Factors affecting digital radiography exposure indices: are radiographers achieving the recommended levels?

Abstract

Introduction With digital radiography (DR), radiographers receive immediate feedback on detector exposure in the form of the exposure index (EI). Purpose To identify whether radiographers were meeting the manufacturer-recommended EI (MREI) value ranges for routine chest, abdomen and pelvis X-ray examinations under a variety of conditions and to examine factors that impacted upon EI levels. Methods Data on 5000 adult X-ray examinations were collected, including: EI values, examination parameters, patients’ gender, dates of birth, dates and times of examinations, grid usages and the presence of metallic implants or prostheses. Descriptive statistics were used to summarise data sets and Mann–Whitney U testing was performed to establish causal agents for EI value variations. Results EI values were often outside the MREI ranges. EI value variations were identified, with significantly higher EI values recorded for female compared with male patients for all manufacturers (P ≤ 0.04), indicating higher detector exposures except for Philips direct digital radiography (DDR), where increased EI values indicated lower exposures (P=0.01). Higher EI values were also noted for the following variables: out of hours radiography (P ≤ 0.02), the absence of secondary radiation grids with Philips DDR chest X-rays (P = 0.03), younger patients with Siemens DDR chest X-rays (P < 0.0001) and higher kVp for Carestream computed radiography (CR) chest X-rays (P = 0.02). Significantly lower median EI values were demonstrated for Carestream CR chest X-rays when a metallic implant or prosthesis was present (P = 0.02). Discussion Non adherence to MREIs has been demonstrated. EI value discrepancies have been identified as being dependent on patient gender, time/day of exposure, grid usage, the presence of a metallic implants or prostheses, patient age and use of ≥ 100 kVp. The value of careful retrospective evaluation of EI databases has been highlighted.