Related Experiment Video
Updated: Jun 24, 2025
Author Spotlight: Integrating High-Resolution Intravital Imaging and MRI to Enhance Stereotactic Body Radiation Therapy Planning
Published on: April 12, 2024
Apparent Diffusion Coefficient (ADC) Differentiates Retinoblastoma from Coats Disease on MRI
Wendy X Zhang1, Joshua S Shimony2, Gregg T Lueder3
1University of Missouri School of Medicine (W.X.Z.), Columbia, Missouri, USA; Department of Ophthalmology and Visual Sciences, Washington University School of Medicine (W.X.Z., G.T.L., M.M.R.), St. Louis, Missouri, USA.
Purpose:
Coats' disease can be difficult to differentiate from retinoblastoma. While MR imaging of retinoblastoma and Coats' disease have been examined for differentiating features such as eye size, vitreous seeding, and shape of retinal detachment, there is a lack of data on apparent diffusion coefficient (ADC). ADC is a measure of the diffusion (of water molecules) within tissue, and is commonly clinically calculated using MRI.
Design:
Retrospective cross-sectional study.
Methods:
Patient or study population: Children < 18 diagnosed with Coats disease or Retinoblastoma between January 1, 2018 and January 8, 2022 who had MRI imaging that was reviewable.
Main Outcome Measure:
Apparent diffusion coefficient (ADC) of the intraocular lesion. Retrospective brain MRIs were obtained from records of 5 eyes of 5 Coats' patients and 29 eyes of 23 patients with retinoblastoma. All MRIs were obtained prior to treatment. The eyedropper tool in Epic's default viewer (Ambra DICOM) was used to measure the ADC of five to eight randomly sampled points within the eye lesions seen on MRI. Average ADC was calculated for each affected eye. Internal reliability was confirmed by re-measuring mean ADC for a random sample of patients masked to their diagnosis and prior measurements. T-test was used to determine if ADC values differ between groups.
Results:
The mean ADC for retinoblastoma patients (442 +/- 210 mm2/s) differed significantly from the mean for Coats' patients (1364 +/- 309 mm2/s), (P < .001). T-test between baseline and repeat measurements was not significantly different. Since ADC values can differ between different scanners and DW MRI pulse sequences, an ADC threshold may be difficult to generalize across institutes, in our data set a threshold of 900 mm2/s was useful in separating the two diagnoses with a high degree of accuracy.
Conclusions:
Clinical features of retinoblastoma and Coats' disease often resemble each other and can lead to misdiagnosis. Since ADCs are derived from diffusion-weighted MRI as an objective parameter, it has the potential to aid in establishing or confirming the diagnosis when retinoblastoma and/or Coats' disease are suspected.
