Neuromyelitis optica (NMO) is an inflammatory, demyelinating syndrome of the central nervous system that is characterized by severe attacks of optic neuritis (ON) and transverse myelitis (TM) (1). Clinical, neuroimaging, and laboratory findings are used to distinguish this clinical entity from multiple sclerosis (MS) (2). Moreover, serum detection of aquaporin-4 immunoglobulin G (NMO-IgG) can help differentiate NMO from other demyelinating disorders with 54%–91% sensitivity and >90% specificity (3,4). The currently used criteria of Wingerchuk et al (5) and the National Multiple Sclerosis Society (6) mainly rely on seropositivity and typical imaging findings. Thus, many patients may go undiagnosed, especially at earlier stages of disease, at which point they may not manifest the full spectrum of signs and symptoms. Early diagnosis and treatment with immunosuppressive agents are important to reduce the risk of further neurological impairment in NMO (7).
Optical coherence tomography (OCT) studies have shown differences in retinal nerve fiber layer (RNFL) values between NMO ON eyes, MS ON eyes, and healthy control eyes (8,9). The primary objective of this study was to analyze RNFL values in NMO patients (with and without history of ON) and to compare these RNFL measurements with those of patients with relapsing–remitting multiple sclerosis (RRMS) and healthy controls. As a secondary objective, we aimed to correlate RNFL thickness in NMO eyes (ON and non-ON eyes) with clinical parameters. Finally, we endeavored to determine whether RNFL thickness could be used as a diagnostic criterion for NMO, and if the pattern or severity of the RNFL deficit could help distinguish NMO patients early in the course of their disease.
PATIENTS AND METHODS:
This was a cross-sectional cohort study. The British Columbia MS (BCMS) database was used to help identify eligible NMO patients. The database has longitudinally collated clinical information on MS and NMO patients registered with 1 of 4 MS clinics across British Columbia since 1980 and is estimated to include 80% of MS and NMO patients in British Columbia (10,11).
Patients were recruited between August 2009 and March 2011. Subjects with NMO spectrum disorder were identified from the BCMS database and were invited to participate by letter. If no answer was received within 4 months, a second invitation letter was sent. All patients were diagnosed by an experienced MS neurologist with special interest in NMO.
Inclusion criteria for NMO spectrum disorders were as follows:
- NMO Diagnostic Criteria positive—fulfill criteria published in 2006 (5).
- NMO Diagnostic Criteria negative—NMO-IgG antibody present with either severe ON or severe TM and clinically suspected NMO (patients with either bilateral simultaneous ON or TM plus normal brain magnetic resonance imaging [MRI] and contiguous spinal cord lesions over at least 3 segments) (12).
Patients with RRMS, based on the modified McDonald criteria (13), and healthy controls were recruited from the University of British Columbia Hospital MS clinic, accompanying persons and hospital staff. We aimed to match for age (±5 years) and refraction (±2 diopters).
Exclusion criteria included patients with a recent history of ON (<6 months), history of ocular disease (including macular degeneration, diabetic retinopathy, uveitis, and glaucoma), history of diseases that could mimic MS or NMO, neurodegenerative conditions that could impact OCT testing (Parkinson disease, Alzheimer disease), and subjects with difficulty maintaining fixation and/or myopic refraction of more than −5.0 diopters. Exclusion criteria were the same for controls as for NMO and MS patients.
This study was approved by the Clinical Research Ethics Board at the University of British Columbia. A patient information sheet was provided, and informed consent was obtained before subject enrolment in the study.
Clinical and demographic data were obtained from the BCMS database, supplemented by chart review, structured questionnaires, and history taking by the study physician (A.L.). Information collected included age, sex, date of symptom onset, history and number of ON events, use of any immunomodulatory drugs (IMDs) or other medications, presence of co-morbidities, MRI findings (all read by a masked radiologist with expertise in NMO and MS), and current Expanded Disability Status Scale (EDSS) scores. If the EDSS score was not recent (within 6 months) or if new MS symptoms were reported, an EDSS examination was performed by an MS neurologist (R.S. or A.T.) at the time of the OCT testing. NMO antibodies were tested in our own laboratory and in at least one other laboratory using ELISA-based techniques (Mayo Laboratories, Rochester, MN; Calgary Mitogen Laboratory, Calgary, Canada) or cell-based techniques (Tohoku Laboratory, Sendai, Japan). If 1 result was positive, antibody status was rated as positive. To minimize misdiagnosis of the antibody-negative cohort, these cases were independently reviewed by 2 other physicians.
