Diagnostic Reference Levels (DRLs) in paediatric radiology

For the application of DRLs in paediatric radiology, medical professionals need to consider the size of children, and hence the dose levels, significantly varies not only by age but also at a given age. Guidelines on DRLs for paediaric examinations have been developed within the European PiDRL project.

Frequently asked questions by the health professionals

» Which dose quantities are used for setting DRLs for paediatric radiology? 
» How should we account for patient size? 
» Where should I start in the absence of well-established national and local support for DRLs (as, for example, a paediatric radiology facility in a less-resourced country)? 
» What are the potential pitfalls in comparing local median values with published values? 

» Which dose quantities are used for setting DRLs for paediatric radiology?

DRLs should be set in terms of the same practical dose quantities used to monitor adult radiology practices.

» How should we account for patient size?

Particular consideration is needed in the grouping of patients for setting paediatric DRLs because the size of children, and hence the dose levels, significantly varies not only by age but also at a given age. Adults usually vary in size by a factor of 4 (40 – 160 kg bodyweight), whereas paediatric patients vary in size from premature babies (e.g., 300-400 g) to obese adolescents (> 80 kg body weight) representing a factor of more than 200.

For all body examinations, and for DRLs based on prospective patient dose surveys, weight should be used as the parameter for patient grouping. Table 1. Recommended grouping of patients for paediatric DRLs. In the transition period, age can be used as an additional parameter for patient grouping and for the purpose of comparison of proposed new, weight-based DRLs with earlier values (trend analysis). For head examinations, age is recommended as the grouping parameter. The groups (intervals) recommended in the PiDRL report are shown in the following table:

Recommended weight groups (intervals) for body examinationsRecommended age groups (intervals) for head examinations
< 5 kg0 - < 3 months
5 - < 15 kg3 months - < 1 y
15 - < 30 kg1- < 6 y
30 - < 50 kg≥ 6 y
50 - < 80 kg 

There is a rough correlation between the average weight and age groups, as can be deduced from the published weight-for-age charts. The following table presents the approximate equivalence of weight and age groups for the purpose of comparing weight-based DRLs with age-based DRLs:

DescriptionWeight groupAge group based on eight-for-age charts Most common age groups
used for the NDRLs (or equivalent)
Neonate< 5 kg< 1 m0 y
Infant, toddler and early childhood5 - < 15 kg1 m - < 4 y1 y
Middle childhood15 - < 30 kg4 - < 10 y5 y
Early adolescence30 - < 50 kg10 – < 14 y10 y
Late adolescence50 - < 80 kg14 - < 18 y15 y

To overcome the problem caused by the need for several patient groups and the general paucity of patient dose data in paediatric imaging, instead of using discrete patient groups, the dosimetric quantity can be presented as a function of the parameter used for patient grouping, i.e. to define a DRL-curve. The comparison of local patient dose data with the DRL-curve can be based e.g. on ten consecutive patients, regardless of their age/size/weight, by inserting these data points in the graph with the DRL-curve.

» Where should I start in the absence of well-established national and local support for DRLs (as, for example, a paediatric radiology facility in a less-resourced country)?

The priority for a medical facility might be to estimate typical patient dose quantities in relation to present practice for a few common examinations on adult patients, according to the following steps:

  • Record displayed values of radiation dose quantity for samples of 10 or more typical adult patients undergoing procedures for common clinical indications. For example, for CT, consider including examinations of head (e.g. in relation to acute stroke), chest (e.g. in relation to lung cancer) and abdomen (e.g. in relation to acute abdominal pain); 
  • Verify the accuracy, and if necessary, apply correction factors, of the displayed values of radiation dose quantities; 
  • Calculate for each type of examination the median values of dose quantities (e.g. CTDIvol and DLP for CT); these are your typical dose levels (but not your local DRLs that are set for a group of imaging systems or a group of hospitals); 
  • Compare your typical dose levels (median values) with published DRLs for a similar practice in the absence of local or national DRLs, in order to provide a broad indication of your relative performance and urgency of need for improvement in your imaging technique;
  • Comparison of typical dose levels (median values) to DRLs is not sufficient, by itself, for optimisation of protection. Image quality or, more generally, the diagnostic information provided by the examination (including the effects of post-processing), must be evaluated as well;
  • If your values are below published DRL, this does not necessarily indicate satisfactory performance. Imaging techniques should always be reviewed for potential reduction in their levels of dose without compromising the clinical purpose of the examination; 
  • If your values are above DRL, there is a more urgent need to investigate whether simple changes can be made to the imaging settings selected for an examination in order to reduce values of radiation dose quantities whilst still providing the required clinical information; 
  • Levels of dose should be reassessed following revision of imaging technique in order to allow further comparisons (see steps above). 

» What are the potential pitfalls in comparing local median values with published values? 

Published DRLs can prove useful in allowing comparison of median dose values in your facility, for a particular imaging system, although potential problems in this process include the following:

  • Published DRLs values from other countries (with potentially different imaging practices and technology) may not be relevant to your particular circumstances;
  • The types of examination or procedure specified for the published DRLs (as being with or without detailed clinical indications) may not be directly relevant to your particular practice; 
  • Published dose values may not have been obtained using the same methodology (e.g. total values or values per projection or per series) or in relation to the same standard condition like CT dosimetry phantom (diameter of 16 cm or 32 cm), or may be given in different dose quantity or unit;
  • Published DRLs values may not be expressed in a different dose quantity or dose unit;
  • The patient sample (number of patients and their body size) in the published survey may be different;
  • Advances in technology, such as post-processing and iterative reconstruction in CT, will need to be taken into account when updating DRLs. 

Read more:

  • European Commission (EC). European Guidelines on DRLs for Paediatric Imaging,  2016. (The draft is available here).