R. Boellaard et al., Characteristics of a new fully programmable blood sampling device for monitoring blood radioactivity during PET, EUR J NUCL, 28(1), 2001, pp. 81-89
Citations number
15
Categorie Soggetti
Radiology ,Nuclear Medicine & Imaging","Medical Research Diagnosis & Treatment
The first performance tests of a new fully programmable blood sampling devi
ce for monitoring blood radioactivity during positron emission tomography (
PET) are described. Blood is withdrawn through 1-mm internal diameter tubin
g using an infusion pump which can be operated at rates varying from 0 to 6
00 ml/h. Activity in blood is measured by a 6-cm-thick bismuth germanate cr
ystal connected to a photomultiplier tube and multichannel analyser (MCA) w
hich are positioned within 6 cm lead shielding. Positioning of the tubing i
s an exact and simple procedure. The minimal readout time of the MCA is 1 s
. Two independent energy windows can be set. Operation of the pump and MCA
is fully controlled by a PC, i.e. sampling time, interval time and pump rat
e can be varied at any time during the PET scan by user-defined scripts. A
number of characteristics of the new system were studied, such as sensitivi
ty, dead time, linearity, effect of background radiation and pump rate as a
function of input pressure. In addition, dispersion was measured as a func
tion of pump rate. Finally, first clinical results were compared with manua
l samples. The sensitivity equalled 0.7 and 0.2 cps/Bq for 511- and 1022-ke
V 30% energy windows, respectively, and the system dead time was 500 ns. Th
e system remained linear within 2% with activity concentrations up to 2.5 M
Bq/cc. Short-term reproducibility was better than 3% for a 1-h period. Long
-term reproducibility was about 5% (1SD), which was mainly caused by variat
ion in the diameter of the tubing. If the device was positioned in such a w
ay that maximum shielding was directed towards the patient, the effects of
background radiation from the patient on the measured activity concentratio
n for clinically relevant conditions was minimal (<3%). Pump rate varied wi
th input pressure, but remained constant for a given pressure. Dispersion c
onstants smaller than 0.14 s(-1) were observed for pump rates higher than 3
00 ml/h, indicating that the system dispersion is small. Clinical data show
ed an excellent agreement to within 3% (1SD) between the results obtained w
ith the new system and manual samples. With the continuous blood sampler ra
dioactivity in blood can be measured accurately during the entire course of
the PET scan. Furthermore, the system is fully programmable allowing adjus
tment of all parameters during a single PET scan.