Sage Journals: Discover world-class research

Abstract

Background

Epilepsy is two to three times more common in patients with multiple sclerosis (pwMS) compared to the general population. Patients with MS and epilepsy without other identifiable causes (MS + E) show greater cortical damage than those without epilepsy (MS-E). However, it's unclear whether MS + E patients exhibit distinct cognitive and neuropsychological features requiring specific management.

Methods

In a cohort of pwMS from three MS centers, MS + E patients were identified and data on MS clinical features, epilepsy history, and treatments were collected. A matched group of MS-E patients was included. Assessments included cognitive and neuropsychiatric tests. Cognitive impairment (CI) was defined as scoring ≥1.5 standard deviations below normative values in ≥1 cognitive domain.

Results

CI was more prevalent in MS + E (n = 33) patients than in MS-E (n = 33). MS + E patients had lower processing speed (p < 0.01) and visuospatial memory (p = 0.03). MS + E was independently associated with CI (odds ratio 3.6, 95% confidence interval 1.21–12). Somatization, phobia, anxiety, and depression were the most affected neuropsychological domains in MS + E, with global psychological distress negatively correlating with processing speed (rho −0.36, p = 0.048).

Conclusions

MS + E is associated with higher CI, particularly in processing speed and visuospatial memory, alongside psychological distress, highlighting the need for targeted multidisciplinary care to improve outcomes and quality of life.

Keywords

Multiple sclerosis damage epilepsy seizure cognitive impairment neuropsychological dysfunction

Introduction

Seizures are a significant clinical manifestation in patients with multiple sclerosis (pwMS), and occurrence rates appear higher by approximately two to three times compared to the general population.¹ Seizures may occur as an initial manifestation of an acute inflammatory demyelinating event, subsequently leading to a diagnosis of MS. Still, in the majority of cases, epilepsy is diagnosed after MS, with a mean latency time of 10 years.^2,3 The prevalence varies across clinical phenotypes: progressive MS exhibits epileptic seizures at a prevalence rate of more than twice that of relapsing-remitting MS.⁴ Lower brain volume, longer disease duration, and higher disability levels are associated with a higher incidence of epilepsy in MS.⁵

Among pathophysiological mechanisms underlying seizures in MS, intracortical lesions, regional cortical atrophy, and cortical microstructural alterations have been considered as the main determinants.^6,7 In MS + E, intracortical lesion distribution shows a predilection for the temporal lobe, particularly the hippocampus.⁸ Localized atrophy is predominantly observed in the temporal lobes, including the para-hippocampal, entorhinal, fusiform gyri, and parietal lobes.^7,8 Damage to cortical gray matter and deep gray matter structures such as the hippocampus and the thalamus are also among the strongest predictors of low cognitive processing speed in MS.⁹ These findings may indicate an association (i) between MS-related damage and epilepsy and (ii) between epilepsy and neuropsychological dysfunction, though causality cannot be inferred.¹⁰

Currently, the cognitive and neuropsychological impact of epilepsy in pwMS remains unexplored, leaving the need for tailored and comprehensive care in this group of patients.

The aim of this study is to investigate the impact of epilepsy-related MS, defined as epilepsy not attributable to causes other than MS (MS + E), on cognitive and psychological outcomes in a cohort of pwMS attending large MS centers in central Italy. To this end, we assessed the cognitive and psychological profiles of MS + E patients and compared them with those of MS patients without epilepsy (MS-E). As a secondary objective, we explored whether MS + E patients exhibited distinct patterns of neuropsychiatric disturbances. Recognizing the impact of epilepsy on cognition and mood may offer ways to identify early and thus fulfill the specific needs of this cohort for improved patient management.

