Brain, Vol. 125, No. 4, 812-822,
April 2002
© 2002 Guarantors of Brain
Contrast-enhanced MRI in acute optic neuritis: relationship to visual performance
1 INN at Beth Israel Medical Center and 2 New York Eye and Ear Infirmary, New York, NY, USA
Correspondence to: Mark J. Kupersmith, MD, INN at Beth Israel North 170 East End Avenue, New York, NY 10128, USA E-mail: mkuper{at}bethisraelny.org
Received July 19, 2001. Revised November 21, 2001. Accepted November 30, 2001.
 |
Summary |
|---|
 Top Summary IntroductionPatients and methodsResultsDiscussionReferences |
|---|
The location and extent of an abnormal signal on MRI of the optic nerve affected by optic neuritis are said to correlate with the severity of initial visual loss and recovery. We used gadolinium-enhanced fat-suppressed MRI to show abnormal enhancement of the optic nerve to determine the sensitivity of this modality in acute optic neuritis and whether the abnormal enhancement correlates with presenting visual deficits or recovery. A total of 107 patients, 93 with follow-up (68 steroid treated), were included; 101 patients had enhancement of the affected optic nerve and no unaffected nerve enhanced. The baseline visual performance was similar between nerves with and without enhancement. Optic nerves with enhancement in the optic canal had poorer colour vision (P = 0.04) and nerves with all segments involved had worse threshold perimetry (P = 0.001) and colour vision (P = 0.008). Nerves with enhancement >10 mm had worse threshold perimetry (P = 0.004), while nerves with enhancing segments >17 mm had poorer baseline visual acuity (P = 0.02), threshold perimetry (P = 0.009) and colour vision (P = 0.01). For all parameters of vision, recovery was similar regardless of location or length of abnormal enhancement. Abnormal contrast enhancement of the optic nerve is a sensitive (94%) finding in acute optic neuritis and is absent in unaffected or previously affected optic nerves. Although lesions involving the canal or longer segments of optic nerve have worse starting vision, the location and length of enhancement are not predictive of recovery.
Keywords: optic neuritis; gadolinium lipid-suppressed MRI; visual performance
Abbreviations: ANOVA= analysis of variance; CV = colour vision; dB = decibel; MD = mean deviation; STIR = short inversion recovery; VA = visual acuity
|
Introduction |
|---|
|
TopSummaryIntroductionPatients and methodsResultsDiscussionReferences |
|---|
MRI is used routinely in the evaluation of patients when optic neuritis is the first demyelinating disease episode (Beck et al., 1993
). Typically, the goal is to identify whether there is evidence of prior demyelinating episodes in the brain. The presence of three or more bright spots in brain white matter on T2 -weighted or fluid attenuation inversion recovery images is associated with the early development of definite multiple sclerosis, and is often used as a criterion for initiating treatment with high dose corticosteroids (Beck et al., 1993) and ß-interferon (Jacobs et al., 2000). In most cases, if rigorous clinical criteria are applied (Optic Neuritis Study Group, 1991), MRI is not required to diagnose and distinguish optic neuritis from other common optic neuropathies. However, on occasion, patients with the disorder non-arteritic anterior ischaemic optic neuropathy or acute compressive neuropathy due to a pituitary tumour or cerebral aneurysm or posterior scleritis may not have the typical features that distinguish these disorders from optic neuritis. Also, 
8% of patients with optic neuritis will not have the expected pain (Optic Neuritis Study Group, 1991). In these situations, MRI should prevent misdiagnosis.
Several small series have demonstrated that following intravenous gadopentetate dimeglumine (gadolinium) administration, there is abnormal enhancement of the optic nerve affected with optic neuritis (Tien et al., 1991; Guy et al., 1992) or vasculitis of the optic nerve (Sklar et al., 1996), but not with non-arteritic anterior ischaemic optic neuropathy, with fat suppression MRI. In this study, we investigated the sensitivity of gadolinium-enhanced MRI in delineating the affected segment of nerve in acute optic neuritis. We also tried to determine whether an unaffected or previously affected optic nerve would enhance, which, if so, would suggest that MRI might not be as useful for distinguishing newly affected optic nerves.
