Estimating the likelihood of epilepsy from clinically noncontributory electroencephalograms using computational analysis: A retrospective, multisite case-control study.

OBJECTIVE: This study was undertaken to validate a set of candidate biomarkers of seizure susceptibility in a retrospective, multisite case-control study, and to determine the robustness of these biomarkers derived from routinely collected electroencephalography (EEG) within a large cohort (both epilepsy and common alternative conditions such as nonepileptic attack disorder). METHODS: The database consisted of 814 EEG recordings from 648 subjects, collected from eight National Health Service sites across the UK. Clinically noncontributory EEG recordings were identified by an experienced clinical scientist (N = 281; 152 alternative conditions, 129 epilepsy). Eight computational markers (spectral [n = 2], network-based [n = 4], and model-based [n = 2]) were calculated within each recording. Ensemble-based classifiers were developed using a two-tier cross-validation approach. We used standard regression methods to assess whether potential confounding variables (e.g., age, gender, treatment status, comorbidity) impacted model performance. RESULTS: We found levels of balanced accuracy of 68% across the cohort with clinically noncontributory normal EEGs (sensitivity =61%, specificity =75%, positive predictive value =55%, negative predictive value =79%, diagnostic odds ratio =4.64, area under receiver operated characteristics curve =.72). Group level analysis found no evidence suggesting any of the potential confounding variables significantly impacted the overall performance. SIGNIFICANCE: These results provide evidence that the set of biomarkers could provide additional value to clinical decision-making, providing the foundation for a decision support tool that could reduce diagnostic delay and misdiagnosis rates. Future work should therefore assess the change in diagnostic yield and time to diagnosis when utilizing these biomarkers in carefully designed prospective studies.

Forecasting stroke-like episodes and outcomes in mitochondrial disease.

In this retrospective, multicentre, observational cohort study, we sought to determine the clinical, radiological, EEG, genetics and neuropathological characteristics of mitochondrial stroke-like episodes and to identify associated risk predictors. Between January 1998 and June 2018, we identified 111 patients with genetically determined mitochondrial disease who developed stroke-like episodes. Post-mortem cases of mitochondrial disease (n = 26) were identified from Newcastle Brain Tissue Resource. The primary outcome was to interrogate the clinico-radiopathological correlates and prognostic indicators of stroke-like episode in patients with mitochondrial encephalomyopathy, lactic acidosis and stroke-like episodes syndrome (MELAS). The secondary objective was to develop a multivariable prediction model to forecast stroke-like episode risk. The most common genetic cause of stroke-like episodes was the m.3243A>G variant in MT-TL1 (n = 66), followed by recessive pathogenic POLG variants (n = 22), and 11 other rarer pathogenic mitochondrial DNA variants (n = 23). The age of first stroke-like episode was available for 105 patients [mean (SD) age: 31.8 (16.1)]; a total of 35 patients (32%) presented with their first stroke-like episode ≥40 years of age. The median interval (interquartile range) between first and second stroke-like episodes was 1.33 (2.86) years; 43% of patients developed recurrent stroke-like episodes within 12 months. Clinico-radiological, electrophysiological and neuropathological findings of stroke-like episodes were consistent with the hallmarks of medically refractory epilepsy. Patients with POLG-related stroke-like episodes demonstrated more fulminant disease trajectories than cases of m.3243A>G and other mitochondrial DNA pathogenic variants, in terms of the frequency of refractory status epilepticus, rapidity of progression and overall mortality. In multivariate analysis, baseline factors of body mass index, age-adjusted blood m.3243A>G heteroplasmy, sensorineural hearing loss and serum lactate were significantly associated with risk of stroke-like episodes in patients with the m.3243A>G variant. These factors informed the development of a prediction model to assess the risk of developing stroke-like episodes that demonstrated good overall discrimination (area under the curve = 0.87, 95% CI 0.82-0.93; c-statistic = 0.89). Significant radiological and pathological features of neurodegeneration were more evident in patients harbouring pathogenic mtDNA variants compared with POLG: brain atrophy on cranial MRI (90% versus 44%, P < 0.001) and reduced mean brain weight (SD) [1044 g (148) versus 1304 g (142), P = 0.005]. Our findings highlight the often idiosyncratic clinical, radiological and EEG characteristics of mitochondrial stroke-like episodes. Early recognition of seizures and aggressive instigation of treatment may help circumvent or slow neuronal loss and abate increasing disease burden. The risk-prediction model for the m.3243A>G variant can help inform more tailored genetic counselling and prognostication in routine clinical practice.

