Cognitive and psychosocial features of childhood and juvenile MS

Multiple sclerosis (MS) is an inflammatory demyelinating disease of the CNS typically affecting young adults. It can cause both physical and cognitive disability, the latter estimated in 40% to 65% of cases.¹ In adult-onset patients, the profile of cognitive impairment shows prominent involvement of episodic memory, complex attention, information processing speed, visuospatial abilities, and executive functions, whereas semantic memory, attentive span, and language functions are relatively preserved.¹ Independently of the degree of physical impairment, cognitive impairment has high functional impact on work and everyday activities.^2,3

In approximately 5% of cases, MS occurs in children or adolescents.⁴ In these patients, the impact on cognitive functioning can be, in hypothesis, even more dramatic than that observed in adult-onset cases, because the inflammatory demyelinating process develops during primary CNS myelinogenesis. It may therefore negatively affect ongoing maturation of white matter pathways and lead to neurodegeneration of neural networks involved in cognition. Moreover, disease-related problems during key formative years may interfere with present and future academic achievements. Conversely, brain plasticity and repair mechanisms may be more efficient in this age range, suggesting greater possibility of recovery and compensation.

Although two recent studies have contributed to rouse the interest of researchers in the field,^5,6 available information on neuropsychological aspects of childhood and juvenile MS remains rather limited.^7–9 It is difficult, moreover, to generalize data from published studies because of a number of limitations, such as the lack of a suitable control group or the absence of a global measure of patient cognitive function in terms of IQ.

In this Italian, multicenter study, we compared a cohort of childhood and juvenile MS cases with a group of demographically matched healthy controls (HCs) to assess the frequency, pattern, and clinical correlates of cognitive impairment and to evaluate psychosocial and academic consequences of the disease in this age range.

METHODS

RESULTS

The study sample consisted of 120 subjects, 63 MS cases, and 57 demographically matched HCs (table 1). Disease onset occurred before age 10 years in 15 subjects (25%). Forty cases (63%) were treated with disease-modifying drugs (35 with interferon beta, 5 with glatiramer acetate), with a mean treatment duration of 1.3 ± 1.7 years (range 0.3–7.8 years). All but 1 patient had normal visual acuity or mild visual impairment which did not preclude valid neuropsychological assessment. A complete neuropsychological assessment was achieved in 61 cases (2 cases had an incomplete assessment because of poor compliance). Because the average IQ score of our HC group was in the superior range, we also repeated the analysis using a subgroup of 40 HC subjects yielding a lower average IQ (106.2 ± 10.4). Table 2 shows mean scores of cases and controls on each neuropsychological test. In comparison with the whole sample of HCs, in the MS group, both verbal and performance IQ were reduced. Seventeen patients (28%) scored in the inferior range on the Wechsler Intelligence Scale for Children–Revised, and 5 (8%) yielded scores < 70. Only 3 patients in the latter group had a previous diagnosis of mental retardation, posed after the beginning of the disease. In the remaining 5 cases, in the absence of a measure of adaptive functioning, a formal diagnosis of mental retardation according to the Diagnostic and Statistical Manual of Mental Disorders, Fourth Edition,²⁵ was not appropriate. No other patient had undergone formal neuropsychological testing before the study.

Criteria for significant cognitive impairment were fulfilled in 19 patients (31%), whereas 32 patients (53%) failed at least two tests. Comparing MS patients with the subgroup of 40 HCs with lower average IQ scores, the main results were confirmed. Although differences in mean IQ scores were no more statistically significant, the proportion of subjects yielding lower IQ scores remained significantly higher in the MS group (table 2). Moreover, using this reference group, 2 cases failed two tests instead of three and were no longer included in the cognitively impaired group, whereas no other patient was reclassified.

Self-report of depression found that only 4 respondents (6%) endorsed significant symptoms. Nine patients were classified as fatigued on the FSS using a cutoff score of 4, proposed for adults.²⁴ However, considering the fifth percentile of HC scores, we also assessed fatigue using a cutoff score of 2. The number of fatigued patients increased to 46 (73%). The proportion of fatigued patients was comparable in the groups with and without cognitive impairment (79% vs 74%, p = 0.67).

