin therapy) and PhIS (thrombolysed before PCI).

1. Level of high-sensitivity cardiac troponin on admission (ng/l);
2. Left ventricle ejection fraction on admission (%).

N – pPCI group, Y – PhIS group.

Figure 4. The angiographic outcomes for the 2 group: pPCI (no fibrin therapy), and PhIS (thrombolysed before PCI). All the measured safety outcomes were similar for the 2 groups, PhIS vs pPCI.

1. The median fluoroscopy time (min);
2. The median air kerma (AK) (mSv);
3. The median dose area product (DAP) (mGycm2);
4. The median contrast volume (ml).

Figure 5. The in-patient hospitalization (days) for MACE and fibrin therapy.

1. For the patients with MACE (right) and without MACE (left);
2. In-patient time duration (days) for patients of the pPCI group (left) and for PhIS group (right).

Figure 6. The predictors of the major adverse cardiac events during hospitalization.

Patients with age over 70 yo, resuscitated before/on admission, with Killip class IV, with LVEF less than 40%, with troponin over 80 ng/ml on admission, with LMCA involved as IRCA, have higher risk of developing MACE during in-hospital stay. Other predictors like active smoking, Killip class I, and length of stents less than 30mm have protective action, decreasing the risk of in-hospital MACE. The male gender, high cholesterol, diabetes mellitus, high blood pressure, family history of CAD, previous CAD, dyspnoea, fibrin therapy before PCI, class Killip II and III, coronary artery except LMCA, operators, need for Gp IIb/IIIa, AK, DAT, amount of contrast, diameter of stents, troponin over 80 ng/ml, total number of vessel disease, in-hospital stay, chest pain to EUH time, irrespective of the cut off (8, 12, 16 or 24 hours), culprit lesion, number of stents over 2, have similar odds for developing MACE.

Figure 7. The predictors of the in-hospital death.

The age over 70 yo, Bucharest area, resuscitation, Killip class IV, troponin over 80 ng/ml, LVEF less than 40% on admission, the total number of vessel disease over 2, LMCA as IRA, length of stents over 33 mm, have higher odds of developing in-hospital death. Other predictors like active smoking, high cholesterol, Killip class I have protective action, decreasing the risk of in-hospital death.

Figure 8. Kaplan-Meier survival curves, with the cumulative incidence of the primary endpoint (in-hospital death) during hospitalization in both treatment arms.

29. Kaplan-Meier curve for in-hospital death shows better in-hospital survival curves in favor of PhIS group. The estimate mean of death in the PhIS group compared with the pPCI group (29.147 days, 95%CI 27.50-30.78 vs 19.776 days, 95%CI 15.907-23.644; p=0.104).
30. The log survival function shows better results for the PhIS group compared with pPCI group.

Figure 9. Kaplan-Mayer curves for in-hospital major cardiac events (MACE).

22. The estimate mean for in-hospital MACE is similar in the PhIS group compared with the pPCI group (22.821 days, 95%CI 16.046-29.597 vs 13.754 days, 95%CI 10.602-16.907; p=0.167). There are similar Kaplan-Meier MACE curves for the first 6 days only, but PhIS group has better in-hospital MACE results afterwards;
23. Kaplan–Meier MACE log survival curve showing disease-free survival by MACE during hospitalization, with similar in-hospital MACE curves for the first 6 days only, afterwards the PhIS curve has better outcome than pPCI curve.