The impact of epilepsy and antiseizure medications on sleep: Findings from a large European survey in adults with epilepsy and matched controls.

OBJECTIVE: To assess the impact of epilepsy and antiseizure medications (ASMs) on sleep quality in people with epilepsy (PWE). METHODS: An online survey was conducted in France, Germany, Italy, Spain and the UK among PWE taking >1 ASM and matched controls. Sleep quality was evaluated using the Pittsburgh Sleep Quality Index (PSQI). Associations between sleep quality (global PSQI) and overall quality of life (QoL; assessed using the 12-Item Short Form Survey [SF-12]) and sleep quality and depressive symptoms (assessed using the Neurological Disorders Depression Inventory for Epilepsy [NDDI-E]) were also evaluated. RESULTS: Overall, 500 PWE and 500 matched controls were included. PWE had significantly greater mean global PSQI scores than controls (9.32 vs 7.56; p < 0.0001), with 80% reporting a score >5 versus 66% of controls (p < 0.001). PWE experienced significantly more problems with most PSQI components than controls. Mean global PSQI scores in PWE receiving 2 versus ≥3 ASMs were 9.03 and 10.18, respectively (p < 0.004); global PSQI scores >5 were reported in 76% versus 90%, respectively (p = 0.001). Regimens containing lamotrigine or phenobarbital were associated with poorer sleep quality than those without these ASMs. In PWE, negative correlations were identified between global PSQI scores and both the SF-12 physical and mental components (Pearson's correlation coefficient [PCC], -0.61 and -0.40, respectively); NDDI-E and global PSQI scores were positively correlated (PCC, 0.6). CONCLUSIONS: PWE experience significantly worse sleep quality than people without epilepsy, with some ASMs contributing to poorer sleep. QoL and physical and mental health were all affected by sleep quality in PWE.

The impact of epilepsy on quality of life: Findings from a European survey.

INTRODUCTION: Epilepsy is a serious neurological disorder affecting the quality of life (QoL) of people with this condition. A survey was conducted in five European countries (France, Germany, Italy, Spain, and the UK) to understand the impact and burden of epilepsy and its treatment on the lives of people with epilepsy (PWE). METHODS: Five hundred PWE (taking >1 antiseizure medication [ASM]) and 500 matched controls completed a 30-minute online questionnaire. The 12-Item Short Form Survey (SF-12) was used to measure QoL and the Neurological Disorders Depression Inventory for Epilepsy (NDDI-E) was used to screen for major depressive disorder (MDD) symptoms. RESULTS: Comorbidities such as migraine, high cholesterol, osteoporosis, and Type 1 diabetes were reported more commonly in PWE, while anxiety disorders, high blood pressure, skin disorders, and mood disorders were more common in controls. However, compared to controls, a significantly higher percentage of PWE had an NDDI-E score of 15-24 (54% vs 35%; p < 0.0001), indicative of MDD symptoms. Significantly more PWE than controls were part-time employed (15% vs 11%; p = 0.03). People with epilepsy had a significantly lower total SF-12 score than controls across the physical and the mental components; compared to controls, a significantly higher proportion of PWE defined their general health as 'poor' or 'fair' and felt limited in carrying out daily and work activities. Among PWE, those taking ≥3 ASMs were more likely to experience difficulties in carrying out these activities than those on two ASMs. Ability to drive, mood, and level of self-esteem were reported as concerns for PWE. CONCLUSION: Epilepsy has a major impact on the physical and mental health of PWE, interfering with their daily and work activities and overall QoL, and its treatment might also contribute to a lower QoL. The impact of epilepsy on mood and mental health might be under-recognized.

