Comparison of 1.5 Tesla and 3.0 Tesla Magnetic Resonance Imaging in the Evaluation of Acute Demyelinating Optic Neuritis

Background: Optic neuritis (ON) is the most common optic neuropathy in young adults. MRI is reported to have a high sensitivity for ON. Higher signal strengths of MRI may enhance resolution and lead to better detection of ON. We sought to compare the sensitivity of 3.0 Tesla (T) MRI to that of 1.5 T MRI in detecting acute demyelinating ON. Methods: A retrospective chart review was performed on patients with a clinical diagnosis of optic neuritis at Mayo Clinic Health System from January 2010 to April 2020. Among 1,850 patients identified, 126 patients met the eligibility criteria. Exclusion criteria comprised questionable or alternative diagnosis, diagnosis of ON before the study period, eye examinations performed elsewhere, or absence of fat-saturated head and orbits MRIs performed locally within 30 days of symptom onset. Gadolinium contrast enhancement, T2 hyperintensity, and the radiologic diagnosis of ON were recorded by a neuro-radiologist who was masked to the clinical history and the magnet strength of the MRI. Results: Fifty-three patients (42.1%) had 3.0 T MRI, and 73 patients (57.9%) had 1.5 T MRI. Overall, 88.9% (112/126) of patients were determined to have a positive MRI for ON. The radiographic sensitivity for ON was higher in the 3.0 T group compared with the 1.5 T group (98.1% vs 82.2%, respectively [ P = 0.004]). The frequency of gadolinium enhancement was found to be greater in the 3 T group compared with the 1.5 T group (98.1% vs 76.7%, respectively [ P < 0.001]). T2 hyperintensity was also more often seen in the 3.0 T group compared with the 1.5 T group (88.7% vs 68.5%, respectively [ P = 0.01]). Conclusions: 3.0 T MRI is more sensitive than 1.5 T MRI in detecting ON. This finding suggests that 3.0 T MRI is a preferred imaging modality for the confirmation of ON.

Original Contribution Section Editors: Clare Fraser, MD Susan Mollan, MD Comparison of 1.5 Tesla and 3.0 Tesla Magnetic Resonance Imaging in the Evaluation of Acute Demyelinating Optic Neuritis Minjun Hur, MD, Ajay A. Madhavan, MD, David O. Hodge, MS, Laurence J. Eckel, MD, Sean J. Pittock, MD, Eoin P. Flanagan, MD, M. Tariq Bhatti, MD, John J. Chen, MD, PhD Background: Optic neuritis (ON) is the most common optic neuropathy in young adults. MRI is reported to have a high sensitivity for ON. Higher signal strengths of MRI may enhance resolution and lead to better detection of ON. We sought to compare the sensitivity of 3.0 Tesla (T) MRI to that of 1.5 T MRI in detecting acute demyelinating ON. Methods: A retrospective chart review was performed on patients with a clinical diagnosis of optic neuritis at Mayo Clinic Health System from January 2010 to April 2020. Among 1,850 patients identified, 126 patients met the eligibility criteria. Exclusion criteria comprised questionable or alternative diagnosis, diagnosis of ON before the study period, eye examinations performed elsewhere, or absence of fat-saturated head and orbits MRIs performed locally within 30 days of symptom onset. Gadolinium contrast enhancement, T2 hyperintensity, and the radiologic diagnosis of ON were recorded by a neuro-radiologist who was masked to the clinical history and the magnet strength of the MRI. Results: Fifty-three patients (42.1%) had 3.0 T MRI, and 73 patients (57.9%) had 1.5 T MRI. Overall, 88.9% (112/126) of patients were determined to have a positive MRI for ON. The radiographic sensitivity for ON was higher in the 3.0 T group compared with the 1.5 T group (98.1% vs 82.2%, respectively [P = 0.004]). The frequency of gadolinium enhancement was found to be greater in