Direct and indirect assessment: comparison of different methods to estimate staff equivalent dose to the lens during different interventional procedures.

Abstract

The reduction of the annual equivalent dose limit to the lens of the eye from 150 to 20 mSv has increased the need for reliable assessment of occupational eye lens exposure, particularly in interventional radiology. Lens absorbed dose is commonly evaluated either using a dosimeter near the eye or indirectly using chest dosimeter readings. Although the indirect approach is well established in interventional cardiology, its validity for procedures with different exposure geometries remains uncertain. This study evaluated the agreement between direct and indirect estimates of lens equivalent dose in haemodynamic, electrophysiological and gastroenterological procedures. Forty-one physicians were monitored using thermoluminescent dosimeters placed close to the eye and on the chest above the lead apron. Direct lens doses were compared with indirect estimates derived from chest dosimeter. Agreement was assessed using linear fitting, Pearson's correlation coefficients and analysis of variance. A total of 449 valid dosimetric pairs were analysed. A statistically significant correlation between direct and indirect lens dose estimates was observed for haemodynamic and electrophysiological procedures. The indirect lens dose estimates provided conservative estimates in approximately 80% of cases and never resulted in exceedance of the 20 mSv annual dose limit. In contrast, no significant correlation was found for endoscopic retrograde cholangiopancreatography procedures, where indirect estimates frequently underestimated lens dose due to non-frontal exposure geometry. These findings confirm that reliable and conservative assessment of eye lens dose estimates in interventional cardiology is possible when chest doses are well below regulatory limits. However, caution is required when extending its use to non-cardiac procedures, where procedure-specific exposure conditions may compromise the reliability of lens dose estimates.