Eslicarbazepine acetate response in intellectual disability population versus general population.

BACKGROUND: A quarter of people with intellectual disability (ID) have epilepsy, compared to approximately one in a hundred across the general population. Evidence for the safe and effective prescribing of antiepileptic drugs (AEDs) for those with ID is, however, limited. AIMS OF STUDY: This study seeks to strengthen the research evidence around Eslicarbazepine Acetate (ESL), a new AED, by comparing response of individuals with ID to those from the general population who do not have ID. METHODS: A single data set was created through retrospective data collection from English and Welsh NHS Trusts. The UK-based Epilepsy Database Research Register (Ep-ID) data collection and analysis method were used. RESULTS: Data were collected for 93 people (36 ID and 57 'no ID'). Seizure improvement of '>50%' was higher at 12 months for 'no ID' participants (56%), compared to ID participants (35%). Retention rates were slightly higher for those with ID (56% compared to 53%). Neither difference was significant. CONCLUSIONS: Tolerance and Efficacy for ID and 'no ID' people in our data set were similar. Seizure improvement and retention rates were slightly lower than that found in other European data sets, but findings strengthen the evidence for the use of ESL in the ID population.

The evidence for switching dibenzazepines in people with epilepsy.

The dibenzazepines particularly carbamazepine are associated with known adverse effects (AEs) and drug to drug interactions. Eslicarbazepine acetate (ESL) is structurally distinct from other members of the dibenzazepine family and has the advantage of once daily dosing. Observational and trial data report successful switching from older dibenzazepines to ESL. The evidence base for doing so is unclear and not standardized. This is a literature review following the PRISMA scoping guidelines identifying the evidence of switching dibenzazepines. Transition methods, ratios, tolerance to change, adverse effects and retention post-change were evaluated. Study quality was assessed using the Oxford Centre for Evidence Based Medicine levels of evidence. Seven studies investigated the outcome of transition between carbamazepine and or oxcarbazepine to ESL, with specific data on the transition dose ratio and scheduling. The available data suggest that the overnight transition between oxcarbazepine and ESL in a 1:1 ratio (most common) is generally well tolerated with high retention rates. The transition showed improvement in adverse events associated with oxcarbazepine across a variety of domains. Almost 60% transitioned because of adverse events experienced no further symptoms at 12 months. There is less data on the transition from carbamazepine to ESL. The evidence available suggests an overnight transition in the ratio of 1:1.3-1.5. The retention rate following transition from carbamazepine to ESL was 69% (follow-up of 4 months) with almost half of those transitioned because of adverse events experiencing no further symptoms. There is Grade C evidence available to help guide clinicians in the transition.

Valproate and epilepsy: for women as well as men.

Sodium valproate remains the best drug for idiopathic generalised epilepsy. For men with the latter diagnosis, this is the drug of choice. Sodium valproate has an unacceptably high level of major fetal malformation and also causes learning disabilities in many children exposed to the drug in utero. Women of reproductive age should not normally be offered this drug. There are many women with refractory epilepsy who would benefit from this drug and who are not planning pregnancy. Individualised epilepsy care is the gold standard, not blanket bans on drug choice based on gender.

Seizures in palliative medicine: brivaracetam.