Methods

Participants and eligibility criteria

Patients with a diagnosis of MS¹¹ and subsequent epilepsy, without any other identifiable epilepsy etiology according to the 2014 International League Against Epilepsy (ILAE) criteria¹² were included. Cases were identified through medical review, based on the documentation of epilepsy diagnosis after MS onset, use of antiseizure medications for epilepsy and neurologists’ report of known cases. The records were drawn from approximately 4000 Italian patients attending large MS centers in central Italy (San Camillo Forlanini Hospital in Rome; SS Annunziata University Hospital in Chieti; Policlinico Campus Biomedico in Rome) between October 2021 and June 2023. Individuals manifesting psychogenic nonepileptic seizures (PNES) were excluded. A cohort of pwMS without epilepsy (MS-E), matched for age, sex, and disease duration, was involved as a control group. Other exclusion criteria included: severe motor and visual impairment that interfered with cognitive testing, incomplete data, and a history of nervous system disorders other than MS.

Data collection and definitions

Clinical data were collected from patients records, including demographics, Expanded Disability Status Scale (EDSS) scores, disease phenotypes, disease-modifying therapy (DMT) at the time of evaluation, epilepsy onset, seizure type and frequency, antiseizure medications and at least a 21-channels electroencephalogram (EEG) recording. If no prior EEG was available, a new one was acquired. Epilepsy onset was categorized as either early (within 1 year after MS onset) or late (more than 1 year after MS onset). Patients were considered seizure-free if they had no seizures in the year before evaluation.

Cognition was evaluated cross-sectionally with the Brief International Cognitive Assessment for MS (BICAMS), which includes the Symbol Digit Modalities test (SDMT), the California Verbal Learning Test Second Edition (CVLT-II), and the Brief Visuospatial Test-Revised (BVMT-R).¹³ For mood assessment, the General Anxiety Disorder 7-item (GAD-7)¹⁴ and the Beck Depression Inventory II (BDI-II) were used.¹⁵ Only epileptic patients underwent the Symptom Checklist-90 (SCL-90) test, a self-administered questionnaire assessing psychiatric alterations.¹⁶ Participants rated psychological symptoms on a five-point scale (0–4), covering nine subscales: somatization, obsessive-compulsive symptoms, interpersonal sensitivity, depression, anxiety, hostility, phobia, paranoid ideation, and psychoticism. Cognitive impairment was defined as scoring ≥1.5 standard deviations below normative values in ≥1 cognitive domain. Raw scores of cognitive and neuropsychological tests were converted to z-scores or T-scores based on normative data.^17,18 The study was performed following the Declaration of Helsinki and was approved by the local Ethics Committees. Informed consent was obtained by all participants.

Statistical analysis

Descriptive statistics (means and medians) were used based on variable distribution. Two-sample t-tests and Mann-Whitney U tests compared demographics, clinical characteristics and cognitive scores between MS + E and MS-E groups, as well as between MS + E patients with early versus late epilepsy onset. For categorical variables, chi-square or Fisher's exact tests were applied as appropriate. Cognitive data comparison effect sizes were reported with the Hedge's g statistic.

To assess the impact of epilepsy on cognitive and psychological outcomes, linear and logistic regression models were performed using BICAMS subtests scores, GAD-7, BDI-II scores and the presence of cognitive impairment (yes/no) as outcome variables. Models were adjusted for demographic variables, years of education, EDSS, and disease duration. In models where cognition was an outcome, a composite psychological burden score, calculated by summing standardized BDI-II and GAD-7 scores, was included as a covariate. The selection of covariates was based from prior literature indicating their relevance to cognitive dysfunction in pwMS.^5,19,20

To explore the relationship bidirectionally, an additional logistic regression was performed by treating epilepsy as the outcome and cognitive impairment (yes/no) as a predictor, using the same covariates. This approach was intended to investigate not only whether epilepsy predicted cognitive outcomes, but also whether cognitive impairment might be associated with the presence of epilepsy.

A one sample z-test was used to compare MS + E SCL-90 scores to normative values. Spearman correlation coefficients were used to assess relationships between cognitive performance and SCL-90 scores.

Statistical analysis was conducted using JAMOVI v2.3.28 and R version 4.4.1 with a significance level set at p < 0.05.

