A Ten-Year Single Practice Experience Correlating Invasive Angiography with High Resolution Coronary CT Angiography Prior to and Following Implementation of CT-FFR

Abstract

Research Question: What is the statistical angiographic correlation between invasive cardiac catheterization results and CT coronary angiographic results using the 512 slice scanners and for heart scans prior to and following use of the FFRCT technique?

Background, Significance, and Rationale: Dr. Sathyamoorthy developed a significant, longitudinal experience with high-resolution coronary CT angiography at Consultants in Cardiovascular Medicine and Science through partnership with the Baylor All Saints Radiology Department over the past decade. Experience over the first five years from 2012 to 2016 was with 64-slice CT scanning technology. Over the last 5 years, our experience has advantaged utilization of the most advanced technology platforms available, the GE lightspeed 512-slice scanner. Approximately two years ago, we added diagnostic capabilities by utilization of a technique called CT FFR. In cases where significant abnormalities are identified on CT angiography, we have proceeded with invasive cardiac catheterization. To date this comprises over 950 CT angiograms and between 120+ cardiac catheterizations.

Materials and Methods: Our retrospective study examined the statistical angiographic correlation between invasive cardiac catheterization results and CT coronary angiographic results using the 512-slice scanner prior to and following use of the FFRCT technique.

Results: A database was developed including all CT coronary angiograms referred from Dr. Sathyamoorthy's practice from the inception of 512-slice CT coronary angiography at Baylor All Saints with subsequent catheterizations performed by Dr. Sathyamoorthy and his associates performed at Baylor Heart and Vascular hospital. Patient clinical and demographic data for all of these cases was derived. Among the 806 patients, 6 were excluded due to PCI such as stenting or CABG leading to 800 patients whose data was included in the final analysis. Diagnostic Accuracy was determined for CCTA vs ICA, FFRCT vs ICA, and of the Entire CCTA Cohort.

Conclusions: In this single-practice, longitudinal, retrospective study we found that CCTA can very accurately identify the presence and severity of obstructive coronary artery disease. This suggests that CCTA can accurately exclude obstructive CAD when patients have stable angina or an abnormal stress test and/or when patients have a moderate pre-test probability angina. When it comes to the diagnostic utility of FFRCT, our study shows that when compared to CCTA, the use of FFRCT can be a diagnostic tool that provides functional information about a lesion to better guide management in cases where the significance of a lesion is unknown by lowering the number of false positive results and increasing the specificity and PPV. However, because of the small sample size of our initial CTFFR cohort (n=22), definitive conclusions cannot be made with respect to our cohort.