Spectral-Domain Optical Coherence Tomography Testing
All cases and controls were assessed by a neuro-ophthalmologist specialized in spectral-domain optical coherence tomography (SD-OCT) (A.L.), using the RNFL protocol of Heidelberg Spectralis SD-OCT (Software version 5.1.2; Heidelberg Engineering, Heidelberg, Germany) in high-resolution mode (axial resolution 3.8 µm, 19,000 scans per second). Sixteen consecutive circular B-scans (each composed of 1,536 A-scans) with a diameter of 3.4 mm were automatically averaged to reduce speckle noise. The online tracking system compensated for eye movements. Several scans were taken without pupil dilation, with the best centered with a quality of at least 25 (14) chosen for analysis. The software algorithm provided objective refraction (spherical equivalent) and the RNFL thickness for the temporal, superior, nasal, and inferior quadrants and the overall mean of these quadrants.
Pairwise Wilcoxon rank sum tests with Holm correction were used to compare RNFL thickness among 6 groups (NMO, RRMS, and healthy control) of patients (without or with 1 or more history of ON). These were performed separately for NMO or RRMS ON eyes and non-ON eyes.
For NMO patients, Spearman rank correlation and Pearson correlation were used to examine the association between RNFL thickness and disability (EDSS) and between RNFL thickness and disease duration, respectively. These were performed for all NMO eyes and then separately for NMO ON eyes and non-ON eyes. Furthermore, a linear mixed-effects model was developed to examine the association between RNFL thickness and patient characteristics, which include sex, current age, disease duration, disability (EDSS) (grouped as: <2, 2.5, 3, 3.5, 4, ≥4.5), history of ON (present, absent), refraction, use of current IMD (ever, never), ethnicity (Asian/Caucasian), and NMO diagnostic criteria positive/negative status, while adjusting for within-patient intereye correlations (i.e., 2 eyes from the same person are correlated). The association between RNFL thickness and patient characteristics initially was assessed univariately. If considered clinically important (i.e., the EDSS) or statistically significant (disease duration) from the univariate analysis, then the model was also adjusted for the history of ON (ever, never). Complementary analyses were performed for each quadrant of the overall RNFL thickness (inferior, superior, temporal, and nasal). Statistical analyses were performed using R: A Language and Environment for Statistical Computing V.2.13.2 (R Foundation for Statistical Computing, Vienna, Austria; 2011).
The primary outcome measures were RNFL thickness in NMO patients, RRMS patients, and healthy controls. Secondary outcomes measures were correlation between RNFL in NMO eyes, EDSS, and disease duration. Furthermore, RNFL values in unilateral ON eyes were compared between the NMO and the RRMS groups.
Patient Demographics and Clinical Characteristics
An invitation letter was sent to 40 NMO patients. Twenty positive responses were received after first contact and an additional 7 after second recruitment letter. These 27 patients with NMO spectrum disorder initially consented and underwent SD-OCT. However, 2 NMO patients had to be excluded due to high myopic refraction (more than −5.0 diopters), leaving 25 patients in the NMO group. Twenty-five gender-, age-, and EDSS-matched patients with RRMS, according to the modified McDonald criteria (13) and 50 healthy controls were included. A demographic summary of our study patients is given in Table 1.
The 3 groups were broadly similar for age and refraction, although differed by disease duration (NMO disease duration shorter than MS) and by gender (with more men included in the control group). Fifteen of the 25 NMO patients fulfilled the NMO diagnostic criteria and 10 patients met our additional criteria for NMO spectrum disorder (3 with pos IgG and ON; 3 with positive IgG and TM; 2 with bilateral simultaneous ON, normal brain MRI and spinal cord lesions; and 2 with TM, normal brain MRI, and spinal cord lesions).