DISCUSSION
To the best of our knowledge, this is the first prospec-tive study in Romania assessing both pharmaco-invasi-ve strategies and primary PCI in acute STEMI. Based on the STEMI Elias Hospital Interventional Registry (STEMI-EIRE), this pilot study compared the strategy of PCI after fibrinolytic therapy at the first local hos-pital, called pharmaco-invasive strategy (PhIS) with the standard therapy of primary PCI.
The following characteristics: age over 70 yo, re-suscitation before admission to EUH, Killip class IV on admission, LVEF (%) less than 40% on admission, troponin over 80 ng/l on admission, total number of vessels disease, LMCA as IRCA, length of stents im-planted over 33mm, are independent risk factors for developing in-hospital death. Similarly, age over 60 yo, resuscitation before EUH, Killip class IV on admission, LVEF less than 40% on admission, LMCA involvement are independent factors for developing in-hospital MACE.
The median treatment times from CP onset to the FMH in our study are in line with the RO-STEMI data (4 hours, IQR: 6, 25 vs 5 hours, IQR: 11.50) for PhIS and pPCI respectively. However, the median time from CP onset to a PCI capable center is higher, especially for patients referred from community hospitals witho-ut fi brinolytic therapy (7 hours, IQR: 12.38). In the RO-STEMI report, the median time for the CP onset to hospitalization was 6 hours across the all groups8.
Our primary outcomes (MACE and death during hospitalization) are in line with different studies car-ried out in countries with similar Cath lab networks: Canada9, Egypt10, India11, USA12, but also in countries like Denmark13, where they compared pharmaco-inva-sive strategy with primary PCI.
In our study, the in-hospital MACE is similar in the two groups, but the other primary endpoint, in-hos-pital death, occurred less frequently in the pharmaco-invasive arm. This may be explained by the fact that we also considered patients who presented more than 12 hours from CP onset to EUH in the pPCI group, exceeding the time stated in guidelines. Nevertheless, a report of the ACC Foundation/AHA Task Force on practice guidelines states that the benefit from revas-cularization can extend up to 24 h, especially if there is ongoing evidence of coronary ischemia like our cases exceeding 12h14.
Data from the Mayo Clinic STEMI database of pa-tients treated with a pharmaco-invasive strategy or pPCI in a regional STEMI network from 2004 to 2012, using a multivariate analyses adjusting for age, gender, and other variables pointed out that there was no sig-nifi cant difference between the 2 strategies for 30-day MACE (RR: 0.66, 95% CI 0.36-1.21) or overall morta-lity (HR: 0.84, 95% CI 0.63-1.12)15.
Paradoxical findings, like lower OR of in-hospital MACE or death for the active smoking patients, have been observed. There is no doubt that diabetes and active smoking are risk factors for ACS/STEMI; it re-mains unclear whether active smoking is a risk factor for STEMI, but a factor with better prognosis for in-hospital MACE and death. High cholesterol seems to be “protective factor” for the in-hospital death.
In a 2016 editorial in the Romanian Journal of Car-diology, Frans van de Werf mentioned that the Stra-tegic Reperfusion Early after Myocardial Infarction (STREAM) trial and the French Registry of Acute ST-elevation or Non-ST-elevation Myocardial Infarction (FAST-MI, 2015) study suggested that pharmaco-inva-sive strategy compares favourably with primary PCI16.
The STREAM study showed that a strategy involving early fibrinolysis with bolus tenecteplase and contem-porary antithrombotic therapy (aspirin, clopidogrel and enoxaparin) offers similar efficacy as the primary PCI in patients with STEMI admitted within 3 hours of symptom onset and who could not undergo primary PCI within 1 hour of first medical contact7.
Regarding in-hospital mortality, our results are si-milar to the FAST-MI study for the PhIS group (2.9% in our study vs 4.3% in French Registry), while for the pPCI the in-hospital mortality is higher (14.80% in our study compared with 5% in FAST-MI)17,18. In our study, reperfusion was performed for the PhIS group by fibrinolytic therapy at FMH after a median time of 4 hours (IQR: 6.25h), while median CP onset-to-door time at EUH for the pPCI group was 7 hours (IQR: 12.38), compared with time to reperfusion therapy in FAST-MI (median 130 minutes and 300 minutes, re-spectively).
There are two explanations for better results in the FAST-MI study: (1) the FAST-MI study took into consideration both STEMI and NSTEMI patients, and by the longer median time to reperfusion in our trial 17,18.

