Content uploaded by Felix Luessi
Author content
All content in this area was uploaded by Felix Luessi on Dec 16, 2017
Content may be subject to copyright.
https://doi.org/10.1177/1352458517748474
https://doi.org/10.1177/1352458517748474
MULTIPLE
SCLEROSIS MSJ
JOURNAL
journals.sagepub.com/home/msj 1
Multiple Sclerosis Journal
1 –6
DOI: 10.1177/
1352458517748474
© The Author(s), 2017.
Reprints and permissions:
http://www.sagepub.co.uk/
journalsPermissions.nav
Introduction
Multiple sclerosis (MS) is a chronic neuroinflamma-
tory disease of the central nervous system (CNS) that
leads to progressive disability. Recent neuroimaging,
immunological, and histopathological studies have
provided valuable insights into the mechanisms of the
disease.1,2 Development of novel biomarkers is, how-
ever, essential since they should serve as optimized
predictors for the disease course. Cortical atrophy
represents an important magnetic resonance imaging
(MRI)-based biomarker that is closely associated with
emerging clinical disability.3 However, there are cur-
rently no unequivocal non-invasive predictors of
ongoing neuroinflammation and damage in the
cortex.
The blood–brain barrier (BBB) plays an important
role in MS pathogenesis, being the selective gate-
keeper for immunoinflammatory responses in the
CNS and the target of present and future drug devel-
opments.4 Cellular and molecular immune system
components can cross the pathologically permeable
BBB and trigger distinct inflammatory activity in dif-
ferent CNS compartments. BBB disruption has been
attested not only in acute lesions, but also in normal-
appearing gray and white matter.5,6
Increased cerebrospinal fluid albumin
and immunoglobulin A fractions forecast
cortical atrophy and longitudinal functional
deterioration in relapsing-remitting multiple
sclerosis
Julia Kroth, Dumitru Ciolac, Vinzenz Fleischer, Nabin Koirala, Julia Krämer,
Muthuraman Muthuraman, Felix Luessi, Stefan Bittner, Gabriel Gonzalez-Escamilla,
Frauke Zipp, Sven G. Meuth and Sergiu Groppa
Abstract
Background: Currently, no unequivocal predictors of disease evolution exist in patients with multiple
sclerosis (MS). Cortical atrophy measurements are, however, closely associated with cumulative disabil-
ity.
Objective: Here, we aim to forecast longitudinal magnetic resonance imaging (MRI)-driven cortical
atrophy and clinical disability from cerebrospinal fluid (CSF) markers.
Methods: We analyzed CSF fractions of albumin and immunoglobulins (Ig) A, G, and M and their CSF
to serum quotients.
Results: Widespread atrophy was highly associated with increased baseline CSF concentrations and quo-
tients of albumin and IgA. Patients with increased CSFIgA and CSFIgM showed higher functional disability
at follow-up.
Conclusion: CSF markers of blood–brain barrier integrity and specific immune response forecast emerg-
ing gray matter pathology and disease progression in MS.
Keywords: Multiple sclerosis, cerebrospinal fluid albumin, cerebrospinal fluid immunoglobulins, cortical
gray matter, atrophy rate
Date received: 15 August 2017; revised: 20 October 2017; accepted: 7 November 2017
Correspondence to:
S Groppa
Department of Neurology,
Focus Program Translational
Neuroscience (FTN),
Research Center for
Immunology (FZI), Rhine-
Main Neuroscience Network
(rmn²), University Medical
Center of the Johannes
Gutenberg University Mainz,
Langenbeckstraße 1, Mainz
55131, Germany.