the 3 T group compared with the 1.5 T group (98.1% vs 76.7%, respectively [P , 0.001]). T2 hyperintensity was also more often seen in the 3.0 T group compared with the 1.5 T group (88.7% vs 68.5%, respectively [P = 0.01]). Conclusions: 3.0 T MRI is more sensitive than 1.5 T MRI in detecting ON. This finding suggests that 3.0 T MRI is a preferred imaging modality for the confirmation of ON. Journal of Neuro-Ophthalmology 2022;42:297–302 doi: 10.1097/WNO.0000000000001559 © 2022 by North American Neuro-Ophthalmology Society A cute demyelinating optic neuritis (ON) is defined as an inflammation of the optic nerve (1). Contrast enhancement and hyperintense T2 signal abnormalities of the optic nerve are usually seen during ON (2). Previous studies have reported high but variable sensitivities of MRI for ON (68.3%–96.7%) (3–10). Currently, 3.0 Tesla (T) and 1.5 T are the most widely used magnetic field strengths of MRI. Because the higher field strength improves the signal-to-noise ratio and increases image quality (11,12), 3.0 T may be superior to 1.5 T, especially for gadolinium enhancing lesions, which has been reported for cortical lesions (13), but a comparison has not been made for detecting ON. We sought to compare the sensitivity of 3.0 T MRI with that of 1.5 T MRI in detecting acute ON and evaluate for other factors that may affect the MRI sensitivity in ON. METHODS Patient Selection and Characteristics Departments of Ophthalmology (MH, MTB, JJC) and Radiology (AAM, LJE), Mayo Clinic, Rochester, Minnesota; Division of Biostatistics (DOH), Mayo Clinic and Mayo Foundation, Rochester, Minnesota; and Department of Neurology (SJP, EPF, MTB, JJC), Mayo Clinic, Rochester, Minnesota. The authors report no conflicts of interest. Address correspondence to John J. Chen, MD, PhD, Department of Ophthalmology, Mayo Clinic, 200 First Street SW, Rochester, MN 55905; E-mail: [e-mail] Hur et al: J Neuro-Ophthalmol 2022; 42: 297-302 The study design was approved by the Mayo Clinic Institutional Review Board (IRB) and adhered to the tenets of the Declaration of Helsinki. A retrospective chart review was performed to identify patients with diagnoses of acute ON at the Mayo Clinic between 2010 and 2020, according to the International Classification of Diseases (ICD) 9 codes: 377.3 (optic neuritis), 377.30 (optic neuritis, unspecified), 377.32 (retrobulbar neuritis), and 377.39 297 Copyright © North American Neuro-Ophthalmology Society. Unauthorized reproduction of this article is prohibited. Original Contribution (other optic neuritis) and ICD 10 codes: H46.10 (retrobulbar optic neuritis, unspecified eye), H46.11 (retrobulbar optic neuritis, right eye), H46.12 (retrobulbar optic neuritis, left eye), H46.13 (retrobulbar optic neuritis, bilateral), H46.8 (other optic neuritis), and H46.9 (unspecified optic neuritis). The study period was selected to balance the number of 1.5 T and 3.0 T because there was a paucity of 3.0 T MRIs before 2010. Inclusion criteria were age older than 18 years, residences in Minnesota or neighboring states (Wisconsin, Iowa, North Dakota, and South Dakota) to minimize the referral bias, and having had 1.5 T or 3.0 T MRIs within 30 days of ON symptoms. A total of 1850 patients were identified. Patients confirmed to have acute demyelinating optic neuritis were included. Patients with optic neuropathy from systemic disorders, such as giant cell arteritis, sarcoidosis, syphilis, or other inflammatory or infectious disease, were excluded. MRIs without contrast or fat saturation of the orbits, MRIs with insufficient image quality, and MRIs performed beyond 30 days after the onset of symptoms were excluded. Only 2 MRIs (both 1.5 T) were considered