Author + information
- Received September 8, 2000
- Revision received December 13, 2000
- Accepted February 13, 2001
- Published online June 1, 2001.
- Ralf Zahn, MD∗,* (, )
- Rudolf Schiele, MD∗,
- Steffen Schneider, PhD∗,
- Anselm K Gitt, MD∗,
- Harm Wienbergen, MD∗,
- Karlheinz Seidl, MD∗,
- Thomas Voigtländer, MD†,
- Martin Gottwik, MD‡,
- Gunther Berg, MD§,
- Ernst Altmann, MD∥,
- Werner Rosahl, MD¶,
- Jochen Senges, MD∗,
- for the Maximal Individual Therapy in Acute Myocardial Infarction (MITRA) and the Myocardial Infarction Registry (MIR) Study Groups
- ↵*Reprint requests and correspondence: Dr. Ralf Zahn, Department of Cardiology, Herzzentrum Ludwigshafen, Bremserstrasse 79, D-67063 Ludwigshafen, Germany
We sought to determine the effectiveness of primary angioplasty compared with thrombolysis in clinical practice.
In clinical practice, primary angioplasty for the treatment of acute myocardial infarction (AMI) has not yet been proven more effective than intravenous thrombolysis, nor have subgroups of patients been identified who would perhaps benefit from primary angioplasty.
The pooled data of two AMI registries—the Maximal Individual TheRapy in Acute myocardial infarction (MITRA) study and the Myocardial Infarction Registry (MIR)—were analyzed. A total of 9,906 lytic-eligible patients with AMI, with a pre-hospital delay of ≤12 h, were treated with either primary angioplasty (n = 1,327) or thrombolysis (n = 8,579).
Despite differences in the patients’ characteristics and concomitant diseases between the two groups, the prevalence of adverse risk factors was balanced. Univariate analysis of hospital mortality showed a more favorable course for patients treated with primary angioplasty: 6.4% versus 11.3% (odds ratio [OR] 0.54, 95% confidence interval [CI] 0.43 to 0.67). This was confirmed by logistic regression analysis (multivariate OR 0.58, 95% CI 0.44 to 0.77). Primary angioplasty was associated with a lower mortality in all subgroups analyzed. We observed a significant correlation between mortality and absolute risk reduction (r = 0.82, p < 0.0001) in the different subgroups: as mortality increased, there was an increase in absolute benefit of primary angioplasty compared with thrombolysis.
These large registry data showed the effect of primary angioplasty to be more favorable than thrombolysis for the treatment of patients with AMI in clinical practice. This effect was not restricted to special subgroups of patients. As mortality increased, the absolute benefit of primary angioplasty also increased.
Primary percutaneous transluminal coronary angioplasty—that is, angioplasty without preceding or concomitant thrombolysis—was introduced for the treatment of patients
with acute myocardial infarction (AMI) in 1982 and 1983 by Meyer et al. (1)and Hartzler et al. (2). During the following 15 years, a series of predominantly small, prospective, randomized trials comparing primary angioplasty with intravenous thrombolysis has been published (3–8). Meta-analysis of these randomized studies showed a superiority of primary angioplasty over thrombolysis in terms of hospital mortality (9).
However, there was concern whether these results, from highly specialized centers treating selected patients, could be transferred into clinical practice. This is especially true, as the largest randomized study—the Global Use of Strategies To Open occluded arteries in acute coronary syndromes (GUSTO IIb) angioplasty substudy (8)—showed only a moderate, nonsignificant reduction in hospital mortality for patients treated with primary angioplasty. Also, three myocardial infarction (MI) registries—the Myocardial Infarction Triage Investigators (MITI) Registry (10), the National Registry of Myocardial Infarction-2 (NRMI-2) (11)and a French registry (12)—failed to show an advantage of primary angioplasty over thrombolysis in the “real world.”
