Angiography-Derived Fractional Flow Reserve Improves Coronary Lesion Assessment
Fractional flow reserve derived from routine coronary angiography may improve coronary lesion assessment.
Fractional flow reserve derived from routine coronary angiography is a noninvasive alternative to the standard pressure wire-derived approach for assessing coronary lesions with high sensitivity, specificity, and accuracy, according to a study published in Circulation.
The authors of this prospective multicenter study sought to evaluate the accuracy of coronary angiography-derived fractional flow reserve vs standard fractional flow reserve, which is more invasive and requires guidewire placement and hyperemic stimulus for each vessel.
The study sample enrolled a total of 301 patients with suspected coronary artery disease from 10 centers in the United States, Europe, and Israel. Participants underwent routine coronary angiography, and fractional flow reserve was measured using the standard approach (a coronary pressure wire and hyperemic stimulus) in vessels with coronary lesions of varying severity.
On-site operators, who were blinded to the pressure-wire derived values, calculated the fractional flow reserve in respective arteries based on at least 2 angiographic projections and using proprietary software to make 3D reconstructions of the coronary tree.
Primary outcomes were the sensitivity and specificity of coronary angiography-derived fractional flow reserve for predicting standard pressure wire-derived fractional flow reserve per vessel; the prespecified cutoff value was 0.80. Secondary outcomes were the diagnostic accuracy of coronary angiography-derived fractional flow reserve and its correlation to pressure-wire derived fractional flow reserve and the device success rate.
The mean pressure wired-derived fractional flow reserve was 0.81; 43% of vessels had a fractional flow reserve of 0.80 or less, while 59% of vessels were between 0.70 and 0.90. The mean coronary angiography-derived fractional flow reserve was 0.80; and 46% of vessels had a value of 0.80 or less, while 64% of vessels were between 0.70 and 0.90.
Exceeding prespecified performance goals, the sensitivity and specificity of the angiography-derived fractional flow reserve predictions were 93.5% (95% CI, 87.8%-96.6%) and 91.2% (95% CI, 86.0%-94.6%), respectively.
The accuracy of angiography-derived fractional flow reserve to diagnose lesions was 92.2% (95% CI, 88.7%-94.8%). When the investigators only considered fractional flow reserve values between 0.75 to 0.85, all values for sensitivity, specificity, and accuracy remained high (89%, 85%, and 87%). They also observed a significant correlation between angiography-derived and pressure wire-derived fractional flow reserve values (r=0.80; Bland Altman confidence limits, -0.14 to 0.12; P <.001). Device success rate for angiography-derived fractional flow reserve was 99%.
Limitations of the study included the lack of processing time and total time to measure coronary-angiography derived fractional flow reserve. Additionally, only lesions amenable to pressure wire-derived measurements were validated, so not all relevant vessels were included. Use of different hyperemic stimulants may have affected the pressure-wire derived results, and investigators did not apply the “smart minimum” algorithm in calculating the standard fractional flow reserve value.
The authors concluded that fractional flow reserve derived from routine coronary angiography alone demonstrated high sensitivity, specificity, and accuracy compared with pressure wire-derived fractional flow reserve. Coronary angiography-derived fractional flow reserve is faster, easier, and noninvasive, and can potentially increase the overall assessment of coronary lesions, thereby improving patient outcomes.
Multiple authors declare associations with the pharmaceutical industry. Please see original reference for a full list of authors' disclosures.
Fearon WF, Achenbach S, Engstrom T, et al. Accuracy of fractional flow reserve derived from coronary angiography [published online September 24, 2018]. Circulation. doi:10.1161/CIRCULATIONAHA.118.037350