Original Article
Stenotic flow reserve derived from quantitative coronary angiography has modest but incremental value in predicting functionally significant coronary stenosis as evaluated by fractional flow reserve
Abstract
Background: Stenotic flow reserve (SFR) derived from quantitative coronary angiography (QCA) has been correlated with myocardial ischaemia as determined by pharmacological stress echocardiography. However, the diagnostic accuracy of SFR in predicting functionally significant coronary stenosis as assessed by the gold standard, fractional flow reserve (FFR), has not been previously characterised.
Methods: Patients who underwent coronary angiography and FFR assessment between January 2010 and February 2012 in a single tertiary centre were retrospectively assessed. QCA parameters such as minimal lumen diameter (MLD), lesion length, diameter stenosis (DS), SFR, turbulent resistance (TR) and Poiseuille resistance (PR) were assessed. Significant FFR was defined as FFR ≤0.8. The diagnostic accuracy of QCA parameters to predict significant FFR was assessed by independent t-test and receiver operator characteristic (ROC) curve. Statistical significance was defined as P value of <0.05.
Results: The study included 272 patients (age: 64±11, 70% males) and 415 vessels. There were 180 (43%) vessels which were FFR significant. The mean FFR value for all vessels was 0.81±0.11. On comparison of AUC for predicting significant FFR, SFR (AUC =0.76) had the highest diagnostic accuracy compared to PR (AUC =0.75), % DS (AUC =0.73), TR (AUC =0.69), MLD (AUC =0.71) and DS >50% (AUC =0.64). Using a retrospectively determined optimal cut-off value of 3.51, the sensitivity of stenotic-flow-reserve was modest at 56% with good specificity of 81%. DS >50% had a sensitivity of 47% and specificity of 82% in predicting significant FFR. There was incremental predictive value when SFR was added to DS >50% on integrated discrimination improvement index (IDI =0.103, P<0.001) and net reclassification index (NRI =0.72, P<0.001).
Conclusions: SFR has modest diagnostic accuracy for predicting significant FFR but adds incremental predictive value to DS >50% for predicting significant FFR.
Methods: Patients who underwent coronary angiography and FFR assessment between January 2010 and February 2012 in a single tertiary centre were retrospectively assessed. QCA parameters such as minimal lumen diameter (MLD), lesion length, diameter stenosis (DS), SFR, turbulent resistance (TR) and Poiseuille resistance (PR) were assessed. Significant FFR was defined as FFR ≤0.8. The diagnostic accuracy of QCA parameters to predict significant FFR was assessed by independent t-test and receiver operator characteristic (ROC) curve. Statistical significance was defined as P value of <0.05.
Results: The study included 272 patients (age: 64±11, 70% males) and 415 vessels. There were 180 (43%) vessels which were FFR significant. The mean FFR value for all vessels was 0.81±0.11. On comparison of AUC for predicting significant FFR, SFR (AUC =0.76) had the highest diagnostic accuracy compared to PR (AUC =0.75), % DS (AUC =0.73), TR (AUC =0.69), MLD (AUC =0.71) and DS >50% (AUC =0.64). Using a retrospectively determined optimal cut-off value of 3.51, the sensitivity of stenotic-flow-reserve was modest at 56% with good specificity of 81%. DS >50% had a sensitivity of 47% and specificity of 82% in predicting significant FFR. There was incremental predictive value when SFR was added to DS >50% on integrated discrimination improvement index (IDI =0.103, P<0.001) and net reclassification index (NRI =0.72, P<0.001).
Conclusions: SFR has modest diagnostic accuracy for predicting significant FFR but adds incremental predictive value to DS >50% for predicting significant FFR.