segroppa@uni-mainz.de
Julia Kroth
Vinzenz Fleischer
Nabin Koirala
Muthuraman Muthuraman
Felix Luessi
Stefan Bittner
Gabriel Gonzalez-Escamilla
Frauke Zipp
Sergiu Groppa
Department of Neurology,
Focus Program Translational
Neuroscience (FTN),
Research Center for
Immunology (FZI), Rhine-
Main Neuroscience Network
(rmn2), University Medical
Center of the Johannes
Gutenberg University Mainz,
Mainz, Germany
Dumitru Ciolac
Department of Neurology,
Focus Program Translational
Neuroscience (FTN),
Research Center for
Immunology (FZI), Rhine-
Main Neuroscience Network
(rmn2), University Medical
Center of the Johannes
Gutenberg University Mainz,
Mainz, Germany/Department
of Neurology, Institute
of Emergency Medicine,
Laboratory of Neurobiology
and Medical Genetics,
Nicolae Testemiţanu State
University of Medicine
and Pharmacy, Chisinau,
Moldova
748474MSJ0010.1177/1352458517748474Multiple Sclerosis JournalJ Kroth, D Ciolac
research-article2017
Original Research Paper
Multiple Sclerosis Journal 00(0)
2 journals.sagepub.com/home/msj
Under physiological conditions, only low amounts of
albumin and certain immunoglobulin (Ig) types cross
the BBB. Albumin quotient (i.e. cerebrospinal fluid
(CSF) albumin/serum albumin ratio) can therefore be
considered a marker of BBB permeability.7,8 Addressing
BBB integrity and longitudinal gray matter processes
could translate inflammatory activity to focal and
global brain atrophy, the latter being an accepted bio-
logical hallmark of the clinical long-term outcome.
Here, we investigate the connection between cortical
atrophy and the presence of CSF markers of BBB per-
meability/integrity. We observed a significant correla-
tion between CSF albumin and IgA and the rate of
cortical atrophy over 12 months. We propose that
elevated baseline levels of CSF albumin and IgA can
serve as markers of early cortical atrophy and rapid
disease progression in MS.
Methods
Patients
Seventy-one relapsing-remitting multiple sclerosis
(RRMS) patients (mean age ± standard deviation
(SD) 31.2 ± 9.4 years, 25 males) were included in this
longitudinal study. RRMS diagnosis was established
accordingly to the 2010 revised McDonald diagnostic
criteria.9 All patients underwent comprehensive clini-
cal, laboratory, and neuroimaging evaluation through
a standardized protocol.10 All patients were rescanned
with the same protocol after 12 ± 1 months. The sus-
tained Expanded Disability Status Scale (EDSS)
score (confirmed after 3 months) at the second time
point has been used as a clinical outcome measure to
quantify clinical disability. We opted to use an EDSS
score of 2 as a cut-off, since an EDSS score of <2.0
with at least 10 years of disease duration seems to be
the most appropriate criterion in identifying patients
with benign MS (Glad et al., 2010). Clinical disease
activity was defined as a clinical relapse, while radio-
logical activity was defined as the appearance of new/
enlarging hyperintense lesions or gadolinium-enhanc-
ing lesions. Of 71 patients, 63 patients (89%) were
receiving different disease-modifying treatment
(DMT) and 8 patients were not on DMT. Informed
consent was obtained from all patients and the study
was approved by the local ethics committee.
CSF variables examination
CSF and serum samples were obtained from each
patient at the time of first clinical event and CSF con-
centrations of albumin (CSFAlb), IgA (CSFIgA), IgG
(CSFIgG), and IgM (CSFIgM) were determined with
immunonephelometry. Quotients of albumin (QAlb),
IgA (QIgA), IgG (QIgG), and IgM (QIgM) were defined
as the ratios of CSF concentrations to the correspond-
ing serum concentrations of these fractions; reference
values were considered according to Berlit.11 Since
QAlb is age-sensitive, increased QAlb was considered
in relation to reference ranges depending on the age of
the patients: 6.5–8 for patients aged <40 years (55
patients) and ≥8.0 for patients aged ≥40 years (16
patients).12 To detect patients with intrathecal synthe-
sis of IgA, IgG, and IgM, hyperbolic functions for
each Ig type were applied as follows
Ig QIgabQ
bc
lo
cA
lb
=
()
−
()
++
22
where Igloc is local Ig synthesis, Q(Ig) is Ig quotient,
QAlb is albumin quotient, and a, b, and c are empirical
values.13
MRI acquisition
Structural MRI images were acquired using a 3T
Magnetom Tim Trio scanner (Siemens Medical
Solutions, Erlangen, Germany) with a 32-channel
head coil, according to a standardized protocol10 at the
Neuroimaging Centre (NIC) Mainz, Germany. This
imaging protocol comprises sagittal three-dimensional
T1- and T2-weighted fluid-attenuated inversion recov-
ery (FLAIR) sequences. T1-weighted magnetization-
prepared rapid gradient echo (MP-RAGE) sequence
included the following acquisition parameters: repeti-
tion time (TR) = 1900 ms, echo time (TE) = 2.52 ms,
inversion time (TI) = 900 ms, echo train length (ETL)
= 1, flip angle = 9°, matrix size = 256 × 256, field of
view (FOV) = 256 × 256 mm, slice thickness = 1 mm,
voxel size = 1 × 1 × 1 mm; FLAIR sequence: TR =
5000 ms, TE = 388 ms, TI = 1800 ms, ETL = 848,
matrix size = 256 × 256, FOV = 256 × 256 mm, slice
thickness = 1 mm, voxel size = 1 × 1 × 1 mm.
T1-weighted, T2-FLAIR, and contrast-enhanced T1
images were analyzed by an experienced neuroradiol-
ogist for the detection of new and contrast-enhancing
lesions.
Initially, lesion maps were drawn on T2-weighted 3D
FLAIR images using the MRIcron software (http://
www.mccauslandcenter.sc.edu/mricro/mricron/).
Using the lesion segmentation toolbox (LST) which is
part of the statistical parameter mapping (SPM8) soft-
ware, 3D FLAIR images were co-registered to 3D T1
images and bias corrected. After partial volume esti-
mation (PVE), lesion segmentation was performed
with 20 different initial threshold values for the lesion
growth algorithm.14 By comparing automatically and
Julia Krämer
Sven G. Meuth
Department of Neurology,
University of Munster,
Munster, Germany
J Kroth, D Ciolac et al.
journals.sagepub.com/home/msj 3
manually estimated lesion maps, the optimal thresh-
old (ĸ value, dependent on image contrast) was deter-
mined for each patient and an average value for all
patients was calculated. Afterward, for automatic
lesion volume estimation and filling of 3D T1 images,
a uniform ĸ value of 0.1 was applied for all patients.
Subsequently, the filled 3D T1 images as well as the
native 3D T1 images were segmented into gray mat-
ter, white matter, and CSF and normalized to Montreal
Neurological Institute (MNI) space. Finally, the qual-
ity of the segmentations was visually inspected.
MRI data processing: cortical thickness
reconstruction
T1-weighted images were analyzed using FreeSurfer
software (v5.3.0, http://surfer.nmr.mgh.harvard.edu/)
for longitudinal cortical thickness (CT) and thalamic
volume estimation in a fully automated fashion. In
brief, after individual surface reconstruction, an unbi-
ased within-subject template was created.15,16 The
unbiased template served for initialization of skull
stripping, normalization, atlas registration, and par-
cellation of individual time points.17 Surface maps of
regional atrophy rates were computed as (thickness2
– thickness1)/(time2 – time1) and smoothed with a 10
mm full width at half maximum (FWHM) Gaussian
kernel for further correlation with CSF variables and
their quotients. Statistical maps of significant correla-
tions were corrected for multiple comparisons using
false discovery rate (FDR, p < 0.05) with a minimum
cluster size of 100 mm2.
Statistical analysis
SPSS 23.0 software (IBM, Armonk, NY, USA) was
used to perform statistical analyses. The data were
checked for normal distribution using the Shapiro–
Wilk test. Summary statistics are presented as mean
± SD, median, and percentage, where applicable.
Standard and stepwise linear regression analyses
were performed to assess the relative contributions of
baseline quotients (QAlb, QIgA, QIgG, and QIgM) in pre-
dicting the rate of cortical atrophy at 1-year follow-
up. Adjusted R2 values are reported. One-sided paired
student’s t-test was used to evaluate the differences in
variables. P-values less than 0.05 were considered
statistically significant. To account for a possible
influence on cortical atrophy rates, age and gender of
the patients were included as covariates.
Results
Patients
Baseline characteristics of the patients are reported in
Table 1. In 68 patients, the CSF was positive for oligo-
clonal bands (OCB); two patients were OCB negative.
During the 1-year follow-up, 18 patients (25.4%) pre-
sented a clinical relapse, 22 (31.0%) had MRI activity,
and 11 (15.5%) presented both; the remaining patients
(28.1%) were clinically and radiologically stable. Two
patients exhibited new cortical lesions (precuneus and
frontal). At follow-up, mean CT (2.52 ± 0.1 mm) and
thalamus volume (8954.9 ± 1255.7 mm3) were signifi-
cantly smaller in comparison to baseline values (2.53
Table 1. Baseline demographic, clinical, and CSF data of RRMS patients.
Parameter Mean ± SD Reference range
Age (years) 31.2 ± 9.4
Male/female: number (%) 25 (35%)/46 (65%)
Age at diagnosis (years) 30.7 ± 9.6
Disease duration (years) 1.5 ± 3.4
EDSS 1.5 ± 1.4
CSF albumin (CSFAlb) 234.1 ± 123.7 mg/L 110–350 mg/L
CSF IgA (CSFIgA) 3.2 ± 3.2 mg/L 1.5–6 mg/L
CSF IgG (CSFIgG) 58.0 ± 36.8 mg/L <40 mg/L
CSF IgM (CSFIgM) 1.1 ± 1.8 mg/L <1 mg/L
Albumin quotient (QAlb) 5.3 ± 2.7 ≤6.5 (<40 years)
≤8 (>40 years)
IgA quotient (QIgA) 1.6 ± 1.4 1.3
IgG quotient (QIgG) 5.1 ± 2.9 2.3
IgM quotient (QIgM) 1.1 ± 1.8 0.3
SD: standard deviation; EDSS: Expanded Disability Status Scale; CSF: cerebrospinal fluid; Ig: immunoglobulin; Q: quotient;
Ig: immunoglobulin.
Multiple Sclerosis Journal 00(0)
4 journals.sagepub.com/home/msj
± 0.1 mm, p = 0.04 and 9030.9 ± 1211.1 mm3, p =
0.006; respectively).
QAlb, QIgA, and cortical atrophy rate
The rate of cortical atrophy over 1 year was highly
associated with QAlb and QIgA. Significant correlations
between atrophy rates and both baseline QAlb and
baseline QIgA were found in the precuneus (PrC),
rostral middle frontal (rMF), precentral (PC), and
inferior parietal (IP) gyri of both hemispheres
(Figure 1). Specific anatomic locations of these clus-
ters are reported in Supplementary Table 1. The
regions with the highest association between baseline
QAlb and cortical atrophy rate were the right PrC (R2 =
0.364, p < 0.001) and left fusiform gyrus (R2 = 0.290,
p < 0.001). Bilateral PrC had the highest associations
with QIgA (R2 = 0.415 for left and R2 = 0.503 for right
hemispheres, p < 0.0001).
QAlb, QIgA, and thalamic atrophy rate
No significant associations between the addressed
quotients and the annual rate of thalamus atrophy
were detected.
QIgM, QIgG, and cortical atrophy rate
Cortical atrophy rates were also associated with QIgG
(R2 = 0.259, p = 0.02, pars opercularis) and QIgM (R2 =
0.358, p = 0.04, precentral cortex), although with a
weaker effect size. Additional associations with QIgG
were found in the right parietal and occipital lobes,
while QIgM correlated with the atrophy rate in the left
parietal region.
