ARTICLE Multiple Sclerosis 2006; 12: 187
195 /

Cognitive impairment in patients with multiple

sclerosis using the Brief Repeatable Battery-
Neuropsychology test
J Sepulcre1, S Vanotti2, R Hernández1, G Sandoval2, F Cáceres2, O Garcea2 and
P Villoslada1

Aims To assess the cognitive impairment of multiple sclerosis (MS) patients using the Brief
Repeatable Battery-Neuropsychology (BRB-N) test.
Methods The performance of 59 patients with MS in the BRB-N test was assessed and compared
with 152 matched healthy controls (HC).
Results In most tests, MS patients performed worse than controls. Age and educational level
strongly influenced the performance of the subjects. The Symbol Digit Modality Test (SDMT) best
correlated with the other individual tests and contributed most to the general BRB-N factor. Further-
more, a correlation between physical disability and performance in some BRB-N tests was observed.
Indeed, patients with progressive MS and greater physical disability performed worse in some tests
than less disabled patients with relapsing MS. By creating a global BRB-N Z score, we found that
patients generally performed 0.7 standard deviation (SD) below the level of controls. We obtained
cut-off values stratified by age and education to determinate cognitive impairment in MS patients.
Conclusions Our data show that cognitive impairment is prevalent amongst MS patients, and that a
single cognitive measurement might be useful for monitoring patients during the progression of this
illness. Multiple Sclerosis 2006; 12: 187
/195. www.multiplesclerosisjournal.com

Key words: Brief Repeatable Battery; cognitive impairment; multiple sclerosis; neuropsychology;
stratified cut-offs

Introduction shown to be of little use in MS [10], the Brief

Multiple sclerosis (MS) is one of the major causes
of disability in young adults [1], with cognitive and
psychiatric disturbances appearing in half of the
cases [2
/4]. The spectrum of clinical manifestations
depends mainly on the location and extension of
the plaques in the central nervous system. The
cognitive functions that are most often affected
are memory, the speed of information processing,
as well as attention and executive functions, visuo-
spatial perception or object naming [4
/7].
Neuropsychological manifestations can be de-
tected in patients even during early stages of the
disease [8,9], highlighting the need to screen for
cognitive deficits. As the Mini-Mental test has been
Repeatable Battery-Neuropsychology (BRB-N) test
was developed as a short and sensitive test to
identify disturbances of cognitive domains in MS
patients [11]. The BRB-N has become the most
widely used neuropsychological battery for MS
[12], and it is now being applied in clinical trials
to monitor cognitive changes [13]. In this study, we
assessed the relationship of the BRB-N with the
physical disability of MS patients and healthy
controls (HC) and obtained the cut-off values
stratified by age and education to determinate
cognitive impairment in MS. Furthermore, we
were interested in developing a unified cognitive
measurement that might be useful for clinical
studies and that would be easier to administer
in practice.

1
Department of Neurology, Multiple Sclerosis Centre, Clinica Universitaria, University of Navarra, Navarra, Spain;
2
Neuroimmunology Clinic, Ramos Mejia Hospital, Argentina
Author for correspondence: Pablo Villoslada, Department of Neurology, Multiple Sclerosis Centre, Clı́nica
Universitaria de Navarra, Pio XII 36, 31008, Pamplona, Navarra, Spain. E-mail: pvilloslada@unav.es
Received 3 September 2004; accepted 6 July 2005

– 2006 Edward Arnold (Publishers) Ltd 10.1191/1352458506ms1258oa

188 J. Sepulcre et al.
Patients and methods
We studied 59 consecutive patients with MS (Poser
criteria [14]) and 152 HC, proportionally matched
by sex, age and education in two MS centres. HC
with cognitive impairment identified with Fol-
stein’s Mini-Mental [15], were also excluded. MS
related disability was evaluated with the Extended
Disability Status Scale (EDSS) [16], and the Multiple
Sclerosis Functional Composite Measure (MSFC)
[17,18], which includes: the Nine Hole Peg Test
(9-HPT); the Timed Walk Test (TWT); and PASAT 3.
Only one patient with complete paralysis of his
dominant hand was excluded for the 9-HPT. In this
study, we included only individuals with no history
of psychiatric or neurological disease other than
MS, no visual or auditory deficits, and no history
of alcohol or drug abuse, or any other major
medical illness. Individuals with psychiatric distur-
bances identified with Cummings’ Neuropsychia-
tric Inventory [19], or Hamilton’s Depression Scale
[20], were excluded from the study.
MS patients were categorized by their EDSS as
follows: low disability (EDSS: 0
/3.0); medium dis-
ability (EDSS: 3.5
/6.5); and high disability (EDSS:
7.0
/9.0). The educational level of the subjects
was classified within one of four levels: low (B/8
years of education: elementary school); middle-low
(between nine and 12 years: high school), middle-
high (between 13 and 16 years: college) and high
(/17 years: advanced degrees). The age of the
individuals was classified within one of three levels:
B/35 years, between 35 and 49 years and /49 years.
Trained personnel carried out the validated
Spanish translation of the BRB-N in both centres.
The two alternative Spanish versions (A or B) of
BRB-N were produced for the European Study
Group on Interferon beta-1b in Secondary Progres-
sive MS [13], and both were administered in alter-
nating order. The BRB-N includes: the Selective
Reminding Test (SRT) to assess verbal memory; the
10/36 Spatial Recall Test (10/36) to assess visual
memory; the Symbol Digit Modalities Test (SDMT)
to assess attention, visual precision search and
executive functions; the Paced Auditory Serial
Addition Task 2 and 3 seconds (PASAT 2-3) to assess
the maintenance of attention; and the Word List
Generation (WLG) to assess associative verbal flu-
ency. These tests were selected to explore most
of the cognitive functions while minimising the
overlap between them.
The SRT-S and SRT-R were performed over six
trials and the 10/36 over three trials. The delayed
SRT-D and 10/36-D were performed 15 minutes
later. The SDMT, the WLG semantic form (A:
animals; B: fruits and vegetables), and PASAT 2-3
were performed only once. The SDMT and the WLG
were each performed in 90 seconds. None of the
Multiple Sclerosis 2006; 12: 187
/195
individuals had been previously studied with this
battery and none of the subjects refused to perform
any of the tests (including PASAT 2-3). The BRB-N
was administered once, thus training-effects are
not applicable.
Statistical analyses were performed using the
SPSS 11.0 software package (SPSS Inc., Chicago, IL,
USA), applying the following tests: ANCOVA (ad-
justed by age and education), unvaried correlation,
multivariate analysis (multiple lineal regression-
stepwise method, P to enter 0.05 and P to exit
0.1), factor analysis (principal components method)
and Z scores. A factor analysis was performed to
assess the principal distribution components of the
battery in HC compared to MS, and not with the
intention to reduce variables. Correlation between
the BRB-N tests and the patient’s disability scales
were assessed using the Pearson correlation test (all
variables had a normal distribution, as assessed by
the Kolmogorov
/Smirnov test). The level of sig-
nificance was set at P B/0.05. To obtain cut-off
values, we used three different criteria: 1 standard
deviation (SD), 1.5 SD and 2 SD below the mean
values of the HC in each test. We also defined the
pathological performance of MS patients at three
levels as the performance below the cut-off in at
least one, two or three tests of the battery. For the
development of cut-off levels, the decimal values
were rounded up in the case of a decimal /0.5 and
down in the case of a decimal 5/0.5.
In order to obtain a global cognitive performance
score, we created a Z score for each cognitive
domain (Z verbal memory (Zv), Z visual memory
(Zvi), Z attention-executive functions (Za) and Z
fluency (Zf)) and Z global cognitive function (BRB-
N Z; see equations below). We calculated two sets of
Z scores, one based on the control group and the
other based on the MS patient group and for each
cognitive domain, the different subtests were
weighted by their Z score to balance the tasks
(e.g., Z SRT-S, Z SRT-R and Z SRT-D to produce the
Zv). We obtained the global cognitive function
(BRB-N Z) by calculating the mean of the Z scores
from the four cognitive domains.
Z-verbal memory (Zv)
(Z SRT-SZ SRT-RZ SRT-D)=3
Z-visual memory (Zvi)(Z 10=36Z 10=36D)=2
Z-attentional=executive (Za)
(Z Pasat3Z Pasat2Z SDMT)=3
Z-fluency (Zf)Z WLG
BRB-N Z or Z-global (Zg)(ZvZviZaZf)=4
The local Ethical Committees of both centres
approved these studies and all participants gave
their informed consent before being included in
the study.
www.multiplesclerosisjournal.com
 Cognitive impairment in MS patients 189