It is interesting to discover that only over half (57.38%) of STEMI patients from remote areas outsi-de of a 2 hours reach of a catheterization laboratory (Cath lab) received thrombolysis at the FMH befo-re transfer to EUH, despite clear ESC guidelines for thrombolysis if the patients cannot reach a PCI-capa-ble facility within 120 minutes2. Therefore, 42.62% of patients assessed at the FMH did not undergo throm-bolysis before referral to a Cath lab for PCI. This may be explained by a lack of resources and skilled cardiologists/AMU physicians trained to administer thrombolytic agents in STEMI patients. The number of patients undergoing thrombolysis in our study (35 of 157 patients, 22.29%) is similar to number of patients receiving thrombolysis in the FAST-MI 2010 registry (291 of 1580 patients, 18.41%)19.
The Canadian trial, at the University of Ottawa Heart Institute regional STEMI system (between April 2009 and May 2011, with 236 and 980 consecutive pa-tients enrolled in pharmaco-invasive and primary PCI strategies, respectively) showed that there was no sig-nificant difference in the primary efficacy outcome (a composite of mortality, re-infarction, or stroke) (odds ratio: 1.54; p=0.21), but with a statistically non-signifi-cant tendency for increased bleeding in the PhIS group (odds ratio: 2.02; p=0.08)10. In our study, the OR for developing MACE is below one, with non-significant p-value, while the odds ratio for in-hospital death is 0.17 (p=0.090).
In a meta-analysis of randomized and controlled cli-nical trials of patients with STEMI, published in 2016, Roule et al. suggest that pharmaco-invasive strategy performed in the early prehospital setting was consis-tently associated with similar rates of short-term death and cardiovascular death and lower rates (decreased risk) of cardiogenic shock compared with pPCI20.
It is debatable whether immediate fibrinolysis followed by timely coronary angiography provides a clinical outcome similar to that of pPCI performed early after acute STEMI. Regarding treatment times, there are conflicting views in literature whether PCI should be performed after a delay from fibrinolytic therapy (immediate stenting vs deferred stenting), in order to decrease the side effects, like distal emboli-zation during stent implantation.
A study published in 2018, by Bendary et al (under-taken from December 2016 to June 2017) randomly assigned 60 patients to undergo either primary PCI (Group I) or immediate fibrinolysis with subsequent coronary angiography with PCI within 3 to 24 hours later (Group II). The results revealed no statistically significant difference in various components of in-ho-spital outcomes (including all-cause death and major bleeding up to 30 days) were found between groups21.
In our trial, there was no correlation between in-hospital MACE and the time from CP onset to FMH (r=0.007, p=0.929) or to EUH (r=0.045, p=0.585). This is similar to a meta-analysis by Lee et al that in-dicated a significant relationship between prolonged total ischemic time and reduced risk of MACE after deferred stenting (OR:0.994, 95% CI: 0.990-0.998; p=0.027)21. Although deferred stenting carried a sig-nificantly lower risk of peri-procedural composite events and abnormal fl ow in patients undergoing pri-mary PCI for STEMI, such benefits had no impact on MACE, which did not differ significantly by the timing of stent placement22.
