Diagnostic image quality of gadolinium-enhanced T1-weighted MRI with and without fat saturation in children with retinoblastoma

Abstract

Background: Gadolinium-enhanced T1-weighted MRI without fat saturation has been recommended for assessment of retinoblastoma.

Objective: The purpose of this study was to compare diagnostic image quality without and with fat saturation following gadolinium administration.

Materials and methods: High-resolution gadolinium-enhanced T1-weighted sequences with and without fat saturation performed in children with subsequently histopathologically confirmed retinoblastoma were included. Image analysis (image quality [1 = poor, 2 = moderate, 3 = good], anatomical detail depiction, tumour extension) was performed by two neuroradiologists in consensus. Enhancement was scored and measured. Signal- and contrast-to-noise ratios were calculated. Image-assessed tumour invasiveness was compared to histopathological findings. Paired sample t-test was used for statistical analysis.

Results: Thirty-six children (mean age, 19.0 ± 16.8 [SD] months) were included. Image quality and anatomical detail depiction were significantly better without fat saturation (P < 0.001). Tumour enhancement was rated higher with fat saturation (P < 0.001). Fat saturation improved detection of (post-)laminar optic nerve infiltration. Detection of choroidal invasion was improved without fat saturation. Combining both sequences was best in the assessment of tumour extension (sensitivity/specificity for (post-)laminar optic nerve infiltration, 75.0%/100.0%, and for choroidal invasion, 87.5%/85.7%).

Conclusion: Combined T1-weighted spin-echo imaging with and without fat saturation improved the image quality for assessment of invasiveness of retinoblastoma.