Evaluation of cognition and fatigue in multiple sclerosis: daily practice and future directions
Abstract
The so-called hidden symptoms or soft signs of multiple sclerosis comprise cognitive dysfunction, fatigue, depression and anxiety. From a patient's perspective, these symptoms are rated as exerting much more negative impact on quality of life and daily functioning than their physical symptoms. Despite this knowledge, the symptoms remain disregarded by many neurologists in clinical practice. This missing awareness can be attributed to several reasons. First, the underlying pathophysiological mechanisms determining occurrence and severity of the different symptoms are still unclear. Second, there is uncertainty in how to reliably assess them. It is undeniable that assessment can be difficult as the hidden symptoms seldom appear isolated but more often highly interact. Third, if standardized, fast and cost-effective assessment to quantify and monitor the evolution of the hidden symptoms would be feasible, the question still remains how to treat these aspects. The present article will give an overview on symptom background and assessment strategies for clinical practice.
1 Hidden and soft?
Fatigue, cognitive dysfunction, depression and anxiety occur among all disease courses in multiple sclerosis (MS) with prevalence rates between 35% and 95% and even in patients with a so-called benign form.1 Due to a strong coincidence and interaction, these symptoms exert a strong negative influence on quality of life (QoL) and vocational status.2 From a patient's perspective, being affected by fatigue and cognitive decline is far from being benign and soft. Concerning the assumption that these symptoms rather remain hidden depends on the eye of the beholder. If you pay enough attention, you will see them very clearly. The only remaining question is how to objectify, classify and quantify them (Fig. 1).
2 Cognitive impairment
Cognition represents a central human ability, which refers to a wide range of higher order brain functions such as processing speed, memory, attention, executive functions and learning abilities. In MS, several cognitive domains can be affected, and once started, this impairment is often persistent and progressive. It is of importance to notice that cognitive problems can occur at early stages of the disease,3 even in patients with clinically isolated syndrome4 or radiologically isolated syndrome5 and in paediatric MS as well6.
The cognitive core deficits in adult MS are verbal and visual short-term memory and learning, cognitive flexibility/attention and processing speed/fluidity.7 As these three cognitive domains are highly relevant for daily functioning, patients experience a significant burden and diminished QoL due to reduced self-esteem, loss of social activities and contacts, and reduction or even loss of work responsibilities.8 Although progression of cognitive decline is generally slow, and in this sense different from other neurodegenerative diseases, the most vulnerable phase for progression seems to occur during the first 5 years after disease onset. This indicates that treatment aimed at protecting against cognitive decline should be initiated during this timeframe.1 However, results of this study are based on a very small sample. Therefore, more longitudinal data on larger samples are urgently needed to get more precise information on the evolution of cognitive functionality in the first years of the disease.
3 MRI measures related to cognitive status in MS
3.1 White matter
There is evidence from several studies that hyperintense T2 lesions correlate much better with the cognitive status of patients with MS than Gd-enhancing lesions or T1 lesions.9, 10 However, it is not the simple number of lesions that determines outcome, but rather their location. It has been shown that particularly the corpus callosum appears to be an important predilection site for cognitive performance as lesions have been found twice as often in cognitively impaired than in cognitively preserved MS patients.11 Further, studies applying magnetization transfer (MT) imaging, in order to get more detailed information on diffuse white matter abnormalities, could show that microstructural changes in the normal appearing white matter (NAWM) are not only predictive for overall cognitive status12 but that they are also predictive for the evolution of cognitive functionality over a period of 7 years.13 The hypothesis that the variety of neuropsychological deficits observed in MS might be the consequence of a “multiple disconnection syndrome” was first driven by neuropsychological data.14 It was further supported by imaging studies showing that the location of lesions in strategic areas of the white matter contributes to this kind of disconnection.11, 15 However, MRI studies combining different techniques revealed that white matter lesions alone cannot be responsible for the diversity of neuropsychological symptoms, but that in addition, damage to the NAWM and gray matter plays a major role.16-18
3.2 Gray matter
One of the first descriptions about cognitive deficits being a result of cortical demyelination derived from a post-mortem study.19 Sensitive immunocytochemical analyses led to the conclusion that the spatial distribution of cortical lesions might be associated with specific cognitive deficits such as information processing speed. In 2009, this assumption was verified by a first application of the double inversion recovery MRI sequence to study distribution of cortical lesions in vivo and to relate these findings to the cognitive profile of the patients.20 Cortical lesion number and volume have been shown to be larger in cognitively impaired patients, and comparable to findings in the white matter, it is not the number of cortical lesions that seems to be relevant for cognitive dysfunction but their location. In addition, T1 hypointense cortical lesions seem to be predictive for decreased memory performance, cognitive speed and verbal fluency.21 Despite the amazing development of new imaging techniques to visualize structural and functional abnormalities, a recent post-mortem study emphasized that existing MRI techniques nowadays are not sensitive enough to detect the real amount and extent of cortical lesions.22
Besides focal lesions, diffuse gray matter damage as measured by magnetization transfer MRI and diffusion tensor MRI contributes not only to overall cognitive impairment but also to decline in specific cognitive subdomains.23
3.3 Brain atrophy