In our study, the door-to-needle time is close to one hour, similar for both groups. A trial published in the European Heart Journal in 2012, by Larson et al regarding safety and efficacy of a pharmaco-invasive reperfusion strategy in rural STEMI patients, less than 20% of STEMI patients transferred for PCI in the USA have a door to balloon time <2h. Moreover, despite a significantly longer door-to-balloon time, there were no signifi cant differences in 30-day mortality (5.5 vs. 5.6%; p=0.94), or major bleeding (1.5 vs. 1.8%; p=0.65), or re-infarction/ischemia (1.2 vs. 2.5%; p=0.088) in pa-tients undergoing PhIS strategy compared with pati-ents presenting directly to the PCI center12.
A meta-analysis from 2012, of seven eligible trials (2961 patients) published by Borgia, Di Mario et al, that compared early routine PCI after successful fi – brinolysis vs standard therapy limiting PCI only to pa-tients without evidence of reperfusion (rescue PCI), found no difference in the incidence of death at 30 days between the two strategies23. Early routine PCI after successful fibrinolysis in STEMI patients signifi cantly reduced, the combined endpoint death/re-infarction (OR:0.65, 95% CI: 0.49-0.88; p=0.004) and recurrent ischemia at 30-day follow-up (OR:0.25, 95% CI: 0.13 0.49; p<0.001), with no significant increase in adverse bleeding events or stroke. In regards to delay after STEMI diagnosis, these trials suggest that all patients receiving fi brinolysis should receive mechanical revas-cularization within 24 h from initial hospitalization23.
Our study is very similar to another observational study, published in 2018, with 138 patients admitted with STEMI within 24 hours of symptom onset, in a single tertiary care center in India, over a 9-month pe-riod, that showed PhIS was as good as primary PCI in STEMI, where primary PCI may be delayed or not pos-sible at all due to fi nancial and logistical constraints. Incidence of composite primary endpoints (mortality within 30 days, cardiogenic shock and re-myocardial infarction) and secondary endpoints (arrhythmias, ble-eding manifestations, ischemic stroke, ejection fracti-on, mechanical complications, and duration of hospital stay) in PhIS was non-inferior to primary PCI at one month after intervention11. In our study, in-hospital mortality was similar for the pPCI arm (14.80% in our study vs 18.90% in the Indian study). In terms of in-hospital mortality of the PhIS group, our study proves better results (2.90% vs 11.60%) than the Indian trial11.
In regards to secondary outcomes (duration of in-hospital stay, safety of Cath lab procedure) no signifi-cant statistical differences was observed between the pharmaco-invasive strategy and primary PCI groups. There were similar amounts of contrast used, X-ray doses, AK and DAP in the two groups. The length of hospitalization was similar in both the PhIS and pPCI groups. We have to mention that we did not take into account local bleeding, as in our study the standard approach was via radial artery, as opposed to femoral approach previously used before 2010 as mentioned in some of the trials.
The conclusions of our research are similar to that of Di Mario and Wijns published in 2012, as it is very clear PhIS is an excellent strategy when STEMI patients face long distances to the Cath lab; although primary angioplasty is the preferred treatment if performed by an experienced team <120 minutes after fi rst medical contact, the fibrinolytic therapy at the local hospital is “buying time” before patient transfer to a PCI capable facility24.

