3.210.
Paragraph 3.168 of GSR Part 3 [3] requires that registrants and licensees of radiology facilities ensure that patient dosimetry be performed in diagnostic radiology and image guided interventional procedures and that typical doses to patients for radiological procedures be determined. Knowledge of the typical doses at a facility forms the basis for applying methods of dose reduction as part of optimization of protection and safety. It also enables the radiology facility to use DRLs (see paras 3.224–3.231) as another tool for the optimization of protection and safety.
3.211.
Clearly, the more radiological procedures at the radiology facility for which typical doses are known, the better the basis for the optimization of protection and safety. GSR Part 3 [3] requires determination of typical doses for common radiological procedures in radiology facilities. The procedures that are considered to fall into this category will vary from facility to facility, and State to State, but common core examinations generally include the following:
Radiography: head, chest, abdomen and pelvis.
CT: head, chest, abdomen and pelvis, for specified clinical indications.
Fluoroscopy: barium swallow and barium enema.
Mammography: craniocaudal and mediolateral oblique.
Dentistry: intraoral, panoramic and CBCT.
Bone densitometry (DXA): spine and hip.
3.212.
For image guided interventional procedures, typical doses for the broad types of procedure performed at the facility should be ascertained. For example, an interventional cardiology facility would characterize typical doses for percutaneous coronary interventions, including percutaneous transluminal coronary angioplasty. A facility performing neurological procedures might characterize typical doses for diagnostic cerebral angiograms and for embolization interventions. Other image guided interventional procedures might include endoscopic retrograde cholangiopancreatography and transjugular intrahepatic portosystemic shunt.
3.213.
The term ‘typical dose’, as used in para. 3.168 of GSR Part 3 [3], is the median or average dose for a representative sample of normal size patients, at clinically acceptable image quality. Patient size has a large influence on dose, so some selection or grouping of patients is recommended. Such groupings include ‘standard adult’, often based on an average mass of 70 kg with a range of ±20 kg. Groupings for children have sometimes been based on age, such as newborn (0 years), infant (1 year), small child (5 years), child (10 years) and teenager (15 years), but more recently size specific groupings are being recommended and used, for example by using body mass intervals [14]. Patient size groupings should be adopted that correspond to the groupings used for the DRLs in the State or region. The sample size used for each patient grouping and radiological procedure should be of sufficient size to assure confidence in the determination of the typical dose. A representative sample of 10–20 patients per procedure type is needed for non-complex examinations such as radiography and CT, preferably 20–30 patients for complex procedures such as fluoroscopy and fluoroscopically guided procedures, and 50 patients for mammography [14] (see also paras 2.39–2.41).
3.214.
The dose in the term ‘typical dose’, as used in para. 3.168 of GSR Part 3 [3], means, for the given radiological procedure, an accepted dosimetric quantity as described in paras 2.40 and 3.202–3.204. For particular reasons (e.g. for risk estimation or for collective dose estimation), the dose to a particular organ or the effective dose can be estimated from the typical dose.
3.215.
Patient dosimetry to determine typical doses should be carried out in conjunction with an assessment of the diagnostic image quality. Exposure alone is not meaningful if it does not correspond to images that are adequate for an accurate diagnosis. Therefore, patients included in the sample used for determining typical doses should only be those whose radiological procedure resulted in acceptable image quality.
3.216.
The results of the surveys used to determine typical doses at the radiology facility should be used as part of the ongoing review of the optimization of protection and safety at the facility, and should be used for comparison with established DRLs (see paras 2.34, 2.45 and 3.224–3.231). The results should also be submitted to the organization in the State or region that is responsible for establishing and reviewing national or regional DRLs. Patient dosimetry surveys, required by GSR Part 3 [3], should take place at intervals of no more than five years and preferably no more than three years. Another trigger for a survey would be the introduction of new equipment or technology into the radiology facility or when significant changes have been made to the protocols or the equipment.
3.217.
Sometimes, patient dosimetry in diagnostic radiology or image guided interventional procedures may be required for specific individual patients, either through measurements or calculations. Reasons might include an unintended or accidental medical exposure, where an estimation of patient doses is required as part of the investigation and report (see para. 3.265), or because there is a need to estimate the dose to an embryo or fetus (see para. 3.161).
3.218.
There are several indirect and direct methods to estimate patient dose in diagnostic radiology and image guided interventional procedures. Methodologies for these determinations are explained in detail in Refs [10–12, 171, 173–178] and are summarized in the following:
Estimations based on incident air kerma or entrance surface air kerma measurements corrected for the techniques used (e.g. X ray tube potential, current and time, and source–skin distance). This approach can be used in radiography (medical and dental), fluoroscopy and mammography.
Estimations based on measured air kerma–area product. This approach can be used in radiography (medical and dental), fluoroscopy and CBCT.
Estimations based on measurements of CT air kerma index and CT air kerma–length product. This approach can be used for CT.
Reported values of dose quantities from DICOM headers or the DICOM radiation dose structured reports. The accuracy of the reported dose quantities should have been validated in acceptance testing and commissioning and by means of quality assurance procedures as explained in para. 3.244. This approach is applicable to all digital modalities.
Direct measurements for selected organs, such as the skin for interventional procedures. For this, thermoluminescent dosimeters and optical stimulated luminescent dosimeters as well as radiochromic or silver halide film can be used.
In the case of CT, size specific dose estimates can be made, where CT air kerma index values are corrected by taking into consideration the size of the patient using linear dimensions measured on the patient or patient images [12, 177].
3.219.
When necessary, organ doses can be derived from the quantities mentioned in para. 3.218 by using conversion coefficients derived from Monte Carlo codes applied to anatomical models. Methods for doing this are described in Ref. [11].