Abstract
Purpose
In people with suspected inflammatory CNS disease, cerebrospinal fluid (CSF) is commonly analyzed. Antibody-associated limbic encephalitis (ab-LE) and anti-NMDAR-encephalitis are recognized as two major syndromes of autoimmune epilepsies. Here, we investigated the diagnostic value of CSF findings in these two entities.
Methods
We reviewed patients from our tertiary epilepsy centre with ab-LE and anti-NMDAR-encephalitis in whom CSF examination including oligoclonal bands (OCB) was performed. Ab-LE patients were subdivided according to antibodies (voltage-gated potassium channels, VGKC; glutamic acid decarboxylase, GAD) or presence of onconeural antibodies/presence of tumour into three groups: VGKC-LE, GAD-LE or paraneoplastic LE (PLE). As controls, patients with CSF investigations in whom autoimmune origin was initially assumed but not confirmed later on were included. In addition, a review of published ab-LE and anti-NMDAR-encephalitis cases with reported CSF data was performed.
Results
55 ab-LE (23 VGKC-LE, 25 GAD-LE, 7 PLE) and 14 anti-NMDAR-encephalitis patients were identified at our centre. OCB were significantly more frequent in ab-LE and anti-NMDAR-encephalitis than in controls. Literature review identified 150 ab-LE and 95 NMDAR cases. Analysis of pooled data confirmed that presence of OCB was significantly more frequent in ab-LE and anti-NMDAR-encephalitis (especially in people with GAD-LE and anti-NMDAR encephalitis) as compared to controls. Sensitivity and specificity of OCB in the pooled ab-LE and anti-NMDAR-encephalitis patients was 34% and 96%, respectively. In patients with ab-LE and anti-NMDAR-encephalitis, the likelihood of OCB in CSF was 8.5-fold higher as compared to controls. Furthermore, in the pooled ab-LE and anti-NMDAR-encephalitis patients, cell counts in CSF were more frequently elevated (especially in those with anti-NMDAR encephalitis) than in controls, whereas protein content of CSF was not different between the groups.
Conclusion
OCB, and to a lesser extent cell counts in CSF, appear to be helpful additional CSF markers in the diagnostic evaluation of people presenting with a constellation suggestive for GAD-LE, PLE and anti-NMDAR-encephalitis, prompting subsequent analysis of specific antibodies.
Keywords
1. Introduction
There is increasing evidence that autoimmune-mediated mechanisms lead to epilepsy in a considerable proportion of patients.
1
Antibody-associated limbic encephalitis (ab-LE) and anti-NMDAR-encephalitis are the two major syndromes. Ab-LE is described as paraneoplastic or non-paraneoplastic syndrome. Several non-paraneoplastic subforms associated with serologically characterized auto-antibodies (abs) were identified including those with abs against voltage-gated potassium channels (VGKC) and glutamic acid decarboxylase (GAD).1
Anti-NMDAR-encephalitis was first described as severe encephalopathic syndrome in young women associated with a higher rate of ovarian teratoma,2
but it also occurs in children, men and in people without teratomas.3
In general, these autoimmune-mediated epilepsies are rare and multifaceted, and unfamiliar initial symptoms may hamper the correct diagnosis at an early stage.
4
In other neurological conditions, analysis of cerebrospinal fluid (CSF) is widely used in routine neurological diagnostics to reveal constellations suggestive for ongoing acute or chronic CNS inflammation. Especially the selective presence of unmatched oligoclonal bands (OCB) in CSF but not in serum is considered as a strong, although unspecific marker of chronic CNS inflammation.5
Here, we asked whether OCB and other CSF features are associated with ab-LE and anti-NMDAR-encephalitis. To this end, we investigated all ab-LE and anti-NMDAR-encephalitis patients at our centre and performed a thorough review of published cases in literature with regard to CSF findings.2. Methods
2.1 Patient selection
All patients from our tertiary epilepsy centre in Bonn with diagnosis of ab-LE and anti-NMDAR-encephalitis and complete CSF investigations (see below) were retrospectively included. In all these patients, we have routinely performed 3 Tesla brain MRI, routine EEG for 20 min or 24 h video-EEG telemetry using scalp electrodes (10–20 system), neuropsychological testing, determination of known abs in serum and CSF (also in order to exclude encephalitis due to infectious agents).