Results

Baseline demographic and clinical characteristics of the whole MS cohort

Out of 64 MS patients with a diagnosis of MS and epilepsy (MS + E), 33 were included in the study. Excluded patients (n = 31) were due to incomplete data (n = 15), motor or visual impairments incompatible with the assessments (n = 7), PNES diagnosis (n = 5), and epilepsy attributed to other causes (n = 4). Thirty-three matched MS-E patients were also recruited. In the whole cohort of patients (MS + E and MS-E), mean ± SD disease duration was 19.4 ± 9.9 years, median EDSS was 3.0 (interquartile range (IQR) 2–5); 68% of patients had relapsing-remitting MS (RRMS) and the rest had secondary progressive (SPMS). CI was observed in 33 out of 66 MS patients (50%). In the MS + E group, 28 out of 33 patients were receiving treatment: 18% were on high-efficacy treatments (natalizumab, ocrelizumab, and ofatumumab), 52% on oral therapies (teriflunomide, dimethyl fumarate, fingolimod, and cladribine), and 15% on platform injectables (interferons and glatiramer acetate). Among the MS-E group, 29 out of 33 patients were treated: 28% were on high-efficacy treatments, 50% were on oral therapies, and 3% were on platform injectables. Overall DMT use was comparable between the two groups (p = 1.00), and the distribution of DMT classes did not differ (p = 0.17).

Demographic and clinical features of both groups (MS + E and MS-E) are reported in Table 1.

Table 1.

Demographic and clinical characteristics of patients with MS-E and MS + E.

Variables	MS-E (n = 33)	MS + E (n = 33)	p-value
Age, mean (SD), y	48.5 (10.2)	48.9 (9.7)	0.84
Female no. (%)	22 (67)	22 (67)	1.00
EDSS score, median (IQR)	2.5 (2.0–4.5)	3.0 (2.0–5.0)	0.48
Disease phenotypes, n (%)	RRMS: 23 (70); SPMS 10 (30)	RRMS: 22 (67); SPMS 11 (33)	0.79
Disease duration, mean (SD), year	19.1 (10.1)	19.7 (9.8)	0.85
DMTs, number of patients (%)	29 (88)	28 (85)	1.00
Years of education, median (IQR)	13 (13–17)	13 (13–13)	0.57
SDMT score, mean (SD)	44.1 (14.5)	35.3 (11.2)	0.01
CVLT-II score, mean (SD)	47.8 (12.4)	44.4 (11.1)	0.25
BVMT-R score, mean (SD)	23.9 (8.9)	19.4 (8.7)	0.03
Cognitive impairment, n (%)	12 (36)	21 (64)	0.03
GAD-7 score, median (IQR)	8.6 (6–8.6)	7 (5–7.6)	0.10
BDI-II score, median (IQR)	8.8 (5–9)	10 (5–16)	0.22

BDI-II: Beck's depression inventory II; BVMT-R: brief visuospatial memory test revised; CVLT-II: California verbal learning test-2^nd edition; DMTs: disease modifying treatments; EDSS: Expanded Disability Status Scale; GAD-7: general anxiety disorder 7-item; IQR: interquartile range; MS: multiple sclerosis; MS + E: multiple sclerosis and epilepsy; MS-E: multiple sclerosis without epilepsy; SDMT: symbol digit modalities test.

Seizure characteristics, EEG findings and timing of epilepsy onset in the MS + E group

Ninety percent of MS + E patients had experienced seizures with focal onset, which generalized to bilateral tonic–clonic seizures in 50% of cases. Among the focal seizures, 57% had presented with nonmotor onset, characterized by behavioral arrest (59%), autonomic manifestations (24%), and sensory symptoms (17%). The remaining 43% had presented with motor onset, which included clonic movements, tonic posturing, and automatisms. Two patients had a history of generalized convulsive status epilepticus (GCSE), which was triggered by infections. Both had an SPMS phenotype. At least one abnormal EEG (spikes and/or focal slowing) was found in 55% of MS + E patients; temporal abnormalities were found in 72% of these patients.

Thirty of the 33 MS + E patients were on antiseizure medications, with 64% on one medication and 18% on two or more antiseizure drugs. The most frequent antiseizure treatment was levetiracetam (55%), followed by carbamazepine (21%), lacosamide (9%), sodium valproate (6%), lamotrigine (3%), and perampanel (3%).