Study limitations. We conducted the study at a tertiary care center equipped with an interventional Cath lab and not at the local hospital where the pa-tients are first seen. Twenty three patients admitted from local hospitals were excluded from the total of 180 patients (12.77%), as they could not be transferred to our facility. We did not process the median time between fibrinolysis and PCI.
Our research did not take into consideration the time from the moment of CP onset to calling the am-bulance or the time to reach the FMH. However, no clear data regarding the onset of chest pain were pro-vided by the patients and patients often delayed calling the ambulance for CP. The long delay from symptom onset to seeking medical assistance may be due to a poor understanding of the relationship of their symp-toms with a severe heart disease. This is one of the explanations of long delays from CP onset to Cath lab, and a possible explanation as to why some patients from local hospitals are not transferred to Cath labs facilities.
Of all the STEMI patients remote areas who could not reach our tertiary center within 120 minutes from the STEMI diagnosis, only 57.37% received fibrinolytic therapy.
Our study did not follow-up the patients after ho-spital discharge; we only took in-hospital endpoints into account. We also did not follow-up in-hospital bleeding, as we performed the procedures via radial approach.

CONCLUSIONS
Our study, based on a small observational cohort is one of the first local prospective studies, comparing PhIS with pPCI in Romania. The study shows that a strategy based on thrombolytic therapy before PCI for patients with STEMI appears to have similar pro-cedural outcomes with slightly better results vs pPCI in terms of in-hospital death. The odds of developing in-hospital MACE are similar in the two groups (PhIS/ pPCI). The treatment lags, for both PhIS and pPCI, are still high compared with other similar interventional studies. The treatment time from CP onset to reper-fusion is longer in the primary PCI group to the PhIS group, and this may explain lower in-hospital death for the pharmaco-invasive group. In conclusion, our study did not show any significant differences between the two strategies, in terms of in-hospital MACE, but with better results for in-hospital death for pharmaco-inva-sive strategy. In-hospital death is considerably higher in patients with STEMI due to LMCA compared with LAD, RCA, or LCx, irrespective of the initial reperfu-sion therapy.
Pharmaco-invasive strategy is an option for patients with STEMI in areas where no facilities for primary PCI are available. It is not inferior to the gold standard (pPCI), and we can consider it as a bailout solution with similar results in terms of efficacy and safety.
Conflict of interest: none declared. Acknowledgment. The authors of this paper would like to thank to their colleagues, medical and technical staff working in the A&E department: Simona Avîrvari, Simona Chirvase, Daniela Cojoacă, Dana Cojocaru, Diana Dobrin, Mihaela Ionescu, Mariana Livadaru, Li-liana Lixăndroiu, Loredana Mălureanu, Ioana Teleanu, Ioana Ţiganiuc, and Cristian Vătafu, as well to our col-leagues from the Catheterization Laboratory/Heart Department of EUH: Nicoleta Agiu, Iulia Dobrin, Mir-cea Echim, Irina Gurau, Georgeta Geambaşu, Adina Hovahinian, Ştefana Iancu, Viorica Mihai, Mihai Potco-varu and Roxana Şerban, for their dedication, support and contribution. We also, express our appreciation and gratitude for the work of our colleagues in the FMHs and ambulances in the south of Romania.

Abbreviations and Acronyms
AMI-PA = Acute Myocardial Infarction-Priority Action FMH = First medical hospital (local or community ho-spital)
EUH = Elias University Hospital
CPR = Cardio-Pulmonary Resuscitation EIRE = Elias Interventional Registry of STEMI PCI = percutaneous coronary intervention
pPCI = primary percutaneous coronary intervention PhIS = pharmaco-invasive strategy
dPCI = delayed percutaneous coronary intervention STEMI = ST-segment elevation myocardial infarction hs-cTn = high-sensitivity cardiac troponin IRCA = infarct-related coronary artery
LMCA = left main coronary artery
LAD = left anterior descending artery
LCx= left circumflex artery
RCA = right coronary artery
MACE = major adverse cardiac events