In addition, other authors have suggested that the length and location of an abnormal signal in the optic nerve affected by optic neuritis on MRI without contrast are associated with the degree of visual loss and might predict visual outcome (Miller et al., 1988; Dunker and Wiegand, 1996). The presence of an abnormal signal in the canalicular portion or in a long segment of optic nerve on MRI appears to be associated with severe visual loss and a poor prognosis for visual recovery. We tried to determine whether the length or location of abnormal enhancement in optic nerve could be related to the baseline visual loss, rate of recovery or degree of visual recovery.
|
Patients and methods |
|---|
|
TopSummaryIntroductionPatients and methodsResultsDiscussionReferences |
|---|
A case series review was performed on the records and MRIs of patients who had acute optic neuritis (Optic Neuritis Study Group, 1991
) with clinical evaluation and MRI performed within 20 days of the visual loss, referred to the Institute of Neurology and Neurosurgery and New York Eye and Ear Infirmary Service during the time period of 1996–2000. Patients were excluded if the affected eye had a prior optic neuritis and vision did not return to 20/20, mean deviation (MD) <–2.00 decibels (dB), and normal colour vision (CV) by pseudoisochromatic plate testing, had optic disc pallor for any cause or if there was a known baseline-corrected visual acuity (VA) <20/40 or visual field loss due to any other ophthalmic disorder. Patients were excluded if corticosteroid treatment for the current event of optic neuritis was begun prior to the clinical or MRI evaluation. None of the patients had corticosteroid therapy within 1 month of the current attack. Patients could have probable or definite multiple sclerosis, and 17 patients had either definite multiple sclerosis or a prior possible demyelinating event and the optic neuritis episode was the second demyelinating event, supporting a diagnosis of probable multiple sclerosis. Prior optic neuritis occurred in six current attack eyes and in six currently unaffected eyes. No patients had known systemic lupus erythematosus, sarcoidosis, syphilis, autoimmune optic neuropathy or other known causes of optic neuropathy. Of the 107 patients who met all of the study entry criteria, 93 patients had follow-up for a minimum of 3 months and most for at least 6 months. There was no follow-up in 14 patients.
Vision analysis
Visual performance was measured by three methods: the best corrected Snellen acuity with the exam room lights off; threshold perimetry using the 24-2 program of the Humphrey perimeter with the ‘STATPAC’; and CV measured with Ishihara plates with the exam room lights on. The VA was measured using the standard Snellen chart and was recorded as decimal equivalent (e.g. 20/20 = 1.0, 20/60 = 0.33, including finger counting = 0.01, hand motion = 0.005, light perception = 0.001, no light perception = 0); the MD of the perimetry recorded in decibels was utilized; and the CV was expressed as a decimal equivalent of the number of correct responses over the total number of test plates. In eyes with VA loss so severe that a spot size III could not be used, the MD was recorded as –35.0 dB. The examinations were repeated at 1 (3–5 weeks), 3 (10–14 weeks) and 6 (5–7 months) months.
MRI examination
The patients had MRI examinations performed with a GE echo speed (General Electric Medical Systems, Milwaukee, Wisc., USA) 1.5 tesla MRI. The examinations consisted of axial T2-weighted and axial fluid-attenuated inversion recovery images of the entire brain (5-mm thick slices, 2-mm spacing) and coronal T1- and T2 -weighted images from the pons to the mid globe (4-mm thick slices, 0.4-mm spacing). Following intravenous injection of gadolinium (Magnevist, Wayne, NJ, USA), axial and coronal views of the enhanced T1-weighted images with fat suppression (Hendrix et al., 1990; Lee et al., 1991) images (3 mm thick, 0.3 mm spacing) through the prechiasmal optic pathways were produced. A neuroradiologist who was aware of the diagnosis but was unaware of the side of the affected optic nerve reviewed each MRI. Both axial and coronal views were used to determine whether abnormal enhancement was present in each optic nerve and whether the orbital, intracanalicular or intracranial segments of the optic nerve were involved. The axial views were used to determine the length of the optic nerve with abnormal enhancement.
Of the 93 patients with follow-up, 68 received treatment with corticosteroids. Forty-nine of these had at least two high signal white matter lesions on the T2-weighted and fluid-attenuated inversion recovery MRI sequences. Fifty-nine patients received intravenous methylprednisolone, 1 g daily in one or two divided doses for 3 days, followed by conventional daily dose (60 mg) oral prednisone for 7–12 days. Nine patients received the equivalent of 500–900 mg oral prednisone daily, divided into three doses over 3 days, followed by 7–12 days of conventional daily dose. Sixteen patients with two or more areas of high signal foci on T2-weighted MRI did not receive any steroids, due either to patient refusal or to medical contraindications.