Ab-LE and anti-NMDAR-encephalitis was defined by new-onset epilepsy (seizure onset not longer than 5 years prior to first assessment at our centre independently of seizure frequency and whether seizures were the sole or predominant clinical symptom) and the presence of abs in serum or CSF. According to additional clinical symptoms and specific antibodies, the following two syndromes were distinguished:
- 1.Ab-LE: Patients with “limbic” signs and symptoms (≥1 of the following: seizures of temporal lobe semiology, disturbance of episodic memory, affective disturbances), brain MRI revealing mesiotemporal encephalitis (T2/FLAIR hyperintensity without atrophy) and presence of specific abs and/or identification of a tumour.1,6In all patients serum was tested for presence of the following abs: VGKC, GAD, onconeural abs, N-methyl-d-aspartate-receptor (NMDAR).
- 2.Anti-NMDA-encephalitis: Patients were assessed according to described clinical features and identification of NMDAR-abs.2
As controls, we included all patients who have undergone CSF investigations for initial suspicion of autoimmune origin which has not been confirmed later on. The diagnosis of non-autoimmune epilepsy was based on the absence of known abs, absence of typical MRI findings (e.g. initial swelling of mesiotemporal structures and subsequent remission or development of atrophy) and the presence of other plausible aetiology for epilepsy.
2.2 Review of the literature
We performed a systematic review of the literature using the following search terms: limbic encephalitis, limbic encephalitis VGKC, limbic encephalitis GAD, anti-NMDAR-encephalitis. Limits were defined as: published in the last 10 years, only items with link to full text, humans, type of article (Clinical Trial, Randomized Controlled Trial, Case reports, Classical article, Journal Article), subsets MEDLINE, language English. All previous publications from our institution in Bonn were excluded to avoid data interferences. Only publications with data of CSF with OCB determination and clearly assigned to individual patients were included in further analysis.
2.3 Comparison of patient cohorts
First, CSF data of the ab-LE and anti-NMDAR-encephalitis patients of our centre were compared to our control group (see Section 2.6). Second, we have pooled all CSF data (including our own patients and the published cases) and compared these pooled data to our local controls. Individual CSF data were not reported in all of the published patients. In those cases missing data were excluded from analysis.
2.4 CSF analysis
All data were obtained in clinical routine diagnostics. Complete CSF analysis included cell count, protein content, and evaluation of blood-CSF-barrier function, intrathecal IgG synthesis by Reiber-scheme and by OCB. White cell counts >5/μl without erythrocytosis and CSF protein >500 mg/l were defined as abnormally elevated. As internationally accepted parameter for blood-CSF function age-dependent albumin-CSF/serum-quotient QAlb was used.
5
Intrathecal IgG synthesis as an indicator for humoral immune response was assessed quantitatively by IgG-CSF/serum-quotient QIgG (values >0.7 indicating for intrathecal synthesis)5
and qualitatively by OCB in isoelectric focusing (IEF) according to international consensus.7
Unmatched OCBs in CSF and not in serum were considered as markers for autochthone intrathecal IgG production. If available, repeated CSF investigations in patients were considered in the same manner.2.5 Ab determination
Ab testing was performed as follows: Serum VGKC-abs were assessed by radioimmunoprecipitation assay (RIA; normal values <100 pmol/l, AV, Weatherall Institute, Oxford, UK and EUROIMMUN laboratory, Luebeck, Germany).
8
Identification of GAD-abs in serum was performed by RIA using 125I-GAD (normal values ≤1 U/ml, AV)6
or by indirect immunofluorescence (IFT, nomal values <1:10, EUROIMMUN). “Well characterized” onconeural abs were also tested by a commercially available routine test using an immune-dot-blot for Hu, Ma, amphiphysin, CV2/CRMP5 abs (Ravo Diagnostika, Freiburg, Germany). NMDAR-abs were tested by specific immunofluorescence test (IFT, AV and EUROIMMUN).9
2.6 Statistics
For nominal data, Fisher's exact test was used. Due to multiple comparisons, p-values were adjusted according to the Holm–Bonferroni stepwise correction procedure, p-value <0.05 were considered significant. Sensitivity and specificity were calculated according to gold-standard test. For assessing the value of performed diagnostic tests, the likelihood ratio was calculated with values >1 indicating positive association with disease.
