1. Fondazione Don Carlo Gnocchi
Disturbi cognitivi e
Sclerosi Multipla
Emilio Portaccio
Fondazione Don Carlo Gnocchi
Firenze

2. Outline Prevalence and neuropsychological profile Clinical correlates
MRI correlates
Assessment
Treatment

3. Outline Prevalence and neuropsychological profile Clinical correlates
MRI correlates
Assessment
Treatment

4. Outline Prevalence and neuropsychological profile Clinical correlates
MRI correlates
Assessment
Treatment

5. Variables that can influence prevalence rates
Study setting
clinic-based versus community-based
Study epoch
Poser’s versus MacDonald’s criteria
Study sample
demographic and clinical characteristics
Assessment tools
Single test
Brief /intermediate/extensive batteries
Computerized batteries
Criteria for defining CI

6. Prevalence of CI in the MS Population
Prevalence range: 40-65%
Nearly 40% in community-based studies
> 65% in clinic-based studies
Review article: Chiaravalloti ND, DeLuca J. Lancet Neurol 2008

8. Percentage of MS group scoring <5th percentile for healthy controls
Prevalence of impairment by cognitive domain
Community-based study: 100 MS pts. versus 100 HCs, extensive NP battery Prevalence of CI in the MS sample 43%, (weak) correlation with EDSS
35%
30%
25%
Percentage of MS group scoring
20%
<5th percentile for healthy controls
22–31%
22–25%
15%
Slide 3
MS-related CD is generally circumscribed, not global. Recent memory and information processing are the most frequently affected domains (with impairment observed in 22%-31% of patients in the study by Rao et al 1991). Memory deficits are reported by patients as forgetfulness, especially with recall of recently learned verbal or visual information. Information processing deficits result in increased distractibility and/or a slowing of mental functioning. This makes it difficult for the patient to focus on more than one task at a time.
Problem solving (ie, executive functions) and visuospatial abilities are the next most commonly affected domains. Patients struggle to complete tasks in an orderly fashion or start several tasks without finishing any one of them. On formal testing they have difficulty developing, testing, and shifting problem-solving strategies. Visual-perceptual, constructional, and visuospatial skills are affected almost as frequently as executive functions.
Lastly, MS less commonly affects language (verbal abilities) and simple attention span. Although patients often describe “inattentiveness,” they are usually referring to an inability to sustain and focus attention over time; often related to information processing deficits and more appropriately termed “complex attention.”
Rao SM, Leo GJ, Bernardin L, Unverzagt F. Cognitive dysfunction in multiple sclerosis. I. Frequency, patterns, and prediction. Neurology 1991;41:
10%
13–19%
12–19%
8–9%
7–8% 5% 0%
Episodic
Memory
Information
Processing
Speed
& Working Memory
Verbal fluency
Conceptual
Reasoning
Visual-spatial
Abilities
Language
Simple
Attention
Adapted from Rao et al. Neurology, 1991

10. The means by which we perceive, process and interpret social information
Linked to poor quality of life, mental health problems, unemployment and loneliness

13. Mean age 39,9 years; Median disease duration 8 years; Median EDSS 2,3; RR course 77%

15. 61,7%
52,8%

20. Outline Prevalence and neuropsychological profile Clinical correlates
MRI correlates
Assessment
Treatment

21. Prevalence of CI in different MS subtypes
time
CIS: range 14-57%
25-30%*
Comparing PP and SPMS, prevalence of CI in PPMS is reported to be inferior, comparable or superior to that reported in SPMS
Early RRMS (< 5 years): range 20-60%
30-40%*
SPMS: range 37-83%
In most of the studies >60%*
PPMS: range 7-87%
In most of the recent studies >50%*
* Criterion for CI : > 2 tests failed,1.5 – 2.0 SDs

22. MINIMUS: an Italian collaborative study (Ruano et al., MSJ 2017)
Cognitive performance was assessed in 1040 patients using a battery validated for MS, the Rao’s Brief Repeatable Battery:
Selective Reminding Test (SRT)
10/36 Spatial Recall Test (SPART)
Paced Auditory Serial Addition Test (PASAT)
Symbol Digit Modalities Test (SDMT)
Word List Generation (WLG)
Stroop Test (ST)
Test failure was defined as a score ≤ 2 SDs, using Italian normative values adjusted for age, sex and education as reference.
Cognitive impairment was defined as impairment in ≥ 2 cognitive domains.
Verbal and visuo-spatial learning
IPS
Cognitive
Executive function