Data analysis
The data were analysed with the intention of exploring the relationship of the baseline and outcome VA, CV and MD in the affected eye with the location and length of abnormal enhancement of the corresponding optic nerve. VA < 0.10, MD < –20.0 dB or CV < 0.10 determined severe visual loss. Major or good recovery was determined when VA was 
0.80, MD was –3.0 dB (Keltner et al., 1994) or CV was 0.80.
The data for each time period for each visual performance measure by categories (which were divided into subgroups) were based on the presence or absence of specific MRI findings. Because the results of the presenting VA, MD and CV in the affected eye did not have a normal distribution, the Mann–Whitney test or the Fisher exact test was used to compare the baseline, 1, 3 and 6 month VA, MD and CV between subgroups within each set of patients in each MRI category. P values 
0.05 were considered significant (Fisher and van Belle, 1993).
An analysis of variance (ANOVA) was used to determine if differences between subgroups in each MRI category for baseline VA, MD or CV occurred because of the interaction between having and not having an MRI factor, such as the location and length of optic nerve enhancement.
The percentage of patients recovering to good visual performance and the percentage of patients with persistent severe visual loss at 1, 3 and 6 months were organized by the baseline MRI categories and subgroups.
Correlation analysis was performed, using the Spearman rank correlation two-tailed test, on the baseline and the 1 month VA, MD and CV with the location of optic nerve enhancement (orbit, canal, intracranial, orbit/canal, canal/intracranial, orbit/canal/intracranial enhancement), length of enhancement, length of enhancement >10 mm and length of enhancement longer than 17 mm.
|
Results |
|---|
|
TopSummaryIntroductionPatients and methodsResultsDiscussionReferences |
|---|
Of the 107 patients who met the inclusion criteria, 83 were women and 24 were men, with a mean age 35.6 years (SD 11.6 years). Visual symptoms were present an average of 7.7 days (SD 4.8; range 1–19 days) prior to the visual testing and 8.7 days (SD 6.0; range 1–19 days) prior to the performance of the MRI. The length of abnormal enhancement did not correlate with the duration of visual loss prior to the MRI. Patients with optic nerve enhancement >17 mm had the same age, gender and duration of symptoms prior to testing as the entire population. There were 93 patients with adequate follow-up, 68 who were treated with steroids after completing the MRI and 25 who received no treatment.
MRI
The affected optic nerve demonstrated abnormal enhancement on MRI in 101 cases (94.4%), and in six cases (5.6%) the affected nerves were not enhanced. No currently unaffected optic nerve had abnormal enhancement. The lengths of abnormal enhancing nerve had an approximately normal distribution (Fig. 1). The mean length of abnormal enhancement was 14.6 mm (SD 9.3 mm) with a maximum of 43 mm. The quartile length was 8 mm for 25%, 13 mm for 50% and 20 mm for 75%. The enhancing segment was >10 mm in 66 affected optic nerves (61.7%) and >17 mm in 37 affected optic nerves (34.6%). Although any segment or combination of contiguous segments of the optic nerve could have abnormal enhancement, the majority of affected nerves included the orbital portion (Table 1 ). The intracranial optic nerve was the segment with the lowest frequency of abnormal enhancement.
|
|
The distribution of the optic nerve enhancement for the 93 patients with follow-up was similar to the entire case series. The steroid-treated and untreated patients had a similar distribution of abnormal segments of enhancement for the affected optic nerve, except that the untreated group had no patients with enhancement of all three segments or both the canalicular and intracranial segments of the affected optic nerves. Steroid-treated and untreated patients had a similar distribution for the total number of millimetres of abnormal enhancement, and the mean length of abnormal enhancement was similar for the steroid-treated (15 mm) and untreated groups (12 mm, P = 0.10). Abnormal enhancement >10 mm occurred with similar frequency in the optic nerves of the steroid-treated patients (60%) and the untreated patients (56%). Abnormal enhancement >17 mm occurred with greater, but not significant, frequency in the optic nerves of the steroid-treated patients (39%) than of the untreated patients (25%, P = 0.16). Abnormal enhancement in the canalicular optic nerve was found in 46% of treated patients and 32% of untreated patients, which was not significantly different.