Regarding the timing of epilepsy onset, 26 out of 33 patients (79%) developed epilepsy a mean ± SD of 10.4 ± 6.7 years after MS onset (late-onset group), while 7 patients (21%) experienced seizures within the first year of MS (early-onset group). At the time of evaluation, the median EDSS score was higher in the late-onset group, compared to the early-onset group (3.0 vs. 1.5, p = 0.03). Among late-onset patients, only 58% had an RRMS phenotype, compared to 100% in the early-onset group. Overall, 21 out of 33 MS + E patients were seizure-free. Among RRMS patients, all individuals in the early-onset group were seizure-free, compared to 47% in the late-onset group (p = 0.02, Fisher's exact test).

Cognitive and neuropsychological findings in MS+ E and MS-E

MS + E patients showed lower scores in processing speed (Hedges’ g = 0.67, p = 0.008) and visuospatial memory (Hedges’ g = 0.54, p = 0.03) compared to MS-E patients (Figure 1). Verbal learning memory appeared comparable between the groups (p = 0.25; Table 1).

Figure 1.

Cognitive scores in MS-E versus MS + E. Mean and 95% confidence interval of SDMT (a) and BVMT-R (b) scores. BVMT-R: brief visuospatial memory test-revised; MS + E: multiple sclerosis and epilepsy; MS-E: multiple sclerosis without epilepsy; SDMT: symbol digit modalities test.

CI was observed in 64% of MS + E patients, compared to 36% of MS-E patients (χ² = 4.9, p = 0.027; Figure 2). When considering only RRMS patients, CI was confirmed to be more frequent in MS + E than in MS-E: 14/22 (64%) versus 7/23 (30%) (χ² = 5, p = 0.026).

Figure 2.

Prevalence of cognitively impaired patients (%) in MS-E versus MS + E. CI: cognitive impairment; MS-E: multiple sclerosis without epilepsy; MS + E: multiple sclerosis and epilepsy.

In MS + E with CI, one domain was impaired in 57% of patients, whilst two or more domains were impaired in 43%. Processing speed was the most impaired domain, with 42% of patients affected, followed by verbal learning memory (36%) and brief visuospatial memory (24%).

Regarding depression and anxiety levels, BDI and GAD-7 scores were comparable among patients with MS + E versus MS-E (p = 0.2 and p = 0.1 respectively).

Predictors of cognitive impairment and bidirectional associations

After adjusting for covariates, a diagnosis of epilepsy was independently associated with lower cognitive performance, with reduced scores on the SDMT (β = –7.1, p = 0.01; adjusted R² = 0.39) and BVMT-R (β = –4.3, p = 0.045; adjusted R² = 0.17). No association was observed between epilepsy status and verbal memory performance as measured by the CVLT-II (p = 0.41). Among the covariates, higher psychological burden was associated with worse cognitive outcomes across multiple domains, including SDMT (β = –1.6, p = 0.029) and CVLT-II (β = –1.8, p = 0.014). Epilepsy was not associated with anxiety (GAD-7) or depression (BDI-II) scores. Within the MS + E group, no correlations were found between SDMT and BVMT-R scores and number of antiepileptic medications (p = 0.81 and p = 0.77, respectively), suggesting that medication burden did not account for the observed cognitive differences.

Epilepsy was independently associated with higher odds of CI (OR = 3.6, 95% CI: 1.21–12, p = 0.027), indicating that MS + E patients were approximately three times more likely to exhibit cognitive impairment compared to MS-E patients. The model's discrimination ability, assessed by Tjur's R², was 0.227. Younger age (OR = 0.92 p = 0.033), and longer disease duration (OR = 1.1, p = 0.048) were also associated with cognitive impairment. In a sensitivity analysis replacing age with age at onset, younger age at onset remained significantly associated with CI (p = 0.03), while disease duration lost significance (p = 0.99), suggesting that age at onset was the primary driver of the observed association. Notably, 8 patients (12%) had a paediatric-onset MS.

When isolating only RRMS patients, epilepsy was confirmed to be an independent predictor of CI (OR = 5.4, p = 0.023).

