Background: Epileptic encephalopathy (EE) is a severe condition in which epileptic activity itself may contribute to severe cognitive and behavioural impairments above and beyond what might be expected from the underlying pathology alone. Next generation sequencing technologies such as whole exome sequencing (WES) can detect underlying genetic causes of in EE. Methods: This report describes genotype-phenotype correlation of 29 subjects with unexplained epileptic encephalopathy, in whom WES, targeting a list of 557 epilepsy-associated genes was performed. Epilepsy phenotyping was done according to current ILAE recommendations. Results: Median age at seizure onset was 14 months (range 1-48). Electroclinical syndromes were applicable for 16/29, 8/16 had a definite/likely diagnosis. 6/8 subjects with West syndrome had variants in ALG13, STXBP1, PAFAH1B1, SLC35A2, CDKL5 and ADSL. 2 patients with Dravet syndrome had variants in SCN1A and PCDH19 respectively. 4/29 had unspecified EE and definite/likely diagnosis due to STXBP1, POLG, and KCNQ2 (2) variants. 4/29 had a possible diagnosis involving GABRB3, ARHGEF9, PCDH19 and SCN3A variants. Conclusions: The high diagnostic yield (definite/likely diagnosis in 11/29 = 38%), involving a broad variety of epilepsy-associated genes in different electroclinical syndromes justifies the diagnostic approach of early onset EE by next generation sequencing.

Background: Autistic regression (AR) accounts for 20-40% of patients with Autism Spectrum Disorder (ASD) .1 Literature demonstrates specific immune changes in AR patients,2 as well as association between AR and autoimmune thyroiditis.3 Our study explores the clinical association between AR and autoimmunity, focusing on possible precipitants and familial autoimmunity, in comparison with patients with infantile autism (IA). Methods: charts of children diagnosed with ASD in 2014 were reviewed, and patients were classified as either AR or IA based on Autism Diagnostic Interview (ADI-R) criteria.4 Information regarding pregnancy, perinatal complications, febrile illness preceding the diagnosis, and family history of autoimmune conditions was collected. Results: 206 children had IA and 33 had AR. No difference was found in prevalence of pregnancy or perinatal complications. The incidence of febrile illness in the 6 months prior to diagnosis and the prevalence of familial autoimmunity, were significantly higher in the AR group (p<0.001). Diabetes type I, celiac disease, autoimmune thyroiditis, and inflammatory bowel disease were more common in families of AR patients (p<0.05). Conclusions: the association between AR and preceding febrile illness, as well as familial autoimmunity, supports the notion of AR as a separate entity within ASD, possibly mediated by autoimmune changes.

Background: Persistent post-concussive symptoms (PCS) have been linked to increased cortical network activation and decreased cerebrovascular reactivity. Decreased cerebral perfusion could help explain PCS and may be a biomarker to track recovery. Methods: Children (ages 8 to 18 years) symptomatic with PCS at one month post-injury were studied. Children who recovered following a mTBI (asymptomatic group) and healthy children acted as controls. Pseudocontinuous arterial spin labeling MRI was used to quantify cerebral blood flow (CBF). All subjects were imaged at approximately 40 days post-injury. Symptomatic group underwent repeat neuroimaging 4-5 weeks later. Results: Seventy-two participants (14.1 years; 95% CIs: 13.5, 14.8) underwent neuroimaging at 40 days post-injury. Global CBF was significantly higher in the symptomatic group compared to healthy controls, and lower in the asymptomatic group (F(2,57) 9.734 p<0.001). Symptomatic children had increased CBF in the frontal and occipital regions, and asymptomatic children had decreased CBF in the temporal regions compared to healthy controls. CBF decreased in symptomatic children over time. CBF was a predictor of cognition (R2=0.235;p=0.001). Conclusions: Cerebral perfusion is altered in children with mTBI and is associated with recovery trajectory. Asymptomatic children had decreased CBF suggesting cerebral recovery is ongoing. Further longitudinal studies are required to determine if these perfusion patterns continue to change over time.

