Sci—Thur PM: YIS — 08: Comparison of New In‐Vivo Measurements for Dosimetry of Rb‐82 with Prior Blood Flow Model Predictions

• Published in: Medical Physics 2010-07, Vol.37 (7), p.3885-3885
• Main Authors: Hunter, C, Ziadi, C, Etele, J, Hill, J, Beanlands, R, deKemp, R
• Format: Article
• Language: English
• Subjects: Anatomy; Biomedical modeling; Blood flow measurement; Cardiac haemodynamics; Computed tomography; Dosimetry; Haemodynamics; Heart; Medical imaging; Positron emission tomography
• ISSN: 0094-2405; 2473-4209
• DOI: 10.1118/1.3476103

Published radiation dose estimates for Rubidium‐82 include two In‐Vivo studies and two theoretical blood flow models, the results of which vary widely. Rb‐82 internal organ and effective doses with PET/CT in humans were determined and compared to the published estimates with emphasis given to the differences between in‐vivo measurements and blood flow model predictions. 26 cardiac patients and 4 normal subjects with no cardiac history were recruited. Dynamic 3D PET scans were acquired (GE Discovery RX/VCT) over 10 minutes following IV injection of 10 MBq/kg Rb‐82. Images were reconstructed using FORE‐OSEM and 8 mm Hann filter. Cardiac scans of the chest were acquired at rest for all 30 subjects, plus one additional scan of the Head, Neck, Abdomen, Pelvis, or Thighs. Mean Rb‐82 residence times were determined in 22 source organs using volumes‐of‐interest drawn on the fused PET/CT images. At least 4 samples were obtained in each source organ. Using ICRP 103, the male and female effective doses were 0.00074 and 0.00092 mSv/MBq respectively. The lungs were found to be the major contributors to the effective dose for all in‐vivo studies, whereas it was the thyroid for both blood flow models. These dose estimates for Rubidium‐82 are the first to be measured directly with PET/CT in humans, and are in agreement with the two in‐vivo measurements but 2 to 3 times lower than blood flow models. The new values derived from human studies suggest a typical effective dose of 0.6 mSv per scan with 3D PET.