Results therefore, were the most important components

of the battery.
Structure of BRB-N and the influence of The values obtained from the HC group (Supple-
demographic variables mentary information S1) were similar to those
found in Dutch and Italian populations [21,22],
The demographic and clinical characteristics of the indicating that language did not influence BRB
HC and MS subjects are shown in Table 1, both performance. Furthermore, we found no differences
groups being homogeneously distributed by age in the performance of the HC subjects with respect
and education. We first assessed the performance of to gender, the BRB-N version used, or between the
BRB-N in the HC group in order to evaluate the two centres in which the tests were performed. We
cognitive performance under normal conditions. As found that age influenced the performance in the
expected, the correlation between tests of the same SRT (three subtests) and SDMT, and that education
cognitive domain was high (r /0.6; between SRT-S, influenced performance in the SRT-S, 10/36 (two
SRT-R and SRT-D in verbal memory, 10/36 and 10/ subtests), SDMT, PASAT 2-3 and WLG tests (P B/
36-D in visual memory, or PASAT 3 and PASAT 2 in 0.05 in all cases).
attention function) and with the exception of the
SDMT, the correlation between tests of different
cognitive domains was low or non-significant. Performance of MS patients in the BRB-N
Indeed, the SDMT had the highest significant
correlation with respect to all other tests of the MS patients performed worse than HC subjects (P B/
different cognitive domains. In HC subjects, we 0.001 in all cases; Supplementary information S1),
found a five-factor solution that explained 90.03% except for WLG, even when assessed by MS sub-
of the total variance and factor 1 was identified as a types (data not shown). We found significant
general factor of the battery. The SRT-S and SDMT differences between patients with relapsing MS
most closely correlated with this factor 1 and, and progressive MS in the three SRT and the
SDMT (P B/0.05 in all cases). We obtained the
Table 1 Demographic and clinical characteristics of both HC BRB-N cut-off scores, stratified by age and educa-
and MS patients tion, using three different criteria ( B/1, B/1.5 or
HC MS patients P values* B/2 SD below the mean values of HC in each test;
Appendix 1
/3) and using such cut-off levels, we
n 152 59
/
Sex (M/F) 44/108 21/38 NS found that cognitive impairment was highly pre-
Age (years) 42.59/11.9 42.69/10.4 NS valent in MS patients. Using a very relaxed criteria
B/35 years 48 17 NS (abnormal result ( B/1 SD in one subtest)), we
35
/49 years 54 25 NS obtained a very high prevalence (89.83%) in the
/17 years 50 16 NS
Education 11.849/3.04 11.989/3.42 NS MS group, but with a big overlap in HC subjects
(years) (64.47% had one abnormal subtest with this cri-
5/8 years 23 11 NS teria). When criteria were set to medium stringency
9
/12 years 53 18 NS
13
/16 years 61 20 NS
(such as abnormal performance (B/1.5 SD) in two
]/17 years 15 10 NS subtests), the prevalence fell to around 60%
EDSS
/ 3.44 (0
/8.5) (14.47% in HC). However, even when the most
EDSS 0
/3.0
/ 34
/ stringent criteria were applied, such as an abnormal
EDSS 3.5
/6.5
/ 19
/
EDSS 7.0
/9.0
/ 6
/ result ( B/2 SD) in three tests, cognitive impairment
MSFC
/ /0.369/1.88
/ was still highly prevalent in MS, around 30%
MS subtype (3.94% in HC; Table 2).
RRMS
/ 37
/
SPMS
/ 9
/
PPMS
/ 11
/ Table 2 Performance of the BRB-N in MS patients according to
PRMS
/ 2
/ the different strength criteria
*P values between HC cohort and MS patients cohort (x2 or Cut-off levels using different strength criteria
Student’s t-test where appropriate).
Results are expressed as number of HC or MS patients, mean9/ Abnormal
SD (or median and rank for EDSS). subtest B/1 SD (%) B/1.5 SD (%) B/2 SD (%)
ns, Not significant; EDSS, Expanded Disability Status Scale; ]/1 subtest 53 (89.83) 46 (77.96) 39 (66.10)
MSFC, Multiple Sclerosis Functional Composite; RRMS, relap- ]/2 subtest 49 (83.05) 34 (57.62) 24 (40.67)
sing remitting multiple sclerosis; SPMS, secondary progressive ]/3 subtest 43 (72.88) 31 (52.54) 18 (30.50)
multiple sclerosis; PPMS, primary progressive multiple sclerosis;
PRMS, progressive relapsing multiple sclerosis. Results are expressed as number of MS patients and percentage.