Background: Perinatal stroke is a common disorder in neonates with unknown etiology. Previous studies have linked anatomic variations in the Circle of Willis to adult stroke. This study aimed to understand the potential relationship between circle anatomy and common forms of perinatal stroke: NAIS, APPIS, and PVI. Methods: 94 subjects (62 NAIS/APPIS, and 32 PVI) were identified from the Alberta Perinatal Stroke Project. Inclusion criteria were: MRI-confirmed perinatal stroke, 3D-TOF MRA, and absence of other disorders. Images were classified as complete, incomplete posterior circulation, incomplete anterior circulation, and incomplete anterior and posterior circulation. Fisher Exact Test compared completeness against stroke type and segment absence ipsilateral to stroke. Mann-Whitney U compared completeness and lesion volume. Results: Completeness was more common in PVI than NAIS/APPIS (p=0.500) and in healthy controls than total stroke population (p=0.251). Ipsilesional absent segments were more frequent in NAIS/APPIS (p= 0.270). NAIS/APPIS patients with complete CoW had larger median lesion volume was compared to those with incomplete circles (p=0.484), with contralateral absence (p=0.943), and with ipsilateral absence (p=1.00). The opposite was found in PVI patients for all lesion volume comparisons (p=0.321, 0.362, 0.739 respectively). Conclusions: Circle anatomy is highly variable in perinatal stroke. Absence of segments is not associated with stroke type, lesion side, and lesion volume.

Background: The therapeutic value of second-generation anticonvulsants such as lamotrigine has not been clearly established in neonates and infants with seizures. To address this issue, we assessed the efficacy of lamotrigine for treating neonatal and infantile seizures, detailed the dosing regimens used, and described its tolerability and safety profile. Methods: This retrospective study included patients (age 0-12 months) diagnosed with seizures and treated with lamotrigine, as monotherapy or adjunctive therapy, by pediatric neurologists at Centre mère-enfant Soleil du CHUQ from 2004 to 2014. The frequency of seizures and EEG patterns were compared before and after introduction of lamotrigine during the first months of life. Data on initial and maintenance doses, rate and magnitude of dosing increments, and adverse effects were collected. Results: Treatment with lamotrigine was initiated in 32 neonates and 13 infants. At first follow-up (mean duration 3 months), 76 % (n = 34) showed a significant (≥50%) reduction of seizures and 64% (n= 29) improvement of EEG pattern compared to baseline. The efficacy in monotherapy and adjunctive therapy was similar. A single case of cutaneous hypersensitivity reaction requiring cessation of treatment was reported. Conclusions: This study suggests that lamotrigine is a useful, safe, and well-tolerated anticonvulsant alternative for the treatment of seizures in neonates and infants.

Background: Ataluren is the first drug to treat the underlying cause of nmDMD. Methods: Phase 2 and 3 studies of ataluren in nmDMD were reviewed, with efficacy and safety/tolerability findings summarized. Results: Ataluren nmDMD trials include: a Phase 2a proof-of-concept study (N=38); a Phase 2b randomized controlled trial (RCT) (N=174); an ongoing US-based open-label safety extension study (N=108); an ongoing non-US-based open-label safety/efficacy extension study (N=94); and a Phase 3 RCT, ACT DMD (N=228), whose primary endpoint was change in six-minute walk distance (6MWD) over 48 weeks. The proof-of-concept study demonstrated increased dystrophin production in post-treatment muscle biopsies from ataluren-treated patients with nmDMD. The Phase 2b results demonstrated an ataluren treatment effect in 6MWD, timed function tests, and other measures of physical functioning, The Phase 3 ACT DMD results demonstrated an ataluren treatment effect in patients with nmDMD in both primary and secondary endpoints, particularly in those with a baseline 6MWD of 300-400m. Ataluren was consistently well-tolerated in all three trials, as well as in the ongoing extension studies. Trial findings will be presented in detail. Conclusions: The totality of the results demonstrates that ataluren enables nonsense mutation readthrough in the dystrophin mRNA, producing functional dystrophin and slowing disease progression.

Supported by: PTC Therapeutics Inc.