For these reasons, primary angioplasty is currently judged, according to official recommendations, to be an alternative to thrombolysis (13,14). Efforts have been made to define subgroups of patients with AMI benefiting most from primary angioplasty. However, these studies did not systematically investigate different subgroups, and they also showed divergent results. In some studies, primary angioplasty was superior to thrombolysis only in high-risk patients (6,15,16), whereas other studies showed a superiority of primary angioplasty in low-risk patients (17). Therefore, with the exception of patients in cardiogenic shock, no general recommendation exists regarding which subgroups of thrombolytic-eligible patients with AMI should be treated with primary angioplasty (18).
To determine the value of primary angioplasty compared with intravenous thrombolysis in different subgroups of patients, we analyzed the pooled data of two German AMI registries—the Maximal Individual TheRapy in Acute myocardial infarction (MITRA) (19)study and the Myocardial Infarction Registry (MIR) (20).
The MITRA and MIR registries were both German, prospective, multicenter, observational studies of the treatment of patients with AMI. Both studies sought to treat more AMI patients with established, prognostically important and recommended therapies. The MITRA study started in 1994 and ended 1997. Fifty-four hospitals participated in the study. The MIR study was a nationwide registry that included patients from 1996 to 1998. A total of 217 hospitals (mainly community hospitals) participated in MIR. The protocols of both studies were almost identical. Therefore, we used the pooled data from both studies for this analysis. All patients presenting within the first 96 h of the onset of pain were registered prospectively, as soon as the diagnosis of AMI had been made.
The following protocols for intravenous thrombolysis were suggested: intravenous application of 1.5 million units of streptokinase over 1 h, or tissue plasminogen activator (t-PA) at a dose of 100 mg over 1.5 h intravenously. Angioplasty was performed according to the standard protocol of each center. Angioplasty facilities were available at 8 (14.8%) of 54 hospitals participating in MITRA and 42 (19.4%) of 217 MIR centers. The decision regarding the type of treatment was left to the discretion of the treating physician, rather than according to the study protocol.
Acute myocardial infarction was diagnosed when the two following criteria were present: 1) persistent angina pectoris lasting ≥20 min and ST segment elevation ≥1 mm in at least two standard leads, or ≥2 mm in at least two contiguous precordial leads; or 2) left bundle branch block. It was later confirmed by cardiac enzyme elevation of more than twice as high as the normal upper range. “Pre-hospital delay” was defined as the time from the onset of symptoms to hospital admission. “In-hospital time to treatment” was defined as the time from admission to the start of primary angioplasty (angiographic needle entry) or the start of infusion of the thrombolytic agent. “Stroke” was defined as a new, persistent cerebral deficiency after hospital admission. Hemorrhagic stroke was not differentiated from nonhemorrhagic stroke or stroke of unknown etiology. A combined clinical end point was determined by the occurrence of death, re-infarction or stroke.
In this analysis, only lytic-eligible patients—that is, patients without contraindications to thrombolysis, treated with either primary angioplasty or intravenous thrombolysis, with a pre-hospital delay of ≤12 h—were included. Contraindications to thrombolysis consisted of stroke within the last three months, surgery or trauma within the last 14 days or active bleeding.
Data on the pre-hospital period and the early intra-hospital period (48 h) were collected within the first two to three days of the patients’ stay in the intensive care unit. Clinical events that occurred during the following hospital period were reported on a separate record form at hospital discharge. Every participating center was committed, during the study period, to include in the study each AMI patient who gave written consent. The patients also gave informed consent for processing of their anonymous data. All data sheets were sent to the central data processing center (Department of Cardiology, Herzzentrum Ludwigshafen) for uniform monitoring and registration.