The strongest correlate for cognitive decline, even at early disease stages, is brain atrophy.20, 24 In the past, studies have primarily focused on global central brain atrophy, which is not only associated with cognitive status per se, but which might even be predictive for the longitudinal development of cognitive performance.13 Although volumetric changes in subcortical structures such as the thalamus have repeatedly been described in the past (e.g.9), it was only recently that the importance of thalamic volume with respect to cognitive functioning was further underlined.25, 26 A recent study by Schoonheim et al.27 indicates that thalamic volume is significantly lower in patients with MS compared with controls, with the lowest volumes found in those patients with the most severe cognitive impairment. Further, thalamic volume was found to be an independent predictor of cognitive impairment, while lesion and whole-brain volumes were not. Thus, irreversible tissue loss, measured by brain atrophy of the white and gray matter, is strongly related to the cognitive status of patients with MS. While white matter atrophy has been shown to be predictive for impairment in processing speed and working memory, gray matter atrophy was not only predictive for a cognitive subdomain (verbal memory), but also predicted neuropsychiatric symptoms such as euphoria and disinhibition.28
4 Neuroanatomy of cognition
However, the all-important question remains: What finally causes cognitive decline in MS? Unfortunately, there is only data on correlations, but not on causality. Thus, our actual explanation is based on a human connectome model that was proposed by Heuvel et al.29 to emphasize the tremendous complexity of cognitive neuroanatomy. Based on that model, cognitive decline is assumed to occur as a consequence of a breakdown or collapse of connectivity in certain “master” neuroanatomical structures, the so-called “rich-club” hubs. If multiple lesions occur in a so-called hub region, that is the highly connected, central neocortical regions that play a key role in global information integration, the consequences for the patient will be grave.29 This explains why a patient can suddenly become cognitively impaired. The question is what kind of therapy we can offer to protect against such network collapse.
5 Cognitive reserve, exercise training and cognitive training
Cognitive reserve describes the concept that intellectual enrichment may be protective against cognitive decline. Applied to a disease like MS, this means that despite the chronic and progressive nature, the burden of the disease and a related negative impact on cognitive functionality can be attenuated by cognitive reserve.30 Cognitive reserve itself can be understood as an individually “filled backpack” with enriching life experiences. The better the quality of the filling, the better the protection against cognitive decline as coping strategies and neuronal networks become more efficient.31 In MS, intellectual enrichment has also been shown to be protective against cognitive decline over 4.5–5 years.30, 32 To further support the individual amount of cognitive reserve throughout lifetime, it is recommended to be engaged in leisure activities that require, for example, attention, speed, memory, flexibility and planning. In addition, exercise has been shown to be linked to cognitive performance33 and, in contrast to former beliefs, it was reported that patients with MS even benefit from high-intensity training.34 Concerning cognitive training, or cognitive rehabilitation, the evidence for treatment efficacy is actually not strong. Reported results, in the few published studies that exist, nevertheless report some benefits for patients.35, 36
6 Assessment of cognitive status
Regular assessment of the cognitive status is as important as monitoring patients via the Expanded Disability Status Scale (EDSS)37. The problem, however, is that application of neuropsychological tests is often not feasible due to a lack of time and expert staff. Therefore, short screening tools of good psychometric quality are required. One of those recommendable short screening batteries is the Brief International Cognitive Assessment for MS (BICAMS)-battery which comprises the Symbol Digit Modalities Test for assessing speed and working memory, the California Verbal Learning Test II to test for verbal short-term memory and learning and the Brief Visuospatial Memory Test Revised to evaluate visuospatial short-term memory and learning.38 Application of BICAMS needs 20 minutes, can be administered by trained non-psychological staff and is suitable for monitoring patients' cognitive performance over time. Additionally, the Stroop Colour Word Interference Test, which measures selective attention, executive function and mental flexibility, can be recommended.39
7 MS fatigue
MS-related fatigue is defined as “a subjective lack of physical and/or mental energy that is perceived by the individual or caregiver to interfere with usual or desired activities”.40 Fatigue symptoms which are present, for any amount of time on 50% of the days for a period of more than six weeks, are indicative for “chronic persistent fatigue”. A new or increased feeling of fatigue in the previous six weeks is defined as “acute fatigue”. An important additional prerequisite for the diagnosis of either acute or chronic persistent fatigue is that fatigue is limiting normal functional activities, hereby significantly affecting QoL. A further important differentiation is the distinction between primary and secondary fatigue. The latter is provoked by factors such as medication, pain or inactivity. Primary fatigue which has its origin in the disease itself might be due to abnormalities in the central nervous system, peripheral nervous system, the immune system, neuroendocrine system or neurotransmitter system. Primary fatigue, for which the underlying pathophysiology is still unknown, is very common and can affect up to 95% of patients.41, 42 For many patients, fatigue represents the most severe MS-related symptom 43 that can occur at all stages of the disease.44 Indeed, it is often present already at onset45 and may even be a red flag for the later development of MS.46
8 MRI measures related to fatigue in MS
The relation between fatigue and brain lesions is similar to that of cognition. It is less the amount of lesions that determines fatigue symptoms but rather their location.47, 48 Similar to what has been described for cognition, it is assumed that fatigue is a result of network disruption, specifically of frontoparietal pathways.49 A diffusion tensor imaging (DTI) study by Pardini et al.50 supports the involvement of the prefrontal cortex in the occurrence of fatigue. Mean fractional anisotropy decreased in those patients with highest fatigue symptoms.