www.multiplesclerosisjournal.com Multiple Sclerosis 2006; 12: 187
/195

190 J. Sepulcre et al.
Relationship between BRB-N and physical
disability
All BRB-N subtests, except for the WLG test, showed
a moderate correlation with the EDSS and MSFC
scales in MS patients (r / /0.395 to /0.598, P B/
0.05 for EDSS; and r /0.276
/0.477, P B/0.05 for
MSFC). Interestingly, the SDMT had the strongest
correlation with both the EDSS (r / /0.598; P B/
0.001) and MSFC (r /0.477; P B/0.001). This finding
was striking because the correlation of the PASAT 3
with the MSFC was slightly lower, despite the fact
that PASAT 3 is included in the MSFC (r /0.423;
P/0.001). This association between SDMT and the
disability scales was not due to hand impairment
because an independent association was main-
tained after adjusting for the dominance of the
9-HPT score (r /0.610, P B/0.001 for EDSS; and r/
0.469, P B/0.001 for MSFC). According to the
categories of the EDSS scale, patients with low
disability (EDSS: 0
/3.0) performed better than
patients with medium disability (EDSS: 3.5
/6.5) in
the SRT and SDMT (P B/0.05 in all cases), and better
than the highly disabled patients (EDSS: 7.0
/9.0)
in the SRT (three subtest), 10/36 (two subtest),
SDMT, PASAT 2 and 3 tests (P B/0.05 in all cases).
Interestingly, MS patients with medium and high
disability performed similarly in all tests. In addi-
tion, the performance of patients with very low
disability (EDSS:5/1.5) was similar to that of HC
subjects, no differences being observed in any test
in the battery.
Developing a global cognitive score
As expected when using the HC reference values
(see Supplementary information 2), the distribution
of the mean BRB-N Z score of MS patients was
skewed 0.7 SD towards a poorer performance
than the HC group. Moreover, the difference was
highest (/0.85 SD) when considering only the
cognitive domains that best discriminated the
groups (abbreviated BRB-N Z score: calculated
with verbal and visual memory tests and the tests
of attention-executive domains). However, when
using reference values from the MS patient group
(see Supplementary information 2), the distribution
of BRB-N from an individual patient reflected
a better or worse performance than an average
patient. This latter approach seems to be more
intuitive and sensitive to clinical changes with
time, and as such, it was adopted for the MSFC
[17]. Finally, the BRB-N Z score correlated signifi-
cantly with the disability scales EDSS (r / /0.540;
P B/0.01) and MSFC (r /0.456; P B/0.01).
Multiple Sclerosis 2006; 12: 187
/195
Discussion
Although fully developed dementia is rare in MS
patients, at least half of MS patients have some
degree of cognitive impairment [2
/4]. We found
that cognitive impairment was indeed highly pre-
valent in our cohort of MS patients, depending on
the degree of disability, and the MS subtype.
Although different cognitive batteries have been
proposed to assess cognitive impairment in MS
[10,11,22
/24], the BRB-N has now become widely
accepted for clinical studies and even in clinical
practice. In our study with the Spanish version of
the BRB-N, we found that the values from HC were
similar to those obtained in Dutch [21], and Italian
[22], populations indicating that this battery is little
influenced by language or cultural differences,
thereby validating its use in different populations.
However, as previously reported, age and educa-
tional level did strongly influence BRB-N perfor-
mance. In this study, we excluded subjects with
depression and therefore this variable did not affect
BRB-N performance. Since, in clinical practice,
patients with mood disorders might perform differ-
ently in this battery, we recommend that cognitive
deficits should not be assessed during this period.
A degree of controversy exists regarding the
correlation between cognitive impairment and
the degree of physical disability measured with
the EDSS scale [4,9,21
/30]. We found a moderate
correlation between the BRB-N and both physical
disability scales analysed, EDSS and MSFC. We
found that the SDMT correlated with the other
tests in the battery, it had the strongest correlation
with disability scales, and weighed most heavily in
the general factor of BRB-N in MS patients. It may
be that the relevance of this test is due to its multi-
modal character [31]. Moreover, this test might be
sensitive to changes in the disease, as reflected by
its ability to discriminate between relapsing and
progressive disease [12,25]. Indeed, SDMT correlates
with the degree of cerebral atrophy measured by
MRI [26]. Taking all these data into account, we
propose that if only one neuropsychological test
were to be used to screen for cognitive impairment,
the SDMT would be the most appropriate test and
could even substitute for the PASAT in the MSFC.
When the MSFC was developed, the inclusion of
the PASAT test was based on the fact that, at that
time, more data was available for the PASAT test
than any other at that time [18].
In our study, all the tests of the BRB-N were
significantly different between HC and MS patients
except for the WLG. This lack of sensitivity of
WLG supports the previously established idea that
language function is more resistant to deterioration
in MS than other cognitive tasks [9,32]. However, it
www.multiplesclerosisjournal.com

