Review of the FRISC trial

Low-molecular-weight heparin during instability in coronary artery disease

Lancet 1996;347:561-68.

Background In most of the prior AMI trials presented here, patients with non_ST-segment myocardial infarction (NSTEMI) and/or unstable angina were included with patients with ST-segment myocardial infarction (STEMI). Patients with NSTEMI usually represented about 1/3 of trial participants.

It had become clear based on subgroup analyses from the previous trials that NSTEMI patients did not benefit from thrombolysis. So, while intravenous anticoagulation with heparin was not found to be beneficial in GISSI-2 or ISIS-3, where thrombolysis was also used, there remained the possibility that it may benefit patients with unstable coronary syndromes who were not candidates for thrombolytic therapy.

This smaller trial represents a departure from the general eligibility criteria for the AMI trials that we have already reviewed. The Fragmin During Instability in Coronary Artery Disease (FRISC) study group sought to test the hypothesis that subcutaneous low-molecular-weight heparin, in combination with aspirin, reduces death and new cardiac events in patients with unstable CAD. For the purposes of this study, unstable CAD represents unstable angina and NSTEMI.

Patients Patients had to be men older than 40 years and women at least 1 year after menopause admitted to the hospital for chest pain within the previous 72 hours. All patients had to have either newly developed or increased angina symptoms during the previous 2 months or persisting chest pain with suspicion for AMI and at least one of the following ECG criteria: transient or persistent ST depression of ≥ 0.1 mV or T-wave inversion of ≥ 0.1 mV in at least 2 adjacent leads without pathological Q waves in the ischemic leads. Essentially, the authors were selecting patients with unstable angina or NSTEMIs and not STEMIs or completed infarcts. There were many exclusion criteria for the trial including the presence of conditions with an increased risk of bleeding, known renal or liver insufficiency, indications for thrombolysis, suspected myocarditis and many others; however, it is worth noting there was no upper age limit.

Baseline characteristics There were 5,137 patients who met eligibility criteria and 1,506 (29%) were randomized. The 3 most common reasons for exclusion were risk of bleeding (20%), compliance problems (15%), and Q waves or bundle branch block (14%). Patients with other severe disease accounted for 4% of exclusions and those with renal or liver insufficiency accounted for 2%.

The median age of participants was 69 years and approximately 65% were men. 20% of participants were active smokers, 13% had diabetes and nearly 30% had a previous heart attack. Patients with NSTEMI accounted for close to 40% of participants and the remainder had unstable angina.

Procedures Treatment was started as soon as possible after admission. During the first 6 days (acute phase), 120 IU per kg bodyweight (maximum dose of 10,000 IU) of dalteparin or placebo was injected every 12 hr. There was then a home treatment phase. For the next 35-45 days—at home— 7,500 IU of dalteparin or placebo was injected once daily. Patients stayed in the hospital during the acute phase for at least 5 days and on day 5-8 were discharged with the lower home dose. On day 40-50, the treatment was stopped and the final follow-up visit was scheduled 5-7 months after trial enrollment.

Endpoints The primary endpoint was the rate of death and new myocardial infarction during the first 6 days. Secondary endpoints were the rates of death and new MI after 40 and 150 days, the frequency of revascularization procedures and need for heparin infusion, and a composite endpoint.

Cause of death and myocardial infarction were verified by the independent endpoint committee who had to differentiate a new event from an inclusion event. Safety endpoints included major and minor bleeding. Major bleeding was defined by a drop of ≥2 g/dL in hemoglobin with associated signs or symptoms of bleeding, and minor bleeding was any other bleeding not meeting the former criteria.

A sample size of 1,500 was based on a power of 0.80 to detect a reduction in the primary composite endpoint from 6% to 3% at a 2-sided alpha of 5%. Subgroup analyses were prespecified and a “high-risk group” was defined by the presence of at least 2 of the following variables: age >70, previous MI, medically treated heart failure, or diabetes.

Results 1,506 patients were included in the final analysis; 757 in the placebo group and 741 in the dalteparin group. At 6 days, dalteparin significantly reduced the occurrence of the primary endpoint (RR 0.37; 1.8% vs 4.8%; 95% CI 0.20-0.68), which was driven by myocardial infarction (RR 0.33; 1.4% vs 4.4%; 95% CI 0.16-0.60). There was no difference in death between groups (RR 0.88; 0.9% vs 1.1%; 95% CI 0.32-2.48) and minor bleeding events were much more common in the dalteparin group (8.2% vs 0.3%).

At 40 (8.0% vs 10.7%) and 150 (14.0% vs 15.5%) days, differences between groups were less pronounced for the primary endpoint and the results, in favor of dalteparin, were not statistically significant. Bleeding events, driven by minor bleeding, continued to be much higher for the dalteparin group.

Subgroup analyses are presented at 6 and 40 days and are generally concordant across subgroups, meaning the treatment effect goes in the same direction, but these analyses should be viewed skeptically as they represent small sample sizes, particularly compared to subgroups in the earlier mega trials. Subgroup data suggests concordant results for patients less than or greater than 70 years of age and for patients meeting high risk criterion compared to those not meeting it.

Conclusions In patients with unstable angina and NSTEMI, dalteparin reduced the primary endpoint of death or nonfatal MI compared to placebo at 6 days and was associated with a NNT of 33 patients. This was driven by a reduction in nonfatal MI. The results were no longer significant at 40 or 150 days. This trial is limited by small sample size and highly selected patient population.