Most diagnostic nuclear cardiology studies involve imaging patients during 2 physiologic conditions: rest and stress. It is expected that the biokinetics of the radiotracer in various organs would differ during these 2 physiologic conditions, resulting in different absorbed doses and EDs. With the exception of 99m
Tc-labeled tracers (10
), dose coefficients and EDs for rest and stress imaging are not generally available.
Pharmacologic agents (e.g., adenosine and dipyridamole) are routinely used in 82
Rb stress myocardial perfusion imaging. These agents are coronary vasodilators and exert their action by stimulating increases in coronary blood flow within the myocardium to maximal or near-maximal levels (11
). However, the increase in coronary blood flow due to pharmacologic stress does not translate into a linear increase in myocardial tracer concentration. The myocardial uptake of 82
Rb shows a leveling-off phenomenon at high flow rates because of a decrease in first-pass extraction (13
). When dipyridamole is used as a stress-inducing agent, the maximum effect occurs 3–7 min after completion of the 4-min infusion of dipyridamole. The hyperemic response is prolonged because the half-life of the drug is on the order of 30 min. This action of pharmacologic stress and the short half-life of 82
Rb make it impossible to obtain source-organ biokinetics using multiple-time-point whole-body imaging. Because of these constraints, it was necessary to obtain the input to the dosimetry calculations by scaling the source-organ TIAC values obtained from a rest-only study by the ratio of stress to rest TIAC.
The organ receiving the highest absorbed dose was the heart wall during stress, as compared with the kidneys at rest. Though differences in TIACs and corresponding differences in absorbed doses were observed between the stress study and the previous rest-only study (2
), these differences were not significant except for heart wall. Comparison of source-organ TIAC at rest calculated using 3 whole-body PET images from this study with that calculated from the previous study (2
) showed comparable results and supports our method. The mean ED (across subjects) calculated using ICRP 60 and ICRP 103 weighting factors during stress in this study was similar to that reported in our previous rest study (2
In our previous report on 82
Rb dosimetry performed under resting conditions (2
), we addressed the discrepancy in the prior dose estimates for 82
Rb between ICRP 53 (3
) and the Cardiogen-82 package insert (14
). Before our report on 82
Rb dosimetry, the 2 major sources of dose estimation were the ICRP 53 and the Radiation Internal Dose Information Center (RIDIC) compendium of dose estimates (15
). The ICRP dose estimates were based on a flow model for the tracer, and the RIDIC dose estimates were based on human data for a limited number of source organs obtained using γ-camera imaging by Ryan et al. (16
). Separate dose estimates for rest and stress conditions were not reported. Similar to our previous observations in the resting state dose estimation, the significant differences between our current stress dose estimation and the dose estimation reported in ICRP 53 were for thyroid and adrenals, which were 25- and 10-fold lower for our study than for ICRP 53. Also, the dose estimates for kidneys and lower large intestine wall were more than 3- and 4-fold lower for our study than for ICRP 53. As explained in our previous report, the conservative blood flow model adopted by the ICRP 53 is the major reason for the differences in dose estimation. On the other hand, the dose estimation by RIDIC using the data of Ryan et al. (16
) was lower than our dose estimation overall.
When we compare our ED calculation based on ICRP 60 (8
) weighting factors with the ICRP 53 data–derived ED (3.4E–03 mSv/MBq) as given in ICRP 80 (17
), our ED for both rest and stress conditions is about 3-fold lower. However, the RIDIC-calculated ED is about 30% lower than our ED. Based on our previously reported dose estimation for 82
Rb at rest and the current dosimetry under pharmacologic stress, for a clinical 82
Rb PET scan with a 1,480-MBq injection at rest and stress (2 × 1,480 MBq) the total ED would be 3.3 mSv and 3.8 mSv based on ICRP 60 and ICRP 103 (9
) weighting factors, respectively. For the same protocol, the organs receiving the highest equivalent doses would be the kidneys (16 mSv), heart wall (13 mSv), and lungs (9 mSv). The additional dose from a transmission CT scan for attenuation correction would need to be added (about 0.3 mSv for the cardiac region in our protocol) to obtain total ED from a clinical PET/CT procedure. The ED from CT strictly depends on the individual protocol, which may vary across institutions and scanners.
As the number of noninvasive cardiovascular imaging procedures continues to increase, the concern about increased radiation dose from these procedures has also grown (1
). An effort is being made to use procedures with the lowest reasonably achievable radiation dose. Dosimetry for PET myocardial perfusion imaging tracers such as 13
N-ammonia and 15
O-water is well defined, and the resulting ED of clinical PET protocols with these tracers is substantially lower than that of SPECT protocols. The results obtained for 82
Rb under both rest (2
) and stress conditions places it in the same lower dose range as alternative PET perfusion tracers.