CSF fractions and cortical atrophy
Higher values of CSFAlb and CSFIgA were mirrored by
increased regional gray matter loss. The annual rate of
cortical atrophy correlated with CSFAlb (R2 = 0.17, p <
0.001, precuneus bilateral), CSFIgA (R2 = 0.12, p =
0.001), and to a lesser extent with CSFIgG (R2 = 0.12,
p = 0.02) and CSFIgM (R2 = 0.21, p = 0.04)
(Supplementary Figure 1).
CSF variables and EDSS
CSFIgA and CSFIgM significantly differed between
patients with mild disability (EDSS 0–1.5) and those
with an EDSS score between 2 and 6 at the second
time point (IgA: 1.67 ± 0.69 mg/L vs 2.03 ± 0.71
mg/L, IgM: 9.87 ± 2.38 mg/L vs 11.5 ± 2.03 mg/L, p
= 0.04 and p = 0.003, respectively). CSFAlb, CSFIgG,
and the quotients showed no further correlations with
the EDSS scores at the first or second time points.
Discussion
Identification of reliable early diagnostic immuno-
logical candidates that mirror ongoing disease activity
and the long-term outcome in patients with MS is
essential for therapeutic decisions in the era of rapidly
increasing options for immune modulation. Here, we
Figure 1. Clusters of significant associations between cortical atrophy rates and (A, B) albumin quotient (QAlb) and (C, D)
IgA quotient (QIgA). Lateral and medial views of right (RH) and left (LH) hemispheres are shown. All labeled clusters were
significant after FDR correction (p < 0.05). The color bar represents the statistical significance of the association.
J Kroth, D Ciolac et al.
journals.sagepub.com/home/msj 5
identified increased CSFAlb and CSFIgA in patients
with RRMS as early markers of cortical atrophy and
emerging clinical disability. As these variables
strongly reflect BBB integrity, we clearly demonstrate
the link of ongoing inflammatory disease activity and
cortical degeneration.
Histopathological analyses have confirmed the
occurrence of BBB damage, inflammatory activity,
and demyelination within the cortex in early MS
patients, even preceding white matter lesions.18
Increased CSFAlb and QAlb have been previously pos-
tulated to be indicators of BBB permeability.19 Our
results show an association between CSFAlb, QAlb,
and cortical atrophy, and an even stronger effect for
CSFIgA and QIgA. Based on the topological similarity
of cortical atrophy patterns for CSFAlb, CSFIgA, and
their quotients, common mechanisms reflecting BBB
disruption can be assumed. A disrupted BBB that
permits albumin transition into the CNS could fur-
ther impact the course of MS given its ability to
induce astrocyte and microglia activation, leading to
an increased synthesis of glutamate or affecting
potassium homeostasis, all potentially leading to
functional and structural alterations of cortical cir-
cuits.20 Hence, higher CSFAlb might not only indicate
BBB permeability alterations, but may also induce a
further deleterious cascade for the long-term out-
come of MS patients. Another IgA-dependent mech-
anism might be driven through meningeal mast cells
that enhance via interleukin-6 the proliferation of B
lymphocytes and induce their differentiation toward
IgA-secreting plasma cells.21
In contrast to our results, a study by Uher et al.8 did not
find any association between QAlb after the first clinical
MS episode and the reduction of global normalized
cortical volume (from 1.5T MRI) 48 months after dis-
ease onset. This discrepancy can be attributed to sev-
eral factors. First, the assessment of global cortical
volume is less sensitive to regional cortical reorganiza-
tion processes. Moreover, the estimation of brain vol-
umes depends on the MRI technical parameters (1.5T
vs 3T) and the type of postprocessing algorithm used,
since the segmentation-based algorithm FSL-SIENAX
applied by Uher et al. shows more heterogeneous
results in brain volume changes than the FreeSurfer-
based analysis of CT22 adapted in our study.