3. Results
3.1 Patient characteristics
A total of 69 patients with ab-LE and anti-NMDAR-encephalitis were identified in Bonn: 55 ab-LE and 14 anti-NMDAR encephalitis patients. Ab-LE patients were subclassified according to ab identification (VGKC-LE, GAD-LE) or onconeuronal abs and detected tumour (paraneoplastic LE, PLE; two patients had onconeuronal ab without detected tumour), for distribution and clinical patient details, see Table 1, Table 2. Twenty-eight control patients were included. They were finally diagnosed as having psychogenic non-epileptic seizures (N = 7), syncope (N = 5), structural epilepsies of other origin than auto-immune mediated mechanisms (N = 6: 2 tumour, 1 infarction, 1 severe head trauma, 2 dysplasia), psychiatric disorders (N = 5: 2 depression, 1 anxiety, 1 psychosis, 1 obsessive–compulsive disorder), dementia (N = 3) or with other conditions (N = 2: 1 transitory ischaemic attacks, 1 normal pressure hydrocephalus) or no identified disease (N = 1). The patient positive for OCB was a 68-year-old woman with a new-onset structural epilepsy due to infarction.
Table 1CSF features of ab-LE and anti-NMDAR-encephalitis in the Bonn patients.
| Bonn patients | Elevated cell counts (>5/μl) | Elevated protein content (>500 mg/l) | Blood-CSF-disturbance | Intrathecal IgG synthesis (QIgG) | Intrathecal IgG synthesis (OCB) |
|---|---|---|---|---|---|
| All (N = 69) | 11 | 16 | 19 | 2 | 25 |
| Ab-LE all (N = 55) | 9 | 11 | 14 | 2 | 18 |
| VGKC-LE (N = 23) | 3 | 6 | 8 | 1 | 0 |
| GAD-LE (N = 25) | 3 | 2 | 4 | 1 | 14 |
| PLE (N = 7) | 3 | 3 | 2 | 0 | 4 |
| NMDAR (N = 14) | 2 | 5 | 5 | 0 | 7 |
| Non-autoimmune controls (N = 28) | 1 | 8 | 9 | 0 | 1 |
| p-Values (all vs. controls) | 0.17 | 0.61 | 0.81 | 1 | <0.001 |
| p-Values (ab-LE vs. controls) | 0.15 | 0.42 | 0.61 | 0.55 | 0.002 |
| p-Values (NMDAR vs. controls) | 0.25 | 0.73 | 1 | 1 | <0.001 |
Ab-LE: antibody associated limbic encephalitis; VGKC: antibodies to VGKC; GAD: antibodies to GAD; PLE: paraneoplastic LE; NMDAR: anti-NMDAR-encephalitis; QIgG: IgG-CSF/serum-quotient according to Reiber
5
; OCB: unmatched oligoclonal bands in CSF, not in serum.a Fisher exact test, significances after Holm–Bonferroni corrections are highlighted in bold.
Table 2Clinical characteristics of patients with ab-LE and anti-NMDAR-encephalitis in Bonn.
| Bonn patients | Gender | Median age at onset in years (range) | Median disease duration in months (range) |
|---|---|---|---|
| All (N = 69) | 41 ♀ | 38 (8–73) | 10 (0–252) |
| Ab-LE all (N = 55) | 28 ♀ | 43 (8–73) | 10 (0.3–252) |
| VGKC-LE (N = 23) | 7 ♀ | 57 (20–73) | 7 (0.3–23) |
| GAD-LE (N = 25) | 16 ♀ | 25 (8–65) | 40 (1–252) |
| PLE (N = 7) | 5 ♀ | 47 (14–60) | 34 (0.4–71) |
| Anti-NMDAR-Enzephalitis (N = 14) | 13 ♀ | 25 (17–58) | 16 (0–168) |
| Non-autoimmune controls (N = 28) | 16 ♀ | 41 (16–73) | 13 (0.6–116) |
For abbreviations see legend in Table 1.
3.2 Review of the literature
Four hundred and fifty-seven articles matched searching criteria, 58 studies matched inclusion criteria: 36 dealt with ab-LE (VGKC: 12; GAD: 8: onconeural: 13; various: 3) and 22 focused on anti-NMDAR-encephalitis. In a total of 245 reported ab-LE and anti-NMDAR-encephalitis patients, CSF data were individually assigned as follows: 150 ab-LE patients (86 VGKC, 24 GAD, 33 PLE, 7 with combination of abs) and 95 NMDAR patients. Details and references of literature are given in Supplementary Table 1.
3.3 CSF findings
3.3.1 Bonn patients
CSF findings are highlighted in Table 1. Only intrathecal IgG synthesis ascertained by OCB was different between the patient groups: it was more frequent in ab-LE and anti-NMDAR-encephalitis patients than in controls (p < 0.001) even when comparing separately anti-NMDAR-encephalitis (p < 0.001) and LE (p = 0.002). Notably, consideration of ab-LE-subforms revealed that presence of OCB in VGKC-LE was not different to controls, whereas OCB was significantly associated with GAD-LE and PLE (VGKC vs. controls p = 1; GAD p < 0.001; PLE p = 0.003). Sensitivity for OCB amounted only to 36% in the Bonn patients with ab-LE and anti-NMDAR-encephalitis (ab-LE: 33%; NMDAR: 50%), whereas specificity amounted to 96%. The likelihood for OCB in people with ab-LE and anti-NMDAR-encephalitis was 9-fold higher as compared to controls (likelihood ratio; ab-LE: 8.25, NMDAR: 12.5).
Serial CSF investigations were performed in 29 (42%) ab-LE and anti-NMDAR-encephalitis patients (19 ab-LE, 10 NMDAR) over a median period of 15 months (range 1–84), but not in the control group. OCB were positive in 13 of these patients (43%). Interestingly, in those patients in whom OCB was not detected in the first CSF examination, OCB was not revealed in the further course (see Supplementary Table 2).
3.3.2 Pooled data of published cases and patients from Bonn
In published ab-LE and anti-NMDAR-encephalitis patients (N = 245), elevated cell counts were reported in 112/235 (48%) patients and elevated protein counts in 73/229 (32%) patients. OCB were reported in CSF of 82/199 (41%) ab-LE and anti-NMDAR-encephalitis patients: in 35/136 (26%) published ab-LE patients (8/77 VGKC, 13/23 GAD, 14/29 PLE) and in 47/63 (75%) published NMDAR patients.
Our CSF findings were pooled with those of previously published cases and compared to our controls (Table 3). Again, presence of OCB was more frequently detected in ab-LE and anti-NMDAR-encephalitis (p < 0.001), and more specifically in the subgroup of anti-NMDAR-encephalitis (p < 0.001) as compared to controls. Furthermore, cell counts were elevated more frequently in ab-LE and anti-NMDAR-encephalitis patients (p < 0.001) than in controls, especially in patients with anti-NMDAR-encephalitis (p < 0.001) but not in patients with ab-LE when compared separately.
Table 3CSF features of ab-LE and anti-NMDAR-encephalitis in the pooled Bonn and published patients.
| Pooled data | Elevated cell counts (>5/μl) | Elevated protein content (>500 mg/l) | Intrathecal IgG synthesis (OCB) |
|---|---|---|---|
| All (N = 314) | 123/304 | 89/298 | 107/268 |
| Ab-LE all (N = 205) | 44/197 | 59/197 | 53/191 |
| VGKC-LE (N = 109) | 23/106 | 37/106 | 8/100 |
| GAD-LE (N = 49) | 3/48 | 5/48 | 27/48 |
| PLE (N = 40) | 18/36 | 15/36 | 18/36 |
| Others (N = 7) | 0/7 | 2/7 | 0/7 |
| Anti-NMDAR-Enzephalitis (N = 109) | 79/107 | 30/101 | 54/77 |
| Non-autoimmune Bonn controls (N = 28) | 1/28 | 8/28 | 1/28 |
| p-Values (all vs. controls) | <0.0001 | 1 | <0.0001 |
| p-Values (ab-LE vs. controls) | 0.02 | 1 | 0.004 |
| p-Values (NMDAR vs. controls) | <0.0001 | 1 | <0.0001 |
For abbreviations see legend in Table 1.
Values of sensitivity, specificity and likelihood ratios for OCB and for elevated cell counts in CSF are given for the pooled Bonn/published ab-LE and anti-NMDAR-encephalitis patients in Table 4A, Table 4B.
Table 4ASensitivity, specificity and likelihood ratio of OCB for the pooled data (our centre and published patients).
| Pooled data OCB | Sensitivity % | Specificity % | Likelihood ratio |
|---|---|---|---|
| All vs. controls | 34 | 96 | 8.5 |
| Ab-LE vs. controls | 28 | 96 | 7 |
| VGKC | 7 | 96 | 1.75 |
| GAD | 59 | 96 | 14.75 |
| PLE | 50 | 96 | 12.5 |
| NMDAR vs. controls | 67 | 96 | 16.75 |
For abbreviations see legend in Table 1.
Table 4BSensitivity, specificity and likelihood ratio of elevated cell counts in CSF for the pooled data (our centre and published patients).
| Pooled data Elevated cell counts | Sensitivity % | Specificity % | Likelihood ratio |
|---|---|---|---|
| All vs. controls | 40 | 96 | 10 |
| Ab-LE vs. controls | 22 | 96 | 5.5 |
| VGKC | 22 | 96 | 5.5 |
| GAD | 6 | 96 | 1.5 |
| PLE | 50 | 96 | 12.5 |
| NMDAR vs. controls | 74 | 96 | 18.5 |
For abbreviations see legend in Table 1.
4. Discussion
We have found that (i) OCB are more frequently observed in people with ab-LE and anti-NMDAR-encephalitis, particularly in people with GAD-LE and anti-NMDAR encephalitis, and that (ii) abnormally elevated cell counts are associated with anti-NMDAR encephalitis. Thus, CSF findings may be a useful marker in the diagnostic evaluation of people in whom autoimmune-mediated epilepsy is suspected, prompting further examinations including detailed search for antibodies typically associated with subforms of autoimmune-mediated epilepsies.
Due to the retrospective design, however, our study has several limitations. First, our clinic is a tertiary reference centre for people with chronic epilepsy, and our data is thus biased by the presence of recurrent seizures. People with autoimmune-mediated processes without seizures or people with seizures due to viral or bacterial encephalitis were not considered. Second, because of diagnostic uncertainties and lacking diagnostic “gold-standard”, people in whom autoimmune-mediated epilepsy was suspected, but in whom known abs (see methods) were not detected, were excluded from the study. Third, we have selectively included patients in whom CSF and OCB investigations have been performed based on clinical judgement. Therefore, our study is biased for patients with CSF samples and initial assumption of CNS inflammation. For clinical purpose, it would be of outstanding interest to know the predictive positive and negative values of CSF findings for ab-LE and anti-NMDAR-encephalitis. This is, however, not possible, as we cannot reliably estimate the prevalence of ab-LE and anti-NMDAR-encephalitis.
Despite these limitations, our study provides extensive data (including literature review) on the usefulness of CSF findings in a clinically well characterized population including a variety of autoimmune-mediated epilepsy syndromes. We have found that OCB are more frequently observed in people with ab-LE and anti-NMDAR-encephalitis with a likelihood of 8.5-fold higher as compared to controls, suggesting that especially OCB are a useful supportive marker in the diagnostic evaluation of autoimmune-mediated epilepsies. Importantly, when OCB are negative at first CSF examination, the probability of OCB detection in the further course in serial lumbar punctures is low. Whereas specificity is relatively high, sensitivity for OCB in ab-LE and anti-NMDAR-encephalitis was low. Based on these results, negative OCB do of course not exclude autoimmune aetiology such as ab-LE and anti-NMDAR-encephalitis and further investigations such as ab testing and tumour search are required. In this context, it is important to note that we have not included people with suspected limbic encephalitis without detection of known abs in this study, as an international consensus on the “gold-standard” of diagnostic work-up has not been reached yet. Thus, there might also be patients with autoimmune-mediated epilepsies without detected abs, but abnormal CSF findings.
Whereas pathological CSF findings were described frequently in “classic” paraneoplastic LE,
10
knowledge about their occurrence in the recently identified non-paraneoplastic ab-positive subforms is scarce. In people with VGKC-LE only slight increases of protein levels and cell counts in CSF were reported11
whereas OCB were found in 60% of patients with anti-NMDAR-encephalitis.3
In line with these reports, none of our patients and only 11% of the published VKGC-LE cases had OCB.The presence of OCB as a measure of intrathecal ab synthesis in anti-NMDAR encephalitis and GAD-LE suggests that these entities may depend on a primary pathological process within the brain or, alternatively, that they develop after migration of ab-producing cells into the CNS.
1
In contrast, the absence of OCB (and abs within the CSF, as shown by others1
, 11
) in people with VGKC-LE is in favour of a primary peripheral synthesis outside of the central nervous system and subsequent diffusion across the blood-CSF barrier where the abs secondarily lead to neuronal hyperexcitability and maybe inflammation.1
, 12
This is, however, pure speculation and further neuropathological examinations of brain tissue in affected patients as well as experimental laboratory investigations are required to address these questions.5. Conclusion
Our data suggest that CSF investigations with determination of OCB are a supportive marker if autoimmune-mediated epilepsies such as GAD-LE, PLE and anti-NMDAR-encephalitis are suspected, whereas it failed to provide significance in VGKC-LE. Ab testings and tumour search should be undertaken to specify the underlying syndrome.
Contributions
Study was designed by MPM and RS. MPM collected the data. Statistical analysis was performed by MPM and RS. MPM, CEE and RS have written the draft and all authors have contributed substantially to the final version.
Disclosure
MPM received payments for congress participation, travel expenses, lecture and manuscript preparation from EISAI and Desitin. CEE received honoraria for consultancy, expert testimony and lectures from UCB Pharma, Desitin and Pfizer. He is employee of the Life and Brain Institute Bonn. RS has received support for congress participation and speaker fees from EISAI; had a consultancy agreement with UCB; is part of PRISM – the Prevention and Risk Identification of SUDEP Mortality Consortium which is funded by the NIH (NBIH/NINDS – 1P20NS076965-01). The funding sources had no role in the design of this report, interpretation, or in the writing of the manuscript.
Acknowledgements
We confirm that we have read the Journal's position on issues involved in ethical publication and affirm that this report is consistent with those guidelines. The authors are grateful for ab determination to the following colleagues and laboratories: Prof. A. Vincent at the Weatherall Institute of Molecular Medicine, University of Oxford, UK; Prof. W. Stoecker and Dr. K. Wandinger at EUROIMMUN laboratory, Luebeck, Germany; Prof. R. Voltz at the Department for Palliative Medicine, University Cologne, Germany; Prof. Seidel at the Institute of Clinical Chemistry, University of Munich (LMU) Germany; Prof. C. Bien at the Department of Epileptology, University of Bonn, Germany (Present address: Epilepsy Centre Bethel, Krankenhaus Mara, Bielefeld, Germany).
Appendix A. Supplementary data
The following are the Supplementary data to this article:
References
- Autoantibodies associated with diseases of the CNS: new developments and future challenges.Lancet Neurology. 2011; 10: 759-772
- Paraneoplastic anti-N-methyl-d-aspartate receptor encephalitis associated with ovarian teratoma.Annals of Neurology. 2007; 61: 25-36
- Clinical experience and laboratory investigations in patients with anti-NMDAR encephalitis.Lancet Neurology. 2011; 10: 63-74
- Autoimmune epilepsies.Current Opinion in Neurology. 2011; 24: 146-153
- Cerebrospinal fluid—physiology, analysis and interpretation of protein patterns for diagnosis of neurological diseases.Multiple Sclerosis. 1998; 4: 99-107
- Antibodies to glutamic acid decarboxylase define a form of limbic encephalitis.Annals of Neurology. 2010; 67: 470-478
- Recommended standard of cerebrospinal fluid analysis in the diagnosis of multiple sclerosis: a consensus statement.Archives of Neurology. 2005; 62: 865-870
- Potassium channel antibody-associated encephalopathy: a potentially immunotherapy-responsive form of limbic encephalitis.Brain. 2004; 127: 701-712
- N-methyl-d-aspartate receptor antibodies in pediatric dyskinetic encephalitis lethargica.Annals of Neurology. 2009; 66: 704-709
- Paraneoplastic limbic encephalitis: neurological symptoms, immunological findings and tumour association in 50 patients.Brain: A Journal of Neurology. 2000; 123: 1481-1494
- CSF findings in patients with voltage gated potassium channel antibody associated limbic encephalitis.Journal of the Neurological Sciences. 2008; 268: 74-77
- Human limbic encephalitis serum enhances hippocampal mossy fiber-CA3 pyramidal cell synaptic transmission.Epilepsia. 2011; 52: 121-131
Article info
Publication history
Published online: January 14, 2013
Accepted:
December 18,
2012
Received in revised form:
December 17,
2012
Received:
October 9,
2012
Identification
Copyright
© 2012 British Epilepsy Association. Published by Elsevier Inc.
User license
Creative Commons Attribution - NonCommercial - NoDerivs | How you can reuse 
Elsevier's open access license policy
Creative Commons Attribution - NonCommercial - NoDerivs
Permitted
For non-commercial purposes:
- Read, print & download
- Redistribute or republish the final article
Not Permitted
- Text & data mine
- Translate the article
- Reuse portions or extracts from the article in other works (except for the author)
- Sell or re-use for commercial purposes
Elsevier's open access license policy
