Knuuti J, et al. - The authors wanted to establish the ranges of pre-test probability (PTP) of coronary artery disease (CAD) in which stress electrocardiogram (ECG), stress echocardiography, coronary computed tomography angiography (CCTA), single-photon emission computed tomography (SPECT), positron emission tomography (PET), and cardiac magnetic resonance (CMR) can reclassify patients into a post-test probability that defines (>85%) or excludes (<15%) anatomically (defined by visual evaluation of invasive coronary angiography [ICA]) and functionally (defined by a fractional flow reserve [FFR] ≤0.8) significant CAD. For the detection of anatomically and functionally significant CAD, the various diagnostic modalities demonstrated different optimal performance ranges. For any given patient, they recommend selecting a diagnostic technique to rule-in or rule-out CAD based on the optimal PTP range for each test and on the presumed reference standard.

Methods

• Researchers performed a broad search in electronic databases until August 2017.
• They included studies on the aforementioned techniques in >100 patients with stable CAD that used either ICA or ICA with FFR measurement as reference.
• Using a hierarchical bivariate random-effects model, study-level data was pooled and likelihood ratios for each technique were attained.
• The PTP ranges were defined for each technique to rule-in or rule-out significant CAD.

Results

• Analysis of a total of 28,664 patients from 132 studies that used ICA as reference and 4,131 from 23 studies using FFR, was performed.
• Anatomically significant CAD can be ruled-in and ruled-out with stress ECG only when PTP is ≥80% (76–83) and ≤19% (15–25), respectively.
• Anatomic CAD was ruled-in with coronary computed tomography angiography at a PTP ≥58% (45–70) and rule-out at a PTP ≤80% (65–94).
• For functionally significant CAD, the corresponding PTP values were ≥75% (67–83) and ≤57% (40–72) for CCTA, and ≥71% (59–81) and ≤27 (24–31) for ICA, demonstrating its poorer performance of anatomic imaging against FFR.
• In contrast, PET, stress CMR, and SPECT were able to rule-in functionally significant CAD when PTP is ≥46–59% and rule-out when PTP is ≤34–57%.