is important to note that we only assessed the
semantic form of this fluency task, and this could
be responsible for the lack of sensitivity.
Although it was not the primary aim of our study
and we might lack power, we found that subjects
with progressive MS performed worse than those
with relapsing MS, which is in agreement with
Huijbregts et al . [12] This poorer performance may
possibly reflect the amount of brain damage, as
highlighted in MRI studies [33].
The creation of a unified cognitive BRB-N scale
could prove to be a useful tool to screen cognitive
impairment in MS patients and to monitor changes
in cognitive functions over time. For this reason, we
created a BRB-N Z score. Through this BRB-N Z
score, we found that patients performed close to 1
SD worse than HC and that the BRB-N Z score also
correlated well with physical disability. One limita-
tion of our study was that our results were obtained
after a single administration of the battery in
each subject, and the psychometric properties of
re-testing were therefore not addressed.
Acknowledgements
We would like to thank the individuals who
participated in our study, the Navarra MS Society
and the Argentina MS Society for their collabora-
tion. We would also like to thank Dr Joseph
Masdeu, Dr Teresa Gomez-Isla, Dr Manuel Alegre
and Dr Isabel Lamet from the Department of
Neurology at the University of Navarra, for their
advice in the design and analysis of this study.
References
1. Compston A, Coles A. Multiple sclerosis. Lancet 2002;
359: 1221
/31.
2. Rao SM, Leo GJ, Bernardin L, Unverzagt F. Cogni-
tive dysfunction in multiple sclerosis. I. Frequency,
patterns, and prediction. Neurology 1991; 41: 685
/91.
3. Rao SM, Leo GJ, Ellington L, Nauertz T, Bernardin
L, Unverzagt F. Cognitive dysfunction in multiple
sclerosis. II. Impact on employment and social function-
ing. Neurology 1991; 41: 692
/96.
4. Rao SM. Neuropsychology of multiple sclerosis. Curr
Opin Neurol 1995; 8: 216
/20.
5. Demaree HA, DeLuca J, Gaudino EA, Diamond BJ.
Speed of information processing as a key deficit in
multiple sclerosis: implications for rehabilitation. J Neurol
Neurosurg Psychiatry 1999; 67: 661
/63.
6. Achiron A, Ziv I, Djaldetti R, Goldberg H,
Kuritzky A, Melamed E. Aphasia in multiple sclerosis:
clinical and radiologic correlations. Neurology 1992; 42:
2195
/97.
7. Laatu S, Hamalainen P, Revonsuo A, Portin R,
Ruutiainen J. Sematic memory deficit in multiple
sclerosis; impaired understanding of conceptual mean-
ings. J Neurol Sci 1999; 162: 152
/61.
www.multiplesclerosisjournal.com
Cognitive impairment in MS patients 191
8. Dujardin K, Donze AC, Hautecoeur P. Attention
impairment in recently diagnosed multiple sclerosis. Eur J
Neurol 1998; 5: 61
/66.
9. Ruggieri RM, Palermo R, Vitello G, Gennuso M,
Settipani N, Piccoli F. Cognitive impairment in
patients suffering from relapsing-remitting multiple
sclerosis with EDSS B/ or / 3.5. Acta Neurol Scand 2003;
108: 323
/26.
10. Beatty WW, Goodkin DE. Screening for cognitive
impairment in multiple sclerosis. An evaluation of the
Mini-Mental State Examination. Arch Neurol 1990; 47:
297
/301.
11. Rao SM, in collaboration with the Cognitive
Function Study Group of the National Multiple
Sclerosis Society. A manual for the Brief Repeatable
Battery of Neuropsychological Tests in multiple sclerosis .
Section of Neuropsychology, Medical College of Wiscon-
sin, 1000 N. 92 Street, Milwaukee, WI 53226, 1990.
12. Huijbregts SCKN, de Sonneville LM, de Groot V,
Reuling IE, Polman CH. Differences in cognitive
impairment of relapsing remitting, secondary, and pri-
mary progressive MS. Neurology 2004; 63: 335
/39.
13. European Study Group on interferon beta-1b in
secondary progressive MS. Placebo-controlled multi-
centre randomised trial of interferon beta-1b in treat-
ment of secondary progressive multiple sclerosis. Lancet
1998; 352: 1491
/97.
14. Poser CM, Paty DW, Scheinberg L, McDonald WI,
Davis FA, Ebers GC et al . New diagnostic criteria for
multiple sclerosis: guidelines for research protocols. Ann
Neurol 1983; 13: 227
/31.
15. Folstein MF, Folstein SE, McHugh PR. ‘Mini-mental
state’. A practical method for grading the cognitive state
of patients for the clinician. J Psychiatr Res 1975; 12:
189
/98.
16. Kurtzke JF. Rating neurologic impairment in multiple
sclerosis: an expanded disability status scale (EDSS).
Neurology 1983; 33: 1444
/52.
17. Fischer JS, Rudick RA, Cutter GR, Reingold SC,
Baier ML, Cookfair DL et al . The Multiple Sclerosis
Functional Composite Measure (MSFC): an integrated
approach to MS clinical outcome assessment. National
MS Society Clinical Outcomes Assessment Task Force.
Mult Scler 1999; 5: 244
/50.
18. Cutter GR, Baier ML, Rudick RA, Cookfair DL,
Fischer JS, Petkau J et al . Development of a multiple
sclerosis functional composite as a clinical trial outcome
measure. Brain 1999; 122 (Pt 5): 871
/82.
19. Cummings JL. The Neuropsychiatric Inventory: asses-
sing psychopathology in dementia patients. Neurology
1997; 48: S10
/16.
20. Hamilton M. A rating scale for depression. J Neurol
Neurosurg Psychiatry 1960; 23: 56
/62.
21. Boringa JB, Lazeron RH, Reuling IE, Ader HJ,
Pfennings L, Lindeboom J et al . The brief repeatable
battery of neuropsychological tests: normative values
allow application in multiple sclerosis clinical practice.
Mult Scler 2001; 7: 263
/67.
22. Solari A, Mancuso L, Motta A, Mendozzi L, Serrati
C. Comparison of two brief neuropsychological batteries
in people with multiple sclerosis. Mult Scler 2002; 8: 169
/
76.
23. Aupperle RL, Beatty WW, Shelton Fde N,
Gontkovsky ST. Three screening batteries to detect
cognitive impairment in multiple sclerosis. Mult Scler
2002; 8: 382
/89.
24. Benedict RH, Fischer JS, Archibald CJ, Arnett PA,
Beatty WW, Bobholz J et al . Minimal neuropsycholo-
gical assessment of MS patients: a consensus approach.
Clin Neuropsychol 2002; 16: 381
/97.
Multiple Sclerosis 2006; 12: 187
/195
192 J. Sepulcre et al.