Absolute numbers and percentages are used to describe the patient population. Median or mean values were computed, as appropriate. Categorical values were compared by chi-square analysis or the Fisher exact test, as appropriate, and the odds ratio (OR) and 95% confidence interval (CI) were computed. Continuous variables were compared by the two-tailed Wilcoxon rank-sum test. Logistic regression analysis was used to analyze the two types of reperfusion therapy. The following variables were examined: availability of angioplasty facilities, age, gender, location of infarction, cardiogenic shock, previous MI, resuscitation, heart failure at hospital admission, pre-hospital delay, renal failure at admission, history of arterial hypertension, and presence of a diagnostic first electrocardiogram (ECG). Logistic regression analyses were also used to adjust for factors influencing hospital mortality for the entire group and for subgroups. The following variables were examined: age, gender, location of infarction, cardiogenic shock, previous MI, resuscitation, heart failure at hospital admission, pre-hospital delay, type of revascularization, and concomitant therapy with beta-blockers and angiotensin-converting enzyme (ACE) inhibitors. Data on the prevalence of left or right bundle branch block were available only in the MITRA registry. Therefore, these variables were not included in the multiple logistic regression model. A further logistic regression analysis was done to compare the primary angioplasty group with the group treated with t-PA as the thrombolytic agent. We used the C statistic to determine the predictive value of each adjusted model. To analyze the effect of primary angioplasty compared with thrombolysis on mortality in different subgroups, the absolute risk reduction (ARR [%] = T death rate − PA death rate), the relative risk (RR = PA death rate/T death rate) (where T = thrombolysis and PA = primary angioplasty) and the number that needed to be treated to save one life (NNT = 100/ARR [%]) were calculated. Pearson’s correlation coefficient was used to calculate the relationship between RR and ARR and mortality in the subgroups. P values <0.05 were considered significant. All p values are the results of two-tailed tests. The tests were performed using the SAS statistical package, version 6.12 (Cary, North Carolina).
Selection and characteristics of patients
The selection of patients from the MITRA and MIR registries is shown in Figure 1. Of 22,749 patients included in both registries, 9,906 lytic-eligible patients with a pre-hospital delay of ≤12 h were treated with either primary angioplasty or thrombolysis. Of the 1,327 patients treated with primary angioplasty, 240 (18.1%) were transferred from other hospitals. Of the 8,579 patients treated with thrombolysis, 4,295 (50%) received streptokinase; 2,369 (27.6%) received t-PA; 1,423 (16.6%) received other thrombolytic agents; and 492 (5.7%) had no available data on the type of thrombolytic agent.
The patients’ characteristics and concomitant diseases are shown in Table 1. The prevalence of adverse risk factors seemed to be balanced between the two groups. Patients treated with primary angioplasty were significantly younger (median 62 vs. 64 years old, p = 0.006), were more often male (74.6% vs. 71.7%, p = 0.033) and showed heart failure less often at hospital admission (2.7% vs. 5.2%, p < 0.001). However, they more often had previous MIs (17.3% vs. 14.1%, p < 0.002), a higher prevalence of renal failure (2.5% vs. 1.3%, p < 0.001) and a history of arterial hypertension (41.2% vs. 35.4%, p < 0.001). The pre-hospital delay (median 150 vs. 120 min, p < 0.001) and the in-hospital time to treatment (median 70 vs. 30 min, p < 0.001) were longer in the primary angioplasty group.
Selection of reperfusion therapy
Logistic regression analysis showed the availability of angioplasty facilities to be the strongest independent predictor of the use of primary angioplasty (OR 24.18, 95% CI 20.50 to 28.65). A pre-hospital delay >6 h (OR 2.67, 95% CI 2.07 to 3.43), a nondiagnostic first ECG (OR 2.11, 95% CI 1.72 to 2.58), cardiogenic shock (OR 1.70, 95% CI 1.16 to 2.47) and a history of arterial hypertension (OR 1.39, 95% CI 1.19 to 1.62) were independently associated with a higher rate, whereas the presence of heart failure at hospital admission was associated with a lower rate (OR 0.46, 95% CI 0.29 to 0.70) of primary angioplasty compared with thrombolysis. No independent association with the use of primary angioplasty was found for age, gender, location of infarction, previous MI, renal failure or resuscitation.
