Seizures, epilepsy, and MS

Any type of injury to the brain – whether focal or diffuse – can increase your risk of developing epileptic seizures. pwMS who have lots of lesions or significant atrophy are probably more at risk of developing epilepsy. The exact risk of epilepsy in MS has been a bit variable in reported studies. Because the absolute numbers of pwMS who also develop epilepsy is quite low, you need a really big cohort to see if there is any increased risk compared to people without MS. 

Another (I know I am obsessed) amazing Swedish registry study has looked into the risk of epilepsy in pwMS. They looked at all of their pwMS (14,545) and matched each person to 3 controls. Over the study period – it’s just abstract at the mo so I can’t see how long – the rate of epilepsy was 3.5% in the MS group vs 1.4% in the control group. Looking at subgroups of the pwMS, they found that the epilepsy rate was much higher (5.5%) in pwProgressiveMS vs 2.2% in pwRelapsingMS. Longer disease duration and higher EDSS were both associated with an increased risk of epilepsy.

There are a couple of caveats and confounders. There may be a diagnostic bias – pwMS may be over or under-diagnosed with seizures. DMTs themselves might increase the risk of having systemic and CNS infections which increase your risk of seizures. There are lots of other possible explanations for why this association might not be causal. I actually find it quite convincing because it makes biological sense and there is a clear dose-response relationship: the longer and the more active the disease, the higher the risk of epilepsy. 

This study shows clearly that pwMS are at a higher risk of epilepsy than age and sex-matched controls. The strong association between risk of epilepsy and both disease duration and EDSS implies that the effect is related to disease activity. I imagine that atrophy and focal lesions both drive the increased risk of seizures. Additionally, chronic inflammation itself may lead to alterations in brain networks which predispose to seizures independently of causing degeneration of focal lesions. 

This work is really important because epilepsy is another chronic disease which can impact substantially on your life and often necessitates long-term medications, many of which have nasty side effects and complex interactions with other drugs. 

These data are all retrospective and so – without full access to the paper yet – I am not sure how many of these people will have been treated in the pre-DMT era. They mention some people who have had MS for over 34 years, so clearly some of their cohort will have been treated initially with no, or low-efficacy, DMTs. If epilepsy risk is directly related to cumulative disease activity and neurodegeneration, I will be very interested to see how these figures change in the next 20 years. I imagine that with the advent of the big hitters – alemtuzumab, natalizumab, fingolimod, and now cladribine – and the move to targeting ‘No Evident Disease Activity’, the rate of epilepsy will drop towards the normal population rate. 




To determine the cumulative incidence of epilepsy in a population-based cohort of patients with multiple sclerosis (MS) and to investigate the association between epilepsy and clinical features of MS.


All available patients in the Swedish MS register (n = 14,545) and 3 age- and sex-matched controls per patient randomly selected from the population register (n = 43,635) were included. Data on clinical features of MS were retrieved from the Swedish MS register, and data on epilepsy and death were retrieved from comprehensive patient registers.


The cumulative incidence of epilepsy was 3.5% (95% confidence interval [CI] 3.17-3.76) in patients with MS and 1.4% (95% CI 1.30-1.52) in controls (risk ratio 2.5, 95% CI 2.19-2.76). In a Cox proportional model, MS increased the risk of epilepsy (hazard ratio 3.2, 95% CI 2.64-3.94). Patients with relapsing-remitting MS had a cumulative incidence of epilepsy of 2.2% (95% CI 1.88-2.50), whereas patients with progressive disease had a cumulative incidence of 5.5% (95% CI 4.89-6.09). The cumulative incidence rose continuously with increasing disease duration to 5.9% (95% CI 4.90-7.20) in patients with disease duration ≥34 years. Patients with an Expanded Disability Status Scale (EDSS) score ≥7 had a cumulative incidence of epilepsy of 5.3% (95% CI 3.95-7.00). Disease duration and EDSS score were associated with epilepsy after multiple logistic regression (odds ratio [OR] 1.03, 95% CI 1.01-1.04 per year, p = 0.001; and OR 1.2, 95% CI 1.09-1.26 per EDSS step, p < 0.0001).


Epilepsy is more common among patients with MS than in the general population, and a diagnosis of MS increases the risk of epilepsy. Our data suggest a direct link between severity of MS and epilepsy.

