DEFER at 15 Years: Postponing PCI Based on FFR in Stable Angina Lowers MI Risk
PARIS, France—Fifteen-year follow-up from the first randomized trial using FFR screening shows no difference in outcomes with stenting compared with deferring PCI of nonischemic lesions in patients with stable angina, according to updated results of the DEFER study presented May 19, 2015, at EuroPCR. In patients in whom PCI was deferred, an advantage was seen in long-term risk of MI.
The study “confirms the long-term benign nature of lesions which are nonischemic and most importantly stable at the time of evaluation… with medical therapy,” Ori Ben-Yehuda, MD, executive director of the CRF Clinical Trials Center (New York, NY), told TCTMD in an email. “These lesions are not prone to rupture so the risk of spontaneous (type I) myocardial infarction is very low. So irrespective of advancement in stenting, since the prognosis is good with medical therapy intervening with PCI is not indicated even for long-term management.”
Nico H. J. Pijls, MD, PhD, of Catharina Hospital Eindhoven (Eindhoven, the Netherlands), and colleagues prospectively randomized 325 patients (average age 62 years; 29% women) with stable angina and > 50% stenosis without evidence of ischemia to PCI (n = 158) or deferral (n = 167) at 14 centers between 1997 and 1998. To avoid bias, all patients underwent FFR after randomization and were grouped as:
- Deferred (n = 91): FFR ≥ .075 and no PCI
- Performed (n = 90): FFR ≥ 0.75 and PCI
- Reference (n = 154): FFR < 0.75 and PCI
In previously published 5-year results, there were no differences in cardiac death or MI between the deferred and performed groups, with outcome risks for both of < 1%. There were also no differences in symptoms.
At 15 years, researchers achieved complete follow-up in 92%
of patients, with mortality data available in 97% and a median follow-up time
of 16.8 years. There were no differences in the incidences of death and any
revascularization between the deferred and performed PCI groups, although there
was a slight MI advantage for those in the former cohort (table 1).
Specifically, there were no differences between the deferred and performed PCI groups with regard to any specific mortality type—cardiac, noncardiac, and unknown. Dr. Pijls emphasized that overall mortality in a healthy cohort of this age group (now 79 years on average) would be expected to be about 28%, so the total mortality rates in the deferred and performed groups (32% and 30%, respectively) are “only somewhat higher than normal.” He also attributed the occurrence of other cumulative events to the advanced age of the patients.
With regard to MI, 1 of each of the events in the deferred arm occurred in a nontarget vessel or for unknown reasons. In contrast, 9 of the 13 MIs in the performed PCI group were attributed to the target vessel.
Because 15 years have passed since the study began, there are limitations to its interpretation, Dr. Pijls said. The biggest of these, he explained, is that DES were not available, and hence their effects cannot be factored into these results. There have also been changes to antiplatelet and optimal medical therapy regimes, he noted, adding that the FFR threshold also has changed, with the value used now typically ≥ 0.80.
“Despite these limitations, the rate of cardiac. unknown death, and MI was so low—about 1% per year—in the DEFER group, that it will be very hard to beat, even today,” he added. “Stenting a nonischemic stenosis has no benefit compared to medical treatment, neither in prognostic nor symptomatic respect.”
Deferral is Not ‘Black and White’
Session moderator William F. Fearon, MD, of Stanford University Medical Center (Stanford, CA), congratulated the authors on their “amazing follow-up,” and called the findings “very impressive.” However, he asked Dr. Pijls to explain the benefit of FFR screening to the naysayers who would argue that the differences in modern PCI compared with 15 years ago might negate these results.
“Maybe there would be no difference,” Dr. Pijls responded. “On the other hand, if you look at the infarction rate in the perform group, it's not really different from the infarction rates we see today with DES. The rate of severe adverse events in the defer group was so low with good medical treatment that it will be really hard to improve it anyway. We can see it as kind of a victory for all the nice work we have done in the last 20 years."
Panelist Chaim Lotan, MD, of the Heart Center at Hadassah Medical Center (Jerusalem, Israel), expressed concern over how to treat patients based on parameters other than FFR, such as lesion location. "I still have a problem with what to do with a patient who has a proximal LAD with 0.81 [on FFR]," he said. "Because still in the defer group, you have a lot of infarcts and mortality.... Just basing your decision on FFR in this case and saying 'yes' or 'no' is a little bit complex."
Dr. Pijls said this concern is "exactly the problem in convincing people" of the value of deferred PCI. "If you look at the evidence, it is safe to defer. There were many proximal LAD lesions in this study [and other previous FFR studies]. The outcome: good."
Still, he said, the issue is not "black and white. There is something grey in between." The data prove, however, that "if you have such a lesion, you can treat it with optimal medical treatment and leave it alone."
The issue is a 2-sided coin, Dr. Pijls observed. "It is safe to defer a nonischemic lesion based upon FFR. With this 15-year follow-up, this is almost a closed discussion. The other side of the coin is that if you have an ischemic lesion as indicated by FFR, it is better to stent it," as evidenced by FAME and FAME 2, he said. While neither of the latter studies have substantially long follow-up yet, 5-year results from FAME are slated to be presented at ESC later this year.
Pijls NHJ. 15-year follow-up of the DEFER trial. Presented at: EuroPCR; May 19, 2015; Paris, France.
- Dr. Pijls reports receiving research grants from St. Jude Medical; serving as a consultant to Boston Scientific and St. Jude Medical; and holding stock in ASML, Heartflow, GE, and Philips.