Seizures occur in around 13% of patients with cancer and can be distressing for family members to witness. In those unable to manage regular antiepileptic medications, midazolam can be administered subcutaneously using a syringe driver, but this may cause sedation. Brivaracetam is a newer drug licensed as an adjunctive therapy in the treatment of partial-onset seizures with or without secondary generalisation and for restricted use in those with refractory epilepsy. It is associated with fewer behavioural or psychiatric side effects than levetiracetam, has a very low incidence of drug interactions and the maximal dose can be accommodated in a single syringe driver. We present three cases from 2019 to 2020 where subcutaneous brivaracetam has been successfully used in a Specialist Inpatient Palliative Care setting to manage seizures. Brivaracetam dosing is 1:1 conversion from oral:subcutaneous, with syringe driver doses ranging from 150 mg to 300 mg/24 hours being successfully used, with no adverse effects observed.

Protocol for a realist and social return on investment evaluation of the use of patient-reported outcomes in four value-based healthcare programmes.

INTRODUCTION: There is growing recognition that in order to remain sustainable, the UK's National Health Service must deliver the best patient outcomes within available resources. This focus on outcomes relative to cost is the basis of value-based healthcare (VBHC) and has led to interest in the recording of patient-reported outcome measures (PROMs) to measure patient perspectives on the impact of a health condition on their lives. Every health board in Wales is now required to collect PROMS as part of routine care. We will evaluate the VBHC programme implemented in a lead health board. The study aim is to understand what works about PROMs collection, for whom, in what contexts and why in a VBHC context. In addition, we will assess the social value of integrating PROMs collection into routine care. METHODS AND ANALYSIS: A three-stage mixed-methods study comprising a realist evaluation integrated with social return on investment (SROI) analysis across four conditions; Parkinson's disease, epilepsy, heart failure and cataract surgery. Workstream 1: Development of logic models, informed by a scoping review, documentary analysis, patient and public involvement (PPI), staff and key stakeholder engagement. Workstream 2: Realist evaluation building on multiple data sources from stages 1 to 3 to test and refine the programme theories that arise from the logic model development. Workstream 3: SROI analysis using interview data with patients, staff and carers, stakeholder and PPI engagement, anonymised routinely collected data, and questionnaires to populate a model that will explore the social value generated by the implementation of PROMs. Findings across stages will be validated with key stakeholders. ETHICS AND DISSEMINATION: The study is approved by Wales Research Ethics Committee #5 (22/WA/0044). Outcomes will be shared with key stakeholders, published in peer-reviewed journals and presented at national and international conferences.This is an open access article distributed in accordance with the Creative Commons Attribution Non Commercial (CC BY-NC 4.0) licence, which permits others to distribute, remix, adapt, build on this work non-commercially, and license their derivative works on different terms, provided the original work is properly cited, appropriate credit is given, any changes made indicated, and the use is non-commercial.

Exploring seizure management in hospitals, unmet need, and the impact of the COVID-19 pandemic on seizure presentations to hospital.

PURPOSE: This study assesses investigations, referrals and admissions in patients presenting to the Emergency Department (ED) with seizures, and the effect of the COVID-19 pandemic on such management. Outcomes in patients with learning disabilities, active significant mental health concerns, and from the most socioeconomically deprived areas were compared to those of the general cohort. METHODS: Investigations, referrals and admissions were recorded for 120 patients across two cohorts; pre-pandemic (September 2019) and during the pandemic (December 2020). Retrospective review of individual patient electronic health care records was used for data collection. RESULTS: There was a decrease in patient numbers from 2019 to 2020. A greater proportion of patients presented with organic cause seizures and fewer presented with non-epileptic attacks. Frequent use of CT heads (45%) is likely to represent improper use of limited resources. There were low referral rates, both to acute neurology (28%) and to the adult epilepsy team (32%). Patients with active significant mental health concerns were significantly less likely to be referred to neurology or admitted. CONCLUSIONS: Despite a greater proportion of admissions during the Covid-19 pandemic, referrals to acute neurology and the epilepsy team remained low. Failure to refer prevents the most vulnerable seizure patients from receiving appropriate support, as seen in patients with active significant mental health concerns. Neurology staff were unaware of a significant number of patients presenting with seizures, which is of concern in an already over-stretched department. This offers an opportunity to improve care for people with epilepsy.

