A BMI-based table (see Figure 1) was proposed to adjust CT CAP acquisition protocol according to patient BMI.
For patients “Small” (BMI < 26 kg/m2),
“Medium” (26 < BMI < 36) and “Large” (BMI > 36 kg/m2),
100 kVp,
120 kVp and 140 kVp should be used,
respectively.
Protocols have been implemented on a GE Discovery 750HD: other minor differences regard bow tie filter and tube rotation time,
complete recap of acquisition parameters can be found in Table 1.
|
Protocol Small
|
Protocol Medium
|
Protocol Large
|
Scan type
|
Helical
|
Helical
|
Helical
|
Detector Configuration
|
64x0.625
|
64x0.625
|
64x0.625
|
Scan FOV
|
Medium Body
|
Large Body
|
Large Body
|
Pitch
|
0.516
|
0.516
|
0.516
|
Rotation time (s)
|
0.4
|
0.5
|
0.5
|
kVp
|
100
|
120
|
140
|
Tube current modulation
|
Smart mA
|
Smart mA
|
Smart mA
|
Noise index
|
21
|
21
|
21
|
Slice thickness (mm)
|
2.5
|
2.5
|
2.5
|
Slice interval (mm)
|
2.5
|
2.5
|
2.5
|
Reconstruction kernel
|
Standard
|
Standard
|
Standard
|
Iterative reconstruction
|
ASiR 50%
|
ASiR 50%
|
ASiR 50%
|
Table 1 - Main acquisitions parameters for the 3 protocols.
Data of unenhanced,
arterial and portal venous phase were retrospectively collected after 6 months (phase I) and 9 months (phase II) with a dose tracking software (RadimetricsTM,
Bayer Healthcare) .
Portal venous phase acquisitions were collected to verify the compliance to the proposed BMI-based chart (portal phase was chosen as it is the only always performed in staging and follow up,
while acquisitions of unenhanced and/or arterial phases depend on the type of tumor) and for dosimetric purposes.
Unenhanced and arterial phase acquisition data were collected for dosimetric purpose only.
Radimetrics collects all data from the modality (technical parameters and dose indices),
in addition provides organ doses and effective doses matching each patient to a corresponding mathematical phantom,
according to patient dimensions.
Dose calculation is performed using libraries with look-up table obtained with MonteCarlo simulations on each phantom.
This method allows to account for patient size,
providing a more accurate dose estimation.
Exposure data were analyzed with R software (version 3.5.1),
comparison among subgroups was assessed with Mann-Witney U-test.
In addition,
image quality of 20 Small (BMI < 26) patients examined with Small Protocol (100 kVp) and of 20 Small patients examined with Medium Protocol (120 kVp) was blinded assessed by 2 radiologists (one resident and one with 10 years of experience in the field) using a 5-points Likert-scale.
Image quality of lung parenchyma,
soft tissue and bones in the portal phase was evaluated with score ranging from 1 (poor,
non-diagnostic) to 5 (excellent),
an additional score was given to overall image quality.
Agreement was evaluated with Cohen’s-k,
while statistical significance of the score between 100 and 120 kVp was evaluated with Mann-Whitney U-test.