The addition of CT-based fractional flow reserve information to CT alone improved the diagnostic accuracy of stenoses, allowing noninvasive assessment of the physiologic consequences of lesions, according to the long-awaited results of the Determination of Fractional Flow Reserve by Anatomic Computed Tomographic Angiography (DeFACTO) study.
However, CT fractional flow reserve (FFR-CT) plus CT narrowly failed to meet the trial’s primary end point – diagnostic accuracy greater than 70% for the lower bound of the 95% confidence interval. Per-patient performance diagnostic accuracy of FFR-CT plus CT was 73% with a 95% CI of 67%-78%.
Nevertheless, the addition of FFR-CT "demonstrated superior diagnostic performance characteristics, as compared with CT stenosis alone, in all patients, in all vessels, and also in vessels of intermediate stenosis severity," lead author Dr. James K. Min said during a press conference.
The results of the study were released in JAMA on Aug. 26th to coincide with the presentation of the study at the European Society of Cardiology meeting (JAMA 2012;308 [doi: 10.1001/2012.jama.11274]).
Fractional flow reserve (FFR) is currently assessed during invasive coronary angiography (ICA) to determine whether a coronary stenosis results in ischemia, and is the currently accepted reference standard for determining lesion-specific ischemia. FFR is the ratio of the mean coronary pressure distal to a coronary stenosis to the mean aortic pressure during maximal coronary blood flow. This value describes coronary flow still attainable despite the presence of a stenotic lesion.
While CT angiography has long been used to accurately and noninvasively assess the anatomic severity of stenoses, the technique has been criticized because it does not yield functional information about the hemodynamic effect of lesions.
Noninvasive calculation of FFR from CT "is a novel method that applies computational fluid dynamics to determine the physiologic significance of CAD [coronary artery disease]. Fractional flow reserve from CT enables calculation of rest and hyperemic pressure fields in coronary arteries without additional imaging, modification of CT acquisition protocols, or administration of medications," the investigators wrote.
"Taken together, these study results suggest the potential of FFR-CT as a promising noninvasive method for identification of individuals with ischemia. The present study findings can be considered proof of concept of the feasibility of this novel technology."
A total of 252 patients were included in the final analysis of the DeFACTO study. These patients had CAD and underwent clinically indicated ICA after CT with no intervening coronary event. Patients were not eligible if they had a history of coronary artery bypass graft (CABG) surgery or percutaneous coronary intervention. About 77% of patients had experienced angina within the past month.
Among 615 study vessels, 271 had less than 30% stenosis and 101 had at least 90% stenosis. In all, 407 vessels were directly assessed by both FFR and FFR-CT.
Computed tomographic angiography was performed on 64- or greater detector scanners with prospective or retrospective electrocardiographic gating.
The investigators evaluated CTs for maximal patient-, vessel-, and segment-based diameter stenosis (characterized as 0%, 1%-29%, 30%-49%, or 50% or larger).
Per-patient and per-vessel CAD stenosis were the maximal stenoses identified in all segments or in all segments within a vessel distribution, respectively. Vessel distributions were categorized for the left anterior descending, left circumflex, and right coronary artery. Computed tomographic angiograms (CTAs) were judged as excellent, good, adequate, or nondiagnostic.
Selective ICA was performed by standard protocol, and FFR was performed at the time of ICA. Fractional flow reserve was considered diagnostic of ischemia at a threshold of 0.80 or less. Computation of FFR-CT was performed in blinded fashion by the FFR-CT core laboratory at HeartFlow, the study’s sponsor.
Per-patient diagnostic accuracy for FFR-CT plus CT was 73%. By comparison, diagnostic accuracy of CT alone was 64%.
FFR-CT also demonstrated greater discriminatory power than CT alone for vessels directly assessed by invasive FFR. For these vessels, the diagnostic sensitivity and specificity of FFR-CT alone were 80% and 61%, respectively.
Importantly, the researchers performed a secondary analysis of patients with an intermediate stenosis ranging from 30% to 70%, "wherein the clinical utility of FFR-CT would be most commonly expected for use." Diagnostic accuracy (73% for FFR-CT and 57% for CT), sensitivity (82% and 37%, respectively), positive predictive value (54% and 34%) and negative predictive value (88% and 68%) were greater for FFR-CT than for CT, though specificity was similar at 66%.
This intermediate group is an important patient population. "We know that patients with 30%-70% stenosis – even though they don’t look high-risk anatomically – actually, some of them experience ischemia and physiologic consequences of their coronary artery disease," said Dr. Min, director of cardiac imaging research and co-director of cardiac imaging at the Cedars-Sinai Heart Institute in Los Angeles.