Cortico-juxtacortical involvement increases risk of epileptic seizures in multiple sclerosis

Ó 2013 John Wiley & Sons A/S Acta Neurol Scand DOI: 10.1111/ane.12064 ACTA NEUROLOGICA SCANDINAVICA Cortico-juxtacortical involvement increases ris...
Author: Leonard Harmon
4 downloads 0 Views 108KB Size
Ó 2013 John Wiley & Sons A/S

Acta Neurol Scand DOI: 10.1111/ane.12064

ACTA NEUROLOGICA SCANDINAVICA

Cortico-juxtacortical involvement increases risk of epileptic seizures in multiple sclerosis Martınez-Lapiscina EH, Ayuso T, Lacruz F, Gurtubay IG, Soriano G, Otano M, Bujanda M, Bacaicoa MC. Cortico-juxtacortical involvement increases risk of epileptic seizures in multiple sclerosis. Acta Neurol Scand 2013: DOI: 10.1111/ane.12064. © 2013 John Wiley & Sons A/S.

E. H. Martínez-Lapiscina1, T. Ayuso1, F. Lacruz1, I. G. Gurtubay2, G. Soriano1, M. Otano1, M. Bujanda1, M. C. Bacaicoa3

Objectives – Previous studies have reported an increased risk for epileptic seizures in multiple sclerosis (MS) patients. However, data on the pathogenesis of seizures remain inconclusive. The aim of our study is to evaluate prevalence, clinical and paraclinical features of epileptic attacks in our MS cohort and to search MS-specific risk factors for epileptic seizures. Materials and methods – In this cohort of 428 MS patients, 13 patients were identified with epileptic seizures occurring at any point during the course of MS including at MS onset. As a control group, we selected 26 MS patients without seizures and matched for gender, age and date of MS onset. We compared demographic features and clinic-radiological findings between the both groups. Results – Thirteen patients (3%) were identified as having epileptic attacks. Ten patients (77%) experienced focal seizures, half of whom had confirmed secondary generalization. We did not find an association between seizures and disease course. Most patients had a single or few (2–5) seizures. MS patients with seizures had a significantly higher number of cortical and juxtacortical lesions on T2weighted/fluid attenuation inversion recovery magnetic resonance imaging than control group [OR = 2.6 CI95% (1.0–6.5); P = 0.047]. Conclusions – Our findings support a credible role of cortical and juxtacortical involvement in the development of epileptic seizures in MS.

1 Neurology, Complejo Hospitalario de Navarra, Pamplona, Spain; 2Neurophysiology, Complejo Hospitalario de Navarra, Pamplona, Spain; 3 Neuroradiology, Complejo Hospitalario de Navarra, Pamplona, Spain

Introduction

The epidemiology of epileptic seizures among multiple sclerosis (MS) patients has been the subject of several studies. A recent review has reported a threefold increased risk for epileptic attacks in MS patients compared with general population (1). All types of seizures have been previously described in MS patients. Some authors have suggested that Focal Seizures (FS) with secondary generalization were the most common types of epileptic attacks experienced by MS patients (1–3) while others found primary generalized tonic–clonic seizures (PGTCS) almost as frequently as FS in a large cohort of 5041 MS patients (4). Epileptic attacks can affect patients at any point during the course of MS (1, 3–5).

Key words: cortico-juxtacortical lesions; epilepsy; epileptic seizures; multiple sclerosis; prevalence Martínez-Lapiscina EH, Neurology, Complejo Hospitalario de Navarra, Irunlarrea Street 3, 31008 Pamplona, Spain Tel.: 0034 948102100 Fax: 0034 948102303 e-mail: [email protected] Accepted for publication November 13, 2012

The pathogenesis of seizures is not well established (1, 4). Anatomopathological findings and magnetic resonance imaging (MRI) metrics have suggested that cortical and juxtacortical inflammation, demyelination and atrophy may be responsible for the development of epileptic seizures (6–8). However, few epidemiological studies have related MRI findings with risk of epileptic seizures in MS (7, 8). Moreover, the multifocal nature of MS and the high variability in MRI acquisition timing relative to seizure activity difficults to establish a causal association between cortical and juxtacortical involvement and seizures (1). In conclusion, although there is much research on demographic and clinical features of epileptic attacks in MS patients, the disease-specific factors involved in the pathogenesis of seizures need to 1

Martınez-Lapiscina et al. be clarified. Our aim was to compare MS patients with and without seizures in order to identify clinical or paraclinical factors related to an increased risk of development epileptic seizures. Materials and methods

and variable quality of MRI acquisitions along the study period. All patients gave informed consent to participate and the local Ethics Committee of the Complejo Hospitalario de Navarra approved the study.

