Benchmarking adult CT-dose levels to regional and national references using a dose-tracking software: a multicentre experience

OBJECTIVES To benchmark CT-dose data for standard adult CT studies to regional and national reference levels using a dose-tracking system. METHODS Data from five CT systems from three hospitals were collected over a 1- to 2.5-year period (2012-2014), using the same type of dose management system. Inclusion criteria were adult patients and standard CT-head, CT-abdomen-pelvis, CT-thorax, CT-lumbar spine, CT-pulmonary embolism, CT-cervical spine and CT-thorax-abdomen studies, with one helical scan. Volumetric CT-dose index (CTDIvol), dose length product (DLP) and scan length from 31,709 scans were analysed statistically. RESULTS After dose optimisation CTDIvol and DLP values were below the national diagnostic reference levels (DRLs) for all CT studies and for all systems investigated. Mostly no significant differences were found between CTDIvol and DLP levels (p values ≥ 0.01) of CT studies performed on different scanners within the same hospital. Significant dose differences (p values < 0.01) were instead observed among hospitals for comparable CT studies. Dose level range and scan length differences for similar CT studies were revealed. CONCLUSIONS Dose-tracking systems help to reduce CT-dose levels below national DRLs. However, dose and protocol data comparison between and within hospitals has the potential to further reduce variability in dose data of standard adult CT studies. KEY POINTS • Retrospective three-centre study on dose levels of standard adult CT procedures. • Dose-tracking systems help hospitals to stay below national dose reference levels. • Dose-tracking systems help to align CT dose levels between scanners within hospitals. • Benchmarking shows CT dose level variability for similar examinations in different hospitals. • Differences in dose level range/scan length for similar CT studies are revealed.