Five year follow up on FFR guided PCI for Stable CAD – What Did We Learn From FAME2?

M. Khurram Afzal, M.D.
By M. Khurram Afzal, M.D. on

In the final follow up of the FAME 2 study,  published in The New England Journal of Medicine, five year outcomes were reported comparing percutaneous coronary intervention (PCI) guided by fractional flow reserve (FFR) to medical therapy alone. The hypothesis had suggested FFR guided PCI would be superior to medical therapy as initial treatment in patients with stable coronary artery disease.


“For the first time in stable CAD patients, PCI – when guided by FFR – is associated with a clear signal towards lower rates of spontaneous MI’s”

– Dr. Bernard De Bruyne  


A total of  1220 patients with angiographically significant stenosis were included in this analysis. Those patients who had a single hemodynamically significant stenosis (FFR, ≤ 0.80) were randomized to PCI plus medical therapy or medical therapy alone. Those with an FFR of greater than 0.80 were given medical therapy and included in a registry. The primary composite endpoint was death, myocardial infarction or urgent revascularization.

The group of 888 patients that were randomized to PCI (447 patients) or medical therapy (441 patients) showed interesting results. These were the five year results; the primary end point occurred in 13.9% of those in the PCI group vs 27% of those in the medical therapy group, showing a significant 54% reduction over 5 years. This finding was primarily driven by a reduction in the occurrence of urgent revascularizations; 6.3% of patients in the PCI group and 21.1% in the medical-therapy group (hazard ratio, 0.27; 95% CI, 0.18 to 0.41). The rate of death was similar between the two groups.

For myocardial infarction the rate was 8.1% and 12% in the PCI group and medical therapy group respectively, with a hazard ratio of 0.66; 95% CI, 0.43 to 1.00.  Looking at previous reports of the FAME 2 trial, at one year the rate of myocardial infarction was similar between the groups (3.4% vs. 3.2%), at three years a numerical reduction of MI was observed in the PCI group compared to medical therapy (5.8% vs. 6.8%).  When asked about the significance of this finding, Dr. De Bruyne told Cardiology Now News, “ It is important to note these were spontaneous MIs, periprocedural MI was not significantly different. The accrual of MI in the medical therapy group seems to happen after 3 years, reinforcing the importance of a long follow up”.

The FAME2 trial also compared the proportion of patients with CCS II-IV Angina between the treatment arms. Patients randomized to PCI demonstrated an immediate and significant drop in the risk of Angina as early as 30 days after randomization. This reduction was sustained through the 3 year report, however, in the current 5 year analysis, the relative risk reduction in angina conferred by PCI was no longer statistically significant.

When asked to comment about the trends in the proportion of patients with CCS II-IV angina over the course of five years, Dr. De Bruyne responded, “At each point in time there was a larger relief of symptoms by PCI than by medical therapy. After 5 years the difference was no longer significant, but at that time 51% of patients initially assigned to receive medical therapy had actually been revascularized (cross over). It is very likely that this high proportion of cross over PCI (half urgently, half electively) contributed to make these patients less symptomatic. In addition, in an intention to treat analysis, this high cross over rate most likely decreased the number of MI and maybe even of death. Therefore the number of ‘hard end-point’ in the medical therapy group is largely underestimated”.

FFR-guided PCI did show to be a brilliant strategy for significantly lowering the rate of the primary composite end point of death, myocardial infarction, urgent revascularization at 5 years in patients with stable coronary artery disease. Although Dr. De Bruyne thought it would be unlikely that we see another follow up on FAME 2 patients, he did say “Would this be possible, I anticipate an increasing difference in the rate of MI and in the need for revascularization, but most likely no difference in all cause mortality”.


Leave a Reply