23. Clinical and demographic characteristics of the study sample
Total
n=1040
CIS
n=167 RR
n=759 SP
n=114 PP
n=40
p-value*
Age (mean, years)
40.1
33.9
39.9
51.6
49.3
<0.001
Sex (% female)
67.7
66.5
69.7
58.1
52.3
0.015
Education (mean, years)
12.2
12.7
12.3
11.0
10.2
Age at onset (mean, years)
29.7
32.5
28.6
32.2
36.4
Disease duration
(mean, years)
10.3
1.4
11.2
19.4
12.8
Relapses in the previous year (mean)
0.9
1.0
1.3
0.3
0.0
EDSS (mean, years)
2.6
1.5
2.4
5.5
5.1
A total of 1040 patients were recruited, including CIS, relapsing remmitign, secondary progressive an primary progressive
The patients in each disease group present the typical clinical and demographic caracteristics
* Student’s t test for independent samples or χ2 test adjusted for multiple comparisons

24. Prevalence of cognitive impairment by clinical subtype
Frequency of Cognitive Impairment
n= 167
n=759
n=74
n=44
n=1040
Significant differences: CIS vs. SP, CIS vs. PP, RR vs. SP and RR vs. PP
(p<0.001, χ2 test adjusted for multiple comparisons)

25. Cognitive impairment by age group (all patients)
OR=1.75 [1.54; 2.00] p-value<0.001
Frequency of Cognitive Impairment
Cognitive impairment increases steadly by age group – this could be an effect of patients with progressive forms having older age
Patient age (years)

26. Cognitive impairment by disability level (all patients)
OR=1.99 [1.68; 2.36] p-value<0.001
Frequency of Cognitive Impairment
In a similliar fashiom, the prevalence of cognitive impairment also increases wiht EDSSS in sample
EDSS score

27. Multivariate logistic regression model for cognitive impairment in MS
Crude OR
[95% CI]
Adjusted OR
p-value
(multivariate)
Age (10 years)
1.75 [1.54; 2.00]
1.62 [1.42; 1.86]
<0.001
EDSS (2 points)
1.99 [1.68; 2.36]*
1.80 [1.51; 2.15]
Variables not in the model
Sex (female)
1.08 [0.82; 1.43]
1.10 [0.81;1.49]
0.55
Fatigue (FSS)
1.01 [1.00; 1.19]
1.00 [0.99; 1.02]
0.32
Depression (MADRS)
1.02 [1.00; 1.15]
1.00 [0.98; 1.03]
0.84
Disease duration (10 years)
1.68 [1.44; 1.97]
1.01 [0.99;1.03]
0.47
Clinical course
0.30
CIS vs. RR
1.84 [1.26; 2.71]
1.10 [0.75;1.63]
0.63
CIS vs. SP
4.80 [2.57; 9.01]
1.42 [0.69;2.92]
0.35
CIS vs. PP
13.11 [3.05; 56.04]
3.64 [0.80;16.57]
0.10

28. Multivariate logistic regression model for cognitive impairment and for each cognitive domain  
p-value
OR [95% CI]
Cognitive impairment
Age (10 years)
<0.001
1.62 [1.42; 1.86]
EDSS (2 points)
1.80 [1.51; 2.15]
Verbal learning
1.56 [1.37; 1.78]
1.44 [1.22; 1.70]
Sex (female)
0.002
0.63 [0.47; 0.85]
Visuospatial learning
0.013
1.47 [1.23; 1.76]
1.32 [1.14; 1.53]
p-value
OR [95% CI]
IPS/WM
Age (10 years)
<0.001
1.50 [1.32; 1.70]
EDSS (2 points)
1.54 [1.30; 1.82]
Executive function
0.007
1.22 [1.06; 1.41]
1.59 [1.28; 1.56]
Sex (female)
1.99 [1.43; 2.77]
Clinical course
CIS vs. RR
0.002
1.84 [1.26; 2.71]
RR vs. SP
0.010
2.64 [1.27; 5.50]
RR vs. PP
0.012
14.46 [1.74; ]
Variable selection was performed using stepwise likelihood ratio method.