Concomitant medications and clinical events
Patients treated with primary angioplasty more often received aspirin, heparin, beta-blockers and ACE inhibitors during the first 48 h after hospital admission (Table 2).
Univariate analysis of clinical events showed a more favorable course for patients treated with primary angioplasty compared with thrombolysis (re-infarction: 2.3% vs. 5.2%, OR 0.43, 95% CI 0.30 to 0.63; hospital mortality: 6.4% vs. 11.3%, OR 0.54, 95% CI 0.43 to 0.67; and the combined end point of death, re-infarction or stroke: 9.1% vs. 15.8%, OR 0.54, 95% CI 0.44 to 0.65) (Table 2). This was confirmed after adjusting for other identified confounding variables (multivariate OR 0.58 for hospital mortality, 95% CI 0.44 to 0.77 in favor of primary angioplasty) (Fig. 2). If we compared the primary angioplasty-treated group with the t-PA-treated group, mortality was 6.4% versus 10% (univariate OR 0.62, 95% CI 0.48 to 0.80; multivariate OR 0.61, 95% CI 0.44 to 0.83).
Primary angioplasty facilities were available at 50 (18.5%) of 271 hospitals. At hospitals with primary angioplasty facilities, thrombolysis was used in 58% of patients. At these hospitals, the mortality rates were 6.6% (72/1,087) for primary angioplasty and 11.3% (169/1,501) for thrombolysis (p < 0.001). At hospitals without primary angioplasty facilities, thrombolysis was used in 96.7% of patients. The remaining patients were treated with primary angioplasty. Mortality was lower at hospitals with angioplasty facilities (9.3% vs. 11.2%, p = 0.009). This difference was no longer significant after adjusting for the presence of angioplasty facilities and other identified confounding variables (OR 1.15, 95% CI 0.94 to 1.42, p = 0.1758).
In Registry patients without contraindications to thrombolysis and with a pre-hospital delay ≤12 h who did not receive reperfusion therapy, hospital was 20.4% (1,174/5,770) and the combined end point was 23.6% (1,364/5,770).
Analysis of different subgroups
Hospital mortality rates in different subgroups are shown in Table 3. Primary angioplasty was associated with lower mortality in all subgroups. The RR showed nearly constant values of ∼0.5 in favor of primary angioplasty over thrombolysis.
These differences in outcome between primary angioplasty and thrombolysis were confirmed by logistic regression analysis after adjusting for confounding variables (Fig. 2). Nearly all subgroups showed an independent association between lower mortality and primary angioplasty, as could be shown for the entire group.
We observed a significant correlation between mortality and the absolute reduction of mortality with primary angioplasty (r = 0.82, p < 0.0001) in the different subgroups: as mortality increased, there was an increase in absolute benefit of primary angioplasty compared with thrombolysis (Fig. 3). However, for the RR, no correlation with overall mortality was found in the different subgroups (r = −0.13, p = NS).
We are presenting, for the first time, registry data from the “real world” showing a superiority of primary angioplasty over intravenous thrombolysis in patients eligible for thrombolysis: 6.4% mortality for primary angioplasty versus 11.3% for thrombolysis (univariate OR 0.54, 95% CI 0.43 to 0.67; multivariate OR 0.58, 95% CI 0.44 to 0.77). This superiority was consistently observed in all subgroups. With an increase in overall mortality for a subgroup, the ARR of dying if treated with primary angioplasty versus thrombolysis increased substantially, expressed by lower NNTs.
At hospitals with angioplasty facilities, primary angioplasty should be the method of choice for reperfusion therapy.
Selection of type of reperfusion therapy in clinical practice
Although the decision of the type of reperfusion therapy was left to the treating physician, multivariate analysis showed some variables to be independently associated with a more frequent use of primary angioplasty. The availability of angioplasty facilities was the strongest independent predictor of the use of primary angioplasty. Nevertheless, thrombolysis was used in 58% of patients at hospitals with angioplasty facilities.