25. De Sonneville LM, Boringa JB, Reuling IE, Lazeron
RH, Ader HJ, Polman CH. Information processing
characteristics in subtypes of multiple sclerosis. Neurop-
sychologia 2002; 40: 1751
/65.
26. Christodoulou C, Krupp LB, Liang Z, Huang W,
Melville P, Roque C et al . Cognitive performance and
MR markers of cerebral injury in cognitively impaired MS
patients. Neurology 2003; 60: 1793
/98.
27. Beatty WW, Goodkin DE, Hertsgaard D, Monson
N. Clinical and demographic predictors of cognitive
performance in multiple sclerosis. Do diagnostic type,
disease duration, and disability matter? Arch Neurol 1990;
47: 305
/308.
28. Foong J, Rozewicz L, Chong WK, Thompson AJ,
Miller DH, Ron MA. A comparison of neuropsycholo-
gical deficits in primary and secondary progressive multi-
ple sclerosis. J Neurol 2000; 247: 97
/101.
29. Vleugels L, Lafosse C, van Nunen A, Nachtergaele
S, Ketelaer P, Charlier M et al . Visuoperceptual
impairment in multiple sclerosis patients diag-
nosed with neuropsychological tasks. Mult Scler 2000; 6:
241
/54.
30. Kujala P, Portin R, Ruutiainen J. The progress of
cognitive decline in multiple sclerosis. A controlled 3-
year follow-up. Brain 1997; 120 (Pt 2): 289
/97.
31. Smith A. The Symbol Digit Modalities Test manual . Los
Angeles: Western Psychological Services, 1982.
32. Filley CM, Heaton RK, Nelson LM, Burks JS,
Franklin GM. A comparison of dementia in Alzheimer’s
disease and multiple sclerosis. Arch Neurol 1989; 46: 157
/
61.
33. Comi G, Filippi M, Martinelli V, Campi A, Rode-
gher M, Alberoni M et al . Brain MRI correlates of
cognitive impairment in primary and secondary progres-
sive multiple sclerosis. J Neurol Sci 1995; 132: 222
/27.

