- Email: [email protected]
Primary percutaneous transluminal coronary angioplasty for Acute Myocardial Infarction in patients not included in randomized studies
Primary Percutaneous Transluminal Coronary Angioplasty for Acute Myocardial Infarction in Patients Not Included in Randomized Studies Ralf Zahn, MD, Rudolf Schiele, MD, Karlheinz Seidl, MD, Caroline Bergmeier, MD, Karl K. Haase, MD, Hans G. Glunz, MD, Karl E. Hauptmann, MD, Thomas Voigtla ¨ nder, MD, Martin Gottwik, MD, and Jochen Senges, MD, for the Maximal Individual Therapy in Acute Myocardial Infarction (MITRA) Study Group* Patients with acute myocardial infarction included in randomized trials comparing primary percutaneous transluminal coronary angioplasty (pPTCA) with thrombolysis represent a special subgroup of patients with a low event rate. Patients excluded from these trials represent a variety of different subgroups, with different patient characteristics and possibly different clinical event rates. Primary PTCA was performed in 491 consecutive patients with acute myocardial infarction in the prospective multicenter observational Maximal Individual Therapy in Acute Myocardial Infarction trial. They were divided into the following groups: group I, patients fulfilling the inclusion criteria of the randomized trials (284 of 491, 58%); group II, patients not included in these trials (207 of 491, 42%). Of group II the following subgroups were defined: group IIa, patients in cardiogenic shock (20 of 491, 4.1%); group IIb, patients with a left bundle branch block (12 of 491, 2,4%); group IIc, patients with contraindications for thrombolysis (42 of 491, 8.6%); group IId, patients with a nondiagnostic first electrocardiogram (95 of 491, 19.3%); group IIe, patients with a prehospital delay of >12 hours (72 of 491, 14.7%); group IIf, patients with an unknown prehospital delay (30 of 491, 6.1%). A comparison of groups I and
II showed similar baseline characteristics but a higher clinical event rate during hospitalization was seen in group II: combined end point of death, reinfarction, heart failure equal to or greater than NYHA class III, any stroke or postinfarction angina, 26.6% versus 18%; p 5 0.022. Hospital deaths were nearly twice as high in these patients, without reaching statistical significance (10.6% vs 6%; p 5 0.06). The subgroups of group II showed quite different rates of clinical events. In-hospital death rates were: IIa, 40% (8 of 20); IIb, 8% (1 of 12); IIc, 12% (5 of 42); IId, 5% (5 of 95); IIe, 6% (4 of 72); and IIf, 13% (4 of 30). The incidence of the combined end point was 60% (12 of 20) in IIa, 33% (4 of 12) in IIb, 29% (12 of 42) in IIc, 16% (15 of 95) in IId, 26% (19 of 72) in IIe, and 33% (10 of 30) in IIf. Thus, in clinical practice, about half of the patients treated with pPTCA would not have been included in randomized trials comparing pPTCA with thrombolysis. These patients represent a population at higher risk for in hospital clinical events. However, they do represent very different nonhomogenous subgroups with different clinical event rates. Q1999 by Excerpta Medica, Inc. (Am J Cardiol 1999;83:1314 –1319)
he randomized trials comparing primary percutaneous transluminal coronary angioplasty (pPTCA) T with intravenous thrombolysis in patients with acute
tients with AMI. High-risk groups such as those in cardiogenic shock, patients receiving prolonged resuscitation, or those with a left bundle branch block were excluded. In addition, patients with contraindications for thrombolysis were not included in such studies. Some of these subgroups such as patients with cardiogenic shock8 –10 seem to profit even more from pPTCA versus treatment with thrombolysis than patients included in randomized trials. For patients with contraindications for thrombolysis, primary angioplasty seems to be the ideal method to reopen the infarctrelated artery and thus reduce mortality and morbidity.11–15 Although large registries have shown the feasibility of pPTCA in a clinical routine setting,16 –21 neither the proportion of these patients excluded in randomized trials nor the outcome of these patients compared with randomized patients has been described in detail. We used data of the Southwest German Maximal Individual Therapy in Acute Myocardial Infarction (MITRA) study22 to analyze the
myocardial infarction (AMI) had strict inclusion and exclusion criteria.1–7 Patients treated with pPTCA in these trials represent a highly selected group of paFrom the Herzzentrum Ludwigshafen, Ludwigshafen, the WestpfalzKlinikum, Kaiserslautern, the Krankenhaus der Barmherzigen Bru¨der Trier, the Johannes Gutenberg Universita¨t Mainz, and the Klinikum, Nu¨rnberg, Germany. This study was supported in part by Zeneca, Bristol Myers-Squibb, Ministerium fu¨r Gesundheit, Arbeit, Soziales des Landes Rheinland-Pfalz, and Lanesversicherungsanstalt RheinlandPfalz, Barmer und Betriebskrankenkassen Rheinland-Pfalz, Mainz, Germany. Manuscript received October 7, 1998; revised manuscript received and accepted December 30, 1998. Address for reprints: Ralf Zahn, MD, Herzzentrum Ludwigshafen, Department of Cardiology, Bremserstrabe 79, D-67063 Ludwigshafen, Germany. *The people and institutions who participated in the MITRA study are listed elsewhere.22
1314
©1999 by Excerpta Medica, Inc. All rights reserved.
0002-9149/99/$–see front matter PII S0002-9149(99)00092-2
FIGURE 1. Selection of patients from the MITRA trial.
proportions and clinical course of these subgroups of patients.
METHODS MITRA is a German, prospective, multicenter observational study of the current treatment of AMI. Fifty-four hospitals in the Southwest of Germany participated in the study, including university hospitals, tertiary care centers, and smaller hospitals. All consecutive patients presenting within the first 96 hours of the onset of pain were registered prospectively as soon as the diagnosis of AMI had been made. AMI was diagnosed in the presence of at least 2 of the 3 following criteria: persistent angina pectoris for $20 minutes; ST-segment elevation of $1 mm in at least 2 standard leads or $2 mm in at least 2 contiguous precordial leads; and elevation of cardiac enzymes twice as high as the normal upper range. AMI was also diagnosed by the presence of a left bundle branch block and persistent angina pectoris for $20 minutes. Final diagnosis of AMI, as well as the decision regarding the type of treatment, was left to the descretion of the treating physician and not to the study protocol. Angioplasty was performed according to the standard protocol of each center. Definitions: Time to treatment was defined as the time from admission to the hospital until the start of pPTCA (angiographic needle entry). Postinfarction angina was diagnosed in the presence of new angina pectoris within 2 weeks after myocardial infarction. Reinfarction was defined as recurrent chest pain lasting .20 minutes with new ST-segment elevation and either emergency angiographic confirmation of an occluded vessel or recurrent elevation of cardiac enzymes. Minor stroke was defined as a transient cerebral ischemia, and major stroke as a persistent cerebral ischemia. A combined clinical end point was defined by the occurrence of death, reinfarction, postinfarction
angina, or heart failure equal to or greater than New York Heart Association class III. For this analysis only patients treated with pPTCA were selected. Only 8 of the 54 participating centers had the facilities to perform pPTCA. Inclusion and exclusion criteria of the Primary Angioplasty in Myocardial Infarction (PAMI) group,1 the Zwolle trial,3 and the Mayo clinic trial2 were used to define patients who were included in trials comparing pPTCA with intravenous thrombolysis. According to these criteria, different groups of patients in the MITRA trial could be defined: group I, patients fulfilling the inclusion criteria of these randomized trials; group II, patients who would not have been included in these trials, with the following subgroups: group IIa, patients with cardiogenic shock; group IIb, patients with a left bundle branch block; group IIc, patients with contraindications for thrombolysis, such as active bleeding, recent (,3 months) cerebral infarction, recent (,14 days) surgery or trauma, or cardiopulmonary resuscitation before admission to the hospital; group IId, patients with a nondiagnostic first electrocardiogram; group IIe, patients with a prehospital delay of .12 hours; and group IIf, patients with an unknown prehospital delay. Also, group II patients were defined as those who would not have been included in the randomized trials and excluded patients in cardiogenic shock. Statistics: DATA COLLECTION: Data concerning the prehospital period and the early in-hospital period (first 48 hours) was collected within the first 2 to 3 days in the intensive care unit. Clinical events afterward were registered on a separate record form upon hospital discharge. All patients gave informed consent for processing their anonymous data. DATA ANALYSIS: Absolute numbers and percentages were computed to describe the patient population. Categorical values were compared using chi-square analysis. Continuous variables were compared by 2-tailed Wilcoxon rank-sum test. A p value ,0.05 was considered significant. All p values are 2-tailed. The tests were performed using the SAS statistical package, version 6.12 (Cary, North Carolina).