Like QAlb and QIgA, QIgG was also associated with corti-
cal atrophy rates, but showing fewer significant clus-
ters. The pathogenic role of IgG in MS remains unclear,
IgG OCB-positive MS patients having higher global
and regional brain atrophy than IgG OCB-negative
patients.23 Rojas et al.24 reported that the presence of
IgG OCB in CSF at MS onset is associated with the
presence of brain volume reduction, mainly in neocor-
tical GM regions, independent of lesion load and other
clinical parameters. Due to the inclusion of a dispro-
portionately large number of OCB-positive patients,
analysis of the regional cortical differences depending
on IgG OCB status was not possible.
Cortical remodeling is an important characteristic of
evolving MS pathology mirroring the long-term pro-
gress of the disease.25 We have recently shown gray mat-
ter network reorganization with a strengthening of local
and modular cortical connectivity.1 Divergent reorgani-
zation patterns of gray and white matter were associated
with clinical impairment. Here, we detected cortical
thinning in a regional pattern with maxima in the precu-
neus and parieto-occipital cortex. The precuneus is a
central region for a large spectrum of neurocognitive
functions, presenting a wide range of connections to
middle frontal gyrus, amygdala, hippocampus, and thal-
amus.26 The depicted structural alterations in these
regions could be linked to emerging functional cognitive
deficits in MS patients with conscious information pro-
cessing or episodic memory alterations.
We did not find any correlation between CSF variables
and the rate of thalamus atrophy. Since thalamic atro-
phy is detectable early in RRMS patients, a different
and possibly BBB-albumin- and IgA-independent alter-
ation of microstructural integrity can be postulated.27
Conclusion
We determined that increased CSF concentrations of
albumin, IgA, and IgM are associated with regional
cortical atrophy and increased disability in patients
with early MS and demonstrated a link between spe-
cific markers of immune response in the CSF, BBB
function, and disease course.
Acknowledgements
J.K., D.C., F.Z., S.G.M., and S.G. contributed equally
to this work. We thank Cheryl Ernest for proofreading
the manuscript.
Declaration of Conflicting Interests
The author(s) declared no potential conflicts of inter-
est with respect to the research, authorship, and/or
publication of this article.
Funding
The author(s) disclosed receipt of the following
financial support for the research, authorship, and/or
publication of this article: This study has been sup-
ported by the German Research Foundation (DFG;
Multiple Sclerosis Journal 00(0)
6 journals.sagepub.com/home/msj
CRC-TR-128) and the Federal Ministry for Education
and Research (BMBF; KKNMS).
References
1. Fleischer V, Gröger A, Koirala N, et al. Increased
structural white and grey matter network connectivity
compensates for functional decline in early multiple
sclerosis. Mult Scler 2017; 23: 432–441.
2. Fleischer V, Kolb R, Groppa S, et al. Metabolic
patterns in chronic multiple sclerosis lesions and
normal-appearing white matter: Intraindividual
comparison by using 2D MR spectroscopic imaging.
Radiology 2016; 281: 536–543.
3. Jacobsen C, Hagemeier J, Myhr K-M, et al. Brain
atrophy and disability progression in multiple
sclerosis patients: A 10-year follow-up study. J
Neurol Neurosurg Psychiatry 2014; 85: 1109–1115.
4. Wiendl H and Meuth SG. Pharmacological
approaches to delaying disability progression in
patients with multiple sclerosis. Drugs 2015; 75:
947–977.
5. Lécuyer M-A, Saint-Laurent O, Bourbonnière L,
et al. Dual role of ALCAM in neuroinflammation and
blood-brain barrier homeostasis. Proc Natl Acad Sci
U S A 2017; 114: E524–E533.
6. Kebir H, Ifergan I, Alvarez JI, et al. Preferential
recruitment of interferon-γ-expressing TH17 cells in
multiple sclerosis. Ann Neurol 2009; 66: 390–402.
7. Alvermann S, Hennig C, Stuve O, et al.
Immunophenotyping of cerebrospinal fluid cells in
multiple sclerosis: In search of biomarkers. JAMA
Neurol 2014; 71: 905–912.