Appendix 1 Proposed BRB-N cut-off scores stratified by age and education using the criteria of 1 SD below the mean values of the
HC in each test (see methods for details)

Age

Education B/35 years 35
/49 years /49 years

5/8 years (low level)
SRT-S 38 40 28
SRT-R 34 32 25
SRT-D 8 8 7
10/36 12 12 12
10/36D 4 4 4
SDMT 32 31 24
PASAT 3 24 24 24
PASAT 2 19 19 19
WLG 15 15 15

9
/12 years (middle-low level)

SRT-S 40 41 33
SRT-R 34 32 25
SRT-D 8 8 7
10/36 15 15 15
10/36D 5 5 5
SDMT 33 36 33
PASAT 3 35 35 35
PASAT 2 26 26 26
WLG 21 21 21

13
/16 years (middle-high level)

SRT-S 43 46 37
SRT-R 34 32 25
SRT-D 8 8 7
10/36 16 16 16
10/36D 5 5 5
SDMT 45 36 36
PASAT 3 40 40 40
PASAT 2 30 30 30
WLG 22 22 22

]/17 years (high level)

SRT-S 50 49 44
SRT-R 34 32 25
SRT-D 8 8 7
10/36 19 19 19
10/36D 7 7 7
SDMT 55 39 40

Multiple Sclerosis 2006; 12: 187
/195 www.multiplesclerosisjournal.com

 Cognitive impairment in MS patients 193

Appendix 1 (Continued)
Age

Education B/35 years 35
/49 years /49 years

PASAT 3 46 46 46
PASAT 2 35 35 35
WLG 25 25 25

SRT-S, Selective Reminding Test Long-Term Storage; SRT-R, Selective Reminding Test Long-Term Retrieval; SRT-D, Selective Reminding
Test Delayed Recall; 10/36, 10/36 Spatial Recall Test; 10/36 D, 10/36 Spatial Recall Test Delayed; SDMT, Symbol Digit Modalities Test;
PASAT 2-3, Paced Auditory Serial Addition Task 2 and 3 seconds; WLG, word list generation.

Appendix 2 Proposed BRB-N cut-off scores stratified by age and education using the criteria of 1.5 SD below the mean values of the
HC in each test (see methods for details)

Age

Education B/35 years 35
/49 years /49 years

5/8 years (low level)

SRT-S 34 36 21
SRT-R 27 26 17
SRT-D 7 7 6
10/36 9 9 9
10/36D 3 3 3
SDMT 25 24 20
PASAT 3 18 18 18
PASAT 2 14 14 14
WLG 10 10 10

9
/12 years (middle-low level)

SRT-S 34 36 27
SRT-R 27 26 17
SRT-D 7 7 6
10/36 13 13 13
10/36D 4 4 4
SDMT 24 30 27
PASAT 3 30 30 30
PASAT 2 21 21 21
WLG 18 18 18

13
/16 years (middle-high level)

SRT-S 37 41 31
SRT-R 27 26 17
SRT-D 7 7 6
10/36 13 13 13
10/36D 4 4 4
SDMT 38 29 31
PASAT 3 35 35 35
PASAT 2 25 25 25
WLG 18 18 18