RESULTS The selection of patients from the MITRA registry is shown in Figure 1. Of 491 patients treated with pPTCA, 207 (42%) would not have been included in the randomized trials. Contraindications for thrombolysis were present in 42 of 491 patients (8.6%): 4 with an active bleeding, 6 with a recent (,3 months) cerebral infarction, 9 with a recent (,14 days) surgery or trauma, and 27 receiving a cardiopulmonary resuscitation before admission to the hospital. In those 72 patients with a prehospital delay of .12 hours, 19 had CORONARY ARTERY DISEASE/PRIMARY ANGIOPLASTY
1315
TABLE I Baseline Characteristics of Patients Included Versus Not Included in Randomized Trials All Patients (n 5 491) Age (yrs) Men Anterior location of infarct Prior infarction History of heart failure Systolic blood pressure (mm Hg) Heart rate (beats/min) RBBB Atrial fibrillation Prehospital delay (min) In-hospital time to treatment (min)
61 363/491 233/486 91/489 12/489 140 79 21/483 23/439 150 88
(53/70) (74%) (47.9%) (18.6%) (2.5%) (120/150) (67/93) (4.4%) (5.2%) (85/395) (55/150)
Randomized Trials Patients (n 5 284) 61 208/284 124/284 50/282 4/282 140 77 12/282 13/247 120 70
(53/70) (73%) (43.7%) (17.7%) (1.4%) (120/150) (65/90) (4.3%) (5.3%) (75/240) (50/107)
Patients Not Included in Randomized Trials (n 5 207) 61 155/207 109/202 41/207 8/207 130 81 9/201 10/192 340 127
(53/70) (75%) (54%)* (19.8%) (3.9%) (120/150) (68/97)* (4.5%) (5.2%) (91/1,280)† (70/382)†
Patients Not Included and Shock Patients Excluded (n 5 187) 60 143/187 100/183 37/187 7/187 135 80 8/185 9/173 412 136
(53.70) (76.5%) (54.6%)* (19.8%) (3.7%) (120/150) (68/94) (4.3%) (5.2%) (120/1,350)† (70/510)†
*p ,0.05; †p ,0.001. Median values are given with interquartile ranges (the 25th and 75th percentiles). RBBB 5 right bundle branch block.
TABLE II Clinical Events During Hospitalization in Patients Included Versus Not Included in Randomized Trials All Patients (n 5 491) Bleeding (gastrointestinal or with need of transfusion) Minor Stroke Major Stroke Heart failure ($NYHA class III) Postinfarction angina Reinfarction CABG/PTCA during hospital stay Death within 48 hours after admission In-hospital death Combined end point (death, reinfarction, heart failure, any stroke, postinfarction angina)
8/491 (1.6%) 5/491 8/491 26/491 36/481 20/491 81/413 18/491
(1%) (1.6%) (5.3%) (6.3%) (4.1%) (19.6%) (3.7%)
39/491 (7.9%) 106/491 (21.6%)
Randomized Trials Patients (n 5 284)
Patients Not Included in Randomized Trials (n 5 207)
Patients Not Included and Shock Patients Excluded (n 5 187)
4/284 (1.4%)
4/207 (1.9%)
3/187 (1.6%)
2/284 3/284 13/284 15/284 9/284 50/242 7/284
(0.7%) (1.1%) (4.6%) (5.3%) (3.2%) (20.7%) (2.5%)
17/284 (6%) 51/284 (18%)
3/207 5/207 13/207 21/207 11/207 31/171 11/207
(1.5%) (2.4%) (6.3%) (10.1%)* (5.5%) (18.1%) (5.3%)
22/207 (10.6%) 55/207 (26.6%)*
3/187 4/187 10/187 21/187 7/187 28/158 6/187
(1.6%) (2.1%) (5.4%) (11.2%)* (3.7%) (17.7%) (3.2%)
14/187 (7.5%) 43/187 (23%)
*p ,0.05. Median values are given with interquartile ranges (the 25th and 75th percentiles). CABG 5 coronary artery bypass grafting; NYHA 5 New York Heart Association.
a prehospital delay of 12 to 18 hours, 15 a delay of 18 to 24 hours, and 38 a delay of .24 hours.
Comparison of patients included and not included in randomized trials: Table I lists baseline characteristics
of the 3 groups. There was no difference in age and distribution of gender between groups. Patients who would have been included in current trials had a lower heart rate at admission, a lower frequency of anterior wall myocardial infarction, shorter prehospital delay times, as well as shorter in-hospital times to treatment than patients who would have not been included. Other baseline characteristics besides the selection criteria were not different between groups. After exclusion of patients in cardiogenic shock, baseline characteristics did not change in patients not included in randomized trials. Clinical events for the groups are listed in Table II. Almost all events occurred more frequently in patients not included in randomized trials. This difference was significant for postinfarction angina (10.1% vs 5.5%; p 5 0,041) and the combined end point (26.6% vs 18%; p 5 0.022). Hospital death 1316 THE AMERICAN JOURNAL OF CARDIOLOGYT
VOL. 83
occurred almost twice as often in these patients, without reaching statistical significance (10.6% vs 6%; p 5 0.061). However, when patients with shock were excluded from analysis, the mortality rates and the rates of the combined end point were similar in both groups. Analysis of subgroups of patients not included in randomized trials: Table III lists baseline characteris-
tics of the patients according to each subgroup. Prehospital delay times and in-hospital time to treatment showed a great variation between groups. Patients with a nondiagnostic first electrocardiogram and patients with a left bundle branch block had the longest prehospital delay times. The range of the median of in-hospital time to treatment was 85 to 205 minutes. Patients with a prehospital delay .12 hours and patients with a nondiagnostic first electrocardiogram had the longest time to treatment. Clinical events during hospitalization are listed in Table IV. Patients with a nondiagnostic first electrocardiogram had the lowest event rates, with a 5% in-hospital death rate and a 16% MAY 1, 1999
TABLE III Baseline Characteristics of Subgroups of Patients Not Included in Randomized Trials Cardiogenic Shock (n 5 20)
LBBB (n 5 12)
Contraindications for Thrombolysis (n 5 42)
Nondiagnostic First ECG (n 5 95)
Prehospital Delay .12 Hours (n 5 72)
Prehospital Delay Unknown (n 5 30)
Age (yrs) 66 (54/73) 77 (72/80) 57 (48/68) 61 (51/69) 61 (54/68) 63 (53/69) Men 12/20 (60%) 7/12 (58%) 33/72 (79%) 75/95 (79%) 50/72 (69%) 25/30 (83%) Anterior location 9/19 (47%) 6/11 (55%) 24/40 (60%) 52/92 (57%) 37/72 (51%) 17/30 (57%) of infarct Prior infarction 4/20 (20%) 4/12 (33%) 5/72 (12%) 27/95 (28%) 7/72 (10%) 7/30 (23%) History of heart 1/20 (5%) 2/12 (17%) 3/42 (7%) 1/95 (1%) 2/72 (3%) 1/30 (%) failure Cardiogenic shock 20/20 (100%) 1/10 (10%) 10/38 (26%) 2/84 (2%) 1/72 (1%) 1/29 (3%) (before treatment) Resuscitation 10/20 (50%) — 11/38 (29%) 5/84 (6%) 2/72 (3%) 3/29 (10%) Systolic blood 130 (90/140)* 150 (125/160) 130 (120/150) 130 (120/160) 135 (120/150) 140 (120/150) pressure (mm Hg) Heart rate 105 (95/128) 88 (73/100) 90 (72/106) 75 (64/94) 80 (68/92) 82 (77/97) (beats/min) LBBB 1/16 (6%) 12/12 (100%) 2/40 (5%) 7/94 (7%) 1/71 (1%) 2/30 (7%) Any BBB 2/16 (13%) 12/12 (100%) 5/40 (13%) 14/94 (15%) 4/71 (6%) 2/30 (7%) Atrial fibrillation 1/19 (5%) 2/10 (20%) 3/38 (8%) 5/82 (6%) 2/73 (2%) 2/29 (7%) Prehospital 75 (40/300) 187 (120/300) 90 (28/150) 232 (90/600) 1,445 (1,032/2,760) — delay (min) In-hospital time to 85 (60/140) 135 (80/1,219) 95 (60/151) 205 (81/660) 180 (90/750) 120 (92/250) treatment (min) *Values most probably obtained from patients receiving intravenous b-adrenoceptor agonists. Median values are given with interquartile ranges (the 25th and 75th percentiles) BBB 5 bundle branch block; ECG 5 electrocardiogram; LBBB 5 left bundle branch block.
