In line with the ESC mission, newly presented content is made available to all for a limited time (4 months for ESC Congress, 3 months for other events). ESC Professional Members, Association Members (Ivory & above) benefit from year-round access to all the resources from their respective Association, and to all content from previous years. Fellows of the ESC (FESC), and Professionals in training or under 40 years old, who subscribed to a Young Combined Membership package benefit from access to all ESC 365 content from all events, all editions, all year long. Find out more about ESC Memberships here.
Minimization of temporal sampling for myocardial blood flow quantification using Rubidium-82 PET
JD Van Dijk1
JA Van Dalen4
1Isala Clinics, Nuclear Medicine - Zwolle - Netherlands (The)
2Isala Clinics, Cardiology - Zwolle - Netherlands (The)
3University of Twente, MIRA: Institute for Biomedical Technology and Technical Medicine - Enschede - Netherlands (The)
4Isala Clinics, Medical Physics - Zwolle - Netherlands (The)
Introduction: Dynamic images are required to quantify myocardial blood flow (MBF) using Rubidium-82 (Rb-82) PET myocardial perfusion imaging (MPI). These dynamic scans are reconstructed from list-mode data using multiple time frames. Both the duration and the number of time frames can influence the measured activity concentrations in the left ventricle (LV) and myocardium and the thereof derived MBF values. Short time frames are crucial to accurately capture the activity concentration peak during the first-pass phase. However, an increasing number of time frames results in time-consuming reconstructions which may hamper clinical throughput.
Purpose: To determine the minimum number of time frames that still results in reliable MBF quantification using Rb-82 PET.
Methods: We retrospectively included 30 consecutive patients (15 with an ejection fraction (EF) <45% and 15 with an EF =45%) who underwent pharmacological induced (regadenoson) stress Rb-82 PET (Discovery 690, GE Healthcare). Data were reconstructed using 26 time frames (12x5s, 6x10s, 4x20s, 4x40s). Activity concentrations during stress were determined for the LV and whole myocardium for each time frame using Corridor4DM software (INVIA, v2016). Next, we simulated the time activity curves for twelve additional temporal sampling protocols varying between 9 and 20 frames, as shown in Figure 1. MBFs were quantified with TracerRkinetic software using using Lortie's one-tissue compartment model. Absolute MBF values where corrected for systemic biases. Temporal sampling protocols were considered reliable if the SD of the relative MBF difference (SD of ?MBF) was =5% compared to that of the reference scan using 26 frames.
Results: Seven protocols were considered reliable as illustrated in Figure 1: the protocols with 20 (12x5s, 6x10s, 2x120s), 18 (12x5s, 2x10s, 2x20s, 2x120s), 17 (12x5s, 3x20s, 2x120s), 16 (12x10s, 2x30s, 1x60s, 1x120s), 14 (9x10s, 3x30s, 1x60s, 1x120s), 14 (12x10s, 2x120s) and 12 (9x10s, 1x30s, 2x120s) frames resulted in a SD of ?MBF =5%.
Conclusion: Temporal sampling protocols with less than 26 frames seem to result in reliable MBF quantification using Rb-82 PET. The minimal number of time frames that can be considered for clinical validation is 12. Moreover, to obtain reliable MBF measurements, the first pass phase (0 to 60 seconds) can be sampled using frame durations of 5 to 10s, the intermediate phase (60 – 120 seconds) with 10 to 20s frames and the tissue phase (120 – 360 seconds) with frame-durations of 120s.
ESC 365 is supported by Bayer, Boehringer Ingelheim and Lilly Alliance, Bristol-Myers Squibb and Pfizer Alliance, Novartis Pharma AG and Vifor Pharma in the form of educational grants. The sponsors were not involved in the development of this platform and had no influence on its content.
Our mission: To reduce the burden of cardiovascular disease