Baseline vision
See Table 2 and 3. The ANOVA results showed equal variances for both subgroups in each MRI category for each visual parameter except: the MD for enhancement length either > or 10 mm (F = 9.11, P = 0.003); MD (F = 6.56, P = 0.01) and CV (F = 7.40, P = 0.008) for enhancement length either > or 17 mm; and MD (F = 4.96, P = 0.03) and CV (F = 5.20, P = 0.02) for the subgroups whether all three segments were abnormally enhanced in contrast to enhancement in fewer segments.
|
|
The baseline VA, MD and CV were similar for patients with and without abnormal enhancement of the optic nerve, although none of the six patients without enhancement had VA
0.80 or MD –3.0 dB. In contrast, 18 eyes had VA 0.80 and two eyes had MD –3.0 dB in the group with enhancing optic nerves. The steroid treatment group had worse mean VA (0.26, P = 0.003) and MD (–24.6 dB, P = 0.003) and a lower percentage of patients with VA 0.80 (P = 0.02) and MD –3.0 dB (P = 0.003), and a trend of a lower percentage of CV 0.80. The steroid treatment group had a significantly higher percentage of patients with severe VA and visual field loss, with a similar trend for CV. Patients with enhancement of the canalicular segment of the optic nerve had worse mean CV (0.25) than those without enhancement of this segment (CV = 0.44, P = 0.03), and a trend of worse MD (–23.2 dB versus –20.3 dB) and VA (0.27 versus 0.36). Patients with >10 mm of optic nerve enhancement had worse MD (–20.6 dB) than those with shorter segments of abnormal enhancement (MD = –17.5 dB, P = 0.004). Patients with >17 mm of enhancement had worse VA (0.23, P = 0.02), MD (–26.1 dB, P = 0.009) and CV (0.21, P = 0.01) than patients with shorter segments of enhancement. If all three optic nerve segments were enhanced, the MD (–30.2 dB) and the CV (0.08) were severely depressed compared with those with enhancement of two or less segments (MD = –18.4 dB, P = 0.001; CV = 0.37, P = 0.008). There was no significant difference in the other subgroups of any MRI category for any visual performance measure. None of the other subgroups of a single segment or combination of segments of abnormal optic nerve enhancement had a significantly higher percentage of patients with severe VA, MD or CV (Table 3).
The baseline VA, MD or CV had no significant correlation with the presence of abnormal enhancement in the intracranial, orbital, combination of intracranial and canal, or canal and orbit portions of the optic nerve. Only CV had a significant correlation (r = 0.23, P = 0.02) with canalicular optic nerve enhancement. If all three segments were involved, there was a weak correlation with reduced VA (r = 0.22, P = 0.03), MD (r = 0.24, P = 0.02) and CV (r = 0.24, P = 0.02). In patients with >10 mm of optic nerve enhancement, only reduced MD was correlated (r = 0.30, P = 0.003). However, for optic nerves with enhancement >17 mm, minor but significant correlations were found for reduction of VA (r = 0.23, P = 0.02), MD (r = 0.27, P = 0.008) and CV (r = 0.27, P = 0.007). There were weak correlations between the total length of enhancement and loss of VA (r = –0.24, P = 0.01), MD (r = –0.35, P = 0.001) and CV (r = 0.22, P = 0.03). There were several weak correlations found for eyes with poor vision. For VA 0.1, r = –0.20 (P = 0.04) for total length of abnormal enhancement and r = 0.27 (P = 0.03) for abnormal enhancement >17 mm. For MD –20.0 dB, r = 0.21 (P = 0.04) for enhancement of the canalicular nerve, r = –0.25 (P = 0.02) for the total length of abnormal enhancement and r = 0.20 (P = 0.05) for optic nerve enhancement >10 mm. There were no other significant correlations between a baseline vision performance measure and the MRI subgroup findings.
Follow-up visual performance
The visual performance improved in the steroid-treated and untreated groups (no difference for all visual parameters), and in all of the subgroups categorized by the baseline MRI findings at 1 (Table 4 and 5), 3 and 6 months (data not shown). At 1 month, there was no significant difference between mean visual performance for each subgroup for each MRI category, including if all three optic nerve segments were enhanced, canalicular optic nerve enhancement and if the enhancing segments were >17 mm. Although a lower percentage of patients with all three segments involved recovered to a VA 0.80, MD –3.0 dB and CV 0.80, it was not significantly less. At 3 and 6 months, the means of each vision performance measure improved and remained similar between the subgroups in each MRI category.
|
|
None of the subgroups had a significantly higher percentage of persistent severe VA, MD or CV loss at 1 (Table 6), 3 and 6 months (data not shown). The only subgroup with a significantly lower percentage of recovery of good vision at 1 month was the group with all three segments enhancing where 33% of patients recovered CV in contrast to the other patients who had 69% recovery (P = 0.02).
