Review of the FAME 2 Trial

FFR guided PCI vs medical therapy in stable coronary artery disease

N Engl J Med 2012;367:991-1001

Background: Percutaneous coronary intervention (PCI) did not improve the outcome of death or nonfatal myocardial infarction in stable coronary artery disease, as seen in the COURAGE trial. Nonetheless, some skepticism remained within the cardiology community regarding the results of the COURAGE trial, with the belief that revascularizing ischemia-causing lesions could improve clinical outcomes (ischemic testing was not required in all patients in the COURAGE trial). While there was data supporting this idea, large randomized trials providing conclusive evidence were still lacking.

Fractional flow reserve (FFR) is a pressure-wire-based measurement performed during coronary angiography to evaluate whether a coronary stenosis is likely to be causing myocardial ischemia.

The investigators of FAME 2 trial sought to test the hypothesis that FFR-guided PCI with drug-eluting stents plus optimal medical therapy (OMT) is superior to OMT alone in patients with stable coronary artery disease.

Patients: Eligible patients had stable angina pectoris and stenosis of 50% or more in at least one major coronary artery that has a diameter of at least 2.5 mm and supplying a viable myocardium. Patients with atypical or no chest pain were included if they had documented ischemia on noninvasive testing. To evaluate the hemodynamic severity of each identified stenosis, FFR was measured using a coronary guidewire during adenosine-induced hyperemia. Patients with at least one stenosis in a major coronary artery with an FFR of 0.80 or less were randomized.

Major exclusion criteria were: Bypass surgery felt to be the preferred treatment strategy, left main disease, myocardial infarction within a week, prior coronary artery bypass surgery, left ventricular ejection fraction <30%, extremely tortuous or calcified coronary arteries or life expectancy less than 2 years.

Baseline characteristics: The trial randomized 888 patients – 447 randomized to the FFR-guided PCI + plus medical therapy and 441 randomized to medical therapy alone.

The average age of patients was 64 years and 78% were men. The average body mass index was 28 kg/m2.  Approximately 78% had hypertension, 76% had hyperlipidemia, 27% had diabetes, 37% had prior myocardial infarction and 20% were current smokers. Sixty one percent had Canadian Cardiovascular Society class II or III angina.

Significant stenosis based on FFR was present in 1 vessel in 76% of the patients, 2 vessels in 21% and 3 vessels in 3%. Among all patients, proximal or middle left anterior descending artery disease was present in 61%.

Procedures: Patients were randomized to undergo PCI using a second-generation drug eluting stent or medical therapy. All patients were prescribed aspirin 80 - 325 mg daily, beta-blockers, ACEi/ ARBs,  statins alone or in combination with ezetimibe, to reduce LDL level to less than 70 mg/dL. Patients who underwent PCI received clopidogrel 600mg followed by 75mg daily for at least 1 year.

Follow up was performed at 1 month, 6 months, 12 months and yearly thereafter.

Endpoints: The primary outcome was a composite of death from any cause, nonfatal myocardial infarction, or unplanned hospitalization leading to urgent revascularization during the first 2 years. Secondary outcomes included individual components of the primary outcome in addition to cardiac death, nonurgent revascularization, and angina class. Outcomes were adjudicated by a committee unaware of treatment assignment. Revascularization was considered urgent when a patient was admitted to the hospital with persistent or increasing chest pain (with or without ischemic EKG changes or elevated cardiac biomarkers) and revascularization was performed during the same hospitalization.

Analysis was performed based on the intention-to-treat principle. The estimated event rate of the primary outcome at 24 months was 12.6% in the PCI group and 18.0% in the medical therapy group, corresponding to 30% relative risk reduction with PCI. Using these event rates, a sample size of 1,632 patients would achieve more than 84% power with a two-sided alpha of 0.05.

Results: The study was stopped early due to significant difference in the primary outcome between the PCI and the medical therapy groups.