Nasal retinal nerve fiber layer attenuation: a biomarker for vigabatrin toxicity.

PURPOSE: To investigate whether nasal peripapillary retinal nerve fiber layer (RNFL) attenuation is associated with visual field loss attributed to the anti-epileptic drug vigabatrin. DESIGN: Prospective cross-sectional observational study. PARTICIPANTS: Twenty-seven individuals with focal-onset epilepsy exposed to vigabatrin and 13 individuals with focal-onset epilepsy exposed to non-GABAergic anti-epileptic drug monotherapy. METHODS: At one visit, suprathreshold perimetry of the central and peripheral field (3-zone, age-corrected Full Field 135 Screening Test) and threshold perimetry of the central field (Program 30-2 and the FASTPAC strategy) were undertaken using the Humphrey Field Analyzer (Carl Zeiss Meditech, Dublin, CA). At a second visit, ocular coherence tomography was undertaken for the right eye using the 3.4 RNFL thickness protocol of the StratusOCT (Carl Zeiss Meditech). MAIN OUTCOME MEASURES: The magnitude, for each individual, of the RNFL thickness, averaged across the 4 oblique quadrants, and for each separate quadrant. RESULTS: Of the 27 individuals exposed to vigabatrin, 11 (group I) exhibited vigabatrin-attributed visual field loss, 15 exhibited a normal field, and 1 exhibited a homonymous quadrantanopia (group II). All 13 individuals exposed to non-GABAergic therapy had normal fields (group III). All individuals in group I exhibited abnormal average and nasal quadrant RNFL thicknesses in the presence of a normal temporal quadrant thickness. Most also exhibited additional RNFL attenuation in either the superior or inferior quadrant, or both. Four individuals in group II exhibited an identical pattern of RNFL attenuation suggesting that nasal RNFL thinning is a more sensitive marker for vigabatrin toxicity than visual field loss. None of the 13 individuals in group III exhibited nasal quadrant RNFL attenuation. CONCLUSIONS: Vigabatrin-attributed visual field loss is associated with a characteristic pattern of RNFL attenuation: nasal quadrant thinning and normal temporal quadrant thickness with, or without, superior or inferior quadrant involvement. Nasal attenuation may precede visual field loss. Ocular coherence tomography of the peripapillary RNFL should be considered in patients previously exposed to vigabatrin. It should also be considered at baseline and follow-up in those commencing vigabatrin for treatment of epilepsy or in trials for anti-addiction therapy. The pattern of RNFL thinning seems to be a useful biomarker to identify vigabatrin toxicity.

In utero exposure to vigabatrin: no indication of visual field loss.

The purpose of the study was to determine whether in utero exposure to vigabatrin caused visual field loss. Three mothers with four children who had been exposed to vigabatrin in utero and who were subsequently formula fed were identified. All seven individuals underwent perimetry and imaging of the retinal nerve fiber layer (RNFL). All individuals yielded reliable outcomes to perimetry and RNFL images of acceptable quality. Two of the three mothers exhibited vigabatrin-attributed visual field loss and an abnormally attenuated RNFL. The third exhibited an upper left quadrantanopia, consistent with previous temporal lobe surgery, and a normal RNFL. All four children yielded normal visual fields and RNFL thicknesses. The presence of the normal findings for the children is reassuring and, if representative, suggests a lack of vigabatrin visual toxicity and therefore obviates the need for ophthalmological examination of those exposed to vigabatrin prenatally.

Effectiveness and Safety/Tolerability of Eslicarbazepine Acetate in Epilepsy Patients Aged ≥ 60 Versus < 60 Years: A Subanalysis from the Euro-Esli Study.