The preference for primary angioplasty in patients with a pre-hospital delay >6 h or those in cardiogenic shock in our registries may have been influenced by studies that suggested a better outcome with primary angioplasty compared with thrombolysis in these subgroups (21–24). For patients with a nondiagnostic first ECG, the availability and use of coronary angiography enable the physician to clarify the diagnosis and may lead to a higher use of primary angioplasty in such patients. In patients with heart failure at hospital admission, the potential hazard of the use of contrast agents in such patients may have led to the restricted use of primary angioplasty.
Primary angioplasty versus thrombolysis
Although the characteristics of patients treated with either primary angioplasty or thrombolysis were different for many variables, these differences were counterbalanced in terms of the distribution of adverse prognostic factors. Patients presenting with a diagnostic first ECG were less frequently observed in the primary angioplasty group than in the thrombolysis group. This might be explained by a less liberal use of thrombolysis in such patients. Tiefenbrunn et al. (11)described the same phenomenon in the NRMI-2 data. Mortality was not different in our patients with a diagnostic or a nondiagnostic first ECG. Primary angioplasty showed a favorable effect in both groups by univariate and multivariate analyses. Therefore, we did not exclude patients with a nondiagnostic first ECG.
In-hospital time to treatment was longer in patients treated with primary angioplasty than in those treated with thrombolysis (median 70 vs. 30 min, p < 0.001). This 40-min difference is well within the range (28 to 69 min) reported in the published data (5,6,8,10,11).
Patients treated with primary angioplasty were more likely to be treated with beta-blockers, compared with patients treated with thrombolysis. The same was found for treatment with ACE inhibitors. Hospitals with primary angioplasty facilities always have cardiologists on call. Treatment by specialized physicians may contribute to better clinical results (25)because they more frequently use recommended drug therapy.
Univariate analysis showed a superiority of primary angioplasty over thrombolysis regarding in-hospital mortality. This was confirmed by logistic regression analysis after adjusting for patient selection and differences in concomitant medications (6.4% vs. 11.3%, univariate OR 0.54, 95% CI 0.43 to 0.67, p < 0.0001; multivariate OR 0.58, 95% CI 0.44 to 0.77, p < 0.0001). These results confirm the data of a meta-analysis of the randomized studies (9)(4.4% mortality for primary angioplasty vs. 6.5% for thrombolysis; OR 0.66, 95% CI 0.46 to 0.94, p = 0.02), but they contrast with the data of three registries (10–12)that showed no benefit of primary angioplasty in clinical practice. In the MITRA and MIR registries, mortality in patients treated with primary angioplasty (6.4%) was similar to that of the other studies (5.2% to 9.2%) (10–12). However, mortality in patients treated with thrombolysis (11.3%) was higher than that reported by the other studies (5.4% to 7.6%) (10–12). These differences may be caused by a different method of patient selection in these studies. Every et al. (10), the NRMI-2 (11), as well as the randomized trials, excluded patients in cardiogenic shock. After exclusion of patients in shock in the MITRA and MIR registries, the mortality rate was 4.7% for primary angioplasty and 9.7% for thrombolysis (p < 0.0001). This 9.7% mortality rate is still higher than those rates previously reported and may be explained by inclusion of sicker patients in the MITRA and MIR registries. In the MITRA and MIR studies, patients were three years older than those in NRMI-2, and anterior wall infarctions occurred more often. In NRMI-2, only t-PA was used, compared with the 27.6% use of t-PA in the MITRA and MIR registries. However, this difference in the type of thrombolytic agent should result in only a small difference in mortality in patients treated with thrombolysis (26). This was confirmed by our data, which showed a mortality rate of 10% in patients treated with t-PA, compared with 11.3% for all patients treated with thrombolysis. The difference in hospital mortality for patients treated with primary angioplasty versus t-PA remained significant. Another factor that might contribute to the better outcome of primary angioplasty in the MITRA and MIR registries could be technical improvement over time. Thrombolysis did not change very much in the last years of the study period. However, primary angioplasty improved through the introduction of concomitant medications, such as glycoprotein IIb/IIIa receptor inhibitors (27,28), and new techniques, such as stents (29–31). Such an improvement in outcome in patients with AMI treated with primary angioplasty, but not with thrombolysis, could be demonstrated in another analysis of the MITRA and MIR data (32).
