21
Cognitive Dysfunction in Multiple Sclerosis
Stephen M. Rao
KEY POINTS FOR CLINICIANS
Cognitive dysfunction …
• Occurs in 43% to 65% of patients with multiple sclerosis (MS)
• Is often under recognized or misdiagnosed as depression, stress, or personality disorder
• Is assessed with neuropsychological testing with the most common deficits observed on measures of recent memory and information processing speed
• Contributes significantly to unemployment, motor vehicle accidents, impairment in activities of daily living, and loss of social contacts
• Is strongly related to the extent of white matter lesion volume and brain atrophy on MRI
• May be treated with disease-modifying and symptomatic treatments with modest success
INTRODUCTION
Cognitive function is often impaired in MS patients (for recent comprehensive reviews of this literature, see (1,2)). Nearly 43% to 65% of MS patients exhibit some degree of impairment on standardized neuropsychological (NP) tests (3). Cognitive impairment can have devastating consequences for the MS patient in the areas of employment (4,5), driving skills and safety (6), social activities (4), personal and community independence (4,5), and the likelihood of benefiting from rehabilitation (7). Not surprisingly, it is a major source of caregiver strain (8).
Cognitive dysfunction occurs in half of all MS patients
Cognitive impairment is the direct result of MS-related cerebral pathology. Brain abnormalities, as visualized by various MRI techniques, correlate moderately with NP tests (for recent reviews (9–11)). Cognitive deficits have been shown to correlate with T2- and T1-weighted white matter lesions as well as lesions in gray matter; brain atrophy; and microscopic pathology, as visualized by magnetization transfer, diffusion tensor, and proton spectroscopy, in both lesions and normal appearing brain tissue. Furthermore, longitudinal studies have shown that deteriorating cognitive function is associated with increased lesion burden and atrophy (12–14).
Secondary progressive MS patients typically perform more poorly on NP testing than do patients with relapsing–remitting or primary progressive MS (15). Surprisingly, NP test scores correlate only weakly with disease duration and neurologic disability (16). The weak cross-sectional correlations between cognitive dysfunction and disease duration may be due to the high variability in symptom presentation in MS: some patients exhibit cognitive dysfunction as an early presentation of the disease, whereas other patients may never exhibit problems with cognition.
Neurologic disability, as typically assessed by the Expanded Disability Standard Scale (EDSS), correlates only modestly with degree of cognitive dysfunction (17). The EDSS tends to emphasize disability associated with ambulation. As a consequence, lesions affecting primarily brain regions associated with higher cognitive functions may not have an impact on the EDSS; likewise, lesions of the spinal cord may impact ambulation and the EDSS score, but have no effect on cognitive functions.
Most common deficits are in the areas of recent memory and information processing speed
192Not all cognitive functions are equally susceptible to disruption by MS. Deficits in learning and recall of new information (episodic memory) and in information processing speed and working memory (i.e., the ability to simultaneously buffer and manipulate information) are the most common (18). Less common, but significant deficits are observed on visuospatial abilities and executive functions (including reasoning, problem solving, and planning/sequencing) (18). In contrast, very few MS patients exhibit deficits on measures of auditory attention span and language abilities, although recent natural history studies suggest that deficits in these domains become evident when cohorts are followed for longer periods of time (5). The severity and pattern of cognitive deficits may vary considerably across individual MS patients (19). This heterogeneity of NP impairment can best be appreciated when large samples of patients are administered a comprehensive NP test battery.
The natural history of MS-related cognitive impairment has been reported from studies (5,20–22) conducted prior to the appearance of disease-modifying drugs (see Table 21.1). Progression rates vary considerably across patients and across cognitive functions, but on average, approximately 5% to 9% of patients will experience deterioration on NP tests annually. In general, cognitive impairment is unlikely to remit to any significant extent, but cognitive deficits may remain stable for long periods of time before worsening. Acute exacerbations and remissions involving cognitive functions can be seen (see Figure 21.1).
In recent years, there has become a greater appreciation that MS can occur in pediatric populations (23). Cognitive deficits have been identified in approximately one third of children and adolescents, with the most prominent deficits in motor and cognitive processing speed as well as in attention, verbal and visual memory, expressive and receptive language, and visuo-motor integration (24–27).
TABLE 21.1 Indicators of Cognitive Dysfunction in MS
• Need for help with activities of daily living not attributable to physical disability |
• Underemployment or unemployment not attributable to physical disability |
• Change in mood or behavior (e.g., increased irritability, disinhibition) not attributable to anxiety or depression |
• Withdrawal from usual social activities not attributable to anxiety or depression |
• MRI showing global atrophy and a high lesion load on T2-weighted imaging |
MS, multiple sclerosis.