TABLE IV Clinical Events During Hospitalization in Subgroups of Patients Not Included in Randomized Trials Cardiogenic Shock (n 5 20) Bleeding (gastrointestinal or needing transfusion) Minor stroke Major stroke Heart failure ($HYHA class III) Postinfarction angina Reinfarction CABG/PTCA during hospital stay Death within 48 hours after admission In-hospital death Combined end point (death, reinfarction, heart failure, any stroke, postinfarction angina)
LBBB (n 5 12)
Contraindications for Thrombolysis (n 5 42)
Nondiagnostic First ECG (n 5 95)
Prehospital Delay .12 Hours (n 5 72)
Prehospital Delay Unknown (n 5 30)
1/20 (5%)
1/12 (8%)
2/42 (5%)
1/95 (1%)
1/72 (1%)
1/30 (3%)
— 1/20 (5%) 3/20 (15%)
1/12 (8%) 1/12 (8%) 2/12 (17%)
2/42 (5%) 2/42 (5%) 1/42 (2%)
— 1/95 (1%) 6/95 (6%)
1/71 (1%) 2/72 (3%) 5/72 (7%)
1/30 (3%) — —
— 4/20 (20%) 3/13 (23%)
1/12 (8%) — 2/11 (18%)
2/42 (5%) 2/42 (5%) 6/30 (20%)
9/72 (13%) 4/72 (6%) 8/60 (13%)
4/30 (13%) 2/30 (7%) 5/27 (19%)
5/20 (25%)
—
2/42 (5%)
2/95 (2%)
2/72 (3%)
3/30 (0%)
1/12 (8%) 4/12 (33%)
5/42 (12%) 12/42 (29%)
5/95 (5%) 15/95 (16%)
4/72 (6%) 19/72 (26%)
4/30 (13%) 10/30 (33%)
8/20 (40%) 12/20 (60%)
9/95 (10%) 2/95 (2%) 18/84 (21%)
Median values are given with interquartile ranges (the 25th and 75th percentiles). Abbreviations as in Tables II and III.
incidence rate for the combined end point. The highest event rate was observed in patients with cardiogenic shock: death rate 40%, combined end point 60%.
DISCUSSION With the exception of the informed consent, all exclusion criteria of the randomized trials comparing pPTCA and intravenous thrombolysis in patients with AMI1–5,7 could be applied to the MITRA registry
patients treated with pPTCA. This population represented 42% (207 of 491) of all patients treated with pPTCA in a clinical routine setting. Unexpectedly, most baseline characteristics, such as age and gender, did not differ between the 2 main groups. Because patients with a prehospital delay .12 hours were excluded from nonrandomized trials, the prehospital delay for patients not included in randomized trials was much longer (340 vs 120 minutes, p 5 CORONARY ARTERY DISEASE/PRIMARY ANGIOPLASTY
1317
0.0001). As expected, the in-hospital time to treatment was also longer (127 vs 70 minutes, p 5 0.0001). Nearly all clinical events occurred more often in patients not included in randomized trials. This difference was significant for postinfarction angina (21 of 207 [10.1%] vs 15 of 284 [5.5%]; p 5 0.041) and the combined end point (55 of 20 [26.6%] vs 51 of 284 [18%]; p 5 0.022). Hospital death occurred almost twice as often in these patients without reaching statistical significance (22 of 207 [10.6%] vs 17 of 284 [6%]; p 5 0.06), because of the small sample size. The observed death rate of 6% in patients treated with pPTCA, who could have been recruited for the randomized studies, is close to the 4.7% death rate of the pooled data of the randomized studies23 and almost the same as the 5.7% death rate of the Global Use of Strategies To Open occluded arteries IIb trial,7 which also represents clinical practice in a broad spectrum of centers performing pPTCA. The differences in clinical event rates between the 2 groups were mainly due to the poor prognosis of patients presenting with cardiogenic shock. After exclusion of these 20 patients, event rates were virtually identical between groups (death rate, 6% vs 7.5%, p 5 NS; combined end point, 18% vs 23%; p 5 NS), with the exception of a still higher rate of postinfarction angina (5.3% vs 11.2%, p 5 0.017). These results are confirmed by the data of Himbert et al,24 who observed an in-hospital death rate of 7.8% (11 of 141) in patients eligible compared with 14.4% (34 of 236, p ,0.05) in patients not eligible for thrombolysis. This difference disappeared when patients in cardiogenic shock were excluded from analysis (7.8% vs 5.8%, p 5 NS). Analysis of the different subgroups of patients not included in randomized trials showed a wide variety in baseline characteristics and clinical events. Some of these subgroups, such as patients in cardiogenic shock8 –10 or contraindications for thrombolysis,11–15 are already described in published reports. The group of patients with a left bundle branch block was very small, representing only 2.4% (12 of 491) of all patients treated with pPTCA, so that conclusions are of limited value. Patients with AMI and contraindications for thrombolysis should be ideal candidates for pPTCA. However, another analysis of the MITRA data15 showed that most of these patients (86.4%) were still treated conservatively and only 13.6% were treated with pPTCA. Hospital mortality was significantly lower in patients who underwent pPTCA (univariate analysis, 2.2% vs 24.7%, p 5 0.001; multivariate analysis, odds ratio 0.46, p 5 0.0230). According to different definitions of contraindications, hospital mortality was 12% (5 of 42 patients) in the current analysis, which is close to the 12.5% observed in the Second National Registry of Myocardial Infarction study25 in this subset of patients. Patients with a nondiagnostic first electrocardiogram represented the largest subgroup of patients (95 of 207, 46%). There are many possible reasons for a nondiagnostic first electrocardiogram: (1) The infarc1318 THE AMERICAN JOURNAL OF CARDIOLOGYT
VOL. 83
tion may be small, so that the changes in the electrocardiogram are only minor. This is supported by the low clinical event rate in this group, which is the lowest of all subgroups. Hospital mortality was 5% (5 of 95) and the combined end point occurred in 16% (15 of 95). (2) The infarct-related artery may not be well represented on the conventional electrocardiogram, as is true for the circumflex branch of the left coronary artery. This may be 1 reason for the low incidence of circumflex occlusions (16% to 18%) in most angiographically controlled studies.21,26 (3) The electrocardiogram may be difficult to interpret because of a history of prior infarctions that left residuals on the electrocardiogram, so it is sometimes difficult to differentiate what is old and what is new. This hypothesis is supported by the high incidence (28%) of previous infarctions in this group compared with an overall incidence of 19% in all patients treated with pPTCA. (4) The physician reading the first electrocardiogram may not have been experienced enough to accurately diagnose subtle changes. (5) The prehospital delay might have been very short, so that the electrocardiogram may show only early nonspecific changes, which will become more obvious in a control electrocardiogram. However, prehospital delay times in our study were very long in this subgroup, with a median of 232 minutes. Patients with a prehospital delay of .12 hours normally were excluded from randomized studies. The reason for this is the time-dependent irreversible loss of myocardial muscle after the occlusion of a coronary artery. However, in patients with persistent angina pectoris or persistent ST-segment elevation, reperfusion therapy is an accepted option despite prehospital delays of .12 hours. The long prehospital delays were associated with a long in-hospital time to treatment (median 180 minutes) and probably reflects a longer decision-making process in these patients. “If the patients come late the doctors do not hurry either.” The clinical event rate was rather low in this subgroup of patients, with 6% in-hospital deaths and a 26% combined end point rate. There may be some reasons why prehospital delay will remain unknown in some patients: (1) They may be resuscitated before admission (10% of our patients), so that they cannot answer this question, because they are mechanically ventilated or under strong sedation. (2) In some patients the beginning of the infarction may be difficult to determine, because there is no clear onset of persistent symptoms. (3) There may be communication problems, especially in patients speaking only a foreign language or patients who are not always orientated in time. (4) No evaluation was performed of the prehospital delay. Baseline characteristics were not different from other subgroups and in-hospital time to treatment was intermediate (median 120 minutes). However, the clinical event rate was rather high in this subgroup, with a hospital death rate of 13% (4 of 30) and a combined end point rate of 33% (10 of 30). Study limitations: We did not collect information on the number of vessels dilated and the rate of technical MAY 1, 1999
success (residual stenosis, Thrombolysis In Myocardial Infarction trial grade 3 flow) of the angioplasty procedures, so we were unable to compare these variables among the different groups. For the given small sample size of 491 patients treated with pPTCA, the power to detect small significant differences is low. However, the primary goal of the current investigation was to demonstrate the spectrum of patients treated with pPTCA beyond patients included in studies comparing pPTCA with intravenous thrombolysis. Acknowledgment: We would like to thank Heinz Hochadel for his assistance with the statistical analysis, and Sine´ad Trainor for her helpful comments on the manuscript.
