Correlation between quantified magnetic resonance imaging (MRI) findings and "MRI score" in Graves' ophthalmopathy.

Abstract

Purpose: To analyze the correlation between a magnetic resonance imaging (MRI) score, we proposed in a previous report, and conventional quantitatively evaluated MRI findings, and to evaluate the clinical utility of the MRI score in patients with Graves' ophthalmopathy (GO).

Methods: Using ImageJ, we retrospectively re-analyzed magnetic resonance images of 25 patients with GO who were evaluated in the previous study. The signal intensity and area measured by ImageJ were normalized for each affected extraocular muscle (EOM) in each patient (normalized quantified inflammation and enlargement). Furthermore, the ratio of the value for the affected EOM to the normal values was calculated in each patient (quantified inflammation and enlargement compared to normal values). We analyzed the correlations between MRI scores and quantitative MRI finding before and after 6 g of intravenous methylprednisolone (IVMP).

Results: The correlation coefficient between the normalized MRI score and normalized quantified inflammation and enlargement was 0.59 before IVMP and 0.69 after IVMP. The correlation coefficient between the MRI score and quantified inflammation and enlargement compared to normal values was 0.58 before IVMP and 0.73 after IVMP. The correlation coefficients for all items were statistically significant (P-value < .05).

Conclusion: This study found a high correlation between the MRI score, which is our proposed semiquantitative method for assessing EOM swelling and inflammation, and the results of conventional quantitative evaluation methods. This suggests that the MRI score, which can be more readily used in clinical practice, may be as reliable as quantitative evaluation methods.