to have insufficient image quality. One scan had significant motion artifacts, and the other had skull base artifacts due to braces. Patients were also excluded if they were not seen at a Mayo Clinic Health System Ophthalmology Department within 30 days of symptom onset. ON was diagnosed with patient history and clinical findings, such as decreased visual acuity, pain with eye movement, visual field defect, decreased color vision, and/or optic nerve examinations (14). Cases with questionable clinical diagnosis of ON were reviewed by a neuro-ophthalmologist (J.J.C.), and patients without enough clinical information to confirm a diagnosis of ON were excluded. Patients without a research authorization were also excluded. Ultimately, 1,724 patients were excluded, and the study consisted of 126 patients (Fig. 1). Most of the patients were excluded because of the incorrect or historical diagnosis codes assigned to the patient visits. The following variables were collected: age at symptom onset, sex, race, etiology of the ON, presence/absence of eye pain, presence/absence of optic disc edema, the duration of the symptoms before MRI, corticosteroid use before MRI, best-corrected visual acuity (BCVA), and prior diagnosis of ON. In cases of bilateral ON, the worse BCVA of the 2 eyes was recorded. Antibodies targeting aquaporin-4 (AQP4) and Myelin Oligodendrocyte Glycoprotein (MOG) were detected through a live cell-based assay using transfected live HEK293 cells (14). MRI Analysis All MRI of the brain and orbits with contrast at the Mayo Clinic Health System used a 1.5 T or 3.0 T magnet and had 2 manufacturers (Siemens Healthcare, Erlangen, Germany and GE Healthcare, Boston, MA, Philips), which were evenly distributed between 1.5 T and 3.0 T. Time to 298 FIG. 1. Patient exclusion tree. Among the 1,850 patients with the diagnosis of acute ON as indicated by ICD-9 and 10 codes, 1,101 patients were excluded because of alternative diagnoses on review of their medical records (e.g., giant cell arteritis, nonarteritic anterior ischemic optic neuropathy, functional vision loss, and optic atrophy) or because of questionable diagnosis. Patients who had on before 2010 were also excluded. Not having had an eye examination within 30 days of symptom onset was considered incomplete medical records. Patients whose MRIs were not performed within the Mayo Clinic Health System within 30 days of symptom onset or whose MRI protocol was without fat saturation and/or intravenous contrast were excluded. MRIs with significant motion artifacts or 7 Tesla MRI were also excluded from this study. ON, optic neuritis. postgadolinium imaging after the contrast injection did not differ among scans. Orbital sequences included precontrast and postcontrast coronal T1 with fat saturation, coronal T2 with fat saturation, axial T1 with fat saturation, axial T2 with fat saturation, and cube reformat images. Conventional T2 fat suppressed images were used in our MR imaging protocols at both 1.5 T and 3.0 T. Short tau inversion recovery (STIR) was not available for review because it is not a part of our standard orbit MRI protocol. Either a 64 or 32 channel head coil was used. A neuroradiologist (A.A.M.), who was masked to the clinical findings, the official imaging reports, and the field of strength of MRI, reviewed the axial and coronal T1W precontrast and postcontrast images of the orbits and axial and coronal T2W images of the orbits. The neuroradiologist documented the presence of gadolinium enhancement and T2 hyperintensity of the optic nerve, and the radiological diagnosis of ON was indicated (positive or negative). Statistical Analysis All statistical analysis was performed using SAS software (Cary, North Carlina). Overall comparisons between the Hur et al: J Neuro-Ophthalmol 2022; 42: 297-302 Copyright © North American Neuro-Ophthalmology Society. Unauthorized reproduction of this article is prohibited. Original Contribution groups were completed with Fisher exact tests for categorical variables, and continuous variables were compared with 2 sample t tests. Logistic regression models were used to assess differences in the positivity outcomes between the field strength groups adjusting for other potential confounders. P values less than 0.05 were considered statistically significant. RESULTS Demographics Participant characteristics are presented in Table 1. Of 126 patients who met the inclusion criteria, 69.8% (88/126) were women and 94.4% (119/126) were Caucasian. The mean age at diagnosis was 40.7 years (±12.9 years). Eleven patients (8.7%) had bilateral ON, and 98 patients (77.8%) reported pain with eye movement. Optic disc edema was present in 41.9% of patients (52/124); the presence or absence of optic disc edema was not documented in 2 patients. A mean of 11.2 days (±7.5 days) elapsed between the symptom onset and the MRI. Nine patients (7.1%) had a previous diagnosis of ON in the affected eye, and 14 patients (11.1%) had received intravenous methylpredniso- lone before the MRI. The median BCVA at nadir was 0.50 logMAR (Snellen visual acuity: 20/63, range: 20/20—no light perception). Idiopathic ON was the most common etiology (42.9%, 54/126) followed by multiple sclerosis (MS), which accounted for 50 (39.7%). Of 126 patients, 53 patients (42.1%) had 3.0 T MRI and 73 patients (57.9%) had 1.5 T MRI. The 2 groups were balanced in age, sex, race, BCVA at nadir, the percentage of patients with eye pain, previous diagnosis of ON, or steroid treatment before the MRI. On average, the 3.0 T group received MRI later (12.6 days ± 6.9 days) than the 1.5 T group (10.2 days ± 7.9 days, P = 0.03). In addition, there was a lower prevalence of 3.0 T performed within 4 days of symptom onset (7.5%, 4/53), compared with 1.5 Tesla (20.5%, 15/73; P = 0.048). MRI Sensitivity in Optic Neuritis Overall sensitivity of MRI for diagnosing ON was 88.9% (112/126). Gadolinium enhancement was present in 85.7% (108/126), T2 hyperintensity in 77.0% (97/126), and 75.4% (95/126) had both gadolinium enhancement and T2 hyperintensity. There was a higher rate of positive radiologic findings in the 3.0 T group (98.1%, 52/53) compared with the 1.5 T group (82.2%, 60/73; P = 0.004), TABLE 1. Comparison of patient characteristics for patients who had 1.5 Tesla MRI and those who had 3.0 Tesla MRI Characteristics Age (mean ± SD) Sex (% female) Race (% white) Bilateral ON (n) Pain with eye movement Disc edema (n) Duration of symptoms before MRI (mean ± SD) Previous diagnosis of ON Previous steroids treatment Best corrected visual acuity at nadir (median) Etiology of optic neuritis (n) Idiopathic Multiple sclerosis MOGAD AQP4+ NMOSD Seronegative NMOSD Unspecified demyelinating disease All patients 1.5 T group (n = 73) 3.0 T group (n = 53) P Difference 95% CI 40.7 years ± 12.9 41.9 years ± 13.3 69.8% (88) 74.0% (54) 94.4% (119) 91.8% (67) 8.7% (11) 12.3% (9) 77.8% (98) 75.3% (55) 41.9% (52*) 40.8% (29†) 11.2 days ± 7.5 10.2 days ± 7.9 39.0 years ± 12.2 64.2% (34) 98.1% (52) 3.8% (2) 81.1% (43) 43.4% (23) 12.6 days ± 6.9 0.19 2.9 (21.6, 7.4) 0.25 0.24 0.12 0.52 0.85 0.03 7.1% (9) 11.1% (14) 0.50 logMAR 9.6% (7) 8.2% (6) 0.5 logMAR 3.8% (2) 15.1% (8) 0.4 logMAR 0.30 5.8 (20.2, 11.8) 0.26 26.9 (215, 1.3) 0.34 0.1 (20.2, 0.4) 42.9% (54) 39.7% (50) 12.7% (16) 2.4% (3) 0.8% (1) 1.6% (2) 42.5% (31) 38.4% (28) 15.1% (11) 1.4% (1) 1.4% (1) 1.4% (1) 43.4% (23) 41.5% (22) 9.4% (5) 