Methods and materials
A list of the most common paediatric procedures in CT, plain radiography and diagnostic fluoroscopy, associated to clinical indications and age/weight bands, was established by the SFIPP and SFR radiation protection boards.
Radiology departments of dedicated paediatric hospitals but also of general healthcare facilities were invited to collect prospective or retrospective data on procedures performed in 2015 or 2016 (Table 1), including:
- Equipment data: manufacturer, model, commissioning year and technical features,
- Examination data: patient age and weight, technical parameters and DRL quantities: volume CT dose index (CTDIvol) and dose-length product (DLP) for CT, air kerma-area product (PKA) for plain radiography and diagnostic fluoroscopy.
Statistical analysis was carried out according to ICRP recommendations [1], which state that DRLs should be established as the rounded value of the 3rd quartile of the distribution of the median values per department. The median values per department were calculated only when the number of patients was at least 10 as recommended by the European guidelines [4].
Since weight was available for less than half the patients, age was used as the parameter for patient grouping.
Results
The study showed that the good practices promoted by the SFIPP were generally applied in the participating departments:
- Use of recent imaging devices with dose reduction options (Table 1),
- Fluoroscopy for patient positioning never or exceptionally used during radiography procedures,
- Single projection examinations in radiography or single phase examinations in CT generally performed.
Table 2 and Table 3 presents the results of the survey for each type of examination according to the age groups recommended by the 2018 European guidelines. The 3rd quartiles of the DRL quantities were calculated to propose new DRL values, as well as the median values which can serve as an additional tool to aid in optimisation, sometimes called “achievable dose level”:
- CTDIvol per sequence and DLP per examination in CT (by reference to the standard 16 cm phantom for head examinations and by reference to the 32 cm phantom for body examinations),
- PKA per projection in plain radiography,
- PKA per examination in diagnostic fluoroscopy.
Discussion
As expected, the 3rd quartiles obtained were far lower than the applicable national DRLs at the time in CT and plain radiography, with the only exception for petrous bone CT DLP [5].
For CT procedures, all the calculated values were lower than the European DRLs published in 2018 [4]. They were at the lower end of the results of the most recent multicentre surveys [6]. The low exposure levels can be explained by the age profile of the CT scanners of the sample, consistent with the installed base in France and by the prevalence of the single sequence procedures.
In plain radiography, the calculated 3rd quartiles were slightly lower than the European DRLs except for pelvis examinations on adolescents and consistent with the most recent published studies and national DRLs [7].
In diagnostic fluoroscopy, the comparison was possible for micturating cysto-urethrography (MCU) only and the results of the study were higher than those of the published studies [7], as well as the EDRLs.
In diagnostic fluoroscopy, a larger spread in patient dose among departments was noted, showing a great need for optimisation and homogenisation of practices. The dose dispersion between departments was low for CT procedures, below those observed in the two other modalities.
Several limitations, common in paediatric DRLs surveys, have to be mentioned: