Association of Retinal Ganglion Cell Layer Thickness With Future Disease Activity in Patients With Clinically Isolated Syndrome | Demyelinating Disorders | JAMA Neurology | JAMA Network
[Skip to Navigation]
Figure 1.  Kaplan-Meier Plots for Percentage of Patients Fulfilling No Evidence Of Disease Activity–3 Criteria
Kaplan-Meier Plots for Percentage of Patients Fulfilling No Evidence Of Disease Activity–3 Criteria

Statistics from log-rank test of ganglion cell and inner plexiform layer grouping (A), peripapillary retinal nerve fiber layer thickness grouping (B), and inner nuclear layer grouping (C). Shaded regions describe the 95% CIs.

Figure 2.  Percentage of Patients Developing Multiple Sclerosis or Disability Worsening
Percentage of Patients Developing Multiple Sclerosis or Disability Worsening

A, Statistics from log-rank test of ganglion cell and inner plexiform (GCIP) layer thickness for multiple sclerosis (MS) diagnosis per McDonald diagnostic criteria, 2010 revision. B, Diagnosis of MS by magnetic resonance imaging criteria only (MS-MRI). C, Diagnosis of clinically definite MS (CDMS) defined by a relapse. D, Disability worsening defined as sustained ≥1 point worsening in Expanded Disability Status Scale (EDSS) score (or ≥1.5 point score-worsening in patients with EDSS of 0 at study inclusion). Shaded regions describe the 95% CIs.

Figure 3.  Cox Proportional Hazard Models for the Association of Ganglion Cell and Inner Plexiform (GCIP) Layer Thickness With Fulfilling No Evidence Of Disease Activity–3 (NEDA-3) Criteria
Cox Proportional Hazard Models for the Association of Ganglion Cell and Inner Plexiform (GCIP) Layer Thickness With Fulfilling No Evidence Of Disease Activity–3 (NEDA-3) Criteria

Forest plots for multiple Cox proportional hazard models, with hazard ratios (circles) with 95% CIs (bars) representing the proportion of patients not meeting NEDA-3 criteria. CIS indicates clinically isolated syndrome; DMT, disease–modifying therapy (at time of event or censoring); IFT, infratentorial; ON, optic neuritis; T2w, T2-weighted.

Table 1.  Cohort Description
Cohort Description
Table 2.  Optical Coherence Tomographic Imaging Parameters for Eyes Without Optic Neuritis of Patients With Clinically Isolated Syndrome and Healthy Control Participants
Optical Coherence Tomographic Imaging Parameters for Eyes Without Optic Neuritis of Patients With Clinically Isolated Syndrome and Healthy Control Participants
1.
Compston  A, Coles  A.  Multiple sclerosis.  Lancet. 2008;372(9648):1502-1517.PubMedGoogle ScholarCrossref
2.
Wingerchuk  DM, Weinshenker  BG.  Disease modifying therapies for relapsing multiple sclerosis.  BMJ. 2016;354:i3518.PubMedGoogle ScholarCrossref
3.
Dörr  J, Paul  F.  The transition from first-line to second-line therapy in multiple sclerosis.  Curr Treat Options Neurol. 2015;17(6):354.PubMedGoogle ScholarCrossref
4.
Hemmer  B, Kerschensteiner  M, Korn  T.  Role of the innate and adaptive immune responses in the course of multiple sclerosis.  Lancet Neurol. 2015;14(4):406-419.PubMedGoogle ScholarCrossref
5.
Polman  CH, Reingold  SC, Banwell  B,  et al.  Diagnostic criteria for multiple sclerosis: 2010 revisions to the McDonald criteria.  Ann Neurol. 2011;69(2):292-302.PubMedGoogle ScholarCrossref
6.
Miller  DH, Chard  DT, Ciccarelli  O.  Clinically isolated syndromes.  Lancet Neurol. 2012;11(2):157-169.PubMedGoogle ScholarCrossref
7.
Huss  AM, Halbgebauer  S, Öckl  P,  et al.  Importance of cerebrospinal fluid analysis in the era of McDonald 2010 criteria: a German-Austrian retrospective multicenter study in patients with a clinically isolated syndrome.  J Neurol. 2016;263(12):2499-2504.PubMedGoogle ScholarCrossref
8.
Hsueh  CJ, Kao  HW, Chen  SY,  et al.  Comparison of the 2010 and 2005 versions of the McDonald MRI criteria for dissemination-in-time in Taiwanese patients with classic multiple sclerosis.  J Neurol Sci. 2013;329(1-2):51-54.PubMedGoogle ScholarCrossref
9.
Fisniku  LK, Brex  PA, Altmann  DR,  et al.  Disability and T2 MRI lesions: a 20-year follow-up of patients with relapse onset of multiple sclerosis.  Brain. 2008;131(pt 3):808-817.PubMedGoogle ScholarCrossref
10.
Novakova  L, Skoog  B, Runmarker  B,  et al.  Clinically isolated syndromes with no further disease activity suggestive of multiple sclerosis at the age of population life expectancy.  Mult Scler. 2014;20(4):496-500.PubMedGoogle ScholarCrossref
11.
Tintore  M, Rovira  À, Río  J,  et al.  Defining high, medium and low impact prognostic factors for developing multiple sclerosis.  Brain. 2015;138(pt 7):1863-1874.PubMedGoogle ScholarCrossref
12.
Kuhle  J, Disanto  G, Dobson  R,  et al.  Conversion from clinically isolated syndrome to multiple sclerosis: A large multicentre study.  Mult Scler. 2015;21(8):1013-1024.PubMedGoogle ScholarCrossref
13.
Zimmermann  H, Oberwahrenbrock  T, Brandt  AU, Paul  F, Dörr  J.  Optical coherence tomography for retinal imaging in multiple sclerosis.  Degener Neurol Neuromuscul Dis. 2014;4:153-162.Google Scholar
14.
Balk  LJ, Cruz-Herranz  A, Albrecht  P,  et al.  Timing of retinal neuronal and axonal loss in MS: a longitudinal OCT study.  J Neurol. 2016;263(7):1323-1331.PubMedGoogle ScholarCrossref
15.
Bennett  JL, de Seze  J, Lana-Peixoto  M,  et al; GJCF-ICC&BR.  Neuromyelitis optica and multiple sclerosis: Seeing differences through optical coherence tomography.  Mult Scler. 2015;21(6):678-688.PubMedGoogle ScholarCrossref
16.
Oertel  FC, Kuchling  J, Zimmermann  H,  et al.  Microstructural visual system changes in AQP4-antibody–seropositive NMOSD.  Neurol Neuroimmunol Neuroinflamm. 2017;4(3):e334.Google ScholarCrossref
17.
Costello  F, Coupland  S, Hodge  W,  et al.  Quantifying axonal loss after optic neuritis with optical coherence tomography.  Ann Neurol. 2006;59(6):963-969.PubMedGoogle ScholarCrossref
18.
Gabilondo  I, Martínez-Lapiscina  EH, Fraga-Pumar  E,  et al.  Dynamics of retinal injury after acute optic neuritis.  Ann Neurol. 2015;77(3):517-528.PubMedGoogle ScholarCrossref
19.
Saidha  S, Sotirchos  ES, Oh  J,  et al.  Relationships between retinal axonal and neuronal measures and global central nervous system pathology in multiple sclerosis.  JAMA Neurol. 2013;70(1):34-43.PubMedGoogle ScholarCrossref
20.
Saidha  S, Al-Louzi  O, Ratchford  JN,  et al.  Optical coherence tomography reflects brain atrophy in multiple sclerosis: A four-year study.  Ann Neurol. 2015;78(5):801-813.PubMedGoogle ScholarCrossref
21.
Martinez-Lapiscina  EH, Arnow  S, Wilson  JA,  et al; IMSVISUAL consortium.  Retinal thickness measured with optical coherence tomography and risk of disability worsening in multiple sclerosis: a cohort study.  Lancet Neurol. 2016;15(6):574-584.PubMedGoogle ScholarCrossref
22.
Knier  B, Schmidt  P, Aly  L,  et al.  Retinal inner nuclear layer volume reflects response to immunotherapy in multiple sclerosis.  Brain. 2016;139(11):2855-2863.PubMedGoogle ScholarCrossref
23.
Oberwahrenbrock  T, Ringelstein  M, Jentschke  S,  et al.  Retinal ganglion cell and inner plexiform layer thinning in clinically isolated syndrome.  Mult Scler. 2013;19(14):1887-1895.PubMedGoogle ScholarCrossref
24.
Knier  B, Berthele  A, Buck  D,  et al.  Optical coherence tomography indicates disease activity prior to clinical onset of central nervous system demyelination.  Mult Scler. 2016;22(7):893-900.PubMedGoogle ScholarCrossref
25.
Metz  I, Beißbarth  T, Ellenberger  D,  et al.  Serum peptide reactivities may distinguish neuromyelitis optica subgroups and multiple sclerosis.  Neurol Neuroimmunol Neuroinflamm. 2016;3(2):e204.PubMedGoogle ScholarCrossref
26.
Zamvil  SS, Slavin  AJ.  Does MOG Ig-positive AQP4-seronegative opticospinal inflammatory disease justify a diagnosis of NMO spectrum disorder?  Neurol Neuroimmunol Neuroinflamm. 2015;2(1):e62.PubMedGoogle ScholarCrossref
27.
Rotstein  DL, Healy  BC, Malik  MT, Chitnis  T, Weiner  HL.  Evaluation of no evidence of disease activity in a 7-year longitudinal multiple sclerosis cohort.  JAMA Neurol. 2015;72(2):152-158.PubMedGoogle ScholarCrossref
28.
Schmidt  P, Gaser  C, Arsic  M,  et al.  An automated tool for detection of FLAIR-hyperintense white-matter lesions in Multiple Sclerosis.  Neuroimage. 2012;59(4):3774-3783.PubMedGoogle ScholarCrossref
29.
Schemper  M, Smith  TL.  A note on quantifying follow-up in studies of failure time.  Control Clin Trials. 1996;17(4):343-346.PubMedGoogle ScholarCrossref
30.
Kappos  L, Edan  G, Freedman  MS,  et al.  The 11-year long-term follow-up study from the randomized BENEFIT CIS trial.  Neurology. 2016;87(10):978-987.Google ScholarCrossref
31.
Gabilondo  I, Martínez-Lapiscina  EH, Martínez-Heras  E,  et al.  Trans-synaptic axonal degeneration in the visual pathway in multiple sclerosis.  Ann Neurol. 2013;75(1):98-107.PubMedGoogle ScholarCrossref
32.
Sinnecker  T, Oberwahrenbrock  T, Metz  I,  et al.  Optic radiation damage in multiple sclerosis is associated with visual dysfunction and retinal thinning – an ultrahigh-field MR pilot study.  Eur Radiol. 2014;25(1):122-131.PubMedGoogle ScholarCrossref
33.
Zimmermann  H, Freing  A, Kaufhold  F,  et al.  Optic neuritis interferes with optical coherence tomography and magnetic resonance imaging correlations.  Mult Scler. 2013;19(4):443-450.PubMedGoogle ScholarCrossref
34.
Calabrese  M, Rinaldi  F, Mattisi  I,  et al.  The predictive value of gray matter atrophy in clinically isolated syndromes.  Neurology. 2011;77(3):257-263.PubMedGoogle ScholarCrossref
35.
Biberacher  V, Boucard  CC, Schmidt  P,  et al.  Atrophy and structural variability of the upper cervical cord in early multiple sclerosis.  Mult Scler. 2015;21(7):875-884.PubMedGoogle ScholarCrossref
36.
Brownlee  WJ, Altmann  DR, Da Mota  PA,  et al.  Association of asymptomatic spinal cord lesions and atrophy with disability 5 years after a clinically isolated syndrome.  Mult Scler J. 2016;23(5):665-674PubMedGoogle ScholarCrossref
37.
Filippi  M, Preziosa  P, Copetti  M,  et al.  Gray matter damage predicts the accumulation of disability 13 years later in MS.  Neurology. 2013;81(20):1759-1767.PubMedGoogle ScholarCrossref
38.
Kalincik  T, Cutter  G, Spelman  T,  et al.  Defining reliable disability outcomes in multiple sclerosis.  Brain. 2015;138(pt 11):3287-3298.PubMedGoogle ScholarCrossref
39.
Gelfand  JM, Nolan  R, Schwartz  DM, Graves  J, Green  AJ.  Microcystic macular oedema in multiple sclerosis is associated with disease severity.  Brain. 2012;135(pt 6):1786-1793.PubMedGoogle ScholarCrossref
40.
Saidha  S, Sotirchos  ES, Ibrahim  MA,  et al.  Microcystic macular oedema, thickness of the inner nuclear layer of the retina, and disease characteristics in multiple sclerosis: a retrospective study.  Lancet Neurol. 2012;11(11):963-972.PubMedGoogle ScholarCrossref
41.
Kaufhold  F, Zimmermann  H, Schneider  E,  et al.  Optic neuritis is associated with inner nuclear layer thickening and microcystic macular edema independently of multiple sclerosis.  PLoS One. 2013;8(8):e71145.PubMedGoogle ScholarCrossref
42.
Oberwahrenbrock  T, Weinhold  M, Mikolajczak  J,  et al.  Reliability of intra-retinal layer thickness estimates.  PLoS One. 2015;10(9):e0137316.PubMedGoogle ScholarCrossref
43.
Outteryck  O, Zephir  H, Defoort  S,  et al.  Optical coherence tomography in clinically isolated syndrome: no evidence of subclinical retinal axonal loss.  Arch Neurol. 2009;66(11):1373-1377.PubMedGoogle ScholarCrossref
44.
Kallenbach  K, Sander  B, Tsakiri  A,  et al.  Neither retinal nor brain atrophy can be shown in patients with isolated unilateral optic neuritis at the time of presentation.  Mult Scler. 2011;17(1):89-95.PubMedGoogle ScholarCrossref
45.
Gelfand  JM, Goodin  DS, Boscardin  WJ, Nolan  R, Cuneo  A, Green  AJ.  Retinal axonal loss begins early in the course of multiple sclerosis and is similar between progressive phenotypes.  PLoS One. 2012;7(5):e36847.PubMedGoogle ScholarCrossref
Original Investigation
September 2018

Association of Retinal Ganglion Cell Layer Thickness With Future Disease Activity in Patients With Clinically Isolated Syndrome

Author Affiliations
  • 1NeuroCure Clinical Research Center, Charité–Universitätsmedizin Berlin, corporate member of Freie Universität Berlin, Humboldt-Universität zu Berlin, and Berlin Institute of Health, Berlin, Germany
  • 2Department of Neurology, Klinikum rechts der Isar, Technische Universität München, Munich, Germany
  • 3Department of Experimental Neuroimmunology, Technische Universität München, Munich, Germany
  • 4Department of Neurology, Charité–Universitätsmedizin Berlin, corporate member of Freie Universität Berlin, Humboldt-Universität zu Berlin, and Berlin Institute of Health, Berlin, Germany
  • 5TUM Neuroimaging Center, Klinikum rechts der Isar, Technische Universität München, Munich, Germany
  • 6Experimental and Clinical Research Center, Max Delbrück Center for Molecular Medicine and Charité–Universitätsmedizin Berlin, corporate member of Freie Universität Berlin, Humboldt-Universität zu Berlin, and Berlin Institute of Health, Berlin, Germany
  • 7Munich Cluster of Systems Neurology (SyNergy), Munich, Germany
JAMA Neurol. 2018;75(9):1071-1079. doi:10.1001/jamaneurol.2018.1011
Key Points

Question  Is optical coherence tomography associated with future disease activity in patients with clinically isolated syndrome?

Findings  In this longitudinal cohort study of 97 patients with clinically isolated syndrome, ganglion cell and inner plexiform layer thickness in eyes without optic neuritis was associated with future disease activity. Patients with thickness in the thinnest tertile had a 3.3-times increased rate for not meeting the no evidence of disease activity-3 criteria compared with patients whose results were in the thickest tertile.

Meaning  Optical coherence tomography might aid in future monitoring and treatment decisions in patients with clinically isolated syndrome.

Abstract

Importance  Clinically isolated syndrome (CIS) describes a first clinical incident suggestive of multiple sclerosis (MS). Identifying patients with CIS who have a high risk of future disease activity and subsequent MS diagnosis is crucial for patient monitoring and the initiation of disease-modifying therapy.

Objective  To investigate the association of retinal optical coherence tomography (OCT) results with future disease activity in patients with CIS.

Design, Setting, and Participants  This prospective, longitudinal cohort study took place between January 2011 and May 2017 at 2 German tertiary referral centers. A total of 179 patients with CIS were screened (80 in Berlin and 99 in Munich). Patients underwent neurological examination, magnetic resonance imaging (MRI), and OCT. Only eyes with no previous optic neuritis were considered for OCT analysis.

Main Outcomes and Measures  The primary outcome was not meeting the no evidence of disease activity (NEDA-3) criteria; secondary outcomes were MS diagnosis (by the 2010 McDonald criteria) and worsening of disability. The primary measure was OCT-derived ganglion cell and inner plexiform layer thickness; the secondary measures included peripapillary retinal nerve fiber layer thickness, inner nuclear layer thickness, and MRI-derived T2-weighted lesions.

Results  A total of 97 of the 179 screened patients (54.2%) were enrolled in the study at a median of 93 (interquartile range [IQR], 62-161) days after a first demyelinating event. The median follow-up duration (Kaplan-Meier survival time) was 729 (IQR, 664-903) days. Of 97 patients with CIS (mean age 33.6 [7.9] years; 61 [62.9%] female), 58 (59%) did not meet NEDA-3 criteria during the follow-up period. A Kaplan-Meier analysis showed a significant probability difference in not meeting NEDA-3 criteria by ganglion cell and inner plexiform later thickness (thinnest vs thickest tertile: hazard ratio [HR], 3.33 [95% CI, 1.70-6.55; P < .001; log-rank P = .001). A follow-up diagnosis of MS was more likely for patients with low ganglion cell and inner plexiform layer thickness (thinnest vs thickest tertile: HR, 4.05 [95% CI, 1.93-8.50]; P < .001). Low peripapillary retinal nerve fiber layer thickness likewise indicated risk of not meeting NEDA-3 criteria (thinnest vs thickest tertile: HR, 2.46 [95% CI, 1.29-4.66]; P = .01; log-rank P = .02). Inner nuclear layer thickness and T2-weighted lesion count were not associated with not meeting NEDA-3 criteria.

Conclusions and Relevance  Retinal ganglion cell and inner plexiform layer thickness might prove a valuable imaging marker for anticipating future disease activity and diagnosis of MS in patients with CIS, which can potentially support patient monitoring and initiation of disease-modifying therapy.

Introduction

Multiple sclerosis (MS) is the most common chronic inflammatory disease of the central nervous system (CNS) and an important cause of sustained disability in young adults.1 Early disease-modifying therapy of MS can improve long-term outcome and reduce the accrual of irreversible disability, which makes timely and accurate diagnosis of MS imperative.2,3 The cause of MS is unknown,4 and in absence of a specific pathologic correlate, current diagnostic criteria rely on clinical presentation and typical MS lesions obtained in magnetic resonance imaging (MRI).5 Either way, dissemination of clinical or radiographic disease activity needs to be proven both in space and time.

The term clinically isolated syndrome (CIS) was coined to describe the monophasic attack that heralds the onset of MS in most of patients.6 In 11% to 34% of patients with CIS, a diagnosis of MS can already be established after this first event using MRI.5 In the remainder, MRI is not indicative of dissemination in both space and time, typically missing the time criterion.7,8 In these patients, a diagnosis of MS cannot yet be established, but more than 80% develop new lesions and more than 60% experience a second clinical attack within 20 years and are then diagnosed with MS.9 Of the total population of patients with CIS, 15% to 20% never have a second clinical attack.10 Known risk factors of future disease activity and a subsequent MS diagnosis in CIS include young age, intrathecal immunoglobulin synthesis, and the presence of brain lesions,11 particularly in the infratentorial region.6,12

Optical coherence tomography (OCT) is a noninvasive tool used to quantify retinal neuroaxonal damage in MS and other neurologic diseases.13-16 Peripapillary retinal nerve fiber layer (pRNFL) and ganglion cell layer thinning occurs in patients with MS after optic neuritis (ON)17,18 but also independently of ON.19,20 In a retrospective study, pRNFL was associated with disability worsening in MS.21 Another study suggests that inner nuclear layer thickness (INL) may be useful for monitoring disease-modifying therapy in the following 12 months.22 In this study, we investigate the value of retinal ganglion cell and inner plexiform layer thickness (GCIP) as a novel parameter of neuronal retinal damage to assess subsequent disease activity in CIS.

Methods
Study Design

Patients were prospectively included from 2 ongoing observational cohort studies at the university hospitals in Berlin and Munich (Cohort Study of Clinically Isolated Syndrome and Early Multiple Sclerosis [CIS-COHORT; ClinicalTrials.gov Identifier: NCT01371071] in Berlin and TUM-MS cohort in Munich) of patients with CIS and early MS who were recruited between January 2011 and September 2015 (in Berlin, from January 2011 to August 2014; in Munich, from August 2013 to September 2015) and followed up until May 2017. Inclusion criteria were (1) an age between 18 and 60 years; (2) a first clinical event suggestive of MS in the previous 365 days after extensive diagnostic workup for alternative causes, and (3) an observation period of at least 365 days after study inclusion. Details on data collection and data screening can be found in the eMethods in the Supplement. The study was censored after a maximal follow-up time of 2.5 years.

The study was approved by the local ethics committees at Charité-Universitätsmedizin Berlin and the Technical University of Munich and was conducted in accordance with the Declaration of Helsinki in its currently applicable version and applicable German laws. All participants gave informed written consent. Fourteen Berlin patients and 12 Munich patients had been included in prior cross-sectional OCT studies.23,24

Exclusion criteria were ON in both eyes, fulfillment of the 2010 revised McDonald diagnostic criteria for MS5 at study inclusion, a diagnosis of other neuroinflammatory disorders (ie, neuromyelitis optica spectrum disorders),25,26 any ophthalmologic comorbidity or refractive error of more than 6 diopters, and a period of less than 30 days between initial event and study inclusion to avoid inclusion of disability scores obtained in the acute disease phase.

Assessments at study inclusion comprised neurological examination including the Expanded Disability Status Scale (EDSS) score, brain MRI at 3 T with quantification of T2-weighted (T2w) lesions and gadolinium-enhancing lesions and retinal OCT. All examinations were completed within 2 weeks of study enrollment. Follow-up visits were performed at least yearly. In Berlin, follow-up visits were carried out in 12-month intervals from the initial event and included evaluation of relapses, EDSS scores, and MRI. In Munich, follow-up visits including MRI were carried out every 12 months, while evaluation of relapses and EDSS scores occurred more frequently (every 3-6 months, depending on patient presentation).

The primary outcome parameter was not meeting the no evidence of disease activity (NEDA-3) criteria, defined as freedom from relapses, absence of sustained disability worsening, and stable radiographic parameters (no change in T2w lesion load and absence of gadolinium-enhancing lesions on brain MRI).27 Additional outcome was MS diagnosis according to the 2010 revision of McDonald criteria (McDonald MS).5 Separate analyses of diagnosis were conducted for MRI measures indicated MS (eg, T2w lesion load, occurrence of gadolinium-enhancing lesions) (MRI-MS) or clinical definite MS (CDMS), defined by a second clinical relapse. Finally, disability worsening was analyzed as a further outcome parameter, defined as an increase in the EDSS score of 1.0 point or more or an increase of 1.5 points or more in patients with an EDSS score of 0 at study inclusion, confirmed by a visit at least 3 months later. Notably, a diagnosis of MRI-MS is synonymous with the development of new lesions in our cohort.

Magnetic Resonance Imaging

The MRI sequences are described in detail in the eMethods of the Supplement. The T2w lesion count was assessed by the lesion prediction algorithm (Berlin) or the lesion growth algorithm (Munich) as implemented in the Lesion Segmentation Tool toolbox, version 2.0.15 for Statistical Parametric Mapping software (Wellcome Trust Centre for Neuroimaging).28 Different lesion segmentation algorithms were used because each center chose the method delivering the best results for their scans based on expert visual inspection.

Optical Coherence Tomography

All participants underwent retinal OCT examination with assessment of the pRNFL and macular GCIP and INL. Details of OCT assessments are described in the eMethods in the Supplement. Seven macular scans of 7 patients and 3 peripapillary ring scans of 3 patients were excluded because of insufficient quality, and 1 eye was excluded because of severe myopia. Only non-ON eyes were included in the analyses (n = 161). A total of 90 healthy control participants were selected from the databases of both institutions (55 from Berlin and 35 from Munich) to provide reference values for OCT images.

Statistical Analysis

Results are presented as either mean (SD) or median with interquartile range (IQR). Characteristics at study inclusion were analyzed using Mann-Whitney U test and Pearson χ2. Differences in retinal layer thickness between patients with CIS and control participants were analyzed using generalized estimation equation models with the correlation matrix “exchangeable” to account for intereye and intrapatient dependencies. To determine the risk of meeting the respective outcomes parameters associated with retinal thickness, Kaplan-Meier estimates and unadjusted Cox regression models were used. We chose tertile dividers for OCT parameters of non-ON eyes and compared risks associated with GCIP thickness values in the lowest or intermediate tertile vs the highest tertile.21 If both eyes of a patient were non-ON eyes, the mean value of both eyes was used. The Kruskal-Wallis rank sum test was used for analyzing differences in time between initial event and study inclusion between the OCT tertile groups. Median follow-up time for censored data was calculated using reverse Kaplan-Meier estimates.29 We then used a multiple Cox proportional hazard regression model including variables that had been identified as risk factors in other CIS cohorts.11 In addition to GCIP, age (median divider), sex, ON as presenting syndrome, presence of 1 or more infratentorial lesion, T2w lesion count (tertile divider), and disease-modifying treatment were included as covariates. We tested the proportional hazards assumption for all variables of the unadjusted and multiple Cox regression models. The assumption was met for all variables; however, the status of disease-modifying treatment can change over time. Therefore, we included this variable as a time-varying covariate, with patients categorized as taking treatment at the time of the visit or not taking treatment at the time of the visit in the multiple Cox model.

We report the hazard ratio (HR) with 95% CIs and the model’s accuracy with Harrell C statistics. Statistical significance was established at P < .05. All analyses were performed in R, version 3.4.2 (R Foundation for Statistical Computing). Data analysis occurred from May to July 2017.

Results
Characteristics at Study Inclusion

Within both centers, 179 patients with CIS were screened for eligibility for the current study (eMethods in the Supplement). Ninety-seven patients with CIS (45 in Berlin and 52 in Munich) met all inclusion criteria and were included (Table 1). There were no significant differences in age and sex between the patients (mean [SD] age, 33.6 [7.9] years; 61 of 97 [63%] women) and healthy control participants (mean [SD] age, 34.6 [8.7] years; P = .46; 62 of 90 [69%] were women; χ21, 1.24; P = .27). Thirty-three eyes were excluded because ON was the presenting clinical symptom. The GCIP and pRNFL thickness measurements of eyes without ON in patients with CIS were in the range of the matched healthy control participants, while the INL was slightly thicker (Table 2).

To assess the value of each OCT parameter at inclusion as a potential risk factor for meeting the respective outcome parameters, we grouped patients into tertiles according to the thickness of GCIP, pRNFL, or INL thicknesses of non-ON eyes at the time of study enrollment (Table 2).

To allow a comparison of our cohort with other cohorts of patients with CIS, we report a comparison of baseline characteristics and events at follow-up in our cohort with 3 independent cohorts from earlier publications11,12,30 (eTable 1 and 2 in the Supplement). While our cohort was comparable with the other cohorts regarding age distribution, sex distribution, disability, and ON frequency, it shows a higher frequency of oligoclonal bands (86 of 97 patients [89%] in this cohort vs 57.2%11 and 74.3%12 in other cohorts; eTable 1 in the Supplement) and more T2w lesions in comparison with 1 study that detected T2w lesions as a risk factor for diagnosis of CDMS.11 Time between CIS onset and study enrollment was longer in our cohort (median [IQR], 93 [61-161] days) compared with the other studies (fewer than 90 days).

Center Differences

There were significant differences in patient characteristics at study inclusion between the centers regarding the time between initial event and study inclusion (Berlin: median [IQR], 122 [86-163] days; Munich: 69.5 [51.75-124.75]; P = .002 days), EDSS score (Berlin: median [IQR], 1.5 [1.0-2.0]; Munich: 0.5 [0.0-1.5]; P = .01), and T2w lesion count (Berlin: median [IQR], 5.5 [1.75, 11]; Munich: 10 [6-18]; P = .01). There were no differences in retinal layer thicknesses between the 2 centers (eTable 3 in the Supplement).

Follow-up

Of 97 patients with CIS, 58 (60%) exhibited disease activity during follow-up, defined as not meeting NEDA-3 criteria. Furthermore, 53 (55%) were diagnosed with McDonald MS; of those, 36 (37%) fulfilled MRI-MS criteria, and 24 (24%) fulfilled CDMS criteria by experiencing a relapse. Eleven patients (11%) exhibited sustained worsening of EDSS score. The 3 patients who did not meet NEDA-3 criteria but did not fulfill diagnostic criteria for MS had experienced EDSS score worsening within the functional system. The median (IQR) duration of clinical follow-up (in Kaplan-Meier analysis, NEDA-3 survival time) was 729 days (664–903) days (censoring occurred either when NEDA-3 status was not met or the end of the clinical follow-up period, whichever came first). There were no differences in follow-up time between both centers. Also, time between initial event and study inclusion prior to OCT examination and follow-up duration between the OCT parameter tertile groups were not different for all retinal layers. Five patients from both centers declined further participation in the study after at least 2 years of follow-up (eMethods in the Supplement).

Factors Associated With Disease Activity, MS Diagnosis, and Disability Worsening

Kaplan-Meier curves revealed a significant increase in the rate of patients with lower GCIP thickness not meeting NEDA-3 criteria (lowest [58.7-69.2 μm], 33 of 97; 34%; intermediate [69.3-74.1 μm], 32 of 97; 33.0%; highest [74.2-84.8 μm]: 32 of 97; 33%; intermediate vs thickest tertile: HR, 2.13 [95% CI, 1.06-4.03]; P = .03; thinnest vs thickest tertile: HR, 3.33 [95% CI, 1.70-6.55; P < .001; Figure 1A). This was also the case for patients with lower pRNFL thickness (lowest tertile [72.5-95.9 μm], 36 of 97; 37%; intermediate [96.0-103.9 μm], 31 of 97; 32%; highest [104.0-126.0 μm], 30 of 97; 31%; intermediate vs thickest tertile: HR, 1.80 [95% CI, 0.924-3.52]; P = .08; thinnest vs thickest tertile: HR, 2.47 [95% CI, 1.29-4.66]; P = .01; Figure 1B) but not for patients with lower INL thickness (Figure 1C). For additional outcome parameters, Kaplan-Meier curves were performed for GCIP thickness only, because it was the primary outcome. Patients with lower GCIP thicknesses had significantly increased rates of a subsequent MS diagnosis by meeting the McDonald 2010 diagnostic criteria (lowest [58.7-69.2 μm], 33 of 97; 33%; intermediate [69.3-74.1 μm], 32 of 97; 33%; highest [74.2-84.8 μm]: 32 of 97; 33%; intermediate vs thickest tertile: HR, 2.69 [95% CI, 1.25-5.78]; P = .01; thinnest vs thickest tertile: HR, 4.05 [95% CI, 1.93-8.50]; P < .001; Figure 2A) and by developing MRI-MS (lowest [58.7-69.2 μm], 33 of 97; 34%; intermediate [69.3-74.1 μm], 32 of 97; 33%; highest [74.2-84.8 μm]: 32 of 97; 33%; intermediate vs thickest tertile: HR, 2.21 [95% CI, 0.90-5.40]; P = .083; thinnest vs thickest tertile: HR, 3.67 [95% CI, 1.56-8.60]; P = .003; Figure 2B).

Rates for diagnosis of CDMS and EDSS score worsening were not significantly different between the GCIP tertile groups (Figure 2C and D). Notably, these analyses were of limited feasibility because of the low number of events.

We then fit a multiple Cox proportional hazard model to further investigate the association of GCIP thickness and other parameters with not meeting NEDA-3 (Figure 3). Only complete data sets were included in the Cox model (n = 89). Unadjusted analysis for each parameters are reported in eTable 4 and eFigure in the Supplement. The distribution of the GCIP tertiles remained the same for the 89 patients with complete data sets. Compared with patients with GCIP thickness in the highest tertile at study enrollment, patients in the intermediate tertile had a statistically nonsignificant trend toward an increased rate of not meeting NEDA-3 (HR, 1.84; 95% CI, 0.85-3.97; P = .12). Patients with a GCIP thickness at inclusion in the lowest tertile had 3.43-fold higher rate of not meeting NEDA-3 criteria compared with the group with the highest GCIP thickness (HR, 3.43; 95% CI, 1.63-7.18; P = .001).

The rate of not meeting NEDA-3 criteria was neither influenced by sex nor age. Patients with ON as their initial symptom had a nonsignificantly lower rate (HR, 0.59; 95% CI, 0.32-1.11; P = .10) of this outcome compared with patients with clinical manifestations in other functional systems. Notably, the presence of at least 1 infratentorial lesion and the number of T2w lesions did not change the rate of not meeting NEDA-3 criteria (Figure 3).

The accuracy of the Cox model was moderate (Harrell C [SE], 0.686 [0.043]). When fitting the same Cox model with continuous GCIP thickness instead of tertiles, a 1 μm increase in GCIP thickness was associated with an approximately 9% reduction in rate of not meeting NEDA-3 criteria (adjusted HR, 0.91; 95% CI, 0.86-0.96; P < .001; Harrell C [SE], 0.685 [0.043]).

To exclude potential confounders, the same models were additionally computed with the covariates time between initial event and study inclusion and center. Neither was significantly associated with not meeting NEDA-3 criteria, nor did including these variables change the results for the association of GCIP with not meeting NEDA-3. Separate analyses (unadjusted models) for both centers showed comparable results (eTable 5 in the Supplement). Oligoclonal bands were not included in the multiple Cox model since oligoclonal bands were absent in only 9 patients; of those, 4 did not meet the NEDA-3 criteria.

Discussion

Retinal GCIP thickness measured by OCT within 12 months after the first clinical event in eyes without former ON is a prognostic factor for subsequent disease activity in patients with CIS. Values of GCIP thickness less than 69.2 μm were a risk factor for not meeting NEDA-3 criteria (at more than a 3-fold increased rate) and a subsequent diagnosis of MS. Conversely, higher GCIP thickness was associated with a higher probability of remaining free from disease activity and of retaining CIS status.

Loss of retinal ganglion cells indicated by GCIP thinning on OCT scan is a frequent finding in patients with MS. The GCIP thickness is reduced by a mean of 11 μm within 6 months after an optic neuritis attack.18 However, retinal ganglion cell loss can also be identified in eyes from patients with MS without prior ON, with retrograde axonal degeneration resulting from subclinical ON,18,23 primary MS–linked retinal neurodegeneration, or transsynaptic degeneration because of central nervous system atrophy as putative causes.31,32 The GCIP thickness in eyes without former ON correlates with cortical gray matter volume19,33 and longitudinal GCIP loss reflects whole brain, gray matter, white matter, and thalamic atrophy.20 Gray matter34 and spinal cord atrophy35 are prevalent in CIS and are associated with disease activity and worsening.34,36,37

As with low GCIP, low pRNFL thickness was associated with an increased rate of subsequent disease activity defined as not meeting the NEDA-3 criteria. A large retrospective multicenter study of 879 patients showed that low pRNFL is a risk factor for sustained EDSS score worsening during a 5-year follow-up period.21 This study also included a subgroup with CIS (n = 74), in which no such association was observed, presumably owing to the low sample size of patients with a sufficiently long duration of observation. Likewise, our data preclude meaningful conclusions with regards to a possible association of OCT measurements with disability worsening, given that we observed only a few events of EDSS score worsening, and owing to the recent observation that short-term follow up assessments may overestimate progression rates.38 Thus, our study confirms in an independent cohort that retinal layer thicknesses do not suggest subsequent EDSS score worsening in patients with CIS. It remains elusive whether this is because both studies were underpowered or whether OCT is just not useful to assess the future possibility of disability worsening in CIS.

The INL thickness has been connected to inflammatory patterns of MS disease activity. Thicker INL and microcystic INL edema are associated with increased paraclinical tissue inflammation in MRI and higher relapse rates.24,39-41 The thickening of INL correlates with larger white matter lesion volumes, and sufficient immunotherapy has been shown to reduce INL volumes in patients with MS.22 In contrast to GCIP and pRNFL thickness, INL measures in our study were different from controls at inclusion but had no association with disease activity. One possible explanation might be that changes to INL induced by disease-modifying therapy confound analysis.22

In our study, not meeting NEDA-3 criteria was largely synonymous with fulfilling the diagnostic criteria of MS according to the 2010 revision of the McDonald criteria.5 Only 3 patients did not meet NEDA-3 but were not diagnosed with MS. In each case, not meeting NEDA-3 was caused by disability worsening, in form of a sustained EDSS score increase but without dissemination of lesions over time.

Several other risk factors for MS conversion and disease activity in patients with CIS have been previously described. A large prospective cohort study on 1015 patients with CIS and a mean follow-up duration of 81 months identified the presence of oligoclonal bands in the cerebrospinal fluid and high lesion counts in cerebral MRI as strong risk factors for progression into CDMS.11 In this study, we did not find an association of T2w lesion counts or presence of infratentorial lesions with subsequent disease activity. This is surprising, given the strong associative power of T2w lesions in previous studies.11,12 The difference presumably stems from cohort differences. Both centers in our study are tertiary referral centers, and our sample included mostly patients in whom the primary neurologist had already determined a potential risk of MS. For example, in the study by Tintore et al,11 31.4% of all patients with CIS had no T2w lesions, whereas this was only the case for 10.1% of the patients in our study. Furthermore, we accepted an interval of as long as 365 days between CIS onset and study enrollment, which was long compared with other studies.11,12 This might have created a potential selection bias toward more benign patients with risk of MS, since patients at higher risk may have already converted to MS during that period. However, despite significant center differences between the patients from Munich and Berlin at study inclusion, these differences had no impact on the associative power of GCIP or pRNFL, which further supports the generalizability of our results. Another explanation could be that the number of variables in the multiple Cox regression may have overstressed the model in light of the limited sample size and number of events. However, a multivariate approach is required to confirm the effect of retinal layer thicknesses and determine their weight in meeting the outcomes. This is why we maintained the multiple Cox model but performed all analyses additionally in an unadjusted manner and reported these in eTables 4 and 5 in the Supplement.

Limitations

An important disadvantage of OCT is that measurements are not easily comparable between devices. While pRNFL analysis has a high level of standardization and reference values are available in many OCT devices, macular intraretinal layer analysis lacks this standardization,42 and we therefore included a reference data set of measurements from healthy control participants. Despite their power, GCIP and pRNFL in patients were comparable with the control participants, whereas INL was mildly increased. Thus, the underlying mechanism behind the association of GCIP and pRNFL with progression to MS remains elusive. In line with our findings, some previous studies reported no evidence of pRNFL reduction in non-ON eyes in patients with CIS when compared with healthy control participants.43,44 Conversely, other studies found pRNFL loss compared with healthy control participants45 or reported a subtle GCIP reduction in patients with CIS or early MS mostly in the form of a focal thickness reduction.23 Moreover, advanced OCT technology might thus be able to unveil subtle changes missed by quantitation of GCIP and pRNFL, where acquisition protocols determine a mean thickness over their respective measurement regions. The GCIP differences between the tertiles were sufficiently large in comparison with the coefficient of repeatability of less than 1 μm,42 which supports the potential clinical utility of GCIP measurements in the suggested context.

These limitations limit the generalizability of our study. It is important to acknowledge that associations on a group level do not necessarily translate into an individual prognosis for a single patient. Ultimately, population-based studies with clear clinically derived inclusion criteria are needed to transform associations on a group level into an individual assessment measure.

Conclusions

In summary, we identified GCIP thickness and to a lesser extent pRNFL as retinal imaging biomarkers for subsequent disease activity in patients with CIS. When obtained routinely during initial workup for suspected central nervous system–demyelinating disease, they can inform early appraisal of future disease activity in patients with CIS. In clinical research, they can support risk-adjusted patient stratification. In clinical care, they can potentially help identifying patients with CIS with worse prognoses and might thus support a risk-adjusted treatment initiation.6

Back to top
Article Information

Corresponding Author: Alexander U. Brandt, MD, NeuroCure Clinical Research Center, Charité–Universitätsmedizin Berlin, Charitéplatz 1, 10117 Berlin, Germany (alexander.brandt@charite.de).

Accepted for Publication: February 23, 2018.

Published Online: April 23, 2018. doi:10.1001/jamaneurol.2018.1011

Author Contributions: Drs Zimmermann and Knier had full access to all the data in the study and take responsibility for the integrity of the data and the accuracy of the data analysis. Drs Zimmermann and Knier share status as co-first authors, and Drs Paul and Brandt share status as co-last authors.

Study concept and design: Zimmermann, Knier, Ruprecht, Hemmer, Korn, Paul, Brandt.

Acquisition, analysis, or interpretation of data: Zimmermann, Knier, Oberwahrenbrock, Behrens, Pfuhl, Aly, Kaminski, Hoshi, Specovius, Gieß, Scheel, Mühlau, Bellmann-Strobl, Ruprecht, Hemmer, Korn, Paul, Brandt.

Drafting the manuscript: Zimmermann, Knier, Oberwahrenbrock, Korn, Paul, Brandt.

Critical revision of the manuscript for important intelellectual content: Behrens, Pfuhl, Aly, Kaminski, Hoshi, Specovius, Gieß, Scheel, Mühlau, Bellmann-Strobl, Ruprecht, Hemmer.

Statistical analysis: Oberwahrenbrock.

Obtained funding: Ruprecht, Hemmer, Korn, Paul, Brandt.

Administrative, technical, or material support: Zimmermann, Knier, Mehrens, Pfuhl, Aly, Kaminsky, Hoshi, Specovius, Gieß, Scheel, Mühlau, Bellmann-Strobl.

Study supervision: Ruprecht, Hemmer, Korn, Paul, Brandt.

Conflict of Interest Disclosures: Dr Zimmermann reports receiving speaking fees from Teva Pharmaceuticals outside the submitted work. Dr Knier reports receiving grants from Kompetenznetz Multiple Sklerose and KKF Technical University of Munich. Dr Oberwahrenbrock reports receiving speaking fees from Teva and Bayer Healthcare. Dr Aly reports receiving grants from Synergy Clinical Scientist program. Dr Hoshi reports receiving grants from Bayer Health Care and personal fees from Novartis and Biogen Idec. Dr Mühlau reports receiving grants from Merck and Novartis. Dr Bellmann-Strobl reports receiving nonfinancial support from Biogen and Bayer and personal fees and nonfinancial support from Teva. Dr Ruprecht reports receiving grants and personal fees from Novartis and personal fees from Bayer Healthcare, Biogen Idec, Merck Serono, Sanofi-Aventis/Genzyme, and Teva Pharmaceuticals. Dr Hemmer reports serving on scientific advisory boards for Novartis, Bayer Schering, F. Hoffmann-La Roche Ltd, Merck Serono, Biogen Idec, and Genentech and having received research support from Chugai Pharmaceuticals, F. Hoffmann-La Roche Ltd, and 5Prime. Dr Hemmer has filed patents for the detection of antibodies and T cells against KIR4.1 in a subpopulation of patients with MS and genetic determinants of neutralizing antibodies to interferon-beta, and has received honoraria for lectures from Teva Pharmaceutical Industries Ltd, F. Hoffmann-La Roche Ltd, Biogen Idec, Novartis, and Medimmune. Dr Paul reports receiving personal fees from Bayer, Teva, Genzyme, Merck, MedImmune, and Novartis. Dr Brandt reports receiving personal fees from Motognosis, Biogen, Teva, Novartis, Bayer, and Nexus. No other disclosures were reported.

Funding/Support: This work is supported by the Kompetenznetz Multiple Sklerose (Drs Knier, Hemmer, Korn, Ruprecht, and Paul), intramural funding from the Technical University Munich (Dr Knier), BMWI EXIST (grant 03EFEBE079, Drs Oberwahrenbrock and Brandt), the German Research Foundation Clinical Scientist Program (Dr Aly), German Research Foundation (grant DFG CRC-TR 128, Dr Hemmer; grants CRC 1054 and CRC-TR 128, Dr Korn; and DFG Exc 257, Dr Paul), the European Research Council ERC (grant CoG 647215, Dr Korn), and the German Federal Ministry of Education and Research (grant N2-ADVISIMS, Dr Paul).

Role of the Funder/Sponsor: The funders had no role in the design and conduct of the study; collection, management, analysis, and interpretation of the data; preparation, review, or approval of the manuscript; and decision to submit the manuscript for publication.

Meeting Presentation: This paper was presented at the American Academy of Neurology (AAN) 2018 Annual Meeting; April 23, 2018; Los Angeles, California.

Additional Contributions: We thank Göran Joost-Zentgraf, technical assistant, Department of Neurology, Technische Universität München, and Janice Schneider, technical assistant, Department of Neurology, Technische Universität München, for OCT support, Claudia Chien, MSc, NeuroCure Clinical Research Center, Charité-Universitätsmedizin Berlin, Jan Kirschke, MD, Department of Neuroradiology, Technische Universität München, and Viola Biberacher, MD, Department of Neurology, Technische Universität München, for MRI support, and Alice Schneider, Dipl Biomath, Institute of Biometry and Clinical Epidemiology, Charité-Universitätsmedizin Berlin, for statistics consulting. They were not compensated for these contributions.