Analysis of subgroups
Analysis of subgroups should improve our knowledge of the principle of action of primary angioplasty compared with thrombolysis. Because of the restricted availability of hospitals performing primary angioplasty, it could help us to define subgroups of patients for whom a transfer to centers with primary angioplasty facilities should be considered.
Previous studies did not examine different subgroups in a systematic way, mainly because those studies involved small numbers of patients. Grines et al. (6)found an advantage of primary angioplasty over thrombolysis only in high-risk patients, but not for low-risk patients. However, Zijlstra et al. (17)reported a reduction in the incidence of the combined end point of death, re-infarction or stroke for primary angioplasty compared with thrombolysis, even in low-risk patients. Ribichini et al. (16)found a reduction in the incidence of the combined end point in patients with large inferior infarctions treated with primary angioplasty.
The large number of patients in the MITRA and MIR registries made it possible to systematically analyze the effects of primary angioplasty compared with thrombolysis in different subgroups of patients. We found a superiority of primary angioplasty in nearly all subgroups of patients in univariate and multivariate analyses. Data on the presence of bundle branch block was available only in the MITRA registry; therefore, no logistic regression analysis was performed in this subgroup. In patients with posterior AMI, heart failure at admission, a nondiagnostic first ECG or a previous MI, multivariate analysis did not reach statistical significance. However, the ORs in these subgroups clearly favored primary angioplasty and were in the same range as those in the other subgroups. Therefore, the missing statistical significance in these subgroups may be due to a lack of power.
Our data showed that the benefit of primary angioplasty over thrombolysis increased in those subgroups of patients who faced an increased risk of death. This was expressed by a strong correlation between the absolute reduction of mortality by primary angioplasty and the overall mortality in the different subgroups (r = 0.82, p < 0.0001). However, we found no decrease in the RR for primary angioplasty versus thrombolysis with increasing mortality (correlation between the RR of dying after primary angioplasty and overall mortality: r = −0.13, p = NS). This is shown especially for the subgroups of patients with different heart rates at hospital admission. Increases in heart rate were associated with increases in mortality: ≤79 beats/min, mortality 7.5%; 80 to 99 beats/min, mortality 9.9%; and ≥100 beats/min, mortality 21.3%. Primary angioplasty was associated with improved outcome in all subgroups. With increasing mortality, the ARR of dying after primary angioplasty increased favorably, as expressed by a decreasing NNT (28, 22 and 11). However, the RR favoring primary angioplasty did not change very much (0.55, 0.57 and 0.60) (Table 5).
These data are confirmed by Holmes et al. (33), who reported data from the GUSTO IIb study on the effect of age on outcome with primary angioplasty versus thrombolysis. They found that primary angioplasty improved outcome better than thrombolysis. However, it did not appear to be more beneficial in older compared with younger patients. Thus, the incremental adverse effect of age did not vary by treatment strategy.