Focusing on the subgroup of patients with cognitive impairment, we found that 15 of 19 subjects had an IQ score < 90. The number and type of tests failed by cognitively impaired cases are listed on table 3. Functions most frequently involved included memory (particularly visuospatial memory), complex attention, verbal comprehension, and executive functions.

Analysis of cognitive impairment was also performed considering the subject performance within each cognitive domain (table 2). Using as the reference criterion the fifth percentile of HC performance, we obtained similar results. The proportion of cognitively impaired cases was 27% (failure in at least three cognitive domains), with a substantially comparable profile of the deficits.

Comparing the two age subgroups, cognitively impaired subjects were 33% in the group aged 8 to 13 years and 30% in the other group (p = 0.83). The profile of deficits did not show any significant difference, with the exception of the Spatial Recall Test, which was more frequently failed by patients aged 14 to 18 years (p = 0.003).

The comparison between cognitively impaired and cognitively preserved patients (table 4) showed a significantly higher proportion of subjects with an IQ score < 90 in the impaired group (15 of 19 subjects, p < 0.01). In a logistic regression analysis including as possible predictors age, sex, educational level, age at onset, number of relapses in the past 2 years, EDSS score, disease-modifying treatment, and IQ, the only significant predictor of cognitive impairment was an IQ score lower than 90 (β 2.90, odds ratio [OR] 18.2, 95% CI 4.6–71.7, p < 0.001). In a further analysis, considering an IQ score < 70 as a dependent variable, the only significant predictor was represented by younger age at onset (β −0.36, OR 0.7, 95% CI 0.5–0.9, p = 0.009).

An interview with the parents was obtained in 41 cases (14 classified as cognitively impaired), from the mother in 30 and from the father in the remaining cases. It revealed that MS had a significant impact on school activities and achievements: 4 of 41 children (10%) had a teacher of support because of cognitive difficulties; 9 (22%) had to repeat a year in school because of multiple missed school days (4 cases) or cognitive difficulties (5 cases). The duration of MS-related absences ranged from 10 to 60 days over the past year (10–30 days: 10 cases; 31–60 days: 10 cases). Absences were most often associated with relapses (6 cases), hospital admissions, medical appointments and exams (14 cases), and therapy side effects (9 cases). Hobbies and sport activities were also negatively affected by the disease in 14 cases (34%), 9 of whom classified as cognitively impaired: 7 cases had to reduce or change their usual sport activities because of disease-related difficulties, and another 7 had to quit sport activities altogether. Behavioral changes were reported by parents in 16 cases (39%), 7 of whom classified as cognitively impaired. Changes included increased levels of anxiety (9 cases), disruptive behavior with aggressiveness toward family members and peers (3 cases), isolation, and feelings of sadness and insecurity (1 case).

DISCUSSION

Acquisition of information on MS-associated cognitive problems in childhood and juvenile cases is of critical importance for helping children and adolescents with MS to manage their difficulties and psychosocial challenges.

Cognitive impairment in this age range has been investigated in two recent uncontrolled studies. In a series of 10 cases,⁶ significant neuropsychological deficits on several cognitive domains were found. In the other study⁵ including 37 cases, cognitive impairment was found in 35% with academic problems in more than a third of the children. A recent longitudinal study of 12 patients also documented further cognitive decline over a 21-month follow-up.²⁶

We found a 31% prevalence of significant cognitive impairment, whereas 53% of the patients exhibited minor degrees of cognitive dysfunction, failing at least two tests. Moreover, 28% of the cases had an IQ score < 90, and 8% had an IQ score < 70.

A multivariate analysis of possible clinical predictors of cognitive impairment revealed that EDSS and number of relapses were not adequate descriptors of the patient cognitive status. A previously published study reported significant relationships with disability levels, number of relapses, and disease duration.⁵ In this study, however, cognitive defects were detected also in subjects with low disability levels. Although differences in methodology and characteristics of the patients may account for these inconsistencies, our findings are in agreement with several reports of cognitive impairment in adults that have shown poor relationships with disability levels and disease duration.^27–31 It is noteworthy that in our cohort the only significant correlate of cognitive impairment was an IQ score in the inferior range and that, in turn, low IQ score was significantly associated with younger age at onset. This finding is consistent with a previous report⁶ of a relationship between cognitive disability and younger age at onset.