References
1.
Compston  A, Coles  A.  Multiple sclerosis.  Lancet. 2008;372(9648):1502-1517.PubMedGoogle ScholarCrossref
2.
Wingerchuk  DM, Weinshenker  BG.  Disease modifying therapies for relapsing multiple sclerosis.  BMJ. 2016;354:i3518.PubMedGoogle ScholarCrossref
3.
Dörr  J, Paul  F.  The transition from first-line to second-line therapy in multiple sclerosis.  Curr Treat Options Neurol. 2015;17(6):354.PubMedGoogle ScholarCrossref
4.
Hemmer  B, Kerschensteiner  M, Korn  T.  Role of the innate and adaptive immune responses in the course of multiple sclerosis.  Lancet Neurol. 2015;14(4):406-419.PubMedGoogle ScholarCrossref
5.
Polman  CH, Reingold  SC, Banwell  B,  et al.  Diagnostic criteria for multiple sclerosis: 2010 revisions to the McDonald criteria.  Ann Neurol. 2011;69(2):292-302.PubMedGoogle ScholarCrossref
6.
Miller  DH, Chard  DT, Ciccarelli  O.  Clinically isolated syndromes.  Lancet Neurol. 2012;11(2):157-169.PubMedGoogle ScholarCrossref
7.
Huss  AM, Halbgebauer  S, Öckl  P,  et al.  Importance of cerebrospinal fluid analysis in the era of McDonald 2010 criteria: a German-Austrian retrospective multicenter study in patients with a clinically isolated syndrome.  J Neurol. 2016;263(12):2499-2504.PubMedGoogle ScholarCrossref
8.
Hsueh  CJ, Kao  HW, Chen  SY,  et al.  Comparison of the 2010 and 2005 versions of the McDonald MRI criteria for dissemination-in-time in Taiwanese patients with classic multiple sclerosis.  J Neurol Sci. 2013;329(1-2):51-54.PubMedGoogle ScholarCrossref
9.
Fisniku  LK, Brex  PA, Altmann  DR,  et al.  Disability and T2 MRI lesions: a 20-year follow-up of patients with relapse onset of multiple sclerosis.  Brain. 2008;131(pt 3):808-817.PubMedGoogle ScholarCrossref
10.
Novakova  L, Skoog  B, Runmarker  B,  et al.  Clinically isolated syndromes with no further disease activity suggestive of multiple sclerosis at the age of population life expectancy.  Mult Scler. 2014;20(4):496-500.PubMedGoogle ScholarCrossref
11.
Tintore  M, Rovira  À, Río  J,  et al.  Defining high, medium and low impact prognostic factors for developing multiple sclerosis.  Brain. 2015;138(pt 7):1863-1874.PubMedGoogle ScholarCrossref
12.
Kuhle  J, Disanto  G, Dobson  R,  et al.  Conversion from clinically isolated syndrome to multiple sclerosis: A large multicentre study.  Mult Scler. 2015;21(8):1013-1024.PubMedGoogle ScholarCrossref
13.
Zimmermann  H, Oberwahrenbrock  T, Brandt  AU, Paul  F, Dörr  J.  Optical coherence tomography for retinal imaging in multiple sclerosis.  Degener Neurol Neuromuscul Dis. 2014;4:153-162.Google Scholar
14.
Balk  LJ, Cruz-Herranz  A, Albrecht  P,  et al.  Timing of retinal neuronal and axonal loss in MS: a longitudinal OCT study.  J Neurol. 2016;263(7):1323-1331.PubMedGoogle ScholarCrossref
15.
Bennett  JL, de Seze  J, Lana-Peixoto  M,  et al; GJCF-ICC&BR.  Neuromyelitis optica and multiple sclerosis: Seeing differences through optical coherence tomography.  Mult Scler. 2015;21(6):678-688.PubMedGoogle ScholarCrossref
16.
Oertel  FC, Kuchling  J, Zimmermann  H,  et al.  Microstructural visual system changes in AQP4-antibody–seropositive NMOSD.  Neurol Neuroimmunol Neuroinflamm. 2017;4(3):e334.Google ScholarCrossref
17.
Costello  F, Coupland  S, Hodge  W,  et al.  Quantifying axonal loss after optic neuritis with optical coherence tomography.  Ann Neurol. 2006;59(6):963-969.PubMedGoogle ScholarCrossref
18.
Gabilondo  I, Martínez-Lapiscina  EH, Fraga-Pumar  E,  et al.  Dynamics of retinal injury after acute optic neuritis.  Ann Neurol. 2015;77(3):517-528.PubMedGoogle ScholarCrossref
19.
Saidha  S, Sotirchos  ES, Oh  J,  et al.  Relationships between retinal axonal and neuronal measures and global central nervous system pathology in multiple sclerosis.  JAMA Neurol. 2013;70(1):34-43.PubMedGoogle ScholarCrossref
20.
Saidha  S, Al-Louzi  O, Ratchford  JN,  et al.  Optical coherence tomography reflects brain atrophy in multiple sclerosis: A four-year study.  Ann Neurol. 2015;78(5):801-813.PubMedGoogle ScholarCrossref
21.
Martinez-Lapiscina  EH, Arnow  S, Wilson  JA,  et al; IMSVISUAL consortium.  Retinal thickness measured with optical coherence tomography and risk of disability worsening in multiple sclerosis: a cohort study.  Lancet Neurol. 2016;15(6):574-584.PubMedGoogle ScholarCrossref
22.
Knier  B, Schmidt  P, Aly  L,  et al.  Retinal inner nuclear layer volume reflects response to immunotherapy in multiple sclerosis.  Brain. 2016;139(11):2855-2863.PubMedGoogle ScholarCrossref
23.
Oberwahrenbrock  T, Ringelstein  M, Jentschke  S,  et al.  Retinal ganglion cell and inner plexiform layer thinning in clinically isolated syndrome.  Mult Scler. 2013;19(14):1887-1895.PubMedGoogle ScholarCrossref
24.
Knier  B, Berthele  A, Buck  D,  et al.  Optical coherence tomography indicates disease activity prior to clinical onset of central nervous system demyelination.  Mult Scler. 2016;22(7):893-900.PubMedGoogle ScholarCrossref
25.
Metz  I, Beißbarth  T, Ellenberger  D,  et al.  Serum peptide reactivities may distinguish neuromyelitis optica subgroups and multiple sclerosis.  Neurol Neuroimmunol Neuroinflamm. 2016;3(2):e204.PubMedGoogle ScholarCrossref
26.
Zamvil  SS, Slavin  AJ.  Does MOG Ig-positive AQP4-seronegative opticospinal inflammatory disease justify a diagnosis of NMO spectrum disorder?  Neurol Neuroimmunol Neuroinflamm. 2015;2(1):e62.PubMedGoogle ScholarCrossref
27.
Rotstein  DL, Healy  BC, Malik  MT, Chitnis  T, Weiner  HL.  Evaluation of no evidence of disease activity in a 7-year longitudinal multiple sclerosis cohort.  JAMA Neurol. 2015;72(2):152-158.PubMedGoogle ScholarCrossref
28.
Schmidt  P, Gaser  C, Arsic  M,  et al.  An automated tool for detection of FLAIR-hyperintense white-matter lesions in Multiple Sclerosis.  Neuroimage. 2012;59(4):3774-3783.PubMedGoogle ScholarCrossref
29.
Schemper  M, Smith  TL.  A note on quantifying follow-up in studies of failure time.  Control Clin Trials. 1996;17(4):343-346.PubMedGoogle ScholarCrossref
30.
Kappos  L, Edan  G, Freedman  MS,  et al.  The 11-year long-term follow-up study from the randomized BENEFIT CIS trial.  Neurology. 2016;87(10):978-987.Google ScholarCrossref
31.
Gabilondo  I, Martínez-Lapiscina  EH, Martínez-Heras  E,  et al.  Trans-synaptic axonal degeneration in the visual pathway in multiple sclerosis.  Ann Neurol. 2013;75(1):98-107.PubMedGoogle ScholarCrossref
32.
Sinnecker  T, Oberwahrenbrock  T, Metz  I,  et al.  Optic radiation damage in multiple sclerosis is associated with visual dysfunction and retinal thinning – an ultrahigh-field MR pilot study.  Eur Radiol. 2014;25(1):122-131.PubMedGoogle ScholarCrossref
33.
Zimmermann  H, Freing  A, Kaufhold  F,  et al.  Optic neuritis interferes with optical coherence tomography and magnetic resonance imaging correlations.  Mult Scler. 2013;19(4):443-450.PubMedGoogle ScholarCrossref
34.
Calabrese  M, Rinaldi  F, Mattisi  I,  et al.  The predictive value of gray matter atrophy in clinically isolated syndromes.  Neurology. 2011;77(3):257-263.PubMedGoogle ScholarCrossref
35.
Biberacher  V, Boucard  CC, Schmidt  P,  et al.  Atrophy and structural variability of the upper cervical cord in early multiple sclerosis.  Mult Scler. 2015;21(7):875-884.PubMedGoogle ScholarCrossref
36.
Brownlee  WJ, Altmann  DR, Da Mota  PA,  et al.  Association of asymptomatic spinal cord lesions and atrophy with disability 5 years after a clinically isolated syndrome.  Mult Scler J. 2016;23(5):665-674PubMedGoogle ScholarCrossref
37.
Filippi  M, Preziosa  P, Copetti  M,  et al.  Gray matter damage predicts the accumulation of disability 13 years later in MS.  Neurology. 2013;81(20):1759-1767.PubMedGoogle ScholarCrossref
38.
Kalincik  T, Cutter  G, Spelman  T,  et al.  Defining reliable disability outcomes in multiple sclerosis.  Brain. 2015;138(pt 11):3287-3298.PubMedGoogle ScholarCrossref
39.
Gelfand  JM, Nolan  R, Schwartz  DM, Graves  J, Green  AJ.  Microcystic macular oedema in multiple sclerosis is associated with disease severity.  Brain. 2012;135(pt 6):1786-1793.PubMedGoogle ScholarCrossref
40.
Saidha  S, Sotirchos  ES, Ibrahim  MA,  et al.  Microcystic macular oedema, thickness of the inner nuclear layer of the retina, and disease characteristics in multiple sclerosis: a retrospective study.  Lancet Neurol. 2012;11(11):963-972.PubMedGoogle ScholarCrossref
41.
Kaufhold  F, Zimmermann  H, Schneider  E,  et al.  Optic neuritis is associated with inner nuclear layer thickening and microcystic macular edema independently of multiple sclerosis.  PLoS One. 2013;8(8):e71145.PubMedGoogle ScholarCrossref
42.
Oberwahrenbrock  T, Weinhold  M, Mikolajczak  J,  et al.  Reliability of intra-retinal layer thickness estimates.  PLoS One. 2015;10(9):e0137316.PubMedGoogle ScholarCrossref
43.
Outteryck  O, Zephir  H, Defoort  S,  et al.  Optical coherence tomography in clinically isolated syndrome: no evidence of subclinical retinal axonal loss.  Arch Neurol. 2009;66(11):1373-1377.PubMedGoogle ScholarCrossref
44.
Kallenbach  K, Sander  B, Tsakiri  A,  et al.  Neither retinal nor brain atrophy can be shown in patients with isolated unilateral optic neuritis at the time of presentation.  Mult Scler. 2011;17(1):89-95.PubMedGoogle ScholarCrossref
45.
Gelfand  JM, Goodin  DS, Boscardin  WJ, Nolan  R, Cuneo  A, Green  AJ.  Retinal axonal loss begins early in the course of multiple sclerosis and is similar between progressive phenotypes.  PLoS One. 2012;7(5):e36847.PubMedGoogle ScholarCrossref
×