Abstract 4147087: Impact of fractional flow reserve measurement on outcomes in the invasive arm of the ISCHEMIA trial

Introduction: Fractional flow reserve (FFR) is an invasive, lesion-specific surrogate for myocardial ischemia. The use of FFR to guide lesion selection for revascularization has been shown to improve outcomes compared to an angiographic approach because it allows revascularization of lesions based on hemodynamic significance. Purpose: To compare the outcomes of patients randomized to the invasive (INV) arm of the ISCHEMIA trial who underwent FFR during initial angiography with those whose treatment was guided by angiography alone. Methods: The ISCHEMIA data set was obtained from the NHLBI. Subjects randomized to the INV arm who underwent FFR were compared to those who underwent angiography alone. Unadjusted cumulative event probabilities were estimated using the Kaplan-Meier method. Multivariable Cox proportional hazards analysis was used to estimate the independent impact of FFR on outcomes. Primary endpoint of interest was cardiovascular death (CV) death or MI. Results: Of the 5,179 patients with chronic coronary syndromes and at least moderate ischemia on stress testing, 2,588 were randomized to the INV strategy, 2,475 underwent angiography and 2,210 had baseline data available for analysis. Of these, 410 (19%) had FFR performed during diagnostic angiograms. Females comprised 24% of FFR patients and 24% of non-FFR patients (P=0.85). FFR patients were older than non-FFR patients (65.7 [8.7] years vs. 64.3 [9.6] years, P=0.006). The incidence of hypertension, diabetes, smoking, or prior MI at baseline did not differ between groups. Fewer FFR patients had severe ischemia at baseline (45% vs. 52%) and more had mild or moderate ischemia (54% vs. 48%) (P=0.009). FFR patients had less extensive disease as manifested by fewer native vessels with >70% stenosis than non-FFR patients (1.0 [0.9] vs. 1.5 [1.0], P<0.001) and less complex disease as assessed by the Duke jeopardy score (P<0.001). After 5 years of follow-up, individuals who underwent FFR had less CV death/MI (89% vs. 84%; log-rank P=0.046). However, on Cox proportional hazards analysis, FFR was not independently associated with CV death/MI (HR 0.89; 95% CI: 0.60-0.98; P=0.58). Conclusion: ISCHEMIA patients randomized to INV strategy who underwent FFR had less complex and extensive coronary disease and, on unadjusted analysis, less CV death/MI. However, after adjustment for differences in baseline characteristics, FFR was not independently associated with CV death/MI.