Background: We determined the association between head circumference (HC) at birth and through neonatal intensive care with neurodevelopmental outcome in preterm neonates, accounting for brain injury on MRI. Methods: 169 neonates born 24-32 weeks gestation were studied prospectively with serial MRI. HC was measured at birth and discharge from neonatal intensive care. Outcome was assessed at 18 months corrected age using Bayley Scales of Infant & Toddler Development III motor and cognitive scores. Using multivariate linear regressions we evaluated the association between HC and outcomes, accounting for severity of brain injury and postnatal infection. Results: 46 neonates had HC <10th percentile at birth (SHC) which predicted poorer motor (~4 points; p=0.001) and cognitive (~4 points; p=0.005) outcomes, relative to those with normal HC at birth. In 9 of these neonates, SHC persisted to discharge; they had dramatically lower motor scores (15 points; p=0.004) and cognitive scores (12 points; p<0.001), even after adjusting for known risk factors Those born with SHC whose HC normalized by discharge did not show significantly poorer outcomes than those born with normal HC. Conclusions: The relationship between small HC at birth and adverse neurodevelopmental outcomes can be attenuated with normalization of head growth through the period of neonatal intensive care.

Background: Dysphagia is a devastating complication of stroke and can lead to malnutrition, immobility, aspiration pneumonia, and death. Guidelines advocate screening all patients with acute stroke for swallowing impairment. However, previous research suggests only 60% are screened, and it is unclear what factors contribute to receiving dysphagia screening. Methods: We used the Ontario Stroke Registry to identify patients who were admitted to Regional Stroke Centres from 2010-2013. We used multivariable regression to identify predictors of receiving a dysphagia screen within 72 hours. Results: Among 7172 patients with acute ischemic stroke, 1705 patients (23.8%) did not undergo screening. Factors increasing the odds of being tested were: Stroke unit admission (adjusted odds ratio aOR 6.5), presenting with speech deficits (aOR 1.9) or weakness (aOR 1.5), or receiving thrombolysis (aOR 1.9). Seizure (aOR 0.49) and mild stroke (aOR 0.59 vs moderate stroke) decreased the odds of being tested. Among those with mild strokes who received a swallowing screen, 33% failed. *All p<0.0001. Conclusions: Patients with mild stroke are at risk of not being screened for dysphagia, despite a significant fail rate among those tested. This may expose untested patients to a higher risk of complications from dysphagia, and suggests a gap in process of care that should be addressed.

Background: Burst-suppression is an electroencephalographic pattern observed during coma and reflects severe encephalopathy. We investigated the reactivity of burst-suppression to photic stimulation in children with acquired brain injury. Methods: Intensive care unit electroencephalographic monitoring recordings containing burst-suppression were obtained from 5 comatose children with acquired brain injury of various etiologies. Intermittent photic stimulation was performed at 1 Hz for 1 minute to assess reactivity. We quantified reactivity by measuring the change in the burst ratio (fraction of time in burst) following photic stimulation. Results: Photic stimulation evoked bursts in all patients, resulting in a transient increase in the burst ratio, while the mean heart rate remained unchanged. The regression slope of the change in burst ratio, referred to as the standardized burst ratio reactivity, correlated with subjects’ Glasgow Coma Scale scores. Conclusions: Reactivity of the burst-suppression pattern to photic stimulation occurs across diverse coma etiologies. Standardized burst ratio reactivity appears to reflect coma severity. Measurement of burst ratio reactivity may represent a simple bedside tool to monitor coma severity in critically ill children.

Background: Planning for neurology training necessitated a reflection on the experience of graduates. We explored practice characteristics, and training experience of recent graduates. Methods: Graduates from 2010-2014 completed a survey. Results: Response rate was 37% of 211. 56% were female. 91% were adult neurologists. 65% practiced in an outpatient setting. 63% worked in academics. 85% completed subspecialty training (median 1 year). 36% work 3 days a week or less. 82% took general call (median 1 night weekly). Role preparation was considered very good or excellent for most; however poor or fair ratings were 17% in advocacy and 8% in leadership. Training feedback was at least “good” for 87%. Burnout a few times a week or more was noted by 5% (6% during residency, particularly PGY1 and 5). 64% felt overly burdened by paperwork. Although most felt training was adequate, it was poor or fair at preparing for practice management (85%) and personal balance (55%). Most conditions were under-observed in training environment. Many noted a need for more independent practice development and community neurology. Conclusions: Although our training was found to be very good, some identified needs included advocacy training, and more training in general neurology in the longitudinal outpatient/community settings.