Study population

Data collection

This retrospective case–control study was conducted between 1 January 2010 and 31 May 2010. The study population was drawn from the electronic database of the MS Clinic at the Complejo Hospitalario de Navarra which serves as a tertiary care reference centre for the evaluation, follow-up and treatment of MS patients for Pamplona City and Navarra region, Spain. Participants need to fulfil the following inclusion criteria: (i) age greater than 18 years; (ii) a definitive diagnosis of MS according to McDonald′s criteria (9, 10) at the time of study inclusion (May 2010); and (iii) seizures occurred at any point during the course of MS including at MS onset (this criteria only applies to MS patients with seizures). Exclusion criteria included: (i) seizures prior to MS onset; (ii) concomitant risk factors for epileptic seizures (head trauma, neoplasia, vascular malformation and cerebrovascular disease); (iii) paroxysmal phenomena that can be misclassified as seizures; and (iv) receiving antiepileptic drugs (AED) as treatment for other MS symptoms (pain, depression and spasm). By May 2010, 428 MS patients included in our database were retrospectively screened for epileptic seizures. We identified 15 patients with epileptic seizures. After reviewing their medical records, we excluded two patients who did not meet the study criteria. One patient was diagnosed with PGTCS 24 years before the MS onset. During these 24 years, he had never had other symptoms, and MRI never showed inflammation. We consider that this patient was predisposed to seizures regardless of MS. Another patient was treated with AEDs for painful spasms of the left lower extremity, and he was erroneously diagnosed with epilepsy. Finally, we included 13 patients with seizures. Then, 26 patients with no history of seizures and matched for gender, age and date of disease onset (range of variability 4 months) were randomly selected as a control group. We took into account the date of MS onset at the first seizure rather than the duration of MS, for the matching controls. We thought that this principle will help us avoid misclassification bias in the MRI findings due to the different protocols

Family and personal medical histories of neurological and autoimmune diseases, demographic and clinical data for each individual case are collected in our database. Clinical data includes: MS disease evolution (date of MS onset and diagnosis, relapses, successive disability scores, type of MS), biological, neurophysiologic and neuroimaging findings and treatments. The disability was estimated according to Kurtzke′s Expanded Disability Status Scale (EDSS) (11). Moreover, the types of MS were classified as relapsing-remitting MS (RRMS), secondary-progressive MS (SPMS) and primary-progressive MS according to Lublin and Reingold consensus (12). Finally, the epileptic seizures were registered as MS symptoms in our database. At the end, we reviewed the patients’ medical records to assess the electroclinical characteristics of their seizures. We classified the epileptic seizures according to the most recent recommendations of International League against Epilepsy (13). We considered a tonic–clonic seizure as a secondary generalized tonic–clonic seizure if there were focal symptoms, focal epileptic activity (spikes and/or sharp waves) or focal slowing in Electroencephalography (EEG). We additionally considered a tonic–clonic epileptic seizure as a PGTCS if neither focal symptoms nor focal epileptic activity or slowing in EEG was presented.

2

Image acquisition

All images were acquired using a 1.5 T AvantoSiemens (Erlangen, Germany). The MRI scans have been acquired based on two different protocols along the study period (protocol 1 dated before 2005 and protocol 2 after 2005). Protocol 1 included three different sets of images: T1weighted (T1W) sagittal image [Time Repetition (TR) 525 ms, Time Echo (TE) 17 ms] with and without gadolinium contrast and Proton-Density T2-weighted (DP-T2W) axial image [TR 2800 ms, TE 14–82 ms, slice thickness 5.0 mm]. Protocol 2 added Fluid Attenuation Inversion Recovery (FLAIR) sagittal image [TR 9000 ms, TE 109 ms, Inversion Time (IT) 2500 ms]. All patients with seizures performed a brain MRI

Multiple sclerosis and risk of epileptic seizures within 24–72 h after epileptic attack. We compared MRI findings of MS patients with seizures with those observed in the MRI of matched-controls performed at the same time (range of variability 4 months). Due to the retrospective design, some cases undergo protocol 1 and others protocol 2. However, as we selected controls matched for date of MS onset, cases and their matched-controls undergo the same MRI protocol in order to avoid misclassification bias due to the different quality of the MRI protocols. Procedure and statistical analysis

First, we describe the demographic and the clinical features, electroencephalographic and neuroimaging findings and management of epileptic seizures in our series. Then, we perform bivariate analyses to compare the demographic features and clinic-radiological findings at the time of the first seizure (or equivalent period for matchedcontrols) between patients with and without seizures. Finally, we select significant variables in these bivariate analyses and evidence from previous studies (5–7, 14) to construct a multivariate model. We fitted a logistic regression analysis adjusted for MS type, number of lesions on T2W/FLAIR MRI, number of cortical and juxtacortical lesions on T2W/FLAIR MRI and frontal and callosal atrophy on T1W MRI at the time of first seizure (or equivalent period for matchedcontrols). We used proportions to describe qualitative results and mean values with simple standard deviations to present quantitative data. Bivariate analyses were made with chi-squared test or a Fisher test for categorical variables and a t-test for independent groups for continuous variables. All P-values are two-tailed at the