Both optic nerve enhancement pattern and white matter lesion distribution as viewed on brain and orbital magnetic resonance imaging (MRI) can aid in differentiation and diagnosis of anterior ischemic optic neuropathy and optic neuritis in patients with an otherwise inconclusive clinical diagnosis, according to research results published in the British Journal of Ophthalmology.
Researchers conducted a retrospective case series review of patients who underwent MRI to determine the role of MRI in assessing optic nerve damage in anterior optic neuropathy or optic neuritis. The goal of the current study was to identify which MRI characteristics — if any — would aid in the differentiation between these diagnoses.
Patients included in the study had presented at an emergency department in Switzerland between 2010 and 2018 with new visual disturbances of an unknown etiology; these patients then underwent an emergency MRI of the brain or dedicated orbital sequences, or both. Included patients further had data available from a concurrent ophthalmic evaluation.
The total cohort included 50 patients (mean age, 53.17±15.87), 29 of whom were diagnosed with arteritic or nonarteritic anterior ischemic optic neuropathy (mean age, 62.47±9.99 years) and 22 of whom were diagnosed with optic neuritis (mean age, 41.34±14.05). The between-group age difference was statistically significant, and women were more likely to be affected by optic neuritis.
Time between symptom onset and MRI evaluation ranged from 5 hours to 30 days. MRI scoring involved 2 independent raters. Good agreement was found for all parameters (Cohen’s kappa, 0.998). Investigators used a stepwise model to automatically select predictive variables, resulting in a model that was dependent on white matter lesions, central bright spot, optic nerve enhancement, diffusion-weighted imaging (DWI) optic nerve, and optic nerve sheath.
A dominance analysis was also performed to evaluate the contribution of each of these different parameters to prediction. Researchers found that optic nerve enhancement played the largest role in disease classification (McFadden index, 0.35). Distribution was a similarly high contributor (McFadden index, 0.32), while central bright spot, DWI optic nerve, and optic nerve sheath had index values of 0.14, 0.07, and 0.07, respectively.
Using the 3 most important parameters — distribution, enhancement, and central bright spot — modelling demonstrated a classification accuracy of 0.92. When using only 2 parameters — distribution and enhancement — classification accuracy was 0.90.
The researchers also tested the diagnostic value of DWI sequences for discrimination between diseases, using only DWI restriction for modeling; prediction accuracy of this model was 0.68. Conversely, the use of all parameters excluding DWI created a model with the same classification accuracy as the full model that included all parameters (0.98).
Study limitations included the retrospective nature of the research and a lack of uniform MRI protocol for the orbits, as well as the recent availability of new DWI techniques to complement the MR sequence spectrum, which were not available at the time of the study.
“Our findings suggest that in patients with vision loss where the clinical diagnosis is equivocal and MRI is being acquired, neuroradiologists should focus on the imaging findings that aid in the diagnosis differentiation,” according to the researchers. “A prospective study of atypical clinical ophthalmological cases should be considered to validate these observations,” they concluded.
Petroulia VD, Brügger D, Hoepner R, et al. MRI signs helpful in the differentiation of patients with anterior ischaemic optic neuropathy and optic neuritis. Br J Ophthalmol. Published online July 19, 2021. doi:10.1136/bjophthalmol-2021-319537