About the author


  • Interesting post thanks Ben and Sweden ahead of the game (again). Wonder if pwMS taking anticonvulsants as pain killers masks comorbidity to some extent? #pregabalin #gabapentin #carbamazepine #oxcarbazepine #phenytoin #valproate #lamotrigine

  • I (EDSS 8.5, 9HPT 90/170) have had two seizures in the past year – both almost fatal, save for Ros'CPR.

    Both were preceded, by a day or so, with haematuria (blood in urine) and some delusional behaviour and I am now in ged care as that has been deemed safer.

    In both cases my, life saving, treatment has been my GP's decision to use very strong anti -biotics based on the assumption that I was suffering the neuro-toxic effects of a UTI. This diagnosis was confounded by the lack of a raised temperature or detected urine pathogens so Richard's hunch is one that I stand (sit!) in awe of.

    Incidentally I had lunch yesterday with a former colleague, Andrew Straneri, of Federation University who is research active in Health Informatics and I would like to introduce you to each other.

  • A known side effect of MS walking drug fampidine is risk of seizures. I wonder if this is mentioned in the paper?

    • But is that cause or effect, i.e. is MS predisposing you to seizures which taking fampridine compounds, or are the seizures a consequence of MS which is incorrectly being assumed to be caused by fampridine? If that makes sense…

      I'm taking fampridine, so this is an area of particular interest (and PPMS with EDSS 6.5 after being diagosed less than 4 years ago, so the epilepsy itself is another cause for concern)

  • METHODS Study design and study cohort. This was a retrospective
    register-based study. The cohort consisted of patients
    with MS (cases) and controls matched for age and sex at the time
    of cohort entry. For each individual present in the Swedish MS
    register (SMSreg) by December 1, 2015 (n 5 15,810), 3 age- and
    sex-matched controls were randomly created from the population
    register per cohort entry date for the case (defined as symptom
    onset or, if missing, diagnosis of MS). Cases for whom controls
    could not be generated (n 5 1,265) because of registration errors
    (n 5 15), absent information on disease onset (n 5 782), or not
    being in the population register at the time of disease onset (n 5
    468) were excluded. The final study cohort consisted of 14,545
    cases and 43,635 controls.
    Registers, clinical data, and definitions. Data on the incidence
    and clinical characteristics of MS were collected from
    SMSreg, which at the time of data export had an estimated coverage
    of 82% ( These data were prospectively
    recorded by treating physicians. Data on epilepsy-related diagnoses
    were collected from the NPR, which is managed by the
    National Board of Health and Welfare and contains all diagnoses
    for in-patient care from 1987, some hospital-based outpatient
    care from 2001, and most hospital-based outpatient care from
    2005 to December 31, 2014. Reporting to NPR is mandatory for
    Swedish caregivers. Epilepsy was defined as ICD-9 code 345
    except Q or ICD-10 code G40. Seizure was defined as ICD-9
    code 780D or ICD-10 code R56.8. Status epilepticus was defined
    as ICD-9 code 345Q or ICD-10 code G41. Cumulative incidence
    refers to the proportion of patients with a diagnosis ever
    being registered in the NPR. For estimation of prevalence, we
    used the definition of Annegers8 and Hauser et al.,9 i.e., patients
    admitted or seen as outpatients with an epilepsy diagnosis within
    the last 5 years before December 31, 2014. Date of death was
    collected from the national death register, to which reporting is
    also mandatory. All data were cross-referenced and anonymized
    by the National Board of Health and Welfare.

  • Table 1 Demographic data and clinical characteristics of patients with MS

    n Years,mean (SD) %
    Women 10,262 71
    Men 4,283 29

    Age at MS onset: 14,524 34 (11)
    Age at end of f.up: 14,545 51 (14)
    Disease duration: 14,167 17 (12)

    n Years,mean (SD) % of subgroup
    MS course, age, and proportions

    All with stated course 13,922 51 (14)
    Primary progressive 1,244 63 (11) 8.9
    Progressive relapsing 193 51 (14) 1.4
    Relapsing remitting 8,408 45 (12) 60
    Secondary progressive 4,077 60 (11) 29


    All with 12,555
    No treatment 1,763 14
    Any treatment 10,792 86
    Ever first-line treatmentb 9,488 76
    Ever second-line treatmentc 4,552 36
    Ever azathioprine, mitoxantrone 520 4.1
    Ever HSCT 90 0.7
    n Median IQR
    EDSS score
    At last examination 8,395 3 1–5

    Sharing is caring

  • Yes my wife is now progressive MS, diagnosed 1998, epilepsy 2014. Under control using Keppra 750mg twice a day.

By DrBenJ



Recent Posts

Recent Comments