INTRODUCTION: Clinical practice studies help guide antiepileptic drug (AED) therapy in patient groups routinely excluded from clinical trials, such as the elderly. The Euro-Esli study investigated the effectiveness and safety/tolerability of eslicarbazepine acetate (ESL) when used in everyday clinical practice in Europe. A subanalysis of data from elderly patients (≥ 60 years) included in the Euro-Esli study was conducted to assess these aspects of ESL use in this population. METHODS: Euro-Esli was a pooled analysis of 14 European clinical practice studies. Effectiveness parameters included responder (≥ 50% seizure frequency reduction) and seizure freedom rates after 3, 6 and 12 months of treatment and at last visit. Safety and tolerability were assessed throughout the follow-up by evaluating adverse events (AEs) and ESL discontinuation due to AEs, respectively. Data were compared for patients aged ≥ 60 versus those aged < 60 years at study entry. RESULTS: Euro-Esli included 2058 patients (mean age 44.0 years). Age at study entry was known for 2057 patients, of whom 358 (17.4%) and 1699 (82.6%) were aged ≥ 60 and < 60 years, respectively. Mean maximum ESL dose was   882.0 and 1008.2 mg/day in patients aged ≥ 60 and < 60 years, respectively  (p < 0.001). At all timepoints, responder and seizure freedom rates were significantly higher in patients aged ≥ 60 versus < 60 years; for example, at 12 months, responder rates were 83.9 and 73.7%, respectively (p = 0.002), and seizure freedom rates were 58.5 and 37.1%, respectively (p < 0.001). The incidence of AEs was significantly higher in patients aged ≥ 60 versus < 60 years (41.4 vs. 32.5%; p = 0.001), but the rate of discontinuation due to AEs was comparable between age groups (16.2 vs 13.1%; p = not significant). The safety/tolerability of ESL in patients aged ≥ 60 years was consistent with its known    profile. CONCLUSION: Eslicarbazepine acetate was efficacious and generally well tolerated when used to treat elderly patients with focal epilepsy in clinical practice, with no new or unexpected safety signals emerging in this setting. FUNDING: Eisai Ltd.

Managing epilepsy in austerity - Evaluating the utility and value of the epilepsy specialist nurse in an open access model of service delivery. Aneurin Bevan Epilepsy Specialist Team (A.B.E.S.T.).

PURPOSE: Healthcare organisations in the UK primarily measure clinical activity from data collected on numbers of attendances at outpatient clinics, inpatient admissions and procedures performed etc. Telephone contacts with patients are not typically measured as clinical activity. This service evaluation examines the utility and value of the Epilepsy Specialist Nurse (ESN) within an innovative 'Open Access Model', giving a breakdown of clinical workload and outcomes. METHODS: A retrospective service evaluation analysing all patient encounters made by the ESN and their outcomes, over a 3-month period from the 01/02/2017 to 30/04/2017. RESULTS: During the 3-month data collection period there were 620 patient encounters with 251 different patients. Nurse-led clinic appointments and telephone calls were the two most common types of encounter. Eighteen percent of ESN time was spent on the phone directly addressing patient concerns (368 encounters). Of these calls, 72% led to prevention of a clinic appointment (268 appointments avoided) while only 22% needed a subsequent clinic appointment. The most common outcome of telephone encounters was 'medication management' (25%). CONCLUSIONS: The evaluation demonstrates that timely intervention by telephone reduces the need for outpatient appointments and leads to treatment changes being implemented quickly to address individual need.

Dibenzazepine Agents in Epilepsy: How Does Eslicarbazepine Acetate Differ?