Five patients (10 eyes) had no previous history of ON, 11 (22 eyes) had a history of bilateral ON, and 9 (18 eyes) had a history of unilateral ON. In the ON group, there were 5 eyes affected by 2 or more ON events, whereas the remaining eyes (n = 26) were affected by a single ON event. Twenty patients had a history of TM. Eight patients who fulfilled NMO diagnostic criteria had positive NMO-IgG titers, 6 were tested negative, and 1 patient was not tested. In the negative group, 6 were tested IgG positive and 4 were negative. Twenty-two patients had MRI spine findings consistent with NMO (spinal cord lesion >3 segments).
In the MS group, 17 patients had no history of ON, 5 had a history of bilateral sequential ON, and 3 had a history of unilateral ON. Thus, 13 eyes (26%) were previously affected by single ON. None of the eyes had a history of 2 or more ON events.
The overall and different quadrant RNFL values of different groups are shown in Table 2, P-values between different groups are shown in Table 3. NMO eyes with history of 1 ON event had 34% lower RNFL thickness compared with NMO non-ON eyes (P < 0.0001). The mean intereye difference in the RNFL thickness (non-ON eyes−ON eyes) for NMO patients with unilateral ON was 36.3 µm (SD ±19.45 µm, P = 0.005) and 16.7 µm (SD ±17.0 µm, P = 0.25) for MS eyes with unilateral ON (Wilcoxon signed rank test). Mean RNFL values in NMO ON eyes tended to be thinner compared with MS ON eyes, but the difference was not statistically significant (P = 0.46). Mean RNFL values in NMO non-ON eyes were not different from healthy controls (P = 0.56). Mean RNFL values in MS non-ON eyes were different from healthy controls (P = 0.03). NMO non-ON eyes were not different from MS non-ON eyes (P = 0.56).
RNFL thickness was not associated with EDSS scores (ρ = −0.21, P = 0.13, Spearman correlation) but lower RNFL values were associated with longer disease duration (ρ = −0.34, P = 0.02, Pearson correlation) when looking at the whole NMO group. No correlation was found when analysis was repeated on the NMO non-ON eyes (EDSS: ρ = −0.13, P = 0.62; disease duration: ρ = 0.17, P = 0.52) or when analysis was repeated on the NMO ON eyes (EDSS: ρ = −0.33, P = 0.06; disease duration: ρ = −0.32, P = 0.07).
Linear Mixed-Effect Model Analysis
The linear mixed-effects model indicated that the mean RNFL measurements were not associated with gender, age, disability (EDSS), refraction, IMD, ethnicity, or NMO diagnostic criteria status (P > 0.1, Table 4), although RNFL measurements were associated with a history of ON (P < 0.001, Table 4) and disease duration (P = 0.03). If the model was adjusted for history of ON, neither EDSS nor disease duration was associated with RNFL thickness (P > 0.1).
Histopathologically, NMO is characterized by acute inflammation associated with neutrophilic and eosinophilic infiltrate, demyelination, and destructive necrotizing cavitation, with vascular hyalinization (thickening) and IgG and complement deposition, which destroys perivascular neural parenchyma (15,16). The intense inflammatory activity in NMO eyes can result in pronounced RNFL atrophy. However, MS lesions exhibit demyelination and inflammation with less axonal loss than in NMO (8), indicating the potential to distinguish NMO from MS, based on severity of RNFL loss after an episode of ON.
Our primary objective was to compare RNFL thickness between NMO, RRMS patients, and healthy controls and to evaluate if severity of RNFL deficit could help distinguish NMO patients early in their disease. We observed RNFL thinning in NMO ON eyes compared with unaffected eyes or healthy control eyes, whereas unaffected NMO eyes were not significantly different from healthy controls.
Ratchford et al (8) reported, in a cross-sectional study on 26 NMO patients, a small difference in RNFL between NMO non-ON eyes (n = 8) and healthy controls (97.9 µm in NMO, 96.3 µm in TM, and 102.4 µm in healthy controls), all of which were not significantly different (Table 5). We used the newer generation SD-OCT with higher resolution and a larger sample size (n = 18) and did not find any significant difference between these groups. This may be due to the more acute course of NMO relapses with no significant disease activity between such episodes (i.e., secondary progressive pattern) (17) compared with gradual RNFL loss observed in MS (18) patients mainly due to on-going brain atrophy that implicates a different pathophysiology.