Finally, a reverse model using epilepsy as the outcome, showed that CI was associated with higher odds of epilepsy (OR = 3.34, 95% CI 1.1–10.6, p = 0.03); however, the model's discrimination ability was modest (Tjur's R² = 0.083) and no other predictors of epilepsy were found.

Neuropsychiatric characteristics of MS + E patients

Out of the nine subscales of the SCL-90 test, which was administered only to MS + E patients, higher scores compared to normative values were found in somatization (p = 0.014), obsessive-compulsive symptoms (p = 0.044), depression (p = 0.02), anxiety (p = 0.007), phobia (p = 0.001), and psychoticism (p = 0.018). In MS + E, somatization (28% of patients), followed by phobia, anxiety, and depression, were the most frequent neuropsychological domains affected (Figure 3). Anxiety (p = 0.041), depression (p = 0.003), and psychoticism (p = 0.031) levels were higher in MS + E patients with at least one abnormal EEG (either epileptic or slow activity) than in MS + E patients with normal EEGs.

Figure 3.

Prevalence of psychiatric comorbidities in the MS + E cohort. In darker colors are indicated subscales of the Symptom Checklist-90 test with higher scores compared to normative values. MS + E: multiple sclerosis and epilepsy.

The aggregate measure of global psychological distress, derived from the sum of all the subscales of the SCL-90, exhibited a negative correlation with processing speed (rho = −0.36, p = 0.048), that is, patients with higher psychological distress exhibited lower processing speed scores.

Discussion

This study assessed the relationship between epilepsy and neuropsychological dysfunction in pwMS, most of whom were on DMTs. We showed that CI, particularly in processing speed and visuospatial memory, is more frequent in MS + E patients. After accounting for relevant covariates, epilepsy emerged as an independent predictor of CI, and was, in turn, predicted by the presence of CI. Anxiety and depression scores did not differ between patients with our without epilepsy. However, detailed neuropsychological assessment in MS + E patients revealed elevated scores, relative to normative data, in several domains including somatization, obsessive-compulsive symptoms, depression, anxiety, phobia, and psychoticism. Characterization of seizure pattern showed that epilepsy onset was mainly focal and generalized to bilateral tonic–clonic seizures in 50% of cases, with temporal abnormalities found in most cases. Epilepsy mainly appeared later in the MS course, though 21% had seizures within a year of MS onset. Among those with later-onset epilepsy, only 58% had a RRMS phenotype. All RRMS patients with early-onset epilepsy were seizure-free, compared to 47% in the late-onset group.

CI prevalence in the whole MS cohort aligns with prior studies on cognitive dysfunction, which have shown a range from 30% to 70%, depending on heterogeneity in methods of ascertaining cognitive decline and different populations involved.²¹ Here, we extended this evidence by showing a higher prevalence of CI in patients with MS-related epilepsy. This finding is unlikely to be explained by antiseizure medication use, as no correlation was observed between cognitive test scores and the number of antiseizure drugs taken. Moreover, the most commonly used medication was levetiracetam, which is not typically associated with cognitive impairment.²²

MS + E patients were approximately three times more likely to exhibit CI than controls, even after adjusting for demographic and clinical covariates, including EDSS, disease duration, depression, and anxiety. This suggests that epilepsy may act as either a risk factor for, or as an indicator of cognitive dysfunction. A recent large-scale study by Freedman et al.²³ similarly reported an association between seizure history and poorer cognitive outcomes in MS. However, their findings were based on self-reported seizures, whereas we confirmed epilepsy diagnoses through clinical assessments, and provided a more detailed characterization of seizure timing, semiology, and psychiatric comorbidities. Another study on pwMS reported an association between worse cognitive symptoms in patients with poor seizure control,⁶ a relationship that however we could not assess, as the majority of our patients (64%) were seizure-free.

Beyond the effect of epilepsy, longer disease duration and younger age at onset were associated with greater cognitive impairment. This finding supports evidence that earlier MS onset, including paediatric-onset MS, may confer increased long-term cognitive risk.²⁴ Our cohort had a mean disease duration of 20 years, and included 8 cases of paediatric-onset MS.

While epilepsy was one of the predictors of CI, it was predicted only by CI. This asymmetric relationship between the two variables, reflecting a difference in frequency of CI and epilepsy in pwMS, suggests that shared underlying mechanisms subserve both clinical expressions in predisposed individuals.

Our cohort of MS + E patients suffered mainly from focal onset seizures, whose clinical characteristics suggested a temporal localization. Also, 72% of MS + E patients with abnormal EEGs exhibited temporal alterations. CI has been described in patients with temporal lobe epilepsy, with elevated seizure frequency and on high number of antiepileptic medications.²⁵ MRI studies in MS patients with epilepsy have demonstrated a predilection for temporal lobe lesions.²⁶ Although temporal-derived seizures may affect cognition through transient behavioral disconnection with surroundings, most of our MS + E patients were seizure-free and only taking one antiepileptic medication. This suggests that factors beyond seizure activity, such as shared structural or inflammatory mechanisms, subserve both epilepsy and CI in this population.

Seizures in MS have been described particularly in patients with intracortical lesions,¹⁰ which may act as triggers for both seizures and cognitive dysfunction.²⁷ While epilepsy is commonly viewed as a cortical disease with epileptogenesis due to a lesional site, connectivity studies on post-stroke lesion-related epilepsy revealed that disruption of deep brain structures might also be implicated through compromised inhibitory control mechanisms.²⁸ Therefore, in MS, lesions occurring in other brain areas, including deep gray matter, could also promote functional diaschisis-related epilepsy.

Also, our study showed that the majority of cases of MS-related epilepsy occur later (rather than at the time of) the onset of MS, confirming evidence from large population based studies²⁹ suggesting that seizures may result from a complex combination of lesion occurrence and neurodegeneration. Yet, most of our epilepsy cases had a relapsing remitting course, indicating that also an inflammatory milieu, rather than a purely degenerative one, may contribute to epileptogenesis. Within RRMS patients, patients with early-onset epilepsy demonstrated higher rates of seizure freedom compared to those with epilepsy occurring later in the MS course. Along with the evidence that only a small minority of MS patients develops epilepsy, this reinforces the possibility of different underlying etiologies influencing clinical manifestations over time: a complex interaction of factors, including specific lesion types and localization together with regional brain microstructural and metabolic alterations, may determine the chances that an individual has of manifesting seizures at different time points along the MS course.

Although our MS + E cohort did not differ from the MS-E cohort in standard measures of depression and anxiety, a more detailed psychological assessment in the epileptic group revealed a broader spectrum of psychopathological alterations. Apart from anxiety and depression, also somatization and phobia were highly prevalent, each affecting approximately 25% of the MS + E cohort. These findings suggest that although epilepsy may not influence overall levels of depression or anxiety in MS, it could contribute to a more complex neuropsychiatric profile not fully captured by conventional screening tools. Yet, we must acknowledge that symptoms such as somatization may also arise from MS itself,³⁰ and therefore may not reflect a profile specific to MS-related epilepsy.

We acknowledge that our study has some limitations. First, we may have underestimated the number of patients with MS + E, but this was not intended as an epidemiological study and our primary aim was to ensure a high level of specificity (over sensitivity) in defining MS-related epilepsy. Second, the BICAMS battery does not evaluate important domains, such as executive functions, which can also be affected in MS. However, it is the recommended minimal cognitive assessment in MS that can be performed also in nonspecialised centers. Third, we did not administer the comprehensive psychological evaluation (SCL-90) to the control group (MS-E), limiting this evaluation to the epileptic cohort. To our knowledge, no psychiatric screening tool, including the SCL-90, has been formally validated in MS populations, which may limit the interpretability of some findings. Fourth, we lacked complete information on symptomatic treatments (e.g. baclofen or fampridine) that may influence seizure threshold. Additionally, data on medical comorbidities other than anxiety or depression were not systematically collected. Finally, although multiple comparisons were performed, we did not apply formal correction methods (e.g. Bonferroni), as this was an exploratory, hypothesis-driven study. However, we acknowledge the increased risk of type I error.