3.8% (2) 0.0% (0) 1.9% (1) 1.00 0.85 0.42 0.57 1.00 1.00 9.8 (21.8, 21.4) 26.3 (211.5, 21.2) 8.6 (1.2, 15) 25.8 (216, 4.4) 22.6 (214.9, 9.8) 22.4 (25, 0.2) 20.9 (213.3, 11.5) 23.2 (215.4, 9.1) 5.6 (22.4, 13.7) 22.4 (26.5, 1.7) 1.4 (20.5, 3.3) 20.5 (23.7, 2.7) *Of 124 patients with documentation of disc edema. † Of 71 patients, 2 patients in the 1.5 T group did not have a documentation of disc edema. AQP4, aquaporin 4; MOGAD, myelin oligodendrocyte glycoprotein antibody-associated disorder; NMOSD, neuromyelitis optica spectrum disorder; ON, optic neuritis. Hur et al: J Neuro-Ophthalmol 2022; 42: 297-302 299 Copyright © North American Neuro-Ophthalmology Society. Unauthorized reproduction of this article is prohibited. Original Contribution shown in Figure 2. The difference remained statistically significant when adjusted for time to MRI (P = 0.02) and for previous steroid use (P = 0.02). Gadolinium enhancement was found in 98.1% (52/53) in the 3.0 T group compared with 76.7% (56/73) in the 1.5 T group (P , 0.001), which remained significantly different when adjusted for time to MRI (P = 0.008) and for previous steroid use (P = 0.008). T2 hyperintensity was seen in 88.7% (47/53) in the 3.0 T group compared with 68.5% (50/73) in the 1.5 T group (P = 0.01), which remained significantly different when adjusted for time to MRI (P = 0.02) and for previous steroid use (P = 0.01). Clinical Factors That Correlate With MRI Findings Other clinical variables that were evaluated for an impact on MRI sensitivity are presented in Table 2. There was no statistically significant difference in age at symptom onset (P = 0.99), sex (P = 0.55), eye pain (P = 0.30), previous steroid treatment (P = 0.65), or prior diagnosis of ON (P = 0.26) between those with positive and negative MRI findings. Patients with positive MRI for ON were more likely to have presented with optic disc edema (45.5%, 50/110) compared with patients with negative MRI (14.3%, 2/14; P = 0.04). When stratified by the magnetic strength of the MRI, the difference was statistically significant only for the 1.5 T group. Optic disc edema was present in 48.3% (28/58) of patients with a positive 1.5 T MRI, compared with 7.7% (1/13) of patients with negative 1.5 T MRI (P = 0.01). The difference was not statistically significant in the 3.0 T group (P = 0.43). Patients with negative MRI findings were more likely to have idiopathic ON (78.6%, 11/14) compared with patients FIG. 2. Radiologic diagnosis of ON, 1.5 Tesla MRI vs 3.0 Tesla MRI. The rates of positive radiologic findings, gadolinium enhancement, and T2 hyperintensity were compared between the 1.5 T group and the 3.0 T group. ON, optic neuritis. 300 with a positive MRI (38.4%, 43/112; P = 0.008). In the 1.5 T group, 35.0% (21/60) of patients with positive MRI had idiopathic ON, compared with 76.9% (10/13) of patients with negative MRI (P = 0.01). Again, the difference was not statistically significant in the 3.0 T group (P = 0.43). The proportions of 3.0 T and 1.5 T were well balanced among the 3 most common causes of ON (idiopathic, MS, and myelin oligodendrocyte glycoprotein antibodyassociated disorder [MOGAD]). In the idiopathic ON group, 42.6% (23/54) of patients had 3.0 T while the other 57.4% (31/54) of patients had 1.5 T (P = 0.28). The radiographic sensitivity was 95.7% (22/23) in the 3.0 T group in contrast to 67.7% (21/31) in the 1.5 T group (P = 0.02). In the MS group, 44.0% (22/50) of patients had 3.0 T while 56.0% (28/50) of patients had 1.5 T (P = 0.40). There was no statistically significant difference in the radiographic sensitivity between the 3.0 T group (100%, 22/22) and the 1.5 T group (89.3%, 25/28; P = 0.25) in patients with MS. In the