Paediatrics (RCH) - Theses

Permanent URI for this collection

http://hdl.handle.net/11343/200

Filters

2016 - 2019 1 2020 - 2023 1
Reset filters

Settings
Statistics
Citations
Type: PhD thesis × Subject: epilepsy ×

Search Results

Now showing 1 - 2 of 2
  • Item
    No Preview Available
    Bottom-of-sulcus dysplasia (BOSD): nidus or network?
    Macdonald-Laurs, Emma Amanda ( 2023-12)
    Bottom-of-sulcus dysplasia (BOSD) is a subtype of focal cortical dysplasia (FCD) type II where imaging, genetic and histopathological abnormalities are maximal at the bottom of a single sulcus. BOSDs are highly epileptogenic but due to their small size and complex seizure manifestations, are challenging to diagnose, and treat. Patients with BOSD often experience diagnostic delays, are thought to be poor surgical candidates, or have larger than necessary resections. This hypothesis tested in this thesis was that the electroclinical and cognitive manifestations of BOSD relate to their connectivity (network) while epileptogenicity and the surgical target are localised to MRI-positive lesion (nidus). Chapters 1-3 review BOSD and FCD literature, and outline thesis aims and methodologies. Chapter 4 describes the clinical, imaging, histopathology, and genetic features of patients with BOSD. BOSDs present with a recognisable clinical syndrome with rapidly-escalating, stereotyped, focal seizures, almost universally becoming drug-resistant and requiring epilepsy surgery. BOSDs were missed on initial MRI in the majority (68%), with diagnosis being aided by 18F-FDG-PET. Genetic testing confirmed BOSD are an “mTORopathy” with pathogenic variants found in 63%. Delay to diagnosis and surgery was important, with low intellect being associated with longer epilepsy duration (network). Chapter 5 examines whether single-stage, limited-resection, epilepsy surgery is effective in BOSD-related epilepsy. All patients had surgery without intracranial EEG monitoring, despite BOSDs frequently involving eloquent cortex (40%). Most patients (68%) had resection of the dysplastic sulcus only; 16% were reoperated for recurrent seizures. At median 6.3 years follow-up 87% of patients were seizure-free. These findings support corticectomy limited to the MRI-positive BOSD (nidus). Chapter 6 seeks to understand the neurobiological basis for the observations made in Chapter 5. Group-level analyses showed that markers of epileptogenicity and dysplasia are mostly confined to the dysplastic sulcus with ECoG and 18F-FDG-PET abnormalities over the gyral crown and adjacent gyri being less and reducing following resection of the dysplastic sulcus only. These findings suggest that intrinsic epileptogenicity is limited to the dysplastic sulcus (nidus), and support limited-resection surgery. Chapter 7 examines the determinants of age at seizure onset in BOSD. Larger BOSD volume, germline mTOR pathway variants, and location within, and increased connectivity to, the default mode network were associated with younger seizure onset. These findings de-emphasise the contribution of postnatal cerebral maturation, for example myelination, to seizure onset age and reveal novel contributions of default mode network connectivity and genetic aetiology. Chapter 8 explores the contribution of 18F-FDG-PET to automated detection of BOSD. This study validated the Multi-centre Epilepsy Lesion Detection group’s machine- learning method in BOSD with the addition of 18F-FDG-PET. Of all surface-based features, cortical and subcortical hypometabolism discriminated BOSD from non-BOSD regions with highest sensitivity. Furthermore, addition of 18F-FDG-PET to the machine- learning method increased automated BOSD detection beyond use of MRI-only. BOSD are a discrete epileptogenic and dysplastic nidus with network-driven electroclinical and cognitive manifestations. Despite growing awareness, BOSD are still frequently missed in clinical practice. Efforts to suspect, detect and effectively treat BOSD are warranted, given their excellent response to surgery.
  • Item
    Thumbnail Image
    The epidemiology and aetiologies of the severe epilepsies of infancy
    Howell, Katherine Brooke ( 2016)
    The severe epilepsies of infancy (SEI) are a group of infantile-onset seizure disorders characterised by frequent seizures, abnormal EEG and pharmacoresistance to anti-epileptic therapy. SEI include well-described epilepsy syndromes such as early infantile epileptic encephalopathy (EIEE), epilepsy of infancy with migrating focal seizures (EIMFS) and West syndrome. Cognitive outcome is often poor, due to effects of seizures, the underlying aetiology, and antiepileptic drugs (AEDs) on the developing brain. There is an urgent need for novel treatments. Where effective therapies are available, such as epilepsy surgery for brain malformations, treatment can be life-changing. Given that developmental outcomes may be significantly improved in the context of optimal seizure control at an early age, determining the underlying cause of SEI early in life is paramount. The aetiologies of SEI are heterogeneous; a large number of acquired and genetic brain disorders are reported. In many infants, the cause remains unknown despite investigation, and is presumed to be genetic. With the emergence of next generation sequencing (NGS) techniques such as whole exome sequencing (WES), a rapidly growing number of genetic causes of SEI is now recognised and gene discovery is ongoing. The genetic epidemiology of SEI has not been studied and the relative importance