Patient radiation doses and references levels in interventional radiology.

Purpose: To set Diagnostic Reference Levels (DRLs) in interventional radiology by means of dose area product (DAP) measurements and the grouping of homogeneous procedures, and to quantify the associated errors in the DRL estimates. To evaluate the Mean Effective Doses per single procedure.
Materials and Methods: Interventional radiology procedures were divided into four main groups: neuroradiological, vascular, extravascular and paediatric. Neuroradiological and vascular procedures were further divided into diagnostic and interventional procedures. Starting from DAP and total fluoroscopy time measurements in 1,256 patients, the DRLs were determined for 17 procedures, together with an estimate of their uncertainty. The correlation between fluoroscopy time and DAP was assessed. Mean effective dose estimates were obtained from measured DAP values and from the analysis of the dosimetric data reported in the literature for similar procedures.
Results: The main features of DAP distributions are long high-dose tails, indicating asymmetric distributions, together with a large interquartile range. Rounded third-quartile values of DAP distributions showed a large range in the procedures taken into consideration. Values of 147, 198, 338 Gy cm(2)were obtained for supra-aortic angiography, cerebral angiography and embolization. Values of 86-101 and 459-438 Gy cm(2)were obtained for diagnostic and interventional vascular procedures on the lower limbs and on the abdomen, respectively. Values of 25-33 Gy cm(2)were obtained for retrograde cystourethrographies and ERCP, and values of 62-158 Gy cm(2)were obtained for nephrostomy and percutaneous transhepatic cholangiography. The correlation between total fluoroscopy time and the DAP values was poor. Mean effective dose estimates showed lower values for extravascular procedures (4.8-28.2 mSv), intermediate values for neuroradiological procedures (12.6-32.9 mSv) and higher values for vascular procedures involving the abdomen (36.5-86.8 mSv).
Discussion: DAP values were generally higher in vascular than in extravascular procedures. In generally, interventional vascular procedures show higher DAP values than the corresponding diagnostic procedures, with the exception of the abdominal region where the values were similar. Extravascular procedures with percutaneous access show significantly higher DAP values than those with endoscopic access. Total fluoroscopy time is a poor predictor of patient doses in interventional radiology.
Conclusions: The systematic recording of DAP values, together with adequate grouping of similar procedures makes it possible to establish stable DRLs on a local basis and to carry out dosimetric evaluations, although on a statistical rather than individual basis. Patient radiation doses during interventional radiological procedures may be high, particularly when the abdominal region is involved.