In conclusion, our findings highlight that patients with MS and epilepsy not explained by causes other than MS show higher prevalence of cognitive dysfunction than patients with MS only. In these patients, pathological distress can be demonstrated, particularly in the domain of somatization. Anatomical substrates shared by epilepsy and CI may play a role in this co-existence. Cognitive testing and rehabilitation strategies should be implemented early in this group of patients. Also, pwMS and cognitive dysfunction should be evaluated carefully when short-lasting symptoms manifest, as they may reflect underestimated features of seizures related to MS.

Footnotes

Acknowledgements

This research was carried out with Prof. Tomassini's institutional support and resources.

Declaration conflicts of interest

CT has the honoraria for speaking and travel grants from Almirall, Bayer, Biogen, Merck, Novartis, Roche, Sanofi and Teva. LP has the personal fees and nonfinancial support from Biogen, Bristol- Mayer Squibb, Merck, Novartis, Roche, Sanofi and Viatris. FC has the travel grants and/or speaking honoraria from Biogen, Merck, Roche and Sanofi and research grants from Merck. AC has the travel funding from Bristol-Meyers Squibb and Sanofi. SH has the travel funding and/or speaker honoraria from Biogen, CSL Behring, Novartis, Roche and Sanofi. SR has the honoraria from Biogen, BMS, Merck, Novartis, Roche for consulting services, speaking and/or travel support. CG has the honoraria for speaking and travel grants from Biogen, Merck, Novartis, Roche, Sanofi, Teva and Viatris. FD has the speaking honoraria from EISAI, Angelini, Jazz Pharmaceutical; and travel support from EISAI, Angelini, Jazz Pharmaceutical, UCB. VT has the honoraria, travel grants, and research funds from Almirall, Alexion, Merck Serono, Biogen, Bristol Myers Squibb, Novartis, Sanofi Genzyme, Janssen, Viatris, Roche, and Lundbeck.

Funding

The authors received no financial support for the research, authorship, and/or publication of this article.

ORCID iDs

Valeria Pozzilli

Carla Tortorella

Luca Prosperini

Maria Chiara Pantuliano

Alessandro Cruciani

Claudio Gasperini

Fedele Dono

Valentina Tomassini

Data availability

Fully anonymised tabulated data will be available from the corresponding author upon reasonable request.

References

Kuntz

Matheson

, et al. Association between multiple sclerosis and epilepsy: a systematic review and meta-analysis. Mult Scler Relat Disord 2023; 69: 104421.

Dagiasi

Vall

Kumlien

, et al. Treatment of epilepsy in multiple sclerosis. Seizure 2018; 58: 47–51.

Striano

Carrieri

, et al. Epilepsia partialis continua as a first symptom of multiple sclerosis: electrophysiological study of one case. Mult Scler 2003; 9: 199–203.

Burman

Zelano

. Epilepsy in multiple sclerosis: a nationwide population-based register study. Neurology 2017; 89: 2462–2468.

Pozzilli

Haggiag

Di Filippo

, et al. Incidence and determinants of seizures in multiple sclerosis: a meta-analysis of randomised clinical trials. J Neurol Neurosurg Psychiatry 2024; 95: 612–619.

Uribe-San-Martín

, et al. Prevalence of epilepsy in a cohort of patients with multiple sclerosis. Seizure 2014; 23: 81–83.

Ciolac

, et al. Altered grey matter integrity and network vulnerability relate to epilepsy occurrence in patients with multiple sclerosis. Eur J Neurol 2022; 29: 2309–2320.

Calabrese

, et al. Epilepsy in multiple sclerosis: the role of temporal lobe damage. Mult Scler 2017; 23: 473–482.

Marzi

, et al. Prediction of the information processing speed performance in multiple sclerosis using a machine learning approach in a large multicenter magnetic resonance imaging data set. Hum Brain Mapp 2023; 44: 186–202.

10.

Calabrese

, et al. Cortical pathology in multiple sclerosis patients with epilepsy: a 3 year longitudinal study. J Neurol Neurosurg Psychiatry 2012; 83: 49–54.

11.