Carbamazepine (CBZ), oxcarbazepine (OXC), and eslicarbazepine acetate (ESL) belong to the dibenzazepine family of antiepileptic drugs and are all thought to primarily act as sodium channel blockers (SCBs). However, ESL is structurally distinct from CBZ and OXC, resulting in differences in metabolism, pharmacokinetics, and pharmacodynamics. Despite a lack of direct comparative data, evidence for potential differences in effectiveness and tolerability within the dibenzazepine family has emerged from studies in which patients being treated with one dibenzazepine agent have received adjunctive treatment with another (having achieved insufficient seizure control with the first) or have transitioned from one dibenzazepine agent to another because of lack of effectiveness or poor tolerability. Most of these studies have been conducted in the real-world clinical practice setting. ESL has been shown to be effective as adjunctive therapy in patients who have previously achieved inadequate seizure control with CBZ, indicating that the use of different dibenzazepine agents in combination can provide additive effectiveness benefits, which may reflect underlying differences in their mechanisms of action. Similarly, ESL monotherapy can be effective in patients who have switched from another dibenzazepine, such as CBZ or OXC, because of inadequate efficacy. There is also considerable evidence to demonstrate that patients transitioning from OXC or CBZ to ESL as a result of adverse events experience improvements in tolerability, which may also be associated with improvements in quality of life, alertness, and/or lipid profiles. Current evidence therefore demonstrates that ESL differs from other dibenzazepine agents in terms of effectiveness and tolerability.Funding: Eisai Ltd.

Epstein-Barr virus-associated cerebellar ataxia.

Cerebellar ataxia is a common neurological presentation. It can be acute, subacute or chronic. Neurological complications of Epstein-Barr virus (EBV) are well-recognised with a variety of presentations. Acute cerebellar ataxia is a rare, but an established complication. It has been described as the sole manifestation of EBV infection without the systemic features of infectious mononucleosis. The pathophysiology is not clear. The course of the illness may last for a few months with a benign outcome, though serious complications can happen. We present a case of a 38-year-old man who presented with an acute cerebellar ataxia owing to EBV infection, along with a review of the literature.

Modelling the risk of visual field loss arising from long-term exposure to the antiepileptic drug vigabatrin: a cross-sectional approach.

BACKGROUND: The antiepileptic drug vigabatrin has been used widely since 1989, but has only been approved for use in the US since 2009. The risk:benefit of vigabatrin is generally predicated upon an assumed frequency of associated visual field loss (VAVFL) of approximately 31 %. This estimate is based upon relatively short-term usage (up to 4-5 years) and it is essential to determine whether the frequency of VAVFL increases with longer-term usage. OBJECTIVE: The aim of this study was to model, from cross-sectional evidence, over greater ranges of treatment duration and cumulative dose than previously evaluated, the risk (frequency) of VAVFL with increasing exposure to vigabatrin. STUDY DESIGN AND SETTING: This was a retrospective cohort study undertaken in a regional hospital epilepsy clinic. PATIENTS: The cohort comprised 147 consecutive patients treated with vigabatrin for refractory complex partial (focal) seizures, who had all undergone ophthalmological examination and who had undertaken perimetry, reliably, according to a standard and robust protocol. The visual field plots were evaluated masked to treatment duration and dose. MAIN OUTCOME MEASURE: The risk (frequency) of VAVFL with increasing exposure to vigabatrin was modelled, from the cross-sectional evidence, by standard and plateau logistic regression. RESULTS: The cohort comprised 80 females and 67 males (mean age 40.3 years, standard deviation 13.7). The median duration of vigabatrin exposure was 7.9 years (interquartile range 3.6-11.0, range 0.2-16.1 years); 46 patients (31 %) had received vigabatrin for over 10 years. Eighty-seven patients (59 %) exhibited VAVFL; the proportion with VAVFL was higher in males (66 %) than females (54 %). The plateau model for duration and for cumulative dose exhibited a better fit than the standard model (both p < 0.001). The modelled frequency of VAVFL increased with increasing exposure up to approximately 6 years duration and 5 kg cumulative dose, and plateaued at approximately 76 % (95 % CI 67-85) and 79 % (95 % CI 70-87), respectively. Severity of VAVFL, classified in terms of the visual field index Mean Deviation, was not significantly associated with either duration or cumulative dose of therapy. CONCLUSION: Clinicians and patients, in enabling informed choice, should be alert to the possible substantial increased risk:benefit for VAVFL with increasing long-term exposure to vigabatrin and the ensuing increased cost:benefit resulting from the necessary additional visual assessments.