|
The cumulative percentage of eyes recovering good VA was similar between MRI subgroups at 1, 3 and 6 months (Fig. 2). The mean time to recover good VA was
3 months in all of the MRI subgroups, with all having a 75% quartile of 1 month. The cumulative percentage of eyes that had visual field recovery to a good MD was similar between MRI subgroups except that no eyes with all three segments of enhancement had MD –3.0 dB until 3 months (Fig. 3 ). The mean time to recover a good MD was between 4 and 5 months in all of the MRI subgroups, with all having a 75% quartile of 3 months. The cumulative percentage of eyes that had good CV recovery was similar between MRI subgroups (Fig. 4). The mean time to recover good CV was 3 months in all MRI subgroups, with all having a 75% quartile of 1 month.
|
|
|
There were no major correlations between the baseline MRI findings and the 1 month visual performance results. Only enhancement of the canalicular optic nerve had significant, albeit weak, correlations; for recovery of VA to
0.80, r = –0.24 (P = 0.03) and of CV to 0.80, r = –0.25 (P = 0.04). |
Discussion |
|---|
|
TopSummaryIntroductionPatients and methodsResultsDiscussionReferences |
|---|
Enhancement of the optic nerve on fat-suppressed T1-weighted MRI after intravenous gadolinium is a sensitive method for demonstrating optic neuritis. The normal optic nerve does not enhance but the optic nerve sheath enhances mildly due to the pial vascular network. On the T1-weighted MRI, abnormal enhancement of the optic nerve affected by acute optic neuritis indicates a blood–optic nerve barrier breakdown in 94.4% of affected optic nerves. There were no false-positive MRI results; no optic nerve without acute clinical symptoms and findings, even in eyes with old optic neuritis-induced visual loss and optic atrophy, demonstrated abnormal enhancement.
Previous MRI studies have demonstrated abnormal signals in the affected optic nerve without the use of intravenous paramagnetic contrast agent. Short inversion recovery (STIR) (Miller et al., 1988) appeared to be more sensitive than conventional spin echo images without fat suppression for showing abnormalities in the optic nerve. The T2-weighted images with fat suppression and STIR MRI without contrast demonstrate an abnormally high signal as a result of optic nerve tissue oedema, inflammation and demyelination with a loss of hydrophobic myelin in acute cases (Stewart et al., 1991), and perhaps axonal loss, extracellular water and gliosis (Barnes et al., 1991) on studies performed on chronic plaques. Abnormal enhancement on contrast MRI in animals with experimental allergic encephalomyelitis (Barnes et al., 1988) and in patients with optic neuritis appears to be due to the same mechanisms that cause enhancement from blood–brain barrier disruption in the acute phase of a demyelinating plaque in the brain (Kermode et al., 1990; Katz et al., 1993). These cause leakage of gadolinium between endothelial cells of capillaries in and around the optic nerve.
When a fat suppression technique is used with gadolinium, the orbital fat is eliminated (Lee et al., 1991) and detection of the abnormal enhancement of the optic nerve is improved (Tien et al., 1991; Guy et al., 1992).
Fat-suppressed gadolinium-enhanced MRI appears to be more useful for detecting acute optic neuritis than STIR since the latter method can also be abnormal in asymptomatic (presumably previously affected) optic nerves ( Miller, 1988), possibly reflecting the permanent injury or change in the optic nerve. Also, STIR has a slightly lower sensitivity since only 77 (Dunker and Wiegand, 1996) to 81 (Miller et al., 1988) to 89% (Kapoor et al., 1998) of optic nerves affected within 30 days have demonstrable lesions. However, STIR was abnormal in 92% of cases when performed 6 weeks after the optic neuritis event (Miller et al., 1988). Newer MRI techniques, such as diffusion-weighted imaging, have failed to demonstrate lesions in acutely affected optic nerves (Iwasawa et al., 1997). The MRI method used in our study might be useful in distinguishing patients with optic neuritis and disc swelling from those with non-arteritic anterior ischaemic optic neuropathy in the situation when in the latter disorder there is some, albeit minor and without eye movement, pain in the affected eye (Swartz et al., 1995).
