CHICAGO, IL—A new, more efficient method of computing fractional flow reserve (FFR) using data from computed tomographic angiography (CTA) not only is better able to detect hemodynamically significant lesions than CTA but does so onsite within a clinically relevant time frame.

The performance of the innovative technology was highlighted in a presentation by Stefan Baumann, MD, of the Medical University of South Carolina (Charleston, SC), on September 17, 2014, at the American Heart Association Scientific Sessions.

In the study, CT-based FFR was performed in 67 coronary artery lesions of 53 patients with suspected CAD (average age 61 years, 64% male). All had CTA and invasive angiography including FFR. Invasive FFR <0.80 defined functionally significant stenosis and served as the reference standard for assessment of the diagnostic performance of CT-based FFR compared with CTA.

Patients with previous CABG, stents in the target vessel, or certain complex anatomies were excluded, as were those whose CTA datasets were not of diagnostic quality. After angiography with FFR, 66% of patients were observed to have a stenosis ≥ 66% and 30% had an FFR < 0.80. The mean calcium Agatston score was 778.4.

No Need for Additional Imaging or Changes to CTA Protocol

First- and second-generation CT systems (Somatom Definition and Somatom Definition Flash; Siemens Healthcare) progressively incorporating various technical advances were used to acquire CTA images. Prototype software (Siemens cFFR; Siemens Healthcare) installed on a regular workstation produced FFR values using regular CTA datasets without additional imaging, modification of the acquisition protocol, or administration of pharmacologic stress agents. The FFR algorithm simulates coronary blood flow from the 3-dimensional model.

Importantly, the mean time for derivation of CT-based FFR including dataset processing and coronary flow computation was 37.5 minutes.

 

Area under the ROC curve was greater for CT-based FFR compared with CTA on a per-lesion basis (0.92 vs 0.72; P = .0049); trend in the same direction was seen on per-patient analysis (0.91 vs 0.78; P = .078).

The study is limited by the fact that it was small, retrospective, and performed at a single center, Dr. Baumann said. In addition, the cohort was biased by inclusion of only patients with at least 1 lesion, leading to a high prevalence of CAD. Also, up to 3 months were allowed between performance of CTA and that of invasive angiography with FFR, and thus any intervening treatment may have impacted the results.

In response to questions from the audience, Dr. Baumann said that bifurcations were generally excluded by direction of Siemens because there is little experience with CT-based FFR in that lesion subset. Highly calcified lesions were also disqualified because definition of stenosis is difficult, he added.

Dr. Baumann explained that development of the prototype algorithm, which is not commercially available, was motivated by the huge computing power required by current versions of CT-based FFR.

 

---

Source: 
Baumann S. Diagnostic value of a novel coronary computed tomography angiography-based approach for assessing fractional flow reserve: comparison with invasive management. Presented at: American Heart Association Scientific Sessions; November 17, 2014; Chicago, IL.

Related Stories:

• DISCOVER-FLOW: CT-Derived FFR Measurements Show High Diagnostic Accuracy
• Noninvasive CT-based FFR More Accurate than CT for Diagnosing Ischemic CAD
• Novel CTA-Based Method Shows Promise in Identifying Functionally Significant Lesions