of each genetic cause is not known. Brain malformations, chromosomal abnormalities, inborn errors of metabolism and some genetic disorders can be diagnosed with technologies currently available in clinical practice. Studies of WES and other NGS techniques in epilepsy populations have shown that these techniques identify the aetiology in 10-50% of undiagnosed patients. No study has specifically looked at the yield in SEI, and no population-based studies have been reported. The yield and cost-effectiveness of NGS for SEI at a population level remains unknown, and access to genetic testing is currently poor in most regions of the world. This population-based study of SEI in Victoria, Australia aimed to study the incidence and determine the aetiologies, electroclinical phenotypes and other phenotypic characteristics of SEI. As part of a particular focus on genetic aetiologies, the study aimed to identify genetic causes in infants with SEI of unknown aetiology using WES, and determine the yield and cost of early genetic testing relative to current standard diagnostic pathways for investigation of SEI aetiology. Infants with SEI born in Victoria during 2011-2013 were identified by a comprehensive state-wide search of multiple sources. Infants with potential SEI were identified by review of all electroencephalogram (EEG) reports on children under two years old during 2011-2015 (n=4505), and search of neonatal intensive care unit (NICU) databases for neonates with seizures born 2011-2013 (n=379). Hospital records of infants with potential SEI from the three main paediatric hospitals in Victoria, The Royal Children’s Hospital (RCH), Monash Health (MMC) and The Austin Hospital, and the two NICUs not co-located with a paediatric hospital, The Royal Women’s Hospital (RWH) and The Mercy Hospital for Women (MHW), were reviewed to confirm clinical and demographic inclusion criteria were met. SEI was defined as epilepsy onset before age 18 months, frequent seizures (> daily for a week or > weekly for a month), epileptiform EEG and pharmacoresistance (failed 2 appropriate anti-epileptic therapies); infantile spasms were automatically included. In infants with confirmed SEI, medical records, EEG recordings and brain magnetic resonance imaging (MRI) were reviewed to determine each infant’s epileptic syndrome, outcome at two year old, and aetiology. Clinical assessment and WES were performed if aetiology or electroclinical phenotype was unknown. 114 infants with SEI were ascertained. The incidence of SEI in Victoria is 51/100,000 live births/year. West syndrome/infantile spasms was the most common epileptic syndrome, with an incidence of 33/100,000 live births/year. EIMFS and EIEE had incidences of 4.5 and 3.6/100,000 live births/year respectively. At two years old, 18 (16%) infants were deceased. 86/98 (90%) survivors had delayed development, and 46/98 (47%) ongoing seizures. All infants whose presenting epileptic syndrome was EIEE, early myoclonic encephalopathy (EME) or EIMFS at epilepsy onset were deceased or had severe developmental impairment. Normal development was seen in 9/64 (14%) infants who presented with West syndromes/infantile spasms or a unifocal epilepsy, and only two were deceased. The aetiology was identified in 76 (67%) and unknown in 38 (33%). Fourteen (12%) infants had an acquired brain insult such as hypoxic-ischaemic encephalopathy or perinatal stroke. The remaining infants had genetic or presumed genetic aetiologies. Brain malformations were identified in 31 (27%), including focal cortical dysplasia (FCD) in 14 (12%). Six (5%) infants had metabolic disorders and nine (8%) had chromosomal abnormalities. Sixteen (14%) had single gene disorders, including 11 (10%) with disorders of ion channel function (channelopathies). Aetiology was known from clinical testing in 61 (54%). Research MRI review identified the cause in a further 4 (4%) and research genetic testing in 11 (10%). Among 86 (75%) infants with no aetiological diagnosis prior to epilepsy onset, the highest yield investigations were MRI and genetic testing, which identified the cause in 25/85 (29%) and 16/50 (32%) respectively. 13/50 (26%) infants had a genetic variant of unknown significance (VOUS) identified on WES; these are being further investigated. Modelling of diagnostic pathways showed that performing WES early in the diagnostic pathway and reducing the amount of metabolic testing increases diagnostic yield for less cost compared with the current standard diagnostic pathway. Approximately 1:2000 infants have SEI, equating to over 150 new cases of SEI in Australia per year. Outcomes for seizure control, development and survival are poor. Brain malformations were the most common cause, were under-recognised, and should be considered in those with unknown aetiology, especially in those with West syndrome or unifocal epilepsy. Channelopathies were the most common group of single gene disorders. Next-generation genetic testing and high quality brain imaging improved diagnostic yield, with implications for treatment and reproductive counselling, and should be implemented early in the diagnostic pathway in clinical practice. Future work will focus on identifying the aetiology in the remaining infants, determining the genetic basis of brain malformations causing SEI, and studying the yield of other NGS and brain imaging techniques to improve the rate of early diagnosis. This work will inform research into development of novel and targeted treatments for these devastating disorders.