Review of the ISIS-3 Trial
A Randomised Comparison of Streptokinase vs TPA vs Anistreplase and of Aspirin Plus Heparin vs Aspirin Alone Among 41,299 Cases of Suspected Acute MI
Background Results from the GISSI and ISIS-2 trials demonstrated a clear benefit for the early opening of acutely blocked arteries with fibrinolytic treatment in patients presenting with STEMIs. ISIS-2 also demonstrated the benefit of aspirin in the acute and subacute phases of MI. However, after opening a blocked artery, concern for re-occlusion persisted, especially in the first several days to week, and experts postulated that antithrombotic therapy with heparin could improve outcomes compared to use of aspirin alone.
The 3rd study from the ISIS group sought to test the hypothesis that therapeutic heparin plus aspirin reduced mortality compared to aspirin alone in patients treated with early fibrinolytic therapy.
The investigators also sought to test whether different fibrinolytic regimens were safer and more effective compared to each other. *Since treatment with fibrinolytic therapy is no longer standard practice in most places and no significant differences in fibrinolytic regimens were ultimately reported, we will focus only on the heparin plus aspirin versus aspirin alone portion of the study.
Patients Patients were eligible for ISIS 3 if they were thought to be within 24 hours of the onset of symptoms of suspected or definite acute MI (with or without ECG changes) and if they had no definite contraindications to fibrinolytic therapy. Contraindications were not specified by protocol but by the responsible physician; standard contraindications to fibrinolysis were suggested.
Patients were eligible even if aspirin or heparin was thought to be clearly contraindicated or if it was thought that aspirin alone was not enough (i.e., that some anticoagulant was clearly indicated). In such cases, patients were still to be randomized and the allocated antithrombotic regimen was then modified as thought appropriate for the particular patient.
Baseline characteristics The majority of patients were men (73%) and under the age of 70 years who presented within 6 hours from the onset of pain (78%). Enrolled patients appeared to be generally stable as only 6% had SBP <100 mmHg and only 15% had a heart rate ≥100 bpm. Patients with anterior and inferior STEMIs were evenly represented (35% and 34%), respectively and just under 2/3rds of patients presented within 6 hours from pain onset and had ST elevation on their presenting ECG.
Procedures The study utilized a 3x2 factorial design. Randomization was done via a 24 hour direct line telephone service which took into account a limited set of baseline characteristics. All patients were randomly assigned to 1 of 3 fibrinolytic treatments as well as to either heparin plus aspirin or aspirin alone. All patients were to receive 162 mg of enteric-coated aspirin given daily for one month, starting immediately with the first tablet crushed or chewed for a rapid antiplatelet effect. Those assigned to heparin plus aspirin were to receive a fixed dose regimen of 12,500 IU of Calciparine (an injectable form of heparin), starting 4 hr after randomization and given subcutaneously twice daily for 7 days or until discharge. For patients assigned to aspirin alone, anticoagulation could still be used if the responsible physician felt it was clearly indicated with no rules governing this.
At discharge, a single-sided form was to be returned to the trial office that provided further identifiers to assist central mortality follow-up after discharge, and brief details of the treatments that were given in hospital, of any apparent side-effects of treatment, and of major events in hospital. Available cause-of-death information was reviewed (blinded to treatment allocation) by the trial coordinator. Causes of death were defined as either “definitely non-vascular” or “vascular.”
Endpoints The primary endpoint of the trial was intended to be vascular mortality within the first 5 weeks (35 days), but the investigators ended up reporting all-cause mortality since non-vascular mortality was rare and divided evenly between groups and did not impact the main results. Secondary endpoints included non-cerebral bleeding, stroke and reinfarction.
Results The protocol specified 3 main comparisons; however, we present only the comparison of heparin plus aspirin vs aspirin alone. This report of ISIS 3 included 41,299 participants, however, this does not represent the total number of patients randomized, which was 55,331. Patient eligibility for the trial specified that patients could either have a “clear indication” for fibrinolysis or an “uncertain” indication. While all of the 36,381 patients with a “clear indication” are included in these results, only half of the 18,950 patients with an “uncertain” indication are included. No great justification is provided for this other than the authors mention that results from those patients will be reported elsewhere.
Compliance with assigned aspirin treatment was over 98% in both the heparin plus aspirin versus aspirin alone group; however, compliance with either intravenous or high-dose subcutaneous heparin was 92% in the heparin plus aspirin group compared to 18% in the aspirin alone group.
Over 5 weeks, there was no significant difference in mortality for patients assigned to heparin plus aspirin versus aspirin alone (10.3% vs 10.6%; p=NS) nor was there any difference at 6 months (13.9% vs 14.0%; p=NS). No additional data on subgroups is provided in the main manuscript.
Non-cerebral bleeding was increased in patients assigned to heparin plus aspirin versus aspirin alone (6.3% vs 3.9%; p<0.00001) and the bleeding risk increased with higher age. Excess bleed risk was 1.3% in patients <70 years and 4.5% in those ≥70, which translates to big differences in numbers needed to harm estimates (77 vs 22). No significant difference in the overall risk of stroke (1.3% vs 1.2%; p=NS) or reinfarction were observed (3.2% vs 3.5%; p=NS); respectively.
Conclusions The ISIS-3 trial does not support the hypothesis that aspirin plus heparin is better than aspirin alone in patients presenting with AMI who undergo revascularization via thrombolysis. However, there are concerns related to trial conduct and reporting. From our standpoint, because the results of the trial were null and the number of patients was very large, it contributes valuable information, which is still relevant today.
Wow . A long time ago . I remember seeing ST elevation as I was talking to the patient - IV tpa given within 5 minutes of ST and down they went and pain resolved. Heparin was given
As an aside - I was an intern when we participated in the STK trial . The resident pulled the "fat" envelope for the anterior MI and the the "thin" one for the inferior MI . Colleagues report similar tactics for ICD placement in other hospitals . I recall it vividly.
I suppose the control of randomization is better now but while this excellent review of early studies is being presented -I wonder if any other geezer docs remember the way I do ? I am 67
Yikes. Embarrassed to say, just like with Gissi 2, this was news to me, and does not reflect how I was taught in training, or the local standard of practice early in my career when lytics were still used. I really should have asked more questions….hopefully not too late to get into that habit.