1. Grines CL, Browne KF, Marco J, Pothbaum D, Stone GW, O9Keefe J, Overlie P, Donohue B, Chelliah N, Timmis GC, Vlietstra RE, Strzelecki M, PuchrowiczOchocki S, O9Neill WW, for the Primary Angioplasty in Myocardial Infarction Study Group. A comparison of immediate angioplasty with thrombolytic therapy for acute myocardial infarction. N Engl J Med 1993;328:673– 679. 2. Gibbons RJ, Holmes DR, Reeder GS, Bailey KR, Hopfenspirger MR, Gersh BJ, for the Mayo Coronary Care Unit and Catheterization Laboratory Group. Immediate angioplasty compared with the administration of a thrombolytic agent followed by conservative treatment for myocardial infarction. N Engl J Med 1993;328:685– 691. 3. Zijlstra F, de Boer MJ, Hoorntje JCA, Reiffers S, Reiber JHC, Suryapranata H. A comparison of immediate coronary angioplasty with intravenous streptokinase in acute myocardial infarction. N Engl J Med 1993;328:680 – 684. 4. O9Neill W, Timmis GC, Bourdillon PD, Lai P, Ganghadarhan V, Walton J Jr, Ramos R, Laufer N, Gordon S, Schork MA. A prospective randomized trial of intracoronary streptokinase versus coronary angioplasty for acute myocardial infarction. N Engl J Med 1986;314:812– 818. 5. Ribeiro, EE, Silva LA, Carneiro R, D9Oliveira LG, Gasquez A, Amino JG, Tavares JR, Petrizzio A, Torossian S, Duprat R, Buffolo E, Ellis SG. Randomized trial of direct coronary angioplasty versus intravenous streptokinase in acute myocardial infarction. J Am Coll Cardiol 1993;22:376 –380. 6. Michels KB, Yusuf S. Does PTCA in acute myocardial infarction affect mortality and reinfarction rates? A quantitative overview (meta-analysis) of the randomized trials. Circulation 1995;91:476 – 485. 7. The Global Use Of Strategies To Open Occluded Coronary Arteries In Acute Coronary Syndromes (GUSTO IIb) Angioplasty Substudy Investigators (1997). A clinical trial comparing coronary angioplasty with tissue plasminogen activator for acute myocardial infarction. N Engl J Med 336:1621–1628. 8. Hochman, JS, Boland J, Sleeper LA, Porway M, Brinker J, Col J, Jacobs A, Slater J, Miller D, Wasserman H, Menegus MA, Talley D, McKimlay S, Sanborn T, LeJemtel T, and the SHOCK Registry Investigators. Current spectrum of cardiogenic shock and effect of early revascularization on mortality. Results of an international registry. Circulation 1995;91:873– 881. 9. Moosvi AR, Khaja F, Villanueva L, Gheorghiade M, Douthat L, Goldstein S. Early revascularization improves survival in cardiogenic shock complicating acute myocardial infarction. J Am Coll Cardiol 1992;19:907–914. 10. Emmerich K, Ulbricht LJ, Probst H, Krakau I, Hoffmeister T, Lu¨rken E, Gu¨lker E. Cardiogenic shock in acute acute myocardial infarction. Improving survival rates by primary coronary angioplasty. Z Kardiol 1995;84(suppl 2):25– 42. 11. Brodie BR, Weintraub RA, Stuckey TD, LeBauer EJ, Katz JD, Kelly TA, Hansen CJ. Outcomes of direct coronary angioplasty for acute myocardial in-
farction in candidates and non-candidates for thrombolytic therapy. Am J Cardiol 1991;67:7–12. 12. Stone GW, Grines CL, Browne KF, Marco J, Rothbaum D, O9Keefe JH, Hartzler GO, Overlie P, Donohue B, Chelliah N, Vlietstra R, PuchromiczOchocki S, O9Neill WW, for the Primary Angioplasty in Myocardial Infarction (PAMI) Investigators. Outcome of different reperfusion strategies in patients with former contraindications to thrombolytic therapy: a comparison of primary angioplasty and tissue plasminogen activator. Cathet Cardiovasc Diagn 1996;39: 333–339. 13. Himbert, D, Juliard J, Steg G, Badaoui G, Baleynaud S, Le Guludec D, Aumont M, Gourgon R. Primary angioplasty for acute myocardial infarction with contraindication to thrombolysis. Am J Cardiol 1993;71:377–381. 14. Waldecker B, Waas W, Heizmann H, Haberbosch W, Voss R, Walker G, Scha¨fer M, Kistler P, Tillmanns H. PTCA for acute myocardial infarction in patients not eligible for iv thrombolysis: in-hospital results. Fibrinolysis 1995; 9(suppl 1):126 –128. 15. Zahn R, Schuster S, Bergmeier C, Gottwik M, for the Maximal Individual Therapy in Acute Myocardial Infarction (MITRA) Study Group. Comparison of primary angioplasty with conservative therapy in patients with acute myocardial infarction and contraindications for thrombolytic therapy. Circulation 1998; 98(suppl I):I-363. 16. O9Neill W, Brodie BR, Ivanhoe R, Knopf W, Taylor G, O9Keefe JO, Grines CL, Weintraub R, Sickinger BG, Berdan LG, Tcheng JE, Woodlief LH, Strzelecki M, Hartzler G, Califf RM. Primary coronary angioplasty for acute myocardial infarction (the primary angioplasty registry). Am J Cardiol 1994;73:627– 634. 17. O9Keefe JO, Bailey WL, Rutherford BD, Hartzler GO. Primary angioplasty for acute myocardial infarction in 1000 consecutive patients. Am J Cardiol 1993;72:107G–115G. 18. Every NR, Parsons LS, Hlatky M, Martin JS, Weaver WD. A comparison of thrombolytic therapy with primary coronary angioplasty for acute myocardial infarction. N Engl J Med 1996;335:1253–1260. 19. Zahn R, Koch A, Rustige J, Schiele R, Wirtzfeld A, Neuhaus K, Kuhn H, Gu¨lker H, Senges J, for the ALKK Study Group. Primary angioplasty versus thrombolysis in the treatment of acute myocardial infarction. Am J Cardiol 1997;79:264 –269. 20. Rogers WJ, Dean LS, Moore PB, Wool KJ, Burgard SL, Bradley EL, for the Alabama Registry of Myocardial Ischemia Investigators. Comparison of primary angioplasty versus thrombolytic therapy for acute myocardial infarction. Am J Cardiol 1994;74:111–118. 21. Zahn R, Vogt A, Seidl K, Schuster S, Gu¨lker H, Heinrich KW, Gottwik M, Neuhaus K, Senges J, fu¨r die ALKK Studiengruppe. Ballondilatation beim akuten Herzinfarkt: Ergebnisse des Registers der Arbeitsgemeinschaft Leitender Kardiologischer Krankenhausa¨rzte (ALKK) bei 4625 Patienten. Z Kardiol 1997;86: 712–721 . 22. Schuster S, Koch A, Burczyk U, Schiele R, Wagner S, Zahn R, Glunz HG, Heinrich F, Stuby K, Berg G, Voigtla¨nder T, Gieseler U, Jakob M, Hauptmann P, Senges J, fu¨r die MITRA Studien-Gruppe. Therapy of acute myocardial infarction, impact of clinical trials on clinical practice: the MITRA pilot phase. Z Kardiol 1997;86:273–283. 23. Grines CL. Primary angioplasty—the strategy of choice. N Engl J Med 1996;335:1313–1315. 24. Himbert D, Feldman LJ, Boudvillain O, Benamer H, Juliard JM, Steg PG. Heterogeneity of prognosis in patient subsets treated by primary coronary angioplasty during acute myocardial infarction. Am J Cardiol 1998;81:1236 –1239. 25. Tiefenbrunn AJ, Chandra NC, French WJ, Gore JM, Rogers WJ. Clinical experience with primary percutaneous transluminal coronary angioplasty compared with alteplase (recombinat tissue-type plasminogen activator) in patients with acute myocardial infarction. A report from the second national registry of myocardial infarction (NRMI-2). J Am Coll Cardiol 1998;31:1240 –1245. 26. Grassman ED, Johnson SA, Krone RJ. Predictors of success and major complications for primary percutaneous transluminal coronary angioplasty in acute myocardial infarction. An analysis of the 1990 to 1994 society for cardiac angiography and interventions registries. J Am Coll Cardiol 1997;30:201–208.