Naismith et al (15) and Nakamura et al (19) have shown more severe RNFL thinning inferiorly and superiorly in NMO eyes compared with MS eyes. In our study, this was not the case. One reason could be that we used the higher resolution SD-OCT compared with the Stratus OCT, which is harder to center and only slight decentration can result in artificial thinning. Other authors also have shown similar differences in RNFL thickness between MS ON eyes and NMO ON eyes (16) and MS ON eyes, NMO eyes, and healthy controls (20,21) using older time domain-OCT technology (Table 5).
Due to a relatively large overlap between RNFL value ranges in NMO eyes with a single episode of ON and MS eyes with ON, it is difficult to use RNFL to differentiate between NMO and MS ON (Table 2). In contrast to MS ON eyes, we found no difference between the unaffected NMO eyes and the healthy controls. Therefore, an alternative might be to use the intereye difference in RNFL values between affected and unaffected eyes in patients with a history of an isolated ON event to distinguish NMO from MS ON. Our data showed a difference of 36.3 µm (n = 10) in the NMO group compared with 16.7 µm (n = 3) in the MS group. Unfortunately, the majority of the MS cohort studied consisted of patients with bilateral sequential ON, and only 3 patients with unilateral ON in the MS group. We found that a difference of more than 20 µm in RNFL was more likely to occur in NMO (70%, 7 of 10) than in MS (33%, 1 of 3) eyes. This could be a better diagnostic marker than RNFL in ON eyes alone because it uses both findings and integrates them in a single number. This was also suggested by Ratchford et al (8), noting that after a first episode of unilateral ON, RNFL difference of more than 15 µm between eyes should prompt consideration of an NMO spectrum disorder. Distinguishing NMO from MS based on the difference in RNFL thickness after unilateral ON requires further studies comparing larger cohorts of NMO and MS patients with a history of unilateral ON to establish a reliable threshold to separate the 2 entities.
Shortcomings of our study primarily were due to a relatively small sample size and type of patients enrolled. As we were trying to find a diagnostic test that helps identifying NMO patients early in the disease stage, diagnosis of NMO spectrum disorders was based on clinical diagnosis of the neurologist (reviewed by 2 other independent neurologists) rather than on seropositivity. This might explain our small number of seropositive patients in the NMO group.
Our data showed that RNFL in NMO ON eyes is significantly reduced compared with the unaffected eyes or eyes of healthy controls, but the unaffected eyes were not significantly different from healthy control eyes. RNFL in NMO is not different enough to distinguish NMO ON from MS ON eyes, but a value of more than 20 µm difference in RFNL between eyes with unilateral history of ON may be a more promising diagnostic marker for NMO but requires further study.
1. Wingerchuk DM, Hogancamp WF, O'Brien PC, Weinshenker BG. The clinical course of neuromyelitis optica (Devic's syndrome). Neurology. 1999;53:1107–1114.
2. Mata S, Lolli F. Neuromyelitis optica: an update. J Neurol Sci. 2011;303:13–21.
3. Lennon VA, Wingerchuk DM, Kryzer TJ, Pittock SJ, Lucchinetti CF, Fujihara K, Nakashima I, Weinshenker BG. A serum autoantibody marker of neuromyelitis optica: distinction from multiple sclerosis. Lancet. 2004;364:2106–2112.
4. Waters P, Vincent A. Detection of anti-aquaporin-4 antibodies in neuromyelitis optica: current status of the assays. Int MS J. 2008;15:99–105.
5. Wingerchuk DM, Lennon VA, Pittock SJ, Lucchinetti CF, Weinshenker BG. Revised diagnostic criteria for neuromyelitis optica. Neurology. 2006;66:1485–1489.
6. Miller DH, Weinshenker BG, Filippi M, Banwell BL, Cohen JA, Freedman MS, Galetta SL, Hutchinson M, Johnson RT, Kappos L, Kira J, Lublin FD, McFarland HF, Montalban X, Panitch H, Richert JR, Reingold SC, Polman CH. Differential diagnosis of suspected multiple sclerosis: a consensus approach. Mult Scler. 2008;14;1157–1174.