Because MITRA and MIR are observational studies, it is not possible to totally control the selection of patients to be treated with one of the two therapies. We did not collect information on the rate of technical success (residual stenosis, Thrombolysis in Myocardial Infarction [TIMI] flow grade 3) of the angioplasty procedures or the use of stents and glycoprotein IIb/IIIa receptor inhibitors. Therefore, we were unable to control the results for these important variables. Only a minority of the participating hospitals had the facilities to perform primary angioplasty. There were great differences in reperfusion strategies at those hospitals with primary angioplasty facilities, as reported elsewhere (34). This fact contributes to potential selection bias. Median in-hospital time to primary angioplasty was 70 min in our study. Longer in-hospital delays are associated with a worse clinical outcome (35,36). Therefore, our data cannot be transferred to clinical settings with exceedingly longer in-hospital delays. This analysis of the MITRA and MIR data suffers from the limitations faced by all registries. Therefore, it is not possible to draw definitive conclusions regarding the mechanism responsible for the differences in short-term outcome between primary angioplasty and thrombolysis. However, the large number of patients at many different centers and the uniform results in different subgroups, which confirm the data of small, randomized studies, strengthen the results in favor of primary angioplasty.
☆ This study was supported in part by AstraZeneca, Bristol Myers-Squibb, Ministerium für Gesundheit, Arbeit, Soziales des Landes Rheinland-Pfalz, Landesversicherungsanstalt Rheinland-Pfalz, Barmer and Betriebskrankenkassen Rheinland-Pfalz.
- angiotensin-converting enzyme
- acute myocardial infarction
- absolute risk reduction
- confidence interval
- myocardial infarction
- Myocardial Infarction Registry
- Myocardial Infarction Triage Investigators registry
- Maximal Individual TheRapy in Acute myocardial infarction study
- National Registry of Myocardial Infarction-2
- odds ratio
- relative risk
- tissue plasminogen activator
- Received September 8, 2000.
- Revision received December 13, 2000.
- Accepted February 13, 2001.
- American College of Cardiology
- Mayo Coronary Care Unit and Catheterization Laboratory Groups,
- Gibbons R.J,
- Holmes D.R,
- Reeder G.S,
- et al.
- Ribeiro E.E,
- Silva L.A,
- Carneiro R,
- et al.
- Global Use of Strategies to Open Occluded Coronary Arteries in Acute Coronary Syndromes (GUSTO IIb) Angioplasty Substudy Investigators
- Tiefenbrunn A.J,
- Chandra N.C,
- French W.J,
- et al.
- Danchin N,
- Vaur L,
- Genes N,
- et al.
- Ryan T.J,
- Anderson J.L,
- Antman E.M,
- et al.
- Ryan T.J,
- Antman E.M,
- Brooks N.H,
- et al.
- Garcia E,
- Elizaga J,
- Perez-Castellano N,
- et al.
- Ribichini F,
- Steffenino G,
- Dellavalle A,
- et al.
- Zijlstra F,
- Beukema W.P,
- van’t Hof A.W,
- et al.
- Scanlon P.J,
- Faxon D.P,
- Audet A.M,
- et al.
- SHOCK Registry Investigators,
- Hochman J.S,
- Boland J,
- Sleeper L.A,
- et al.
- Moosvi A.R,
- Khaja F,
- Villanueva L,
- et al.
- ReoPro and Primary PTCA Organization and Randomized Trial (RAPPORT) Investigators,
- Brener S.J,
- Barr L.A,
- Burchenal J.E,
- et al.
- Neumann F.J,
- Blasini R,
- Schmitt C,
- et al.
- Antoniucci D,
- Santoro G.M,
- Bolognese L,
- et al.
- Suryapranata H,
- van’t Hof A.W,
- Hoorntje J.C,
- et al.
- ↵Zahn R, Schiele R, Schneider S, et al. Decreasing hospital mortality between 1994 and 1998 in patients with acute myocardial infarction treated with primary angioplasty, but not in patients treated with intravenous thrombolysis: results from the pooled data of the Maximal Individual TheRapy in Acute myocardial infarction (MITRA) Registry and the Myocardial Infarction Registry (MIR). J Am Coll Cardiol 2000;36:2064–71.
- Holmes D.R.J,
- White H.D,
- Pieper K.S,
- et al.
- Zahn R,
- Schiele R,
- Seidl K,
- et al.
- Berger P.B,
- Ellis S.G,
- Holmes D.R,
- et al.