CORONARY ARTERY DISEASE/PRIMARY ANGIOPLASTY
1319
II showed similar baseline characteristics but a higher clinical event rate during hospitalization was seen in group II: combined end point of death, reinfarction, heart failure equal to or greater than NYHA class III, any stroke or postinfarction angina, 26.6% versus 18%; p 5 0.022. Hospital deaths were nearly twice as high in these patients, without reaching statistical significance (10.6% vs 6%; p 5 0.06). The subgroups of group II showed quite different rates of clinical events. In-hospital death rates were: IIa, 40% (8 of 20); IIb, 8% (1 of 12); IIc, 12% (5 of 42); IId, 5% (5 of 95); IIe, 6% (4 of 72); and IIf, 13% (4 of 30). The incidence of the combined end point was 60% (12 of 20) in IIa, 33% (4 of 12) in IIb, 29% (12 of 42) in IIc, 16% (15 of 95) in IId, 26% (19 of 72) in IIe, and 33% (10 of 30) in IIf. Thus, in clinical practice, about half of the patients treated with pPTCA would not have been included in randomized trials comparing pPTCA with thrombolysis. These patients represent a population at higher risk for in hospital clinical events. However, they do represent very different nonhomogenous subgroups with different clinical event rates. Q1999 by Excerpta Medica, Inc. (Am J Cardiol 1999;83:1314 –1319)
he randomized trials comparing primary percutaneous transluminal coronary angioplasty (pPTCA) T with intravenous thrombolysis in patients with acute
tients with AMI. High-risk groups such as those in cardiogenic shock, patients receiving prolonged resuscitation, or those with a left bundle branch block were excluded. In addition, patients with contraindications for thrombolysis were not included in such studies. Some of these subgroups such as patients with cardiogenic shock8 –10 seem to profit even more from pPTCA versus treatment with thrombolysis than patients included in randomized trials. For patients with contraindications for thrombolysis, primary angioplasty seems to be the ideal method to reopen the infarctrelated artery and thus reduce mortality and morbidity.11–15 Although large registries have shown the feasibility of pPTCA in a clinical routine setting,16 –21 neither the proportion of these patients excluded in randomized trials nor the outcome of these patients compared with randomized patients has been described in detail. We used data of the Southwest German Maximal Individual Therapy in Acute Myocardial Infarction (MITRA) study22 to analyze the
myocardial infarction (AMI) had strict inclusion and exclusion criteria.1–7 Patients treated with pPTCA in these trials represent a highly selected group of paFrom the Herzzentrum Ludwigshafen, Ludwigshafen, the WestpfalzKlinikum, Kaiserslautern, the Krankenhaus der Barmherzigen Bru¨der Trier, the Johannes Gutenberg Universita¨t Mainz, and the Klinikum, Nu¨rnberg, Germany. This study was supported in part by Zeneca, Bristol Myers-Squibb, Ministerium fu¨r Gesundheit, Arbeit, Soziales des Landes Rheinland-Pfalz, and Lanesversicherungsanstalt RheinlandPfalz, Barmer und Betriebskrankenkassen Rheinland-Pfalz, Mainz, Germany. Manuscript received October 7, 1998; revised manuscript received and accepted December 30, 1998. Address for reprints: Ralf Zahn, MD, Herzzentrum Ludwigshafen, Department of Cardiology, Bremserstrabe 79, D-67063 Ludwigshafen, Germany. *The people and institutions who participated in the MITRA study are listed elsewhere.22
1314
©1999 by Excerpta Medica, Inc. All rights reserved.
0002-9149/99/$–see front matter PII S0002-9149(99)00092-2
FIGURE 1. Selection of patients from the MITRA trial.
proportions and clinical course of these subgroups of patients.
METHODS MITRA is a German, prospective, multicenter observational study of the current treatment of AMI. Fifty-four hospitals in the Southwest of Germany participated in the study, including university hospitals, tertiary care centers, and smaller hospitals. All consecutive patients presenting within the first 96 hours of the onset of pain were registered prospectively as soon as the diagnosis of AMI had been made. AMI was diagnosed in the presence of at least 2 of the 3 following criteria: persistent angina pectoris for $20 minutes; ST-segment elevation of $1 mm in at least 2 standard leads or $2 mm in at least 2 contiguous precordial leads; and elevation of cardiac enzymes twice as high as the normal upper range. AMI was also diagnosed by the presence of a left bundle branch block and persistent angina pectoris for $20 minutes. Final diagnosis of AMI, as well as the decision regarding the type of treatment, was left to the descretion of the treating physician and not to the study protocol. Angioplasty was performed according to the standard protocol of each center. Definitions: Time to treatment was defined as the time from admission to the hospital until the start of pPTCA (angiographic needle entry). Postinfarction angina was diagnosed in the presence of new angina pectoris within 2 weeks after myocardial infarction. Reinfarction was defined as recurrent chest pain lasting .20 minutes with new ST-segment elevation and either emergency angiographic confirmation of an occluded vessel or recurrent elevation of cardiac enzymes. Minor stroke was defined as a transient cerebral ischemia, and major stroke as a persistent cerebral ischemia. A combined clinical end point was defined by the occurrence of death, reinfarction, postinfarction
angina, or heart failure equal to or greater than New York Heart Association class III. For this analysis only patients treated with pPTCA were selected. Only 8 of the 54 participating centers had the facilities to perform pPTCA. Inclusion and exclusion criteria of the Primary Angioplasty in Myocardial Infarction (PAMI) group,1 the Zwolle trial,3 and the Mayo clinic trial2 were used to define patients who were included in trials comparing pPTCA with intravenous thrombolysis. According to these criteria, different groups of patients in the MITRA trial could be defined: group I, patients fulfilling the inclusion criteria of these randomized trials; group II, patients who would not have been included in these trials, with the following subgroups: group IIa, patients with cardiogenic shock; group IIb, patients with a left bundle branch block; group IIc, patients with contraindications for thrombolysis, such as active bleeding, recent (,3 months) cerebral infarction, recent (,14 days) surgery or trauma, or cardiopulmonary resuscitation before admission to the hospital; group IId, patients with a nondiagnostic first electrocardiogram; group IIe, patients with a prehospital delay of .12 hours; and group IIf, patients with an unknown prehospital delay. Also, group II patients were defined as those who would not have been included in the randomized trials and excluded patients in cardiogenic shock. Statistics: DATA COLLECTION: Data concerning the prehospital period and the early in-hospital period (first 48 hours) was collected within the first 2 to 3 days in the intensive care unit. Clinical events afterward were registered on a separate record form upon hospital discharge. All patients gave informed consent for processing their anonymous data. DATA ANALYSIS: Absolute numbers and percentages were computed to describe the patient population. Categorical values were compared using chi-square analysis. Continuous variables were compared by 2-tailed Wilcoxon rank-sum test. A p value ,0.05 was considered significant. All p values are 2-tailed. The tests were performed using the SAS statistical package, version 6.12 (Cary, North Carolina).