Thompson

, et al. Diagnosis of multiple sclerosis: 2017 revisions of the McDonald criteria. Lancet Neurol 2018; 17: 162–173.

12.

Fisher

, et al. ILAE official report: a practical clinical definition of epilepsy. Epilepsia 2014; 55: 475–482.

13.

Benedict

RHB

, et al. Brief International Cognitive Assessment for MS (BICAMS): international standards for validation. BMC Neurol 2012; 12: 55.

14.

Spitzer

Kroenke

Williams

JBW

, et al. A brief measure for assessing generalized anxiety disorder: the GAD-7. Arch Intern Med 2006; 166: 1092–1097.

15.

Sacco

, et al. Psychometric properties and validity of Beck Depression Inventory II in multiple sclerosis. Eur J Neurol 2016; 23: 744–750.

16.

Lignier

, et al. The structure of the symptom checklist-90-revised: global distress, somatization, hostility, and phobic anxiety scales are reliable and robust across community and clinical samples from four European countries. Psychiatry Res 2024; 331: 115635.

17.

Sarno

Preti

Prunas

, et al. SCL-90-R, Adattamento Italiano . 2011.

18.

Parmenter

Testa

Schretlen

, et al. The utility of regression-based norms in interpreting the minimal assessment of cognitive function in multiple sclerosis (MACFIMS). J Int Neuropsychol Soc 2010; 16: 6–16.

19.

Ruano

, et al. Age and disability drive cognitive impairment in multiple sclerosis across disease subtypes. Mult Scler 2017; 23: 1258–1267.

20.

Benedict

RHB

Amato

DeLuca

, et al. Cognitive impairment in multiple sclerosis: clinical management, MRI, and therapeutic avenues. Lancet Neurol 2020; 19: 860–871.

21.

Meca-Lallana

, et al. Cognitive impairment in multiple sclerosis: diagnosis and monitoring. Neurol Sci 2021; 42: 5183–5193.

22.

Foster

, et al. Antiepileptic drugs are not independently associated with cognitive dysfunction. Neurology 2020; 94: e1051–e1061.

23.

Freedman

Meza

, et al. Seizure history and cognitive dysfunction in people with multiple sclerosis. Mult Scler 2025; 668–676.

24.

McKay

, et al. Long-term cognitive outcomes in patients with pediatric-onset vs adult-onset multiple sclerosis. JAMA Neurol 2019; 76: 1028–1034.

25.

Novak

Vizjak

Rakusa

. Cognitive impairment in people with epilepsy. J Clin Med 2022; 11: 267.

26.

Saridas

Mesut

Ozpar

, et al. Coexistence of epilepsy or seizure and multiple sclerosis; review of the literature with a single center experience. Mult Scler Relat Disord 2024; 92: 105948.

27.

Calabrese

, et al. Cortical lesions and atrophy associated with cognitive impairment in relapsing-remitting multiple sclerosis. Arch Neurol 2009; 66: 1144–1150.

28.

Schaper

FLWVJ

, et al. Mapping lesion-related epilepsy to a human brain network. JAMA Neurol 2023; 80: 891–902.

29.

Benjaminsen

Myhr

K-M

Alstadhaug

. The prevalence and characteristics of epilepsy in patients with multiple sclerosis in nordland county, Norway. Seizure 2017; 52: 131–135.

30.

Bruce

Arnett

. Longitudinal study of the symptom checklist 90-revised in multiple sclerosis patients. Clin Neuropsychol 2008; 22: 46–59.

Epilepsy in MS and its association with cognitive and psychological burden

Abstract

Background

Methods

Results

Conclusions

Keywords

Introduction

Methods

Participants and eligibility criteria

Data collection and definitions

Statistical analysis

Results

Baseline demographic and clinical characteristics of the whole MS cohort

Seizure characteristics, EEG findings and timing of epilepsy onset in the MS + E group

Cognitive and neuropsychological findings in MS+ E and MS-E

Predictors of cognitive impairment and bidirectional associations

Neuropsychiatric characteristics of MS + E patients

Discussion

Footnotes

Acknowledgements

Declaration conflicts of interest

Funding

ORCID iDs

Data availability

References