Local diagnostic reference levels for paediatric non-cardiac interventional radiology procedures

Purpose: To establish local diagnostic reference levels (DRLs) for non-cardiac interventional procedures in paediatrics. Methods: The type of procedure, the patient's weight and age and dose-related data from 279 interventions was recorded in a database completed by interventional radiologists, radiographers and technicians of the Medical Physics department. These procedures were classified into 14 categories and 6 weight ranges. Local DRLs were proposed for those ranges in which a sample of at least 15 patients could be gathered and were calculated as the third quartile (Q3) of the air kerma-area product (P-KA) values. The Q3 of the fluoroscopy time (FT) and number of digital subtraction angiography (DSA) images were also obtained. Finally, the correlation between P-KA and weight was analysed. Results: Local DRLs are proposed for three types of procedures: hepatic/biliary interventions (5-15 kg, 1304 cGy.cm(2); 15-30 kg, 2121 cGy.cm(2)), sclerotherapy procedures (15-30 kg, 704 cGy.cm(2); 30-50 kg, 4049 cGy.cm(2); 50-80 kg, 3734 cGy.cm(2)) and central venous catheter (CVC) procedures (5-15 kg, 84 cGy.cm(2)). Hepatic/biliary interventions showed a moderate correlation (r = 0.61), while sclerotherapy procedures presented a poor correlation (r = 0.34) between P-KA and weight, possibly due to the P-KA dependence on the complexity level. Regarding CVC procedures, a clearly higher correlation was found when the fluoroscopy P-KA value was normalised to the FT (r = 0.85 vs r = 0.35). Conclusions: The results support the feasibility of establishing DRLs for the most common procedures (sclerotherapy, hepatic/biliary and CVC interventions) despite the small number of paediatric interventions.