RESULTS The selection of patients from the MITRA registry is shown in Figure 1. Of 491 patients treated with pPTCA, 207 (42%) would not have been included in the randomized trials. Contraindications for thrombolysis were present in 42 of 491 patients (8.6%): 4 with an active bleeding, 6 with a recent (,3 months) cerebral infarction, 9 with a recent (,14 days) surgery or trauma, and 27 receiving a cardiopulmonary resuscitation before admission to the hospital. In those 72 patients with a prehospital delay of .12 hours, 19 had CORONARY ARTERY DISEASE/PRIMARY ANGIOPLASTY
1315
TABLE I Baseline Characteristics of Patients Included Versus Not Included in Randomized Trials All Patients (n 5 491) Age (yrs) Men Anterior location of infarct Prior infarction History of heart failure Systolic blood pressure (mm Hg) Heart rate (beats/min) RBBB Atrial fibrillation Prehospital delay (min) In-hospital time to treatment (min)
61 363/491 233/486 91/489 12/489 140 79 21/483 23/439 150 88
(53/70) (74%) (47.9%) (18.6%) (2.5%) (120/150) (67/93) (4.4%) (5.2%) (85/395) (55/150)
Randomized Trials Patients (n 5 284) 61 208/284 124/284 50/282 4/282 140 77 12/282 13/247 120 70
(53/70) (73%) (43.7%) (17.7%) (1.4%) (120/150) (65/90) (4.3%) (5.3%) (75/240) (50/107)
Patients Not Included in Randomized Trials (n 5 207) 61 155/207 109/202 41/207 8/207 130 81 9/201 10/192 340 127
(53/70) (75%) (54%)* (19.8%) (3.9%) (120/150) (68/97)* (4.5%) (5.2%) (91/1,280)† (70/382)†
Patients Not Included and Shock Patients Excluded (n 5 187) 60 143/187 100/183 37/187 7/187 135 80 8/185 9/173 412 136
(53.70) (76.5%) (54.6%)* (19.8%) (3.7%) (120/150) (68/94) (4.3%) (5.2%) (120/1,350)† (70/510)†
*p ,0.05; †p ,0.001. Median values are given with interquartile ranges (the 25th and 75th percentiles). RBBB 5 right bundle branch block.
TABLE II Clinical Events During Hospitalization in Patients Included Versus Not Included in Randomized Trials All Patients (n 5 491) Bleeding (gastrointestinal or with need of transfusion) Minor Stroke Major Stroke Heart failure ($NYHA class III) Postinfarction angina Reinfarction CABG/PTCA during hospital stay Death within 48 hours after admission In-hospital death Combined end point (death, reinfarction, heart failure, any stroke, postinfarction angina)
8/491 (1.6%) 5/491 8/491 26/491 36/481 20/491 81/413 18/491
(1%) (1.6%) (5.3%) (6.3%) (4.1%) (19.6%) (3.7%)
39/491 (7.9%) 106/491 (21.6%)
Randomized Trials Patients (n 5 284)
Patients Not Included in Randomized Trials (n 5 207)
Patients Not Included and Shock Patients Excluded (n 5 187)
4/284 (1.4%)
4/207 (1.9%)
3/187 (1.6%)
2/284 3/284 13/284 15/284 9/284 50/242 7/284
(0.7%) (1.1%) (4.6%) (5.3%) (3.2%) (20.7%) (2.5%)
17/284 (6%) 51/284 (18%)
3/207 5/207 13/207 21/207 11/207 31/171 11/207
(1.5%) (2.4%) (6.3%) (10.1%)* (5.5%) (18.1%) (5.3%)
22/207 (10.6%) 55/207 (26.6%)*
3/187 4/187 10/187 21/187 7/187 28/158 6/187
(1.6%) (2.1%) (5.4%) (11.2%)* (3.7%) (17.7%) (3.2%)
14/187 (7.5%) 43/187 (23%)
*p ,0.05. Median values are given with interquartile ranges (the 25th and 75th percentiles). CABG 5 coronary artery bypass grafting; NYHA 5 New York Heart Association.
a prehospital delay of 12 to 18 hours, 15 a delay of 18 to 24 hours, and 38 a delay of .24 hours.
Comparison of patients included and not included in randomized trials: Table I lists baseline characteristics
of the 3 groups. There was no difference in age and distribution of gender between groups. Patients who would have been included in current trials had a lower heart rate at admission, a lower frequency of anterior wall myocardial infarction, shorter prehospital delay times, as well as shorter in-hospital times to treatment than patients who would have not been included. Other baseline characteristics besides the selection criteria were not different between groups. After exclusion of patients in cardiogenic shock, baseline characteristics did not change in patients not included in randomized trials. Clinical events for the groups are listed in Table II. Almost all events occurred more frequently in patients not included in randomized trials. This difference was significant for postinfarction angina (10.1% vs 5.5%; p 5 0,041) and the combined end point (26.6% vs 18%; p 5 0.022). Hospital death 1316 THE AMERICAN JOURNAL OF CARDIOLOGYT
VOL. 83
occurred almost twice as often in these patients, without reaching statistical significance (10.6% vs 6%; p 5 0.061). However, when patients with shock were excluded from analysis, the mortality rates and the rates of the combined end point were similar in both groups. Analysis of subgroups of patients not included in randomized trials: Table III lists baseline characteris-
tics of the patients according to each subgroup. Prehospital delay times and in-hospital time to treatment showed a great variation between groups. Patients with a nondiagnostic first electrocardiogram and patients with a left bundle branch block had the longest prehospital delay times. The range of the median of in-hospital time to treatment was 85 to 205 minutes. Patients with a prehospital delay .12 hours and patients with a nondiagnostic first electrocardiogram had the longest time to treatment. Clinical events during hospitalization are listed in Table IV. Patients with a nondiagnostic first electrocardiogram had the lowest event rates, with a 5% in-hospital death rate and a 16% MAY 1, 1999
TABLE III Baseline Characteristics of Subgroups of Patients Not Included in Randomized Trials Cardiogenic Shock (n 5 20)
LBBB (n 5 12)
Contraindications for Thrombolysis (n 5 42)
Nondiagnostic First ECG (n 5 95)
Prehospital Delay .12 Hours (n 5 72)
Prehospital Delay Unknown (n 5 30)
Age (yrs) 66 (54/73) 77 (72/80) 57 (48/68) 61 (51/69) 61 (54/68) 63 (53/69) Men 12/20 (60%) 7/12 (58%) 33/72 (79%) 75/95 (79%) 50/72 (69%) 25/30 (83%) Anterior location 9/19 (47%) 6/11 (55%) 24/40 (60%) 52/92 (57%) 37/72 (51%) 17/30 (57%) of infarct Prior infarction 4/20 (20%) 4/12 (33%) 5/72 (12%) 27/95 (28%) 7/72 (10%) 7/30 (23%) History of heart 1/20 (5%) 2/12 (17%) 3/42 (7%) 1/95 (1%) 2/72 (3%) 1/30 (%) failure Cardiogenic shock 20/20 (100%) 1/10 (10%) 10/38 (26%) 2/84 (2%) 1/72 (1%) 1/29 (3%) (before treatment) Resuscitation 10/20 (50%) — 11/38 (29%) 5/84 (6%) 2/72 (3%) 3/29 (10%) Systolic blood 130 (90/140)* 150 (125/160) 130 (120/150) 130 (120/160) 135 (120/150) 140 (120/150) pressure (mm Hg) Heart rate 105 (95/128) 88 (73/100) 90 (72/106) 75 (64/94) 80 (68/92) 82 (77/97) (beats/min) LBBB 1/16 (6%) 12/12 (100%) 2/40 (5%) 7/94 (7%) 1/71 (1%) 2/30 (7%) Any BBB 2/16 (13%) 12/12 (100%) 5/40 (13%) 14/94 (15%) 4/71 (6%) 2/30 (7%) Atrial fibrillation 1/19 (5%) 2/10 (20%) 3/38 (8%) 5/82 (6%) 2/73 (2%) 2/29 (7%) Prehospital 75 (40/300) 187 (120/300) 90 (28/150) 232 (90/600) 1,445 (1,032/2,760) — delay (min) In-hospital time to 85 (60/140) 135 (80/1,219) 95 (60/151) 205 (81/660) 180 (90/750) 120 (92/250) treatment (min) *Values most probably obtained from patients receiving intravenous b-adrenoceptor agonists. Median values are given with interquartile ranges (the 25th and 75th percentiles) BBB 5 bundle branch block; ECG 5 electrocardiogram; LBBB 5 left bundle branch block.
