Validation of a novel ‘wireless’ fractional flow reserve measurement during coronary angiography

Background: Fractional flow reserve (FFR) has become the gold standard method of ischaemia evaluation in the catheter laboratory. However, adoption of this technique remains limited, due to its associated cost and potential risk. Therefore, a technique to calculate FFR without an intracoronary wire would be invaluable in clinical practice. Methods: A prospective validation study was performed to test a novel in-house-developed software application that enables derivation of FFR without use of an intracoronary wire or induction of hyperaemia: the wireless FFR (wFFR). This method incorporates three-dimensional angiography, formulas based on Navier-Stokes equations, and machine learning algorithms. A total of 104 lesions from 84 patients who had a clinical indication for FFR measurement were included in this study. All patients underwent wFFR and actual FFR measurements. Results: Mean age was 66 ± 8 years and mean FFR was 0.82 ± 0.08. Time taken for wFFR calculation ranged between 3–5 minutes. There was a strong correlation between wFFR and FFR (r = 0.88, p < 0.001). Bland-Altman analysis demonstrated good agreement between wFFR and FFR (mean difference 0.001 ± 0.040, p = 0.81). A wFFR cutoff ≤0.80 was able to predict a positive FFR (≤0.80) with accuracy, sensitivity, specificity, positive predictive value, and negative predictive value of 92%, 91%, 93%, 91%, and 93%, respectively (AUC 0.96, 95% CI 0.929–0.997) (see Figure). Conclusions: The wFFR had high diagnostic accuracy in this single-centre pilot study. Compared to other virtual FFR methods, wFFR appeared to be highly accurate and clinically applicable. External validation of these results will be required.