References
1. Jordan M, Caesar J. Improving door-to-needle times for patients pre-senting with ST-elevation myocardial infarction at a rural district gen-eral hospital. BMJ Quality Improvement Reports. 2016;5:u209049. w6736. doi:10.1136/bmjquality.u209049.w6736.
2. Ibanez B, James S, Agewall S, Antunes MJ, Bucciarelli-Ducci C, Bue-no H, Alida L P Caforio, Filippo Crea, John A Goudevenos, Sigrun Halvorsen, Gerhard Hindricks, Adnan Kastrati, Mattie J Lenzen, Prescott E, Roffi M, Valgimigli M, Varenhorst C, Vranckx P, Widim-ský 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). European Heart Journal. 2017;2018;39 (2):119–177. doi: 10.1093/eurheartj/ ehx393.
3. O’Gara PT, Kushner FG, Ascheim DD, Casey DE Jr, Chung MK, de Lemos JA, Ettinger SM, Fang JC, Fesmire FM, Franklin BA, et al. American College of Emergency Physicians, Society for Cardiovas-cular Angiography and Interventions. 2013 ACCF/AHA guideline for the management of ST-elevation myocardial infarction: executive summary: a report of the American College of Cardiology Founda-tion/American Heart Association Task Force on Practice Guidelines. J Am Coll Cardiol. 2013; 61(4):485-510.
4. Carville S, Harker M, Henderson M, Gray H. Acute management of myocardial infarction with ST-segment elevation: summary of NICE guidance. BMJ 2013; 347-f4006. doi: 10.1136/bmj.f4006.
5. Steg PG, James SK, Atar D, Badano LP, Blomstrom-Lundqvist C, Borger MA, Di Mario C, Dickstein K, Ducrocq G, Fernandez-Aviles F, Gershlick AH, Giannuzzi P, Halvorsen S, Huber K, Juni P, Kastrati A, Knuuti J, Lenzen MJ, Mahaffey KW, Valgimigli M, van’t Hof A, Wi-dimsky P, Zahger D. ESC Guidelines for the management of acute myocardial infarction in patients presenting with ST-segment eleva-tion. Eur Heart J 2012;33:2569e2619.
6. Vora AN, Holmes DN, Rokos I, Roe MT, Granger CB, French WJ et al. 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 (2): 207-215.
7. Armstrong PW, Gershlick AH, Goldstein P, Wilcox R, Danays T, Lambert Y, Sulimov V, Rosell Ortiz F, Ostojic M, Welsh RC, Carv-alho AC, Nanas J, Arntz HR, Halvorsen S, Huber K, Grajek S, Fres-co C, Bluhmki E, Regelin A, Vandenberghe K, Bogaerts K, Van de Werf F; STREAM Investigative Team. Fibrinolysis or primary PCI in ST-segment elevation myocardial infarction. N Engl J Med 2013; 368:1379-1387. doi: 10.1056/NEJMoa1301092.
8. Romanian Society of Cardiology, Acute Cardiac Care Working Group, Coordinator Tatu-Chiţoiu G. RO-STEMI, the Romanian reg-istry for ST-elevation myocardial infarction (1997-2009), final re-port. Amaltea Medical Publishing House. 2010. ISBN 978-973-162-068-8.
9. 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 Pharma-coinvasive Strategy in ST-Segment Elevation Myocardial Infarction: A Patient Population Study Comparing a Pharmacoinvasive Strategy With a Primary Percutaneous Coronary Intervention Strategy With-in a Regional System. JACC Cardiovasc Interv. 2016; 10;9(19):2014-2020. doi: 10.1016/j.jcin.2016.07.004.
10. Helal AM, Shaheen SM, Elhammady WA, Ahmed MI, Abdel-Ha-kim AS, Allam LE. Primary PCI versus pharmacoinvasive strategy for ST elevation myocardial infarction. Int J Cardiol Heart Vasc. 2018;27;21:87-93. doi: 10.1016/j.ijcha.
11. Alex AG, Lahiri A, Devika., Geevar T, George OK. Observational study comparing pharmaco-invasive strategy with primary percuta-neous coronary intervention in patients presenting with ST elevation myocardial infarction to a tertiary care center in India. J Postgrad Med. 2018; 64(2):80-85.
12. Larson DM, Duval S, Sharkey SW, Garberich RF, Madison JD, Stok-man PJ, Dirks TG, Westin RK, Harris JL, Henry TD. Safety and effica-cy of a pharmaco-invasive reperfusion strategy in rural ST-elevation myocardial infarction patients with expected delays due to long-dis-tance transfers. Eur Heart J. 2012 May;33(10):1232-40. doi: 10.1093/ eurheartj/ehr403.