]/17 years (high level)

SRT-S 45 45 41
SRT-R 27 26 17
SRT-D 7 7 6
10/36 17 17 17
10/36D 6 6 6
SDMT 52 33 36
PASAT 3 42 42 42
PASAT 2 30 30 30
WLG 22 22 22

SRT-S, Selective Reminding Test Long-Term Storage; SRT-R, Selective Reminding Test Long-Term Retrieval; SRT-D, Selective Reminding
Test Delayed Recall; 10/36, 10/36 Spatial Recall Test; 10/36 D, 10/36 Spatial Recall Test Delayed; SDMT, Symbol Digit Modalities Test;
PASAT 2-3, Paced Auditory Serial Addition Task 2 and 3 seconds; WLG, word list generation.

www.multiplesclerosisjournal.com Multiple Sclerosis 2006; 12: 187
/195

194 J. Sepulcre et al.

Appendix 3 Proposed BRB-N cut-off scores stratified by age and education using the criteria of 2 SD below the mean values of the
HC in each test (see methods for details)

Age

Education B/35 years 35
/49 years /49 years

5/8 years (low level)
SRT-S 30 33 14
SRT-R 21 20 11
SRT-D 6 6 5
10/36 7 7 7
10/36D 2 2 2
SDMT 19 18 16
PASAT 3 12 12 12
PASAT 2 10 10 10
WLG 5 5 5

9
/12 years (middle-low level)

SRT-S 28 32 22
SRT-R 21 20 11
SRT-D 6 6 5
10/36 11 11 11
10/36D 3 3 3
SDMT 15 24 22
PASAT 3 25 25 25
PASAT 2 16 16 16
WLG 15 15 15

13
/16 years (middle-high level)

SRT-S 32 37 25
SRT-R 21 20 11
SRT-D 6 6 5
10/36 10 10 10
10/36D 3 3 3
SDMT 32 26 27
PASAT 3 31 31 31
PASAT 2 20 20 20
WLG 15 15 15

]/17 years (high level)

SRT-S 41 41 39
SRT-R 21 20 11
SRT-D 6 6 5
10/36 15 15 15
10/36D 5 5 5
SDMT 49 29 32
PASAT 3 39 39 39
PASAT 2 26 26 26
WLG 19 19 19

SRT-S, Selective Reminding Test Long-Term Storage; SRT-R, Selective Reminding Test Long-Term Retrieval; SRT-D, Selective Reminding
Test Delayed Recall; 10/36, 10/36 Spatial Recall Test; 10/36 D, 10/36 Spatial Recall Test Delayed; SDMT, Symbol Digit Modalities Test;
PASAT 2-3, Paced Auditory Serial Addition Task 2 and 3 seconds; WLG, word list generation.

Multiple Sclerosis 2006; 12: 187
/195 www.multiplesclerosisjournal.com

 Cognitive impairment in MS patients 195

Supplementary information S1 BRB-N mean scores of HC and MS patients

HC MS P values*
SRT-S 50.919/11.34 40.129/14.5 B/0.001
SRT-R 42.459/13.24 28.469/15.91 B/0.001
SRT-D 9.79/2.04 7.759/2.61 B/0.001
10/36 20.869/5.43 17.059/5.98 B/0.001
10/36D 7.169/2.23 5.669/2.39 B/0.001
SDMT 49.239/14.64 36.869/16.87 B/0.001
PASAT 3 45.969/10.71 36.339/13.93 B/0.001
PASAT 2 36.979/10.69 26.839/11.43 B/0.001
WLG 28.159/7.32 26.159/8.47 NS

Results are expressed as mean9/SD.

*P values between HC cohort and MS patients cohort using ANCOVA test (adjusted by age and education). NS, not significant; SRT-S,
Selective Reminding Test Long-Term Storage; SRT-R, Selective Reminding Test Long-Term Retrieval; SRT-D, Selective Reminding Test
Delayed Recall; 10/36, 10/36 Spatial Recall Test; 10/36 D, 10/36 Spatial Recall Test Delayed; SDMT, Symbol Digit Modalities Test;
PASAT 2-3, Paced Auditory Serial Addition Task 2 and 3 seconds; WLG, word list generation.

www.multiplesclerosisjournal.com Multiple Sclerosis 2006; 12: 187
/195