29. Predictors of worse cognitive performance (linear regression):
Study design: cross-sectional, mono-centric
Assessment through a computerized tool, the Mindstream GAB (NeuroTrax Corp., Bellaire, TX, USA)
1500 MS patients
In a cluster analysis, CIS and RR patients and, respectively, SP and PP patients had a similar cognitive pattern
Predictors of worse cognitive performance (linear regression):
Older age
Higher EDSS score
Longer disease duration
Sharper decline after 5 years from onset
with no particular domain disproportionately represented.
Achiron et al., PlosOne 2013

30. Cognitive performance as a function of MS duration
Cognitive impairment was evident only at five years from onset suggesting a therapeutic window
during which patients may benefit from intervetions to maintain cognitive health
Achiron et al., PlosOne 2013

31. CI at baseline 29% CI after 5 years 54% JNNP 2011
with no particular domain disproportionately represented.

32. A 10-year longitudinal study
Amato et al. Arch Neurol. 1995, 2001
50 patients with early MS (untreated) compared with 70 HCs
10 years
CI= 26%
Study onset
4.5 years
CI 49%
CI 56%
Predictors of a worse cognitive outcome after
10 years:
Older age
Higher EDSS
Shift from RRMS to SPMS
No impairment (0-2 failed subtests)
Mild impairment (3-5 failed subtests)
Moderate impairment (>5 failed subtests)

33. RIS None of the RIS had a strictly normal cognitive
function and 10/26 failed at least 1 test
(PASAT or Digit Span)
RIS
«It could be suggested that these patients
are MS patients with an undiagnosed isolated
symptom presenting as cognitive dysfunction»

34. CI in Radiologically Isolated Syndrome (RIS)
(Amato et al., Neurology 2012)
with no particular domain disproportionately represented.
Rao’s battery

35. Cognitive impairment in “Benign MS”
163 patients with “benign” MS
(disease duration >15 years and EDSS <3.0):
45% cognitive impairment
49% fatigue
54% depression
In 38% of cases, cognitively impaired patients had reduced their social and work activities measured on the Environmental Status Scale (ESS)
A reliable definition of BMS should include the preservation
of cognitive functioning as an additional requisite
(Rovaris et al., Neurology 2008)

36. CI in «BMS» is associated with MRI metrics and may have a prognostic role
Cognitive impairment in «benign» MS is associated with
Greater T1 lesion loads in the WM
More pronunced cortical tissue changes (reduced volumes, total and regional MTR values)
Higher risk of progression to a no longer benign status after 5 years
Results confirmed in a 12-year follow-up (unpublished data)
Amato et al., 2008, Portaccio et al., 2009

37. CI: a factor in prognosis - Longitudinal studies
In CIS subjects CI predicted more rapid conversion to CDMS after a 3 year follow-up (Zipoli et al. Mult Scler 2010)
In early RRMS patients (mean duration at baseline 23 months) baseline IPS and verbal memory impairments predicted higher EDSS score after 5 and 7 years (Deloire et al., Mult Scler 2010)
In newly diagnosed MS patients CI predicted faster progression to EDSS 4 and shift to SPMS in a 10-year follo-up (Moccia et al., Mult Scler 2015)
Implications for the therapeutic conduct?
with no particular domain disproportionately represented.

38. A «cortical variant» of MS?
Clinica findings
prominent cognitive and/or psychiatric disorders at presentation
most often PP course
Positive CSF OB
Neuropsychological findings
Severe cognitive deficits, sometimes with a «cortical pattern» (aphasia, apraxia)
MRI findings
both discrete and confluent, diffuse WM abnormalities
severe brain atrophy at presentation
higher reduction of GM fraction compared with WM fraction
high number of CLs with DIR

39. Major areas of impairment
CI in pediatric onset MS (Amato et al, Neurology 2016 Suppl 2)
Study
Region
MS/HC
Definition of CI
% CI
Major areas of impairment
(>20% of MS pts.)
MacAllister 2005
U.S.
MS 37
> 2 test scores >1.5 SDs below normative data
35.1%
Complex attention, Verbal memory [Naming 18.9%]
Amato 2008
Italy,
multi-center
MS 63
HC 63
performance on > 3 tests
< 5th percentile of HC performance
31%
Verbal and visual memory
Complex attention, Executive functions, Expressive and receptive language
IQ reduced in 8% (younger age at onset)
Till 2011
Canada
MS 35
HC 33
≥3 test scores <1.5 SDs below normative data
29.4%
Attention, IPS, Visuomotor integration, Verbal fluency, Spelling abilities
Julian 2013
U.S.,
MS 187
CIS 44
≥33% of test scores <1 SD below normative data
35%
18%
Fine motor speed, Visuomotor integration, IPS
with no particular domain disproportionately represented.

40. Evolution of CI in pediatric onset MS: five-year follow-up (Amato et al, Neurology 2012)
Change in Cognitive Impairment Index
CCI Year 0-5 (same versions of the tests):
56.3% presented deterioration
25% improvement
18.7% stability
Functions more prone to deteriorate:
visual spatial learning
verbal fluency
expressive language
In the univariate analysis, CI was associated with:
younger age at MS onset
lower education
with no particular domain disproportionately represented.

41. How do cognitive problems affect the lives of people with MS?
Activity and Participation
Safety
Disease
Employment
Driving
Medical decisions
Relationships
Falls
Medication adherence
Social function
Rehabilitation benefit
Daily activities
Symptom management
Physical independence
Coping
Leisure activities
Risk appraisal
Mood
General
Life satisfaction, Quality of Life
Review article: Langdon DW, Curr Opin Neurol Jun;24(3):244-9.

42. What about Pediatric MS? Results of an interview with the parents
(30 patients, 25 classified as having CI)
School activities
Negative impact in 10 out of 30 cases (33%), all classified as having CI
3 cases had a teacher of support
7 cases had to repeat a year in school
Daily living activities
Negative impact on hobbies and sport activities in 16 out of 30 cases (53%)
8 cases had to reduce or change their usual sport activities
8 cases had to quit sport activities altogether
Family and social relationships
Negative impact in 11 out of 30 cases (37%)
(behavioural changes, anxiety, aggressiveness, isolation…)
with no particular domain disproportionately represented.
Amato et al., Neurology 2012

43. Outline Prevalence and neuropsychological profile Clinical correlates
MRI correlates
Assessment
Treatment

44. MRI correlates of CI Role of t1 and t2 lesion volumes and location
with no particular domain disproportionately represented.

45. MRI correlates of CI Role of t1 and t2 lesion volumes and location
Damage in NAWM
Role of grey matter involvement
with no particular domain disproportionately represented.

50. Cognitive impairment in MS: the grey (does) matter – cortical volumes50

51. Cognitive impairment in MS: the grey (does) matter – cortical volumes
NC and MS subgroups 5
SIENAX ns
ZNCV -5
-10
p=0.02
-15
Amato MP, et al Neurology 2004 NC MS
Cognitively
Preserved MS
Cognitively
Impaired
No significant difference in terms of whole brain volumes or lesion loads 51

52. Amato MP, et al. Arch Neurol 200752

53. 53

54. Cognitively preserved
MS patients
Cognitively impaired
MS patients
Healthy controls
Calabrese M, et al Neurology 2010 54

55. NGMV was the strongest MRI predictor of:
EDSS (OR: 0.67; p<0.001)
PASAT (beta: 0.19; p<0.001)

56. In RIS subjects, the number of tests failed was related with
Higher T1LV (rho=0.526;p=0.025)
Lower NCV (rho=-0,481;p=0,043)
Amato MP, et al. Neurology 2012

57. Cognitive impairment in MS: the grey (does) matter – cortical lesions
Calabrese M, et al. Arch Neurol 2009

58. Calabrese M, et al. Arch Neurol 2009

60. Cortical lesions were found in 6 of 15 RIS subjects
Giorgio A, et al. Neurology 2011

61. Minagar A, et al. Neurology 2013

62. Houtchens MK, et al. Neurology 2007

63. Houtchens MK, et al. Neurology 2007

67. MEN
WOMEN

68. Significant correlations only in men

69. WOMEN
MEN

71. «…most evident when attempts to correlate a non- specific marker for focal brain damage (T2) with a crude clinical measure (EDSS)»
Barkhof D, Curr Opin Neurol 2008

75. GRAZIE PER L'ATTENZIONE!!