It is therefore possible to hypothesize that the more precocious the neuropathological damage in the CNS is, the more disruptive the global impact is on the development of intellectual faculties and, consequently, on the subject general intelligence.

In our sample, the proportion of patients treated with disease-modifying drugs was comparable in the cognitively impaired and preserved groups. However, it is possible that the therapy may positively influence the cognitive outcome of the subject in the long term.³²

Regarding the pattern of cognitive impairment, as in previous studies,^5,6,27,33 we observed prominent defects of verbal and visuospatial memory, complex attention, and aspects of executive functions. Extrapolation of these findings would predict a high functional impact, because skills involving these faculties are increasingly emphasized in the higher academic degrees. In particular, for their critical role, executive function deficits deserve a more detailed evaluation in further studies, using a broader range of assessment tools. We also documented language problems in 20% to 40% of the cases, particularly in terms of verbal comprehension and, to a lesser extent, verbal fluency. Deficits of receptive language have been reported in a recent study,⁶ whereas verbal fluency was normal in another series.⁵ Involvement of linguistic function is perhaps the most peculiar finding in childhood and juvenile compared with adult cases. With the exception of verbal fluency, linguistic problems are rarely reported in adults, and it is believed that they may derive from impairment in other cognitive domains.^27,34,35 Because in pediatric cases the disease occurs during a critical phase for the development of linguistic faculties, language may prove to be particularly vulnerable in this age range. Even subtle language difficulties are likely to have important functional consequences. Therefore, systematic assessment of language function in pediatric MS deserves particular attention in future studies.

Fatigue is one of the most frequent and invalidating symptoms in adult MS patients.³⁶ In our study, we used the FSS, and we applied a cutoff score of 2 on the basis of the HC assessment. With this approach, we documented significant fatigue in 73% of the cases. It is also possible that the FSS is not fully adequate to capture symptoms in a pediatric population and that some items may require an adaptation or modification. However, this finding points to the necessity to systematically assess and treat fatigue also in this age range.

Depression is another highly prevalent disturbance in MS and one of the most important determinants of the subject’s quality of life.^37,38 In a previous study, using a structured psychiatric evaluation, depression was reported in 50% of childhood and juvenile cases.⁵ Using self-assessment on the CDI, we found a 6% prevalence. In our study, the prevalence of depression can be underestimated because of low sensitivity in assessment. In this respect, one of the major limitations of the study is the lack of a systematic referral to psychiatric examination for a formal diagnosis of depression or behavioral disorder.

Finally, using an interview with parents, we confirmed that the disease had a great functional impact and that, beyond the extent of physical disability, cognitive problems play a relevant role negatively affecting school, everyday, and social activities.

On the whole, our findings emphasize the importance of systematic assessment of cognitive and psychosocial problems in childhood and juvenile MS patients. They also provide a few clues for planning future research in the field and developing comprehensive intervention strategies.

ACKNOWLEDGMENT

The authors thank Orietta Picconi, MScStat (Public Health Agency of Lazio, Italy), for the supervision of the statistical analysis and Faustino Colombo and the Fencing Society of Prato for help in the recruitment of healthy controls.

APPENDIX

Multiple Sclerosis Study Group of the Italian Neurological Society: M.P. Amato, B. Goretti, V. Zipoli, E. Portaccio, S. Centorrino, V. Contri (Department of Neurology, University of Florence); S. Lori (Neurological Unit, Meyer Hospital, Florence); M. Falcini (Neurological Unit, Hospital of Prato); G. Comi, L. Moiola, M. Falautano (Department of Neurology, San Raffaele Scientific Institute, Milan); M. Trojano, M.F. De Caro, M. Lopez (Department of Neurology, University of Bari); F. Patti, R. Vecchio (Department of Neurology, University of Catania); P. Gallo, P. Grossi (Department of Neurology, University of Padua); A. Bertolotto, S. Giampaolo (MS Centre, Hospital S Luigi Gonzaga, Orbassano, Turin); C. Pozzilli, V. Bianchi (Department of Neurological Sciences, “La Sapienza” University, Rome); I. Manca, C. Masia (Hospital of Sassari); R. Bergamaschi, P. Veggiotti (Multiple Sclerosis Center, Neurological Institute C. Mondino, Pavia); A. Ghezzi, M. Roscio (MS Centre, Hospital of Gallarate, Italy).