Background A cerebral cavernous malformation (CCM) is a small cluster of thin-walled, dilated blood vessels within the brain which is prone to bleed. Although the quantity of blood leaking tends to be small, even a small intracranial haemorrhage (ICH) can result in a clinically significant neurological deficit. Because some focal neurological deficits (FND) may in fact be haemorrhages that were undetected by imaging, FND were also included in the analysis wherever possible. In Scotland, between 2006 and 2010, the annual CCM detection rate was 0.8 per 100,000 people. Since estimates of prognosis inform decisions about whether to treat CCM, it is crucial that the untreated clinical course of the disease is fully understood. Aim The aims of this thesis are (i) to quantify the risk of ICH (or ICH or FND, referred to as ‘clinical event’) for an untreated adult within five years of CCM diagnosis, (ii) to identify prognostic factors for ICH (clinical event), and (iii) to create a model to predict, at the time of diagnosis, an individual’s risk of a subsequent ICH (clinical event). Methods Initially, a literature review was undertaken. Then data from adults diagnosed with CCM in the Scottish Intracranial Vascular Malformation Study (SIVMS) were analysed. SIVMS is a prospective, population-based cohort study: it includes all adults resident in Scotland at the time of diagnosis of a first-ever intracranial vascular malformation during the two five-year periods 1999–2003 and 2006–2010. Time-to-event methods were employed to compare the estimated risk of ICH (clinical event) for those who experienced a first ICH (clinical event) during untreated five-year follow-up with those who experienced a second ICH (clinical event). A statistical challenge when analysing clinical outcomes from patients with CCM is that the outcome event of ICH or FND is comparatively rare; therefore a larger cohort of CCM patients was required to identify more robustly potential predictors of ICH (clinical event) and to create a prognostic model to predict, at the time of diagnosis, an individual’s risk of a subsequent ICH (clinical event). Three research groups agreed to contribute their data to enable an individual patient data meta-analysis (IPDMA) to be undertaken. Results In the two SIVMS cohorts, 136 (1999–2003) and 165 adults (2006–2010) were diagnosed with CCM. In the earlier cohort, the estimated risk of a first ICH within five years of presentation (2.4%, 95% CI 0.0% to 5.7%) was significantly lower (p < 0.0001) than the risk of a recurrent ICH (31.9%, 95% CI 4.5% to 59.3%), but the annual risk of a recurrence declined over the five-year period. In the same cohort, women had an increased risk of a second clinical event (log-rank χ2(1) = 6.2, p = 0.01). The IPDMA was based on 988 adults, 62 of whom suffered a first ICH within five years of CCM diagnosis. When the data were pooled, the estimated adjusted hazard ratio for first ICH for clinical presentation (ICH/FND vs other presentation) was 4.5 (95% CI 1.5 to 13.4) and for brainstem location (brainstem vs other location) the adjusted hazard ratio was 3.3 (95% CI 1.5 to 7.2); age, sex and CCM multiplicity did not add any additional prognostic information. Conclusion In this thesis two risk factors have been identified that are independently associated with increased likelihood of experiencing an ICH (or clinical event) within five years of diagnosis. A prognostic model has been built and evaluated, based on these factors. Other areas to be explored in the future include external validation of the model and investigating the effects of (i) antithrombotic therapy and (ii) pregnancy on the progression of the disease.

Objectives: Intracranial subdural empyema (ICSDE) can be a devastating condition, with a sequelae ranging from epilepsy, focal deficits to death. Factors affecting the outcome in subdural empyema range from level of consciousness, the extend of subdural pus at the time of diagnosis and the type of surgical procedure performed. Previous studies have conflicting results of unfavourable prognostic factors associated with ICSDE. The outcome of this condition at Steve Biko Academic Hospital (SBAH) is reported, as well as factors influencing the outcome. Methods: A retrospective analysis of all the patients admitted at neurosurgery unit of SBAH during 2006 – 2010 period with confirmed subdural empyema on brain CT scan and at surgery. Data sheet was used to collect all clinical information from patients’ records. Glasgow Outcome Scale and Henk W. Mauser grading were used to report on the outcome. Results: A total of 34 patients (20 males and 14 females) with mean age of 16.1 years were admitted with a diagnosis of ICSDE. The common presenting features were headache (58.8%), fever and seizures (47.0% each). Over 61% of patients had hemiplegia at presentation. CT scan confirmed subdural collections with 70.6% over the convexity, 23.5% at the convexity and parafalx and only 5.9% had bilateral collections. Complicated paranasal sinusitis was the origin of infection in 82.3%, followed by meningitis with 8.8%. Burr hole washout was done in 52.9% of patients, while 38.2% had burr holes with drains in situ and 8.8% had craniotomy to evacuate the subdural pus. All the patients were given empiric triple antibiotic therapy. Streptococci species were the most cultured organisms in the 19 (56.0%) patients who had positive cultures, however 15 (44.0%) patients had negative cultures. Resistance to penicillin was noted in 5.0% of cases only. Sixty-five percent of patients had good outcome with no seizures nor neurological deficits. The overall mortality was 15.0% in this study, with none from patients who had craniotomy. Conclusion: Clinical presenting features and organisms cultured seems to be the same internationally, particularly those due to complicated sinusitis. Empiric triple antibiotic therapy of 3rd generation cephalosporin plus vancomycin plus metronidazole is still relevant at SBAH. Factors associated with favourable outcome were ages between 11 and 20 years, and craniotomy as the surgical procedure of choice. / Dissertation (MMEd)--University of Pretoria, 2013. / Neurology / Unrestricted

Objective: This study aims to derive a clinically-applicable decision rule to predict the risk of symptomatic vasospasm, a neurological deficit primarily due to abnormal narrowing of cerebral arteries supplying an attributable territory, in aneurysmal subarachnoid hemorrhage (SAH). Methods: SAH patients presenting from 2002 to 2011 were analyzed using logistic regression and recursive partitioning to identify clinical, radiological, and laboratory features that predict the occurrence of symptomatic vasospasm. Results: The incidence of symptomatic vasospasm was 21.0%. On multivariate logistic regression analysis, significant predictors of symptomatic vasospasm included age 40-59 years, high Modified Fisher Grade (Grades 3 and 4), and anterior circulation aneurysms. Conclusion: Development of symptomatic vasospasm can be reliably predicted using a clinical decision rule created by logistic regression. It exhibits increased accuracy over the Modified Fisher Grade alone and may serve as a useful clinical tool to individualize vasospasm risk once prospectively validated in other neurosurgical centres.

Thesis (MMED. (Forensic Medicine)) --University of Limpopo, 2019 / Traumatic intracranial haemorrhages are common, carry a high mortality rate and are therefore commonly known in the practice of forensic pathology as unnatural deaths. Studies have demonstrated a significant decrease in mortality rate among patients who received surgical interventions compared to patients who were treated medically. Missed or mis-diagnoses, which may be apparent during an autopsy procedure, present possible missed treatment opportunities