Background: Distal/asymmetric presentations of myasthenia gravis (MG) are uncommon and occur in 3-7% of patients with MG. This pattern of weakness is often not recognized as a manifestation of MG, leading to inappropriate investigations, delayed diagnosis and potentially missed opportunities for treatment. Our knowledge about this atypical presentation is limited to small case series and individual case reports. This study therefore aims to expand our understanding by describing the clinical course, diagnosis and treatment of a larger series of patients with this presentation. Methods: We conducted a retrospective chart review of patients with definite MG (either acetylcholine receptor [AChR] or MuSK antibody positive or clear evidence of postsynaptic neuromuscular junction dysfunction on electrodiagnostic studies), who attended the MG Clinic in London. Details of the clinical course, electrodiagnostic studies, antibody testing and response to treatment are reported. Results: 5.9% (54/921) of patients with definite MG had distal/asymmetric limb involvement, 56% at onset and 4% developing more than 10 years later. Males predominated (2:1). Finger extensors were most affected. 83% were AChR antibody positive. 7% had thymomas. On repetitive nerve stimulation most patients showed the most significant decrement distally on the more affected side. Almost all patients improved with treatment. Conclusions: This study expands our understanding of distal/asymmetric presentations of MG.

Background: In British Columbia, neuromuscular disease accounts for 31% of IVIg use, at a cost of $10.1 M. In addition to the new screening pathway, the BC Neuromuscular IVIg Program developed the Chronic High User Project to identify areas for improvement in utilization. Methods: Utilizing CTR data, all patients on IVIg maintenance therapy for approved neuromuscular conditions between April 1, 2013 and March 31, 2014 were identified. Patients receiving higher than usual IVIG treatments (CIDP and MG >1110 grams/year, MMNCB > 1400 grams/year) were evaluated. Following panel review, utilization data was compared with a second cohort (2014 to 2015) to determine impact. Following review, appropriateness of treatment was determined by consensus from a 3-member panel, and recommendations were made. Results: Of 377 patients, 38 “High Users” were identified. 29 cases were determined to be appropriate; 9 were not. There was a reduction in mean grams/episode in CIDP (1135g to 990g) and MG (1099 g to 1022g) between cohorts. The mean grams/episode for MMNCB did not change. Conclusions: In specific cases, the IVIg High User Program identified patients in whom the treatment could be optimized. However, the vast majority of use of IVIg for Neuromuscular Disease in BC is appropriate, including in patients requiring higher that “usual” doses.

Background:DEPDC5 gene, mapped to 22q12.2-q12.3, has been associated with a variety of familial epilepsies, including FFEVF, autosomal dominant nocturnal frontal lobe epilepsy, and familial TLE. Notably, DEPDC5 has never been linked to increased risk of sudden unexpected death in epilepsy (SUDEP). Methods: Cases review. Results: We studied a three-generation, non-consanguineous, French-Canadian family with nine clinically affected individuals. The index case is a 39-year-old man who started having seizures (as 2rily GTCS) at the age of 13 years. EEGs showed interictal discharges over the right anterior-temporal region. Brain MRI was unremarkable. Two individuals in this family suffered definite autopsy-confirmed SUDEP, at the ages of 58 and 50 years, respectively. Overall, seizure-history in this family can be summarized by an onset before reaching adulthood followed by subsequent progressive decrease in seizure frequency. Seizures were predominantly nocturnal 2rily GTC. Genetic analysis revealed a pathogenic heterozygous variant in the DEDPC5 gene (p.Gln216, c.646C>T), which results in a premature stop codon, in all affected family members plus on heatlhy relative. Importantly, all the subjects were cognitively intact, and there was no history of cardiac symptomatology/cardiovascular risk factor. Conclusions: The finding in this family suggests that DEPDC5 mutations may be a risk factor for SUDEP.