A total of 1,220 patients underwent FFR, and among them 888 met the inclusion criteria and were randomized. Among the 447 patients assigned to receive PCI, 435 (97.3%) patients underwent the intervention. Among the 441 patients assigned to the medical therapy group, 2 (0.5%) patients erroneously underwent PCI. The mean follow up time was approximately 7 months. Kaplan-Meier curves were provided up to 1 year since the trial was terminated early. Only 13% of the patients completed 1 year follow up.

FFR-guided PCI plus medical therapy was associated with lower rate of the primary composite outcome compared to medical therapy alone (4.3% vs. 12.7%; HR: 0.32; 95% CI: 0.19 - 0.53; p<0.001). This difference was driven by lower rate of urgent revascularization in the PCI group (1.6% vs 11.1%, HR: 0.13, 95% CI 0.06 – 0.30; p< 0.001). There was no significant difference in death from any cause (0.2% vs 0.7%; p= 0.31) or non-fatal myocardial infarction (3.4% vs 3.2%; p= 0.89). Cardiac death was similar (0.2% in both groups). Non-urgent revascularization was lower with PCI (1.6% vs 8.6%, HR: 0.17, 95% CI: 0.08 – 0.39; p< 0.001).

Among the 56 patients who underwent urgent revascularization, 12 (21.4%) patients had the procedure for myocardial infarction, 15 (26.8%) patients had it for unstable angina accompanied by ischemic EKG changes, and 29 (51.8%) patients had it for unstable angina without ischemic EKG changes.

At 12 months, patients in both treatment groups had significant improvement in symptoms. The percentage of patients with angina class II-IV based on the Canadian Cardiovascular Society classification was numerically lower in the PCI group at 1 year, but this did not reach statistical significance.

Subgroup analysis demonstrated one significant interaction. Patients with FFR <0.65 had more benefit with PCI compared to patients with FFR of 0.65 or more (3.1% vs 17.1% and 5.4% vs 8.9%; p for interaction= 0.01).

Conclusion: In patients with stable coronary artery disease and lesions with FFR of 0.80 or less, PCI compared to medical therapy reduced the need for urgent revascularization with a number needed to treat of approximately 11 patients. There was no significant difference in death or nonfatal myocardial infarction.

Despite what seems to be a strong advantage of the FFR-guided PCI strategy, we believe this trial does not justify its routine use for several reasons. First, the trial did not show any significant improvement in hard outcomes, such as death or nonfatal myocardial infarction. Second, the trial was terminated early, with only 13% of patients completing 1-year follow up. Most importantly, the endpoint of urgent revascularization was not applicable to most patients in the PCI group, as these patients had already undergone revascularization. This, combined with the awareness of both patients and their physicians regarding who underwent PCI, biased the revascularization endpoint in favor of PCI.

We would like to introduce our readers to the terms “faith healing” and “subtraction anxiety” which are well-explained in relation to revascularization by Rajkumar CA et al. In “Faith healing” patients experience a reduction in symptoms after being reassured by a negative test result. For instance, in the FAME 2 trial, patients with an FFR greater than 0.80 had significant symptom improvement, even without PCI and no significant changes to their medications. In “subtraction anxiety,” physicians knowledge of untreated lesions will trigger series of actions when patients present with symptoms even if they were atypical. For instance, there were 3 reasons a patient could have urgent revascularization—biomarkers, ECG-positivity or symptoms alone. In the control arm (PCI-subtracted), the vast majority of urgent revascularization procedures were due to “symptoms only” vs positive ECGs or enzymes.

As practicing cardiologists, we often see patients who come to the emergency department with chest pain but without myocardial infarction. If these patients have coronary stenosis, it is often assumed to be the cause of their chest pain and is treated. However, if the lesion was already treated using PCI, there is nothing left to intervene on. In the medical arm of this trial, 19.5% of the patients underwent revascularization on follow up compared to 3.1% in the PCI arm, including both urgent and non-urgent cases. In our opinion, this demonstrates a success of medical therapy since around 80% of patients were spared an expensive procedure without increasing their risk of death or myocardial infarction. As we discussed previously in the COURAGE trial, performing PCI on all patients to reduce the future need for revascularization by 15-20% is not cost-effective or practical.