MOGAD group, 31.3% (5/16) of patients had 3.0 T while 68.8% (11/16) of patients had 1.5 T (P = 0.13). All 16 patients with MOGAD had positive radiologic interpretation, so there was no statistically significant difference in the radiographic sensitivity between the 2 magnet strengths. The average time to MRI was 11.2 days (±7.2 days) in the positive MRI group compared with 11.4 days (±10.5 days) in the negative MRI group, which was not statistically significant (P = 0.57). However, there was a lower radiographic sensitivity of MRI performed within 4 days of onset of symptoms (Fig. 3). MRIs performed within 2 days of symptom onset had a sensitivity of 40.0% (2/5), compared with MRIs performed after 2 days (90.9%, 110/121; P = 0.01). MRIs performed within 3 days had a sensitivity of 55.6% (5/9) compared with MRIs performed later than 3 days (91.5%, 107/117; P = 0.009), and MRIs performed within 4 days had a sensitivity of 73.7% (14/19), compared with MRIs performed later than 4 days (91.6%, 98/107; P = 0.04). There was no statistically significant difference in sensitivity between MRIs performed within 5 days (80.7%, 25/31) and MRIs performed later than 5 days (91.6%, 87/95; P = 0.11). There was no significant difference for any of the later time points. Because there was a higher prevalence of 1.5 T performed within 4 days of symptom onset, a further analysis was performed to include only MRIs performed 5 or more days after symptom onset. There was still a statistically significant difference in the radiographic sensitivity between the 3.0 T group (100.0%, 45/45) and the 1.5 T group (84.0%, 42/50; P = 0.006). CONCLUSIONS To the best of our knowledge, this is the first study to compare the sensitivity of 3.0 T vs 1.5 T MRI in ON, which found that 3.0 T was more sensitive than 1.5 T for Hur et al: J Neuro-Ophthalmol 2022; 42: 297-302 Copyright © North American Neuro-Ophthalmology Society. Unauthorized reproduction of this article is prohibited. Original Contribution TABLE 2. Other variables that could affect MRI sensitivity Age (mean) Sex (female) Eye pain Presence of disc edema 1.5 T 3.0 T Etiology of ON (% of MS) Etiology of ON (% of idiopathic ON) 1.5 T 3.0 T Previous steroids treatment Prior diagnosis of ON Days to MRI from symptom onset Best-corrected visual acuity at nadir (median) Positive Negative P 40.7 years ± 12.9 68.8% 79.5% 45.5% 48.3% 42.3% 42.0% 38.4% 35.0% 42.3% 10.7% 6.3% 11.2 days ± 7.2 0.3 logMAR 40.9 years ± 12.9 78.6% 64.3% 14.3% 7.7% 100.0% 21.4% 78.6% 76.9% 100.0% 14.3% 14.3% 11.4 days ± 10.5 0.3 logMAR 0.99 0.55 0.30 0.04 0.01 0.43 0.16 0.008 0.01 0.43 0.65 0.26 0.57 0.74 MS, multiple sclerosis; ON, optic neuritis. detecting ON. In addition, MRI sensitivity was lower when performed within the first 4 days of the onset of symptoms. This suggests that there may be a delay of a few days before the optic nerve develops radiographic signs of ON. By contrast, a previous study reported that the median time to MRI was shorter in patients with gadolinium enhancement of the optic nerve (5 days), compared with patients without enhancement (8 days), but this difference was not statistically significant (P = 0.06) (3). In the 1.5 T group, patients with a positive MRI were more likely to have optic disc edema and less likely to have idiopathic ON. We can postulate that there may be more prominent intraorbital signal on MRI in ON that causes clinical optic disc edema, although a recent study found no statistically significant difference in sensitivity depending on the presence of optic disc edema (4). We are not aware of any previous studies that have investigated the difference in MRI sensitivity depending on the underlying cause of ON. It