8. Uher T, Horakova D, Tyblova M, et al. Increased
albumin quotient (QAlb) in patients after first clinical
event suggestive of multiple sclerosis is associated with
development of brain atrophy and greater disability 48
months later. Mult Scler 2016; 22: 770–781.
9. Polman CH, Reingold SC, Banwell B, et al.
Diagnostic criteria for multiple sclerosis: 2010
revisions to the McDonald criteria. Ann Neurol 2011;
69: 292–302.
10. Droby A, Lukas C, Schänzer A, et al. A human post-
mortem brain model for the standardization of multi-
centre MRI studies. Neuroimage 2015; 110: 11–21.
11. Berlit P. Klinische Neurologie. Berlin: Springer,
2011.
12. Reiber H and Felgenhauer K. Protein transfer at the
blood cerebrospinal fluid barrier and the quantitation
of the humoral immune response within the central
nervous system. Clin Chim Acta 1987; 163: 319–328.
13. Reiber H and Albaum W. Statistical evaluation of
intrathecal protein synthesis in CSF/Serum quotient
diagrams. Acta Neuropsychiatr 2008; 20: 47–48.
14. 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: 3774–3783.
15. Reuter M, Rosas HD and Fischl B. Highly accurate
inverse consistent registration: A robust approach.
Neuroimage 2010; 53: 1181–1196.
16. Hanganu A, Groppa SA, Deuschl G, et al. Cortical
thickness changes associated with photoparoxysmal
response. Brain Topogr 2015; 28: 702–709.
17. Reuter M, Schmansky NJ, Rosas HD, et al. Within-
subject template estimation for unbiased longitudinal
image analysis. Neuroimage 2012; 61: 1402–1418.
18. Lucchinetti CF, Popescu BF, Bunyan RF, et al.
Inflammatory cortical demyelination in early multiple
sclerosis. N Engl J Med 2011; 365: 2188–2197.
19. Freedman MS, Thompson EJ, Deisenhammer F,
et al. Recommended standard of cerebrospinal
fluid analysis in the diagnosis of multiple sclerosis:
A consensus statement. Arch Neurol 2005; 62:
865–870.
20. LeVine SM. Albumin and multiple sclerosis. BMC
Neurol 2016; 16: 47.
21. Merluzzi S, Frossi B, Gri G, et al. Mast cells enhance
proliferation of B lymphocytes and drive their
differentiation toward IgA-secreting plasma cells.
Blood 2010; 115: 2810–2817.
22. Durand-Dubief F, Belaroussi B, Armspach J,
et al. Reliability of longitudinal brain volume loss
measurements between 2 sites in patients with
multiple sclerosis: Comparison of 7 quantification
techniques. Am J Neuroradiol 2012; 33:
1918–1924.
23. Ferreira D, Voevodskaya O, Imrell K, et al.
Multiple sclerosis patients lacking oligoclonal
bands in the cerebrospinal fluid have less global and
regional brain atrophy. J Neuroimmunol 2014;
274: 149–154.
24. Rojas JI, Patrucco L, Tizio S, et al. Oligoclonal
bands in the cerebrospinal fluid and increased
brain atrophy in early stages of relapsing-remitting
multiple sclerosis. Arq Neuropsiquiatria 2012; 70:
574–577.
25. Calabrese M, Rinaldi F, Mattisi I, et al. Widespread
cortical thinning characterizes patients with MS with
mild cognitive impairment. Neurology 2010; 74:
321–328.
26. Cavanna AE and Trimble MR. The precuneus: A
review of its functional anatomy and behavioural
correlates. Brain 2006; 129: 564–583.
27. Deppe M, Krämer J, Tenberge JG, et al. Early silent
microstructural degeneration and atrophy of the
thalamocortical network in multiple sclerosis. Hum
Brain Mapp 2016; 37: 1866–1879.
Visit SAGE journals online
journals.sagepub.com/
home/msj
SAGE journals
