Purpose
Investigate whether image quality of the cervical spine after 40-80% CT radiation dose reduction can be maintained by the use of Iterative Model Reconstruction (IMR,
Philips).
Methods and Materials
In this prospective randomized clinical pilot-study performed in our level I trauma center,
patients aged ≥50 years suspected of cervical spine injury (based on the National Emergency X-Radiography Utilization Study-criteria) were scheduled for a standard of care CT scan (120kV; Effective Dose 1.16±0.42mSv) reconstructed with hybrid iterative reconstruction (iDose4 level 3) and an additional low-dose CT scan with a 40%,
60% or 80% (figure 1) reduced radiation dose reconstructed with IMR level 2.
Three radiologists assessed the subjective image quality using a four-point Likert scale....
Results
32 patients (mean age 70 years,
SD ± 2 years; 66% male) were distributed evenly over the different dose reduction groups.
A significantly lower subjective image quality was found for dose reduction IMR scans (p<0.005; figure 2).
Dose reduction IMR scans showed significant higher SNRs and CNRs,
with the highest scores in the -40% and -60% groups (figure 3).
Conclusion
In this interim analysis of our pilot study,
loss of image quality of cervical spine CT due to 40-80% radiation dose reduction could not be overcome by the use of IMR.
The observed quality with the low radiation doses raises doubts whether a dose reduction of 40% or more can be compensated with the currently available reconstruction techniques.
Personal Information
B.Y.M.
van der Kolk and L.N.
Buijteweg: Department of Emergency Medicine,
Isala hospital,
Zwolle,
The Netherlands.
B.
Korte,
B.A.A.M.
van Hasselt,
H.J.
Boelhouwers and M.F.
Boomsma: Department of Radiology,
Isala hospital,
Zwolle,
The Netherlands.
M.A.
Edens: Department of Innovation and Science,
Isala hospital,
Zwolle,
The Netherlands.
M.
Podlogar: Department of Neurosurgery,
Isala hospital,
Zwolle,
The Netherlands.
M.
Maas: Department of Radiology,
Academic Medical Center,
Amsterdam,
The Netherlands.