The mean length of the optic nerve that enhanced was 14.6 mm, which approximates to the 10 mm reported by Miller et al. (1988) and 17.5 mm reported by Dunker and Wiegand (1996) using STIR. STIR signal abnormality 17 mm and longer seemed to occur in patients with a longer duration between the onset of visual loss and performing the MRI (Kapoor et al., 1998). However, our patients with gadolinium enhancement >17 mm had a duration of visual loss prior to the MRI similar to the patients with shorter affected segments, and the length of abnormal enhancement did not correlate with the number of days of visual symptoms prior to the MRI. As with STIR, gadolinium-enhancing MRI revealed lesions most commonly in the orbital portion of the optic nerve, but the canalicular segment, often in association with involvement of other segments, was seen in almost 45% of nerves. Gadolinium enhancement revealed abnormal enhancement in the intracranial portion of the optic nerve, alone in almost 6%, and together with other segments of the optic nerve in almost 16% of our 107 patients. In prior studies, STIR showed lesions in the intracranial segment of the optic nerve in 9% of 22 patients (Dunker and Wiegand, 1996) or 4.5% of 37 patients (Miller et al., 1988). This difference probably reflects a lack of sensitivity for STIR to demonstrate lesions of the intracranial optic nerve.
In general, the baseline visual performance was not worse in patients with a particular segment of optic nerve enhancement. In fact, the few patients without abnormal enhancement had visual deficits similar to those of patients with enhancement. Only patients with involvement of the canalicular optic nerve had worse vision, which was significant in only one modality, colour vision. There were no major correlations with the VA, MD or CV and a specific abnormal segment of optic nerve. Even patients with two segments of optic nerve enhancement did not have worse vision. Only patients with three segments of abnormal enhancement had worse vision, as measured by visual field and CV, but not VA. This worse vision most probably relates to having a longer segment of optic nerve affected.
Eyes with >10 mm of optic nerve enhancement had worse visual fields but not worse VA or CV. All three measures of vision were significantly worse for eyes with enhancement of the optic nerve >17 mm. The length of abnormal enhancement correlated weakly with worse VA and CV and only modestly with worse MD. This weak correlation differs from the significant negative correlation between the length of STIR lesion and the presenting visual acuity reported by Dunker and Wiegand (1996) but is similar to the results of Kapoor et al. (1998). Although disruption of the blood–optic nerve barrier impairment leads to oedema and an increase in lymphocyte and macrophage infiltration into the nerve, all of which are associated with reduced signal conduction, the presence of abnormal contrast enhancement probably does not necessarily directly reflect the compromise of the optic nerve function.
By 1 month, the steroid-treated patients, who had worse baseline vision, improved the visual performance so that by 1 month the previous difference between the steroid-treated and the untreated group was no longer present. Prior studies have shown that corticosteroids do not affect the visual outcome except to speed recovery (Beck et al., 1992; Keltner et al., 1994) even in patients with STIR-demonstrated lesions of the canalicular optic nerve (Kapoor et al., 1998). Similarly to previous studies (Beck et al., 1992, 1994), the visual performance improved in all groups by 1 month regardless of steroid use. This is contrasted with reports that suggest that abnormal signal length, >15 or 17.5 mm, and intracanalicular location of an optic nerve lesion on MRI are associated with poor or slow recovery from optic neuritis even if treated with steroids (Miller et al., 1988; Dunker and Wiegand, 1996). In a previous study using STIR, of patients with lesions <17.5 mm, 36% had VA of 20/25 at 1 month in contrast to none of the patients with lesions >17.5 mm (Dunker and Wiegand, 1996). This contrasts with our study where 65% of optic nerves with lesions >17 mm recovered to 20/25 at 1 month. Also, 79% of the eyes that did not improve had STIR lesions >17.5 mm (Dunker and Wiegand, 1996). In a later more extensive study using STIR, the initial site or length of abnormal optic nerve did not correlate with visual recovery, but longer length and a persistence of abnormal signal in the canalicular portion of the optic nerve at 26 weeks were seen in patients with poor visual recovery (Kapoor et al., 1998). The persistence of the lesion on STIR MRI might suggest that in these optic nerves there was more extensive gliosis, axonal loss and extracellular water. Thus, STIR and gadolinium-enhanced T1-weighted MRI do not reflect identical processes.
Recovery of VA, MD and CV was similar for our patients with and without abnormal enhancement, and no specific location of abnormal enhancement caused significantly worse visual performance by 1 month. However, patients with canalicular enhancement had a trend of worse VA, MD and CV at 1 month. In addition, of patients with optic nerves that had enhancement of all three segments, a lower percentage had good recovery of vision. Also, patients with canalicular involvement had weak inverse correlation with recovery of VA and CV. As previously reported (Beck et al., 1992), our patients with optic neuritis further improved VA, visual field and CV over 3 and then 6 months. Between subgroups, the percentage of eyes reaching a good level of visual recovery was almost the same at 3 months and was the same by 6 months.