TABLE IV Clinical Events During Hospitalization in Subgroups of Patients Not Included in Randomized Trials Cardiogenic Shock (n 5 20) Bleeding (gastrointestinal or needing transfusion) Minor stroke Major stroke Heart failure ($HYHA class III) Postinfarction angina Reinfarction CABG/PTCA during hospital stay Death within 48 hours after admission In-hospital death Combined end point (death, reinfarction, heart failure, any stroke, postinfarction angina)
LBBB (n 5 12)
Contraindications for Thrombolysis (n 5 42)
Nondiagnostic First ECG (n 5 95)
Prehospital Delay .12 Hours (n 5 72)
Prehospital Delay Unknown (n 5 30)
1/20 (5%)
1/12 (8%)
2/42 (5%)
1/95 (1%)
1/72 (1%)
1/30 (3%)
— 1/20 (5%) 3/20 (15%)
1/12 (8%) 1/12 (8%) 2/12 (17%)
2/42 (5%) 2/42 (5%) 1/42 (2%)
— 1/95 (1%) 6/95 (6%)
1/71 (1%) 2/72 (3%) 5/72 (7%)
1/30 (3%) — —
— 4/20 (20%) 3/13 (23%)
1/12 (8%) — 2/11 (18%)
2/42 (5%) 2/42 (5%) 6/30 (20%)
9/72 (13%) 4/72 (6%) 8/60 (13%)
4/30 (13%) 2/30 (7%) 5/27 (19%)
5/20 (25%)
—
2/42 (5%)
2/95 (2%)
2/72 (3%)
3/30 (0%)
1/12 (8%) 4/12 (33%)
5/42 (12%) 12/42 (29%)
5/95 (5%) 15/95 (16%)
4/72 (6%) 19/72 (26%)
4/30 (13%) 10/30 (33%)
8/20 (40%) 12/20 (60%)
9/95 (10%) 2/95 (2%) 18/84 (21%)
Median values are given with interquartile ranges (the 25th and 75th percentiles). Abbreviations as in Tables II and III.
incidence rate for the combined end point. The highest event rate was observed in patients with cardiogenic shock: death rate 40%, combined end point 60%.
DISCUSSION With the exception of the informed consent, all exclusion criteria of the randomized trials comparing pPTCA and intravenous thrombolysis in patients with AMI1–5,7 could be applied to the MITRA registry
patients treated with pPTCA. This population represented 42% (207 of 491) of all patients treated with pPTCA in a clinical routine setting. Unexpectedly, most baseline characteristics, such as age and gender, did not differ between the 2 main groups. Because patients with a prehospital delay .12 hours were excluded from nonrandomized trials, the prehospital delay for patients not included in randomized trials was much longer (340 vs 120 minutes, p 5 CORONARY ARTERY DISEASE/PRIMARY ANGIOPLASTY
1317
0.0001). As expected, the in-hospital time to treatment was also longer (127 vs 70 minutes, p 5 0.0001). Nearly all clinical events occurred more often in patients not included in randomized trials. This difference was significant for postinfarction angina (21 of 207 [10.1%] vs 15 of 284 [5.5%]; p 5 0.041) and the combined end point (55 of 20 [26.6%] vs 51 of 284 [18%]; p 5 0.022). Hospital death occurred almost twice as often in these patients without reaching statistical significance (22 of 207 [10.6%] vs 17 of 284 [6%]; p 5 0.06), because of the small sample size. The observed death rate of 6% in patients treated with pPTCA, who could have been recruited for the randomized studies, is close to the 4.7% death rate of the pooled data of the randomized studies23 and almost the same as the 5.7% death rate of the Global Use of Strategies To Open occluded arteries IIb trial,7 which also represents clinical practice in a broad spectrum of centers performing pPTCA. The differences in clinical event rates between the 2 groups were mainly due to the poor prognosis of patients presenting with cardiogenic shock. After exclusion of these 20 patients, event rates were virtually identical between groups (death rate, 6% vs 7.5%, p 5 NS; combined end point, 18% vs 23%; p 5 NS), with the exception of a still higher rate of postinfarction angina (5.3% vs 11.2%, p 5 0.017). These results are confirmed by the data of Himbert et al,24 who observed an in-hospital death rate of 7.8% (11 of 141) in patients eligible compared with 14.4% (34 of 236, p ,0.05) in patients not eligible for thrombolysis. This difference disappeared when patients in cardiogenic shock were excluded from analysis (7.8% vs 5.8%, p 5 NS). Analysis of the different subgroups of patients not included in randomized trials showed a wide variety in baseline characteristics and clinical events. Some of these subgroups, such as patients in cardiogenic shock8 –10 or contraindications for thrombolysis,11–15 are already described in published reports. The group of patients with a left bundle branch block was very small, representing only 2.4% (12 of 491) of all patients treated with pPTCA, so that conclusions are of limited value. Patients with AMI and contraindications for thrombolysis should be ideal candidates for pPTCA. However, another analysis of the MITRA data15 showed that most of these patients (86.4%) were still treated conservatively and only 13.6% were treated with pPTCA. Hospital mortality was significantly lower in patients who underwent pPTCA (univariate analysis, 2.2% vs 24.7%, p 5 0.001; multivariate analysis, odds ratio 0.46, p 5 0.0230). According to different definitions of contraindications, hospital mortality was 12% (5 of 42 patients) in the current analysis, which is close to the 12.5% observed in the Second National Registry of Myocardial Infarction study25 in this subset of patients. Patients with a nondiagnostic first electrocardiogram represented the largest subgroup of patients (95 of 207, 46%). There are many possible reasons for a nondiagnostic first electrocardiogram: (1) The infarc1318 THE AMERICAN JOURNAL OF CARDIOLOGYT
VOL. 83
tion may be small, so that the changes in the electrocardiogram are only minor. This is supported by the low clinical event rate in this group, which is the lowest of all subgroups. Hospital mortality was 5% (5 of 95) and the combined end point occurred in 16% (15 of 95). (2) The infarct-related artery may not be well represented on the conventional electrocardiogram, as is true for the circumflex branch of the left coronary artery. This may be 1 reason for the low incidence of circumflex occlusions (16% to 18%) in most angiographically controlled studies.21,26 (3) The electrocardiogram may be difficult to interpret because of a history of prior infarctions that left residuals on the electrocardiogram, so it is sometimes difficult to differentiate what is old and what is new. This hypothesis is supported by the high incidence (28%) of previous infarctions in this group compared with an overall incidence of 19% in all patients treated with pPTCA. (4) The physician reading the first electrocardiogram may not have been experienced enough to accurately diagnose subtle changes. (5) The prehospital delay might have been very short, so that the electrocardiogram may show only early nonspecific changes, which will become more obvious in a control electrocardiogram. However, prehospital delay times in our study were very long in this subgroup, with a median of 232 minutes. Patients with a prehospital delay of .12 hours normally were excluded from randomized studies. The reason for this is the time-dependent irreversible loss of myocardial muscle after the occlusion of a coronary artery. However, in patients with persistent angina pectoris or persistent ST-segment elevation, reperfusion therapy is an accepted option despite prehospital delays of .12 hours. The long prehospital delays were associated with a long in-hospital time to treatment (median 180 minutes) and probably reflects a longer decision-making process in these patients. “If the patients come late the doctors do not hurry either.” The clinical event rate was rather low in this subgroup of patients, with 6% in-hospital deaths and a 26% combined end point rate. There may be some reasons why prehospital delay will remain unknown in some patients: (1) They may be resuscitated before admission (10% of our patients), so that they cannot answer this question, because they are mechanically ventilated or under strong sedation. (2) In some patients the beginning of the infarction may be difficult to determine, because there is no clear onset of persistent symptoms. (3) There may be communication problems, especially in patients speaking only a foreign language or patients who are not always orientated in time. (4) No evaluation was performed of the prehospital delay. Baseline characteristics were not different from other subgroups and in-hospital time to treatment was intermediate (median 120 minutes). However, the clinical event rate was rather high in this subgroup, with a hospital death rate of 13% (4 of 30) and a combined end point rate of 33% (10 of 30). Study limitations: We did not collect information on the number of vessels dilated and the rate of technical MAY 1, 1999
success (residual stenosis, Thrombolysis In Myocardial Infarction trial grade 3 flow) of the angioplasty procedures, so we were unable to compare these variables among the different groups. For the given small sample size of 491 patients treated with pPTCA, the power to detect small significant differences is low. However, the primary goal of the current investigation was to demonstrate the spectrum of patients treated with pPTCA beyond patients included in studies comparing pPTCA with intravenous thrombolysis. Acknowledgment: We would like to thank Heinz Hochadel for his assistance with the statistical analysis, and Sine´ad Trainor for her helpful comments on the manuscript.
