Background/introduction
National dose reference levels (NDRLs) are normally set on the basis of the typical doses observed in wide scale (national) surveys [1].
NDRLs represent the dose values averaged over a broad range of healthcare facilities and do not take into account any dose reduction technique (i.e.
automatic exposure control,
low tube voltage scanning,
etc.) applied by specific hospital.
Additionally,
majority of the existing NDRLs are based on anatomical regions and do not consider clinical indications and desired image quality.
Clinical-based DRLs may overcome these limitations....
Description of activity and work performed
Data collection
Dose data of various CT procedures performed within the last 12 months were electronically collected by dose tracking software (DOSE,
Qaelum NV,
Belgium) using CT DICOM radiation dose structure reports.
The dose metrics together with patient-specific parameters were collected from five chest CT protocols used for the following indications: pulmonary embolism (PE),
lung cancer (LC),
lung nodule follow-up (LNFU) and coronary CT angiography (CCTA).
Data analysis
The data was filtered to remove the scout view,
monitoring and pre-monitoring acquisitions.
The dose length product...
Conclusion and recommendations
Analysis included 1280 CT procedures (PE=290; LC=595; LNFU=360; CCTA=95).
The clinical DRLs for the investigated indications are shown in Table 1.
The clinical DRLs for PE and LC were similar to the recently revised NDRLs (CTDIvol=7mGy; DLP=300 mGy*cm) for the chest.
The clinical DRLs defined for CCTA were higher than NDRLs,
while the values for LDFU were significantly lower.
Within the same type of the procedure,
a strong correlation between CTDIvol and patient size was observed for LE and LNFU (r-values 0.77 and 0.79,
respectively);...
Personal/organisational information
Contact details: Natalia Saltybaeva,
University Hospital Zurich,
Zurich ,
Switzerland; email:
[email protected]
References
1. ICRP,
2017.
Diagnostic reference levels in medical imaging.
ICRP Publication 135.
Ann.
ICRP 46(1).