Our study confirms the utility of the gadolinium-enhanced fat-suppressed MRI of the orbit reported in a small series (Guy et al., 1992). However, the abnormal gadolinium enhancement is not diagnostic of demyelinating optic neuritis since this finding has been demonstrated in neoplastic infiltration, cytomegalovirus, radiation vasculopathy (Guy et al., 1991), systemic lupus erythematosus and rheumatoid arthritis-associated optic neuropathy (Sklar et al., 1996; Sartoretti-Schefer et al., 1997). Gadolinium-enhanced MRI should be useful in cases where there are no prior clinical demyelinating episodes or MRI evidence of brain white matter lesions. Gadolinium-enhanced MRI and STIR should be helpful to distinguish acute optic neuritis from acute non-arteritic anterior ischaemic optic neuropathy (Gass et al., 1995). None of the optic nerves in our study with prior optic neuritis enhanced, suggesting that abnormal gadolinium enhancement should not occur as a result of residual damage to the optic nerve from optic neuritis. Although longer segments of abnormal optic nerve enhancement, and, to a lesser degree, canalicular involvement are associated with worse presenting vision, no specific finding on gadolinium MRI is a predictor of a poor or better visual outcome.
|
References |
|---|
|
TopSummaryIntroductionPatients and methodsResultsDiscussionReferences |
|---|
Barnes D, McDonald WI, Landon DN, Johnson G. The characterization of experimental gliosis by quantitative nuclear magnetic resonance imaging. Brain 1988; 111: 83–94.
Barnes D, Munro PM, Youl BD, Prineas JW, McDonald WI. The longstanding MS lesion. A quantitative MRI and electron microscopic study. Brain 1991;114: 1271–80.
Beck RW, Cleary PA, Anderson MM, Keltner JL, Schults WT, Kaufman DI, et al. A randomized, controlled trial of corticosteroids in the treatment of acute optic neuritis. The Optic Neuritis Study Group. N Engl J Med 1992; 326: 581–8.[Abstract]
Beck RW, Cleary PA, Backlund JC. The course of visual recovery after optic neuritis. Experience of the Optic Neuritis Treatment Trial. Ophthalmology 1994; 101: 1771–8.[ISI][Medline]
Beck RW, Cleary PA, Trobe JD, Kaufman DI, Kupersmith MJ, Paty DW, et al. The effect of corticosteroids for acute optic neuritis on the subsequent development of multiple sclerosis. The Optic Neuritis Study Group. N Engl J Med 1993; 329: 1764–9.
Dunker S, Wiegand W. Prognostic value of magnetic resonance imaging in monosymptomatic optic neuritis. Ophthalmology 1996; 103: 1768–73.[ISI][Medline]
Fisher LD, van Belle G. Biostatistics. A methodology for the health sciences. New York: John Wiley; 1993. p. 138–450
Gass A, Barker GJ, MacManus D, Sanders M, Riordan-Eva P, Tofts PS, et al. High resolution magnetic resonance imaging of the anterior visual pathway in patients with optic neuropathies using fast spin echo and phased array local coils. J Neurol Neurosurg Psychiatry 1995; 58: 562–9.[Abstract]
Guy J, Mancuso A, Beck R, Moster ML, Sedwick LA, Quisling RG, et al. Radiation-induced optic neuropathy: a magnetic resonance imaging study. J Neurosurg 1991; 74: 426–32.[ISI][Medline]
Guy J, Mao J, Bidgood D, Mancuso A, Quisling RG. Enhancement and demyelination of the intraorbital optic nerve. Fat suppression magnetic resonance imaging. Ophthalmology 1992; 99: 713–19.[ISI][Medline]
Hendrix LE, Kneeland JB, Haughton VM, Daniels DL, Szumowski J, Williams AL, et al. MR imaging of optic nerve lesions: value of gadopentetate demeglumine and fat-suppression technique. AJNR Am J Neuroradiol 1990; 11: 749–54.[Abstract]
Iwasawa T, Matoba H, Ogi A, Kurihara H, Saito K, Yoshida T, et al. Diffusion-weighted imaging of the human optic nerve: a new approach to evaluate optic neuritis in multiple sclerosis. Magn Reson Med 1997; 38: 484–91.[ISI][Medline]
Jacobs LD, Beck RW, Simon JH, Kinkel RP, Brownscheidle CM, Murray TJ, et al. Intramuscular interferon beta-1a therapy initiated during a first demyelinating event in multiple sclerosis. CHAMPS Study Group. N Engl J Med 2000; 343: 898–904
Kapoor R, Miller DH, Jones SJ, Plant GT, Brusa A, Gass A, et al. Effects of intravenous methylprednisolone on outcome in MRI-based prognostic subgroups in acute optic neuritis. Neurology 1998; 50: 230–7.