1. Grines CL, Browne KF, Marco J, Pothbaum D, Stone GW, O9Keefe J, Overlie P, Donohue B, Chelliah N, Timmis GC, Vlietstra RE, Strzelecki M, PuchrowiczOchocki S, O9Neill WW, for the Primary Angioplasty in Myocardial Infarction Study Group. A comparison of immediate angioplasty with thrombolytic therapy for acute myocardial infarction. N Engl J Med 1993;328:673– 679. 2. Gibbons RJ, Holmes DR, Reeder GS, Bailey KR, Hopfenspirger MR, Gersh BJ, for the Mayo Coronary Care Unit and Catheterization Laboratory Group. Immediate angioplasty compared with the administration of a thrombolytic agent followed by conservative treatment for myocardial infarction. N Engl J Med 1993;328:685– 691. 3. Zijlstra F, de Boer MJ, Hoorntje JCA, Reiffers S, Reiber JHC, Suryapranata H. A comparison of immediate coronary angioplasty with intravenous streptokinase in acute myocardial infarction. N Engl J Med 1993;328:680 – 684. 4. O9Neill W, Timmis GC, Bourdillon PD, Lai P, Ganghadarhan V, Walton J Jr, Ramos R, Laufer N, Gordon S, Schork MA. A prospective randomized trial of intracoronary streptokinase versus coronary angioplasty for acute myocardial infarction. N Engl J Med 1986;314:812– 818. 5. Ribeiro, EE, Silva LA, Carneiro R, D9Oliveira LG, Gasquez A, Amino JG, Tavares JR, Petrizzio A, Torossian S, Duprat R, Buffolo E, Ellis SG. Randomized trial of direct coronary angioplasty versus intravenous streptokinase in acute myocardial infarction. J Am Coll Cardiol 1993;22:376 –380. 6. Michels KB, Yusuf S. Does PTCA in acute myocardial infarction affect mortality and reinfarction rates? A quantitative overview (meta-analysis) of the randomized trials. Circulation 1995;91:476 – 485. 7. The Global Use Of Strategies To Open Occluded Coronary Arteries In Acute Coronary Syndromes (GUSTO IIb) Angioplasty Substudy Investigators (1997). A clinical trial comparing coronary angioplasty with tissue plasminogen activator for acute myocardial infarction. N Engl J Med 336:1621–1628. 8. Hochman, JS, Boland J, Sleeper LA, Porway M, Brinker J, Col J, Jacobs A, Slater J, Miller D, Wasserman H, Menegus MA, Talley D, McKimlay S, Sanborn T, LeJemtel T, and the SHOCK Registry Investigators. Current spectrum of cardiogenic shock and effect of early revascularization on mortality. Results of an international registry. Circulation 1995;91:873– 881. 9. Moosvi AR, Khaja F, Villanueva L, Gheorghiade M, Douthat L, Goldstein S. Early revascularization improves survival in cardiogenic shock complicating acute myocardial infarction. J Am Coll Cardiol 1992;19:907–914. 10. Emmerich K, Ulbricht LJ, Probst H, Krakau I, Hoffmeister T, Lu¨rken E, Gu¨lker E. Cardiogenic shock in acute acute myocardial infarction. Improving survival rates by primary coronary angioplasty. Z Kardiol 1995;84(suppl 2):25– 42. 11. Brodie BR, Weintraub RA, Stuckey TD, LeBauer EJ, Katz JD, Kelly TA, Hansen CJ. Outcomes of direct coronary angioplasty for acute myocardial in-
farction in candidates and non-candidates for thrombolytic therapy. Am J Cardiol 1991;67:7–12. 12. Stone GW, Grines CL, Browne KF, Marco J, Rothbaum D, O9Keefe JH, Hartzler GO, Overlie P, Donohue B, Chelliah N, Vlietstra R, PuchromiczOchocki S, O9Neill WW, for the Primary Angioplasty in Myocardial Infarction (PAMI) Investigators. Outcome of different reperfusion strategies in patients with former contraindications to thrombolytic therapy: a comparison of primary angioplasty and tissue plasminogen activator. Cathet Cardiovasc Diagn 1996;39: 333–339. 13. Himbert, D, Juliard J, Steg G, Badaoui G, Baleynaud S, Le Guludec D, Aumont M, Gourgon R. Primary angioplasty for acute myocardial infarction with contraindication to thrombolysis. Am J Cardiol 1993;71:377–381. 14. Waldecker B, Waas W, Heizmann H, Haberbosch W, Voss R, Walker G, Scha¨fer M, Kistler P, Tillmanns H. PTCA for acute myocardial infarction in patients not eligible for iv thrombolysis: in-hospital results. Fibrinolysis 1995; 9(suppl 1):126 –128. 15. Zahn R, Schuster S, Bergmeier C, Gottwik M, for the Maximal Individual Therapy in Acute Myocardial Infarction (MITRA) Study Group. Comparison of primary angioplasty with conservative therapy in patients with acute myocardial infarction and contraindications for thrombolytic therapy. Circulation 1998; 98(suppl I):I-363. 16. O9Neill W, Brodie BR, Ivanhoe R, Knopf W, Taylor G, O9Keefe JO, Grines CL, Weintraub R, Sickinger BG, Berdan LG, Tcheng JE, Woodlief LH, Strzelecki M, Hartzler G, Califf RM. Primary coronary angioplasty for acute myocardial infarction (the primary angioplasty registry). Am J Cardiol 1994;73:627– 634. 17. O9Keefe JO, Bailey WL, Rutherford BD, Hartzler GO. Primary angioplasty for acute myocardial infarction in 1000 consecutive patients. Am J Cardiol 1993;72:107G–115G. 18. Every NR, Parsons LS, Hlatky M, Martin JS, Weaver WD. A comparison of thrombolytic therapy with primary coronary angioplasty for acute myocardial infarction. N Engl J Med 1996;335:1253–1260. 19. Zahn R, Koch A, Rustige J, Schiele R, Wirtzfeld A, Neuhaus K, Kuhn H, Gu¨lker H, Senges J, for the ALKK Study Group. Primary angioplasty versus thrombolysis in the treatment of acute myocardial infarction. Am J Cardiol 1997;79:264 –269. 20. Rogers WJ, Dean LS, Moore PB, Wool KJ, Burgard SL, Bradley EL, for the Alabama Registry of Myocardial Ischemia Investigators. Comparison of primary angioplasty versus thrombolytic therapy for acute myocardial infarction. Am J Cardiol 1994;74:111–118. 21. Zahn R, Vogt A, Seidl K, Schuster S, Gu¨lker H, Heinrich KW, Gottwik M, Neuhaus K, Senges J, fu¨r die ALKK Studiengruppe. Ballondilatation beim akuten Herzinfarkt: Ergebnisse des Registers der Arbeitsgemeinschaft Leitender Kardiologischer Krankenhausa¨rzte (ALKK) bei 4625 Patienten. Z Kardiol 1997;86: 712–721 . 22. Schuster S, Koch A, Burczyk U, Schiele R, Wagner S, Zahn R, Glunz HG, Heinrich F, Stuby K, Berg G, Voigtla¨nder T, Gieseler U, Jakob M, Hauptmann P, Senges J, fu¨r die MITRA Studien-Gruppe. Therapy of acute myocardial infarction, impact of clinical trials on clinical practice: the MITRA pilot phase. Z Kardiol 1997;86:273–283. 23. Grines CL. Primary angioplasty—the strategy of choice. N Engl J Med 1996;335:1313–1315. 24. Himbert D, Feldman LJ, Boudvillain O, Benamer H, Juliard JM, Steg PG. Heterogeneity of prognosis in patient subsets treated by primary coronary angioplasty during acute myocardial infarction. Am J Cardiol 1998;81:1236 –1239. 25. Tiefenbrunn AJ, Chandra NC, French WJ, Gore JM, Rogers WJ. Clinical experience with primary percutaneous transluminal coronary angioplasty compared with alteplase (recombinat tissue-type plasminogen activator) in patients with acute myocardial infarction. A report from the second national registry of myocardial infarction (NRMI-2). J Am Coll Cardiol 1998;31:1240 –1245. 26. Grassman ED, Johnson SA, Krone RJ. Predictors of success and major complications for primary percutaneous transluminal coronary angioplasty in acute myocardial infarction. An analysis of the 1990 to 1994 society for cardiac angiography and interventions registries. J Am Coll Cardiol 1997;30:201–208.
CORONARY ARTERY DISEASE/PRIMARY ANGIOPLASTY
1319