7. Wingerchuk DM. Evidence for humoral autoimmunity in neuromyelitis optica. Neurol Res. 2006;28:348–353.
8. Ratchford JN, Quigg ME, Conger A, Frohman T, Frohman E, Balcer LJ, Calabresi PA, Kerr DA. Optical coherence tomography helps differentiate neuromyelitis optica and MS optic neuropathies. Neurology. 2009;73:302–308.
9. Lange AP, Sadjadi R, Alkabie S, Traboulsee AL. Spectral domain optical coherence tomography of retinal nerve fiber layer thickness in NMO patients. Poster presented at: NANOS; 2011; Vancouver, Canada.
10. Sweeney VP, Sadovnick AD, Brandejs V. Prevalence of multiple sclerosis in British Columbia. Can J Neurol Sci. 1986;13:47–51.
11. Sadovnick AD, Ebers GC, Wilson RW, Paty DW. Life expectancy in patients attending multiple sclerosis clinics. Neurology. 1992;42:991–994.
12. Wingerchuk DM, Lennon VA, Lucchinetti CF, Pittock SJ, Weinshenker BG. The spectrum of neuromyelitis optica. Lancet Neurol. 2007;6:805–815.
13. Polman CH, Reingold SC, Edan G, Filippi M, Hartung HP, Kappos L, Lublin FD, Metz LM, McFarland HF, O'Connor PW, Sandberg-Wollheim M, Thompson AJ, Weinshenker BG, Wolinsky JS. Diagnostic criteria for multiple sclerosis: 2005 revisions to the “McDonald Criteria”. Ann Neurol. 2005;58:840–846.
14. Langenegger SJ, Funk J, Toteberg-Harms M. Reproducibility of retinal nerve fiber layer thickness measurements using the eye tracker and the retest function of spectralis SD-OCT in glaucomatous and healthy control eyes. Invest Ophthalmol Vis Sci. 2011;52:3338–3344.
15. Naismith RT, Tutlam NT, Xu J, Klawiter EC, Shepherd J, Trinkaus K, Song SK, Cross AH. Optical coherence tomography differs in neuromyelitis optica compared with multiple sclerosis. Neurology. 2009;7212:1077–1082.
16. Green AJ, Cree BA. Distinctive retinal nerve fibre layer and vascular changes in neuromyelitis optica following optic neuritis. J Neurol Psychiatry Neurosurg. 2009;90:1002–1005.
17. Wingerchuk DM, Pittock SJ, Lucchinetti CF, Lennon VA, Weinshenker BG. A secondary progressive clinical course is uncommon in neuromyelitis optica. Neurology. 2007;68:603–605.
18. Talman LS, Bisker ER, Sackel DJ, Long DA Jr, Galetta KM, Ratchford JN, Lile DJ, Farrell SK, Loguidice MJ, Remington G, Conger A, Frohman TC, Jacobs DA, Markowitz CE, Cutter GR, Ying GS, Dai Y, Maguire MG, Galetta SL, Frohman EM, Calabresi PA, Balcer LJ. Longitudinal study of vision and retinal nerve fiber layer thickness in multiple sclerosis. Ann Neurol. 2010;67:749–760.
19. Nakamura M, Nakazawa T, Doi H, Hariya T, Omodaka K, Misu T, Takahashi T, Fujihara K, Nishida K. Early high-dose intravenous methylprednisolone is effective in preserving retinal nerve fiber layer thickness in patients with neuromyelitis optica. Graefes Arch Clin Exp Ophthalmol. 2010;248:1777–1785.
20. De Seze J, Blanc F, Jeanjean L, Zephir H, Labauge P, Bouyon M, Ballonzoli L, Castelnovo G, Fleury M, Defoort S, Vermersch P, Speeg C. Optical coherence tomography in neuromyelitis optica. Arch Neurol. 2008;65:920–923.
21. Merle H, Olindo S, Donnio A, Richer R, Smadja D, Cabre P. Retinal peripapillary nerve fiber layer thickness in neuromyelitis optica. Invest Ophthalmol Vis Sci. 2008;49:4412–4417.