Background: The effect of the single seizure clinic (SSC) model on patient diagnosis, work-up, wait-times, and clinical care is poorly characterized. This study assesses patient characteristics and evaluates the impact of a SSC model on wait-times and access to care. Methods: A prospective study of all patients (n=200) referred to our SSC for first-seizure evaluation. Demographic, clinical, and paraclinicial variables were analyzed against a historical cohort. Binary logistic regression analysis was performed to predict impact of dichotomized variables on diagnosis of epilepsy. Diagnostic concordance between SSC nurses and epileptologists was also assessed. Results: Predominant referral sources were emergency department physicians and general practitioners. Mean wait-time for first assessment was significantly reduced by 70.5% employing the SSC model versus historical usual care. A diagnosis was established at first-contact in 80.5% of cases while 16.0% of patients required a second visit. Eighty-two patients (41.0%) were diagnosed with epilepsy. The most common non-seizure diagnosis was syncope (24.0%). An abnormal EEG was found in 93.9% of patients diagnosed with epilepsy. Sixty-three patients were started on anti-epileptic drugs. In 18% of cases driving restrictions were initiated by the SSC. There was moderate correlation between SSC nurses and physicians (kappa=0.54; p<0.001) diagnoses. Conclusions: The SSC model reduces wait-times, streamlines assessments, and impacts clinical care decisions.

Background: Despite the ease of intraoperative injection, intrathecal morphine (ITM) is rarely provided in lumbar spine surgery. We therefore sought to demonstrate the safety and efficacy of ITM following lumbar fusion. Methods: In this double-blind trial, 150 patients undergoing elective instrumented lumbar fusion were randomly assigned to receive a single injection of ITM (0.2 mg) or placebo (saline) prior to wound closure. Primary outcomes were postoperative pain on the visual-analog scale during the inital 24 hours after surgery and respiratory depression. Secondary outcomes included related adverse events, opioid requirements, and length of stay. Outcome curves were estimated in an intention-to-treat, repeated-measures analysis. Results: Age, disability, operative times, and pre-operative pain were similar in both groups. ITM was associated with less pain both at rest (p<0.002) and with movement (p<0.02) during the initial 24 hours following surgery. ITM did not increase the cumulative incidence of respiratory depression (hazard ratio 0.86, p=0.66). While ITM reduced postoperative opioid requirements (p<0.03), there was no significant difference in length of stay (p=0.67). Adverse events did not significantly differ between groups. The early benefits of ITM on postoperative pain were no longer apparent after 48 hours. Conclusions: A single ITM injection safety reduces postoperative pain following lumbar fusion. (ClinicalTrials.gov NCT01053039)

Background: The pathophysiology of subarachnoid hemorrhage (SAH) is complex and includes disruption of the blood-brain barrier (BBB). We freshly isolated BBB endothelial cells (BECs) by 2 distinct methods after experimental SAH and then interrogated their gene expression profiles with the goal of uncovering new therapeutic targets. Methods: SAH was induced using the prechiasmatic blood injection mouse model. BBB permeability studies were performed by administering intraperitoneal cadaverine dye injections at 24h and 48h. BECs were isolated either by sequential magnetic-based sorting for CD45-CD31+ cells or by fluorescence-activated cell sorting (FACS) for Tie2+Pdgfrb- cells. Total RNA was extracted and analyzed using Affymetrix Mouse Gene 2.0 ST Arrays. Results: BBB impairment occurred at 24h and resolved by 48h after SAH. Analysis of gene expression patterns in BECs at 24h reveal clustering of SAH and sham samples. We identified 707 (2.8%) significant differentially-expressed genes (403 upregulated, 304 downregulated) out of 24,865 interrogated probe sets. Many significantly upregulated genes were involved in inflammatory pathways. These microarray results were validated with real-time polymerase chain reaction (RT-PCR). Conclusions: This study is the first to investigate in an unbiased manner, whole genome expression profiling of freshly-isolated BECs in an SAH animal model, yielding targets for novel therapeutic intervention.

Background: It remains difficult to predict which patients will experience ongoing seizures or neuropsychological deficits following Temporal Lobe Epilepsy (TLE) surgery. MRI allows measurement of brain structures, such as the contralateral (non-resected) hippocampus (cHC) after TLE surgery. Preliminary evidence suggests that the cHC atrophies following surgery, however, the time course of this atrophy, relation to cognitive deficits and seizure outcome remains unclear. Methods: T1-weighted MR imaging and hippocampal volumetry in 26 TLE patients pre- and post-TLE surgery (and 12 controls) as: 1) two-scan group (TSG) (pre- and post-operatively at 5.4 years) and 2) longitudinal group (LG; pre- and on post-operatively on day 1,2,3,6,60,120 and at an average 2.4 years. Seizure outcome and pre- and post-operative neuropsychological assessment was performed. Results: The TSG had significant atrophy by 12% of the unresected cHC (p<0.0001) most pronounced (27%) in the hippocampal body alone. The LG revealed that this atrophy occured rapidly over the first week (1.3%/day; 3%/day cHC body). Significantly greater cHC atrophy was observed in those with ongoing seizures versus the seizure free (p=0.048). Conclusions: Significant cHC atrophy following TLE surgery that begins immediately, progresses over the first week, and remains signficantly depressed. The severity postoperative cHC atrophy may represent an early biomarker of the propensity for delayed seizure recurrence.