13. Terkelsen CJ, Lassen JF, Nørgaard BL, Gerdes JC, Poulsen SH, Ben-dix K, Jens Peter Ankersen, Gøtzsche LB-H, Rømer FK, Nielsen TT, Andersen HR. Reduction of treatment delay in patients with ST-elevation myocardial infarction: impact of pre-hospital diagnosis and direct referral to primary percutaneous coronary intervention. Eu-ropean Heart Journal. 2005;26(8):770–777. doi: 10.1093/eurheartj/ ehi100.
14. O’Gara PT, Kushner FG, Ascheim DD, Casey DE, Jr, Chung MK, de Lemos JA. 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:e362–425. doi: 10.1161/ CIR.0b013e3182742cf6.
15. Siontis KC, Barsness GW, Lennon RJ, Holmen JL, Wright RS, Bell MR, Gersh BJ. Pharmaco-invasive and Primary Percutaneous Cor-onary Intervention Strategies in ST-Elevation Myocardial Infarction (from the Mayo Clinic STEMI Network). Am J Cardiol. 2016; 117(12):1904-10.
16. Frans van de Werf. The advantages of a pharmaco-invasive strategy. Romanian Journal of Cardiology. Vol 26, No. I, pg 6-8, 2016.
17. Danchin N, Coste P, Ferrières J, Steg PG, Cottin Y, Blanchard D, et al. Comparison of thrombolysis followed by broad use of percuta-neous coronary intervention with primary percutaneous coronary intervention for ST-segment-elevation acute myocardial infarction: Data from the french registry on acute ST-elevation myocardial in-farction (FAST-MI) Circulation. 2008;118:268–76.
18. Danchin N, Puymirat E, Steg PG, Goldstein P, Schiele F, Belle L, Cot-tin 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 reperfu-sion therapy. The French registry on Acute ST-elevation and non-ST-elevation Myocardial Infraction (FAST-MI) 2005 cohort. Circula-tion 2014;129: 1629-1636.
19. Puymirat E, Caudron J, Steg PG, Lemesle G, Cottin Y, Coste P, Schiele F, de Labriolle A, Bataille V, Ferrières J, Simon T1, Danchin N; FAST-MI investigators. Prognostic impact of non-compliance with guide-lines-recommended times to reperfusion therapy in ST-elevation
myocardial infarction. The FAST-MI 2010 registry. Eur Heart J Acute Cardiovasc Care. 2017;6(1):26-33. doi: 10.1177/2048872615610893.
20. Roule V, Ardouin P, Blanchart K, Lemaitre A, Wain-Hobson J, Legal-lois D, Alexandre J, Sabatier R, Milliez P, Beygui F. Prehospital fibri-nolysis versus primary percutaneous coronary intervention in ST-el-evation myocardial infarction: a systematic review and meta-analysis of randomized controlled trials. Crit Care. 2016 Nov 5;20(1):359.
21. Bendary A, Tawfek W, Mahros M, Salem M. Primary PCI versus Pharmaco-Invasive Strategy in Patients with ST-Elevation Myocardial Infarction; a Randomized Clinical Study. Journal of Cardiovascular Disease Research. 2018;9(1):28-31.
22. Lee JM, Rhee TM, Chang H, Ahn C, Park TK, Yang JH, Song YB, Choi SH, Gwon HC, Hahn JY. Deferred versus conventional stent implan-tation in patients with acute ST-segment elevation myocardial infarc-tion: An updated meta-analysis of 10 studies. Int J Cardiol. 2017 Mar 1;230:509-517. doi: 10.1016/j.ijcard.2016.12.071.
23. Borgia F, Goodman SG, Halvorsen S, Cantor WJ, Piscione F, Le May MR, Fernández-Avilés F, Sánchez PL, Dimopoulos K, Scheller B, Armstrong PW, Di Mario C. Early routine percutaneous coronary intervention after fibrinolysis vs. standard therapy in ST-segment elevation myocardial infarction: a meta-analysis. Eur Heart J. 2010 Sep;31(17):2156-69. doi: 10.1093/eurheartj/ehq204.
24. Di Mario C, Syrseloudis D, James S, Viceconte N, Wijns W. STEMI guidelines: from formulation to implementation. EuroIntervention 2012;8:P11-P17. doi: 10.4244/EIJV8SPA4.