Review of the Primary Angioplasty in Myocardial Infarction Study Group trial
Angioplasty versus Thrombolytic Therapy for Acute Myocardial Infarction
Background: Previous trials established that thrombolysis improves mortality in patients with acute myocardial infarction, as seen in the GISSI-1 and ISIS-2 trials. However, thrombolysis has limitations, including an increased risk of bleeding and the inability to achieve arterial patency in approximately 20% of the cases. As a result, there was a growing interest in the use of percutaneous transluminal coronary angioplasty (PTCA).
The Primary Angioplasty in Myocardial Infarction Study Group sought to test the hypothesis that PTCA compared to thrombolysis, improves outcomes and reduces bleeding in patients with acute myocardial infarction.
Patients: Eligible patients presented within 12 hours of ischemic chest pain and had ST elevation of at least 1 mm in two or more contiguous electrocardiographic leads. Patients were excluded if they had dementia, LBBB, cardiogenic shock or elevated bleeding risk.
Baseline characteristics: The study enrolled 395 patients – 195 assigned to the PTCA arm and 200 assigned to the thrombolysis arm. The average age of patients was 60 years with 73% being men. Approximately 14% had prior myocardial infarction, 43% had hypertension, 12% had diabetes and 2% had congestive heart failure. The average ejection fraction 52%.
The infarct was anterior in 34% of the patients, inferior in 59% and lateral in 8%.
Procedures: All patients were given 325 mg of aspirin plus 10,000-unit bolus of intravenous heparin. After that, patients were randomly assigned to thrombolytic therapy or PTCA. The thrombolytic agent used was tissue plasminogen activator (t-PA) at a dose of 100 mg (or 1.25 mg/kg of body weight for patients weighing less than 65 kg) over three hours. Patients randomly assigned to PTCA underwent immediate diagnostic catheterization.
Angiographic criteria for exclusion from PTCA included left main stenosis of more than 70%, infarct-related vessel was patent, three-vessel disease, morphologic features of the lesion known to indicate high risk, small infarct-related vessels or stenosis <70%.
Bypass surgery was recommended for high-risk patients.
In both treatment groups, intravenous heparin was administered for 3 - 5 days, targeting PTT level 1.5 - 2 times the control value or activated clotting time 160 - 200 seconds.
Endpoints: The primary end point was all-cause death or recurrent ischemia. Recurrent ischemia was defined as ischemic chest pain of more than 20 minutes despite nitrate therapy plus new ST- T-wave changes, new pulmonary edema, a holosystolic murmur, or hypotension. Secondary endpoints were reinfarction, and left ventricular function at 6 weeks using radionuclide ventriculography. Reinfarction was defined as recurrent chest pain longer than 30 minutes with new ST-segment elevation and either coronary angiography confirming an occluded vessel or recurrent elevation of cardiac enzymes.
Statistical analysis was performed based on the intention-to-treat principle. The estimated sample size to provide 80% power was 370 patients, assuming an event rate of the primary outcome of 25% in the t-PA arm and 12% in the PTCA arm. The intended follow up time for the primary endpoint was not clearly defined but was provided for in-hospital and again at 6 months.
Results: The average time from the onset of chest pain to treatment was approximately 4 hours in both groups. All of the patients assigned to the PTCA arm underwent coronary angiography but 10% did not undergo angioplasty based on the exclusion criteria mentioned previously. PTCA was successful in 97% of the patients who underwent the procedure.
Of note, the sample size was based on the expected number of death and recurrent ischemia but authors focused on reporting death and reinfarction (rather than recurrent ischemia).
The endpoint of in-hospital death or reinfarction was lower in the PTCA group (5.1% vs 12.0%; p= 0.02). In-hospital death and reinfarction were numerically lower with PTCA “number of events was the same for both endpoints” (2.6% vs 6.5%; p= 0.06). All 5 deaths in the PTCA group were due to cardiac causes. Among the 13 deaths in the t-PA group, 4 were due to intracranial bleeding and 9 were due to cardiac causes.
Unscheduled coronary angiogram was significantly less common in the PTCA group (13.3% vs 63.0%; p< 0.001). Recurrent ischemia was the reason for unscheduled coronary angiogram in 5.1% of the patients in the PTCA arm and 23.5% of the patients in the t-PA arm.
Death or infarction at 6-months was lower in the PTCA group (8.5% vs 16.8%; p= 0.02). Death at 6-months was numerically lower with PTCA (3.7% vs 7.9%; p= 0.08).
In-hospital hemorrhagic strokes were more frequent in t-PA arm (2.0% vs 0.0%; p = 0.05). Non-CABG related bleeding requiring transfusion was not significantly different between both groups (6.2% with PTCA vs 5.0% with t-PA; p= 0.62).
At 6 weeks, radionuclide ventriculography was performed in 65% of the survivors. The average left ventricular ejection fraction at rest was similar in both groups at 53 ± 13%.
Subgroup analysis for mortality was performed based on “low-risk” vs “not low risk”. Not low risk was defined as anterior infarction, age >70 years or admission heart rate > 100 bpm. PTCA reduced in-hospital mortality in the “not low risk” group (2.0% vs 10.4%; p= 0.01) but not in the low risk group (3.1% vs 2.2%; p= 0.69).
Conclusion: In patients with ST-elevation myocardial infarction, PTCA compared to t-PA reduced death and reinfarction at the hospital and at 6 months with a number needed to treat of approximately 14 and 12, respectively.
This was one of the trials that established the foundation for the use of PTCA in patients with acute myocardial infarction. While the treatment effect was large, there are important considerations to keep in mind. First, the sample size was small. In comparison, GISSI-1 had almost 12,000 patients and ISIS-2 had over 17,000. The results of small trials are not always replicated in larger pragmatic trials. Second, the use of aspirin + heparin + t-PA likely increased bleeding in the t-PA arm as heparin plus thrombolysis compared to thrombolysis without heparin increased bleeding without improving outcomes, as seen in the GISSI-2 and ISIS-3 trials. Third, two thirds of the patients had inferior or lateral infarcts and these subgroups did not benefit from thrombolysis in the GISSI-1 trial. Finally, standalone angioplasty is infrequently performed nowadays and patients often receive a stent which has improved vessel patency.
In the current era, patients with ST-elevation myocardial infarction receive early revascularization with stent placement, which improved outcomes in these patients. We discussed the limitations above to help readers and learners appraise clinical trials, as these limitations were important at the time of this trial's publication.