Development, diagnostic performance, and interobserver agreement of a 18F-flurpiridaz PET automated perfusion quantitation system

Abstract

Background: Computerized methodologies standardize the myocardial perfusion imaging (MPI) interpretation process.

Methods: To develop an automated relative perfusion quantitation approach for ¹⁸F-flurpiridaz, PET MPI studies from all phase III trial participants of ¹⁸F-flurpiridaz were divided into 3 groups. Count distributions were obtained in N = 40 normal patients undergoing pharmacological or exercise stress. Then, N = 90 additional studies were selected in a derivation group. Following receiver operating characteristic curve analysis, various standard deviations below the mean normal were used as cutoffs for significant CAD, and interobserver variability determined. Finally, diagnostic performance was compared between blinded visual readers and blinded derivations of automated relative quantitation in the remaining N = 548 validation patients.

Results: Both approaches yielded comparable accuracies for the detection of global CAD, reaching 71% and 72% by visual reads, and 72% and 68% by automated relative quantitation, when using CAD ≥ 70% or ≥ 50% stenosis for significance, respectively. Similar results were observed when analyzing individual coronary territories. In both pharmacological and exercise stress, automated relative quantitation demonstrated significantly more interobserver agreement than visual reads.

Conclusions: Our automated method of ¹⁸F-flurpiridaz relative perfusion analysis provides a quantitative, objective, and highly reproducible assessment of PET MPI in normal and CAD subjects undergoing either pharmacological or exercise stress.

Keywords: Flurpiridaz; PET MPI; automated relative quantitation; diagnostic performance; interobserver variability.

Figure 1.

Study Patients.

Figure 2.. Polar Maps of Flurpiridaz in Normal Patients.

Polar maps of the mean flurpiridaz count distribution with relative color intensity and corresponding numerical values in a 17-segment model. Results at rest and with pharmacological and exercise stress in normal patients are presented. The Rest row depicts results obtained from the rest studies of the same patients who underwent either pharmacological or exercise stress. Flurpiridaz myocardial counts were reduced in the basal septum due to the membranous septum. The distal apex demonstrates lower intensity values, reflecting the ability of flurpiridaz to provide images with high resolution in this anatomically thin segment. Flurpiridaz count distributions were uniform in the remaining segments.

Figure 3A.. Diagnostic Performance in Pharmacological and Exercise Stress.

Comparison of performance metrics of visual reads vs. automated relative quantitation with the reference standard of angiographically determined CAD. * P <0.05.

Figure 3B.. Diagnostic Performance in Pharmacological Stress.

* P <0.05.

Figure 3C.

Diagnostic Performance in Exercise Stress.

Figure 4A.. Interobserver Agreement in Pharmacological Stress.

Interobserver agreement in visual reads vs. automated relative quantitation in normal and CAD territories following pharmacological stress. The dotted horizontal line indicates the cutoff above which an ‘almost perfect’ (0.80–1.00) agreement was characterized.

Figure 4B.

Interobserver Agreement in Exercise Stress.

Figure 5A.. Example of a discordant interpretation between visual read and automated relative quantitation of flurpiridaz MPI in a normal patient undergoing pharmacological stress.

65 year old female with atypical angina and an intermediate pre-test probability of CAD. Invasive coronary angiography demonstrated 0% stenosis in all territories. The inferior wall was interpreted as stress-inducible ischemia by visual reads, constituting a false-positive, whereas the automated relative quantitation approach correctly identified a normal perfusion scan (1% deficit). The defect extent shows the raw perfusion polar map with pixels that fall below the normal criteria set to 0, i.e. blacked-out.

Figure 5B.. Example of a concordant interpretation between visual read and automated relative quantitation of flurpiridaz MPI in a CAD patient undergoing exercise stress.

42 year old male with typical angina and a high pre-test probability of CAD. Invasive coronary angiography was significant for a 76% stenosis in the LCx artery, and 0% stenosis in the other territories. The LCx territory/lateral wall was correctly interpreted as demonstrating stress-inducible ischemia by both visual reads and automated relative quantitation (14% deficit).