Optimal diagnostic approach for using CT-derived quantitative flow ratio in patients with stenosis on coronary computed tomography angiography

Jonathan N. Dahl*, Laust D. Rasmussen, Daixin Ding, Shengxian Tu, Jelmer Westra, William Wijns, Evald Høj Christiansen, Ashkan Eftekhari, Guanyu Li, Simon Winther, Morten Bøttcher

*Corresponding author for this work

Research output: Contribution to journal/Conference contribution in journal/Contribution to newspaperJournal articleResearchpeer-review


Background: Coronary computed tomography angiography (CCTA)-derived quantitative flow ratio (CT-QFR) is an on-site non-invasive technique estimating invasive fractional flow reserve (FFR). This study assesses the diagnostic performance of using most distal CT-QFR versus lesion-specific CT-QFR approach for identifying hemodynamically obstructive coronary artery disease (CAD). Methods: Prospectively enrolled de novo chest pain patients (n ​= ​445) with ≥50 ​% visual diameter stenosis on CCTA were referred for invasive evaluation. On-site CT-QFR was analyzed post-hoc blinded to angiographic data and obtained as both most distal (MD-QFR) and lesion-specific CT-QFR (LS-QFR). Abnormal CT-QFR was defined as ≤0.80. Hemodynamically obstructive CAD was defined as invasive FFR ≤0.80 or ≥70 ​% diameter stenosis by 3D-quantitative coronary angiography. Results: In total 404/445 patients had paired CT-QFR and invasive analyses of whom 149/404 (37 ​%) had hemodynamically obstructive CAD. MD-QFR and LS-QFR classified 188 (47 ​%) and 165 (41 ​%) patients as abnormal, respectively. Areas under the receiver-operating characteristic curve for MD-QFR was 0.83 vs. 0.85 for LS-QFR, p ​= ​0.01. Sensitivities for MD-QFR and LS-QFR were 80 ​% (95%CI: 73–86) vs. 77 ​% (95%CI: 69–83), p ​= ​0.03, respectively, and specificities were 73 ​% (95%CI: 67–78) vs. 80 ​% (95%CI: 75–85), p ​< ​0.01, respectively. Positive predictive values for MD-QFR and LS-QFR were 63 ​% vs. 69 ​%, p ​< ​0.01, respectively, and negative predictive values for MD-QFR and LS-QFR were 86 ​% vs. 85 ​%, p ​= ​0.39, respectively). Conclusion: Using a lesion-specific CT-QFR approach has superior discrimination of hemodynamically obstructive CAD compared to a most distal CT-QFR approach. CT-QFR identified most cases of hemodynamically obstructive CAD while a normal CT-QFR excluded hemodynamically obstructive CAD in the majority of patients.

Original languageEnglish
JournalJournal of Cardiovascular Computed Tomography
Pages (from-to)162-169
Number of pages8
Publication statusPublished - Mar 2024


  • Computed tomography derived quantitative flow reserve
  • Coronary artery disease
  • Coronary computed tomography angiography
  • Fractional flow reserve
  • Non-invasive cardiac imaging
  • Quantitative coronary analysis


Dive into the research topics of 'Optimal diagnostic approach for using CT-derived quantitative flow ratio in patients with stenosis on coronary computed tomography angiography'. Together they form a unique fingerprint.

Cite this