Ru-82 3D PET/CT MPI training leads to 94% inter-reader agreement

The Rubidium-ARMI (alternative radiopharmaceutical for myocardial imaging) multicenter trial has established a bar for assessing myocardial perfusion imaging (MPI) that meets or even exceeds previous SPECT MPI reader scores after thorough consensus review, according to a study published Nov. 18 in the Journal of Nuclear Medicine.

A team of researchers, including Jennifer M. Renaud, MSc, a scientist from the University of Ottawa Heart Institute, standardized imaging and interpretation protocols associated with the Rubidium-ARMI trial and then assessed them against previous studies. Results showed that standardized interpretation and optimized imaging protocols lead to 94 percent inter-reader agreement, which meets or exceeds that of previous SPECT studies that showed a range of agreement from 68 to 97 percent agreement.

“Rb-82 myocardial perfusion imaging protocols were implemented with highly repeatable interpretation in centers using 3D PET/CT technology, through an effective standardization and quality assurance program,” wrote Renaud et al.

The main objective of the trial is to evaluate 3D PET/CT for MPI and its associated effectiveness, health outcomes and cost. This aspect of the trial establishes consensus review for the procedure. Standardization was achieved using 17-segment visual interpretation with a discretized color map. For this study, cardiac phantom qualifying scans were performed across seven recruiting centers. At every site, scores were accumulated for 25 patients from low-dose, 10 megabecquerel per kilogram (MBq/kg) rest and pharmacologic stress scans using Ru-82 3D PET/CT. Scores for stress, rest and the difference between were summed and additionally interpreted by the University of Ottawa Heart Institute core lab to evaluate imaging quality, clinical diagnosis and scoring agreement. A third review was then implemented to further improve scoring agreement.  

Results of the study showed that qualifying scores for stress, rest and the difference were uniform across all seven sites and nine different scanners. Overall interpretation revealed “excellent” agreement of 94 percent after statistical analysis. Agreement scores between studies determining normal versus abnormal and nonischemia versus ischemia was also high, at 0.90 and 0.88, respectively.

“The improvement in scoring agreement after consensus review demonstrates the added value of this process in improving consistency and training experience,” wrote the authors. “Significant improvement in half of the sites demonstrates that they required the consensus review process to improve their technique and experience to accurately score more difficult cases, whereas the other sites already had adequate understanding of the scoring methods after initial training.”