is plausible that the optic nerve enhancement or T2 hyperintensity might be more prominent in demyelinating diseases. Because MOGAD is known for its prominent optic nerve enhancement on MRI, it was not surprising to note that all MOGAD-associated ON cases were noted to have positive radiological findings in our study. The proportions of 1.5 T and 3.0 T were well balanced in this group, so this did not contribute to the increased sensitivity in the 3.0 T MRI. There was also no difference in the proportions of 1.5 T and 3.0 T for MS-associated ON or idiopathic ON. Therefore, having a demyelinating disease unlikely contributed to the difference between the MRI magnetic strengths. Obtaining dedicated MRI orbits is now recommended in clinically suspected ON because the presence of enhancement can confirm ON (15). This is important because up to 60% of patients referred to a neuro-ophthalmologist do not have ON (16). In addition, the pattern of enhancement can Hur et al: J Neuro-Ophthalmol 2022; 42: 297-302 aid in the differentiation of various demyelinating conditions (15). For example, MOGAD and neuromyelitis optica spectrum disorder tend to cause longer segments of enhancement and MOGAD optic neuritis demonstrates perineural enhancement in 50% of cases (17). There were some limitations of this study. The retrospective nature of this study could have allowed for selection bias, which we tried to limit by only including patients who reside in Minnesota and neighboring states. Furthermore, excluding MRIs without dedicated orbital cuts or fat saturation could have affected our radiographic sensitivities. The expertise level of the interpreting neuro-radiologist also could have influenced the outcomes, although consistency was maintained with only 1 neuroradiologist reviewing the scan. In addition, the differences in the vendor and the age of the FIG. 3. Increased sensitivity with the increasing number of days to MRI. The rate of positive radiologic findings for MRIs increased as the symptom duration before MRI increased, up to 5 days. The sensitivity of MRIs performed within 2 days of symptom onset was compared with that of the MRIs performed after 2 days. The same was performed for 3, 4, and 5 days. 301 Copyright © North American Neuro-Ophthalmology Society. Unauthorized reproduction of this article is prohibited. Original Contribution MRI magnets are potential confounders. We attempted to minimize these differences by only including MRIs performed within the Mayo Clinic Health System. In fact, some variability may help the generalizability of our results. Finally, although our main variables underwent logistic regression models to adjust for time to MRI and previous steroid use, the comparison between the MRI positive group and the negative group was not performed by a multivariate analysis. Therefore, a larger study with multivariate analysis may be needed in the future to confirm our findings. Our study suggests that 3.0 T MRI is better than 1.5 T MRI in detecting optic nerve enhancement or T2 hyperintensity in ON. In addition, there may be a decrease in sensitivity for MRIs performed within 4 days of symptom onset of ON. 6. 7. 8. 9. 10. 11. STATEMENT OF AUTHORSHIP Category 1: a. Conception and design: J. J. Chen and L. J. Eckel; b. Acquisition of data: M. Hur and A. A. Madhavan; c. Analysis and interpretation of data: J. J. Chen, D. O. Hodge, M. Hur, A. A. Madhavan, and L. J. Eckel. Category 2: a. Drafting the manuscript: J. J. Chen, M. Hur, and A. A. Madhavan; b. Revising it for intellectual content: J. J. Chen, M. Hur, A. A. Madhavan, S. J. Pittock, E. P. Flanagan, M. Tariq Bhatti, D. O. Hodge, and L. J. Eckel. Category 3: a. Final approval of the completed manuscript: J. J. Chen and M. Hur. REFERENCES 1. Voss E, Raab P, Trebst C, Stangel M. Clinical approach to optic neuritis: pitfalls, red flags and differential diagnosis. Ther Adv Neurol Disord. 