Background: Electrical stimulation (ES) to promote corticospinal tract (CST) repair has been recently examined, though remains under investigated. We examine the role of motor cortex ES on axonal re-growth and functional recovery in a spinal cord injury (SCI) rat model. Methods: A partial transection was performed at C4 in 48 rats. Animal groups included: ES333 rats (n=14; 333Hz, biphasic pulse, 0.2ms every 500ms), ES20 (n=14; 20Hz, biphasic pulse, 0.2ms every 1ms), SCI only (n=10), and sham (n=10; electrode insertion without ES). Rats were trained in stairwell-grasping with subsequent SCI and ES. Post-injury reaching scores were recorded weekly, and histology completed quantifying axonal re-growth. Results: Post-SCI grasping (p<0.01, ANOVA) and well reached were lower than baseline values (p<0.01, ANOVA) for all groups. ES20 animals had lower grasping scores (p=0.03, ANOVA) and farthest well reached scores post-SCI than controls (p=0.03, ANOVA). ES333 rats had more axonal collaterals (axonal sprouts rostral to lesion) compared to control animals (p<0.01, M-W). No difference was found between groups with respect to axonal regeneration into the lesion (p=0.13, ANOVA). Conclusions: Cortical ES of the injured CST results in greater axonal outgrowth, and influences functional outcomes depending on ES parameters. ES is a potentially promising SCI therapy, but further investigation is required.

Background: The Canadian Neurosurgery Research Collaborative (CNRC) is a trainee-led multi-centre collaboration made up of representatives from 12 of 14 neurosurgical centres with residency programs. To demonstrate the potential of this collaborative network, we gathered administrative operative data from each centre in order to provide a snapshot of the operative landscape in Canadian neurosurgery. Methods: Residents from each training program provided adult neurosurgical operative data for the 2014 calendar year, including the number of surgeries in the subcategories cranial, spinal, and peripheral nerve. Because some residency programs have surgeries distributed among more than one hospital, we calculated mean case load per residency program and per hospital. Results: Interim results from 6 neurosurgery residency programs are presented (with data from other programs forthcoming). Overall, there were on average 2,352 operative cases per residency program (n=6) and 1,176 operative cases per adult hospital (n=12). Among 5 programs with more detailed operative data, the mean numbers of cranial, spinal, peripheral nerve, and miscellaneous surgeries per residency program were 757 (47%), 487 (30%), 47 (3%), and 319 (20%) respectively. Conclusions: We show as a proof-of-concept that a trainee-led nation-wide research collaborative can generate meaningful data in a Canadian context.

Background: Aggressive surgical resections of posterior fossa tumours result in tremendous neurological sequelae as a result of damage to the brainstem. As such we sought to re-evaluate the role of aggressive surgical resections in the molecular era. Methods: 820 posterior fossa ependymoma and 787 medulloblastoma were genomically profiled and correlated with pertinent clinical variables. Results: Across 787 medulloblastoma cases, the value of extent of resection was greatly dampened when accounting for molecular subgroup. Near-total resections are equivalent to gross total resections across all four subgroups even when correcting for treatment. The prognostic value of a gross total resection as compared to a subtotal resection (>1.5cm2 residual) was restricted to Group 4 tumours (HR 1.26). Across 820 posterior fossa ependymoma PFA ependymoma was a very high risk group compared to PFB ependymoma, and a subtotal PFA ependymoma conferred an extremely poor prognosis. Gross totally resected PFB ependymoma could be cured with surgery alone. Prognostic nomograms in both medulloblastoma and ependymoma revealed molecular subgroup to be the most important predictor of outcome. Conclusions: The prognostic benefit of EOR for patients with medulloblastoma is marginal after accounting for molecular subgroup affiliation. In both molecular subgroups of posterior fossa ependymoma, gross total resection remains an important predictor of outcome.