Concerning brain atrophy, white and gray matter atrophy have been reported to be more prominent in fatigued than non-fatigued patients,51, 52 and specifically the volume of the corpus callosum seems to play a major role. A study including 70 RRMS patients, with or without fatigue, showed that corpus callosum atrophy was more pronounced in fatigued patients, independent of EDSS, disease duration, gender and age.53 Further, the thalamus is also discussed to be of high relevance in fatigue. When compared to healthy controls, a diffusion MRI study reported stronger alterations in the thalamic region of RRMS patients with fatigue than in RRMS patients without fatigue symptoms.54 A recent study on cognitive fatigue revealed changes in cortico-cortical and cortico-subcortical connectivity with thalamic involvement after performance of a cognitive demanding task.55 In patients with cognitive fatigue, a hyperconnectivity in a frontal–temporal–occipital circuit was detectable 30 minutes after performance of the Paced Auditory Serial Addition Test (PASAT), at the same time as a thalamic hypoconnectivity was observed. Involvement of the prefrontal cortex and subcortical structures, such as thalamus and basal ganglia, has been described in many former studies using different imaging approaches such as positron emission tomography,56 functional magnetic resonance imaging57 and diffusion tensor imaging (DTI)50. A disruption of cortico-subcortical pathways was already at that time assumed. However, with the emergence of more sensitive and sophisticated imaging and post-processing methods, proper investigation of the assumed disconnection hypothesis only recently became feasible.
9 Assessment of fatigue
The major problem with the assessment of MS fatigue is that no measurements are currently available that could help to clearly objectify this purely subjectively perceived symptom. Therefore, assessment of fatigue is entirely dependent on the patient's perception and report. In addition, assessment is made even more difficult by confounding variables such as depression, sleep disturbances and cognitive deficits. Thus, besides application of an appropriate fatigue assessment tool, a reliable process of fatigue diagnosis also requires consideration and proper evaluation of possible confounding conditions.
The state-of-the-art assessment strategy to evaluate fatigue in daily clinical routine is based on self-reports. A large number of fatigue scales have been developed over the past 15 years. Major reasons for renewal were serious methodological problems of the older scales and the exclusion of mental fatigue symptoms.
One scale which differentiates between the two main components of fatigue (cognitive and motor fatigue) is the Modified Fatigue Impact Scale (MFIS)40. The MFIS consists of 21 items. The total score is the sum of the scores for the 21 items. Another structured self-report which can be recommended is the Fatigue Scale for Motor and Cognitive Functions (FSMC)58. This scale consists of 20 items of which 10 items assess cognitive fatigue and 10 items assess motor fatigue. The FSMC not only does differentiate between cognitive and motor fatigue, but also allows for further graduation of fatigue severity in the two subdomains by offering cut-off values.
10 Future perspectives
It is evident that symptoms like cognitive deficits and fatigue, which are often additionally accompanied by depression and anxiety, are much more relevant for the patient's daily life than what is assumed by many treating physicians. Thus, to increase awareness for these aspects is, from a patient's perspective, an unmet need. A German initiative proposed a new treatment decision model (MS decision model, MSDM) by adding neuropsychological aspects such as cognitive performance, fatigue, depression, anxiety and QoL to the well-known disease factor relapses, disease progression and MRI activity.59 This first proposal tries to meet the multifactorial complexity of the disease by integrating neuropsychological issues to the treatment decision process. With respect to NEDA (no evidence of disease activity), MSDM more realistically pictures the complex nature of the disease by considering factors with a high potential to influence disease progression. It would therefore be relevant to complement future NEDA concepts by these factors.
Further, due to time constraints in clinical practice, the use of electronic devices will most likely become increasingly relevant in assessing and treating cognition and fatigue in the future. Nevertheless, the patient–doctor and patient–psychologist interaction cannot be replaced. A possible scenario might be that patients will be screened by an electronic neuropsychological test, and whenever impairment turns out, patients will be referred to a neuropsychologist for a more comprehensive testing and for a recommendation on how to proceed therapeutically.
Acknowledgements
None.
Conflicts of interest
The author has no conflict of interest to disclose with respect to this review. I confirm that I have read the Journal's position on issues involved in ethical publication and affirm that this report is consistent with those guidelines.