2011;4:123–134. 2. Gala F. Magnetic resonance imaging of optic nerve. Indian J Radiol Imaging. 2015;25:421–438. 3. Berg S, Kaschka I, Utz KS, Huhn K, Lämmer A, Lämmer R, Waschbisch A, Kloska S, Lee DH, Doerfler A, Linker RA. Baseline magnetic resonance imaging of the optic nerve provides limited predictive information on short-term recovery after acute optic neuritis. PLoS One. 2015;10:e0113961. 4. Bursztyn LLCD, De Lott LB, Petrou M, Cornblath WT. Sensitivity of orbital magnetic resonance imaging in acute demyelinating optic neuritis. Can J Ophthalmol. 2019;54:242–246. 5. Hickman SJ, Toosy AT, Miszkiel KA, Jones SJ, Altmann DR, MacManus DG, Plant GT, Thompson AJ, Miller DH. Visual recovery following acute optic neuritis–a clinical, electrophysi- 302 12. 13. 14. 15. 16. 17. ological and magnetic resonance imaging study. J Neurol. 2004;251:996–1005. Kupersmith MJ, Alban T, Zeiffer B, Lefton D. Contrastenhanced MRI in acute optic neuritis: relationship to visual performance. Brain. 2002;125:812–822. McKinney AM, Lohman BD, Sarikaya B, Benson M, Lee MS, Benson MT. Accuracy of routine fat-suppressed FLAIR and diffusion-weighted images in detecting clinically evident acute optic neuritis. Acta Radiol. 2013;54:455–461. Rizzo JF III, Andreoli CM, Rabinov JD. Use of magnetic resonance imaging to differentiate optic neuritis and nonarteritic anterior ischemic optic neuropathy. Ophthalmology. 2002;109:1679–1684. Soelberg K, Skejoe HPB, Grauslund J, Smith TJ, Lillevang ST, Jarius S, Wildemann B, Paul F, Asgari N. Magnetic resonance imaging findings at the first episode of acute optic neuritis. Mult Scler Relat Disord. 2018;20:30–36. Wan H, He H, Zhang F, Sha Y, Tian G. Diffusion-weighted imaging helps differentiate multiple sclerosis and neuromyelitis optica-related acute optic neuritis. J Magn Reson Imaging. 2017;45:1780–1785. Kilsdonk ID, de Graaf WL, Barkhof F, Wattjes MP. Inflammation high-field magnetic resonance imaging. Neuroimaging Clin N Am. 2012;22:135–157, ix. Wattjes MP, Barkhof F. High field MRI in the diagnosis of multiple sclerosis: high field-high yield? Neuroradiology. 2009;51:279–292. Sicotte NL, Voskuhl RR, Bouvier S, Klutch R, Cohen MS, Mazziotta JC. Comparison of multiple sclerosis lesions at 1.5 and 3.0 Tesla. Invest Radiol. 2003;38:423–427. Hassan MB, Stern C, Flanagan EP, Pittock SJ, Kunchok A, Foster RC, Jitprapaikulsan J, Hodge DO, Bhatti MT, Chen JJ. Population-based incidence of optic neuritis in the era of aquaporin-4 and Myelin Oligodendrocyte Glycoprotein antibodies. Am J Ophthalmol. 2020;220:110–114. North American Neuro-Ophthalmology Society Practice Support Committee. North American Neuro-Ophthalmology Society (NANOS) Statement on the Justification for MRI Orbits With and Without Contrast in Addition to MRI Brain With and Without Contrast in Optic Neuritis. North American Neuro-Ophthalmology Society, 2021. Available at: https://www.nanosweb.org/files/ Committees/Practice%20Support/NANOS_Justification_ for_MRI_Orbits_in_Optic_Neuritis_Statement_3_30_21_ (15).pdf. Accessed November 13, 2021. Stunkel L, Kung NH, Wilson B, McClelland CM, Van Stavern GP. Incidence and causes of overdiagnosis of optic neuritis. JAMA Ophthalmol. 2018;136:76–81. Chen JJ, Pittock SJ, Flanagan EP, Lennon VA, Bhatti MT. Optic neuritis in the era of biomarkers. Surv Ophthalmol. 2020;65:12–17. Hur et al: J Neuro-Ophthalmol 2022; 42: 297-302 Copyright © North American Neuro-Ophthalmology Society. Unauthorized reproduction of this article is prohibited.