Katz D, Taubenberger JK, Cannella B, McFarlin DE, Raine CS, McFarland HF. Correlation between magnetic resonance imaging findings and lesion development in chronic, active multiple sclerosis. Ann Neurol 1993; 34: 661–9.[ISI][Medline]
Keltner JL, Johnson CA, Spurr JO, Beck RW. Visual field profile of optic neuritis. One-year follow-up in the Optic Neuritis Treatment Trial. Arch Ophthalmol 1994; 112: 946–53.[Abstract]
Kermode AG, Thompson AJ, Tofts P, McManus DG, Kendall BE, Kingsley DP, et al. Breakdown of the blood–brain barrier precedes symptoms and other MRI signs of new lesions in multiple sclerosis. Brain 1990; 133: 1477–89.
Lee DH, Simon JH, Szumowski J, Feasby TE, Karlik SJ, Fox AJ, Pelz DM. Optic neuritis and orbital lesions: lipid-suppressed chemical shift MR imaging. Radiology 1991; 179: 543–6.
Miller DH, Newton MR, van der Poel JC, du Boulay EP, Halliday AM, Kendall BE, et al. Magnetic resonance imaging of the optic nerve in optic neuritis. Neurology 1988; 38: 175–9.
Optic Neuritis Study Group. The clinical profile of optic neuritis. Experience of the Optic Neuritis Treatment Trial. Arch Ophthalmol 1991; 109: 1673–8.[Abstract]
Sartoretti-Schefer S, Wichman W, Valvanis A. Optic neuritis: characteristic magnetic resonance imaging features and differential diagnosis. Int J Neuroradiol 1997; 3: 417–27.
Sklar EM, Schatz NJ, Glaser JS, Post JD, ten Hove M. MR of vasculitis-induced optic neuropathy. AJNR Am J Neuroradiol 1996; 17: 121–8.[Abstract]
Stewart WA, Alvord ECr, Hruby S, Hall LD, Paty DW. Magnetic resonance imaging of experimental allergic encephalomyelitis in primates. Brain 1991; 114: 1069–96.
Swartz NG, Beck RW, Savino PJ, Sergott RC, Bosley TM, Lam BL, et al. Pain in anterior ischemic optic neuropathy. J Neuroophthalmol 1995; 15: 9–10.[Medline]
Tien RD, Hesselink JR, Szumowski J. MR fat suppression combined with Gd-DTPA enhancement in optic neuritis and perineuritis. J Comput Assist Tomogr 1991; 15: 223–7.[ISI][Medline]
CiteULike 
Connotea 
Del.icio.us What's this?
This article has been cited by other articles:
|
|
 |
|
  J de la Cruz and M J Kupersmith Clinical profile of simultaneous bilateral optic neuritis in adults Br. J. Ophthalmol., May 1, 2006; 90(5): 551 - 554. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 F. Barkhof, U. Held, J. H. Simon, M. Daumer, F. Fazekas, M. Filippi, J. A. Frank, L. Kappos, D. Li, S. Menzler, et al. Predicting gadolinium enhancement status in MS patients eligible for randomized clinical trials Neurology, November 8, 2005; 65(9): 1447 - 1454. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 M A Rocca, S J Hickman, L Bo, F Agosta, D H Miller, G Comi, and M Filippi Imaging the optic nerve in multiple sclerosis Multiple Sclerosis, October 1, 2005; 11(5): 537 - 541. [Abstract] [PDF]
|
|
|
|
 |
|
 S. J. Hickman, A. T. Toosy, S. J. Jones, D. R. Altmann, K. A. Miszkiel, D. G. MacManus, G. J. Barker, G. T. Plant, A. J. Thompson, and D. H. Miller A serial MRI study following optic nerve mean area in acute optic neuritis Brain, November 1, 2004; 127(11): 2498 - 2505. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
S. J. Hickman, A. T. Toosy, S. J. Jones, D. R. Altmann, K. A. Miszkiel, D. G. MacManus, G. J. Barker, G. T. Plant, A. J. Thompson, and D. H. Miller Serial magnetization transfer imaging in acute optic neuritis Brain, March 1, 2004; 127(3): 692 - 700. [Abstract] [Full Text] [PDF]
|
|
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||