Retrospective review of CT brain image quality, diagnostic adequacy and radiation dose in a paediatric population imaged at a non-paediatric tertiary hospital.

Abstract

Minimising radiation exposure in paediatric imaging examinations whilst maintaining acceptable diagnostic quality continues to present a challenge. The aims of this study were to assess institutional compliance of paediatric CT brain (CTB) examinations performed in an adult hospital with ARPANSA radiation dose recommendations and to compare qualitative CTB diagnostic acceptability with objective imaging parameters and radiation dose. A retrospective review of 115 consecutive paediatric CTB examinations was undertaken at an adult tertiary referral centre in Australia over a 2-year period. Dose length product (DLP) was compared with the ARPANSA standards. CTB image quality was subjectively classified by two neuroradiologists independently, with discordant results resolved by consensus. Objective assessment of image quality included measurements of signal-to-noise (SNR) and contrast-to-noise ratios (CNR) of grey and white matter. All patient scans complied with ARPANSA DLP recommendations; however, 10 out of 115 scans were classified as being of diagnostically suboptimal image quality. These scans had significantly lower mean DLP values compared with diagnostically adequate examinations (105.1 vs 379.2 mGy.cm; P < 0.0001). CTB scans of adequate diagnostic quality, when compared to suboptimal scans, had significantly higher CNR (1.8 vs 1.1; P < 0.0001) and SNR in grey (7.1 vs 4.6; P < 0.0001) and white matter (5.6 vs 3.8; P < 0.0001). All CTB examinations in this series complied with the ARPANSA DLP recommendations; however, 9% were of suboptimal diagnostic image quality. While it is important to minimize unnecessary radiation exposure, our results suggest that excessively low DLP values can lead to suboptimal diagnostic image quality.