EVALUATION
Cognitive dysfunction is typically evaluated by a board-certified clinical neuropsychologist. The purposes of such an evaluation can be varied and involve questions of differential diagnosis (depression vs. cognitive dysfunction), disability assessment (e.g., Social Security), design of cognitive rehabilitation interventions, and clinical management with symptomatic and disease-modifying drugs. Comprehensive assessment will necessarily involve several measures that capture different cognitive domains. An example of a comprehensive for monitoring MS patients is the Minimal Assessment of Cognitive Function in MS (MACFIMS) (28) (see Table 21.2). The MACFIMS takes approximately 90 minutes to administer and was created on the basis of a consensus panel of neuropsychologists. The Brief Repeatable NP Battery (BRB) (29) and the Brief International Cognitive Assessment for MS (BICAMS) (30,31) are alternative batteries that focus exclusively on the assessment of processing speed and episodic memory and take less than 30 minutes to administer.
FIGURE 21.1 Female RRMS patient (age 37) had normal cognitive functions at baseline NP testing. At 5 weeks, she developed a selective impairment in verbal episodic memory in association with the development of a large white matter hyperintensity in the left temporal lobe on T2-weighted MRI imaging. The white matter lesion had decreased in size at 13 weeks postbaseline with an improvement in verbal episodic memory, although her test performance did not return to baseline. Her performance on NP tests of visuospatial episodic memory and information processing speed were unchanged during the three examinations.
NP, neuropsychological; RRMS, relapsing–remitting MS.
193TABLE 21.2 Minimal Assessment of Cognitive Function in MS
TEST | ESTIMATED ADMINISTRATION TIME |
Information Processing Speed |
|
Paced Auditory Serial Addition Test (PASAT) | 10 min |
Symbol Digit Modalities Test (SDMT) | 5 min |
Episodic Memory |
|
California Verbal Learning Test-II (CVLT-2) | 25 min |
Brief Visuospatial Memory Test-Revised (BVMT-R) | 10 min |
Executive Functions |
|
California Sorting Test (CST) | 25 min |
Visuospatial Perception |
|
Judgment of Line Orientation Test (JLO) | 10 min |
Language/Other |
|
Controlled Oral Word Association Test (COWAT) | 5 min |
MS, multiple sclerosis.
Source: Jennekens-Schinkel A, Laboyrie PM, Lanser JBK, et al. Cognition in patients with multiple sclerosis. After four years. J Neurol Sci. 1990;99:229–247.
NP evaluations are needed to gauge the severity and pattern of cognitive deficits
A major challenge in clinical practice is the identification of MS patients who would benefit from a comprehensive NP evaluation. Numerous studies have shown that self-report of cognitive dysfunction is frequently inaccurate. Patients with depression may over report cognitive symptoms but perform normally on NP testing. Conversely, patients with significant cognitive dysfunction determined from objective NP testing may lose self-awareness and minimize the report of their cognitive deficits. Likewise, because the cognitive deficits in MS typically do not involve language or communication deficits, physicians who specialize in MS are not typically able to identify cognitive dysfunction from the mental status examination. Furthermore, screening examinations for dementia, such as the Mini Mental State Examination, are insensitive to the cognitive dysfunction in MS.
Self-report of cognitive impairment can be unreliable
Single test screening examinations using measures of processing speed, like the Paced Auditory Serial Addition Test (PASAT) or the Symbol Digit Modalities Test (SDMT), are capable of identifying 45% to 74% of patients diagnosed with cognitive dysfunction relative to results derived from a comprehensive NP battery (18,32). This approach, however, will be insensitive to MS patients with cognitive dysfunction in other domains (e.g., episodic memory). In addition, even brief tests like the 90-second SDMT are infrequently administered in busy clinical practices due to personnel time constraints associated with administration, scoring, conversion of raw scores using normative data, and entering results into the electronic medical record.
To address these concerns, we developed the Multiple Sclerosis Performance Test (MSPT) (33), a self-administered, iPad-based computerized battery of cognitive, vision, and motor tasks that is currently integrated in several large MS clinics. The cognitive measure from the MSPT is called the Processing Speed Test (PST), which closely resembles the SDMT. The PST has excellent retest reliability, demonstrates a strong correlation with the SDMT, was slightly more sensitive than SDMT in discriminating MS from healthy control groups, and correlated better with cerebral T2 lesion load than did the SDMT (34). Finally, PST performance was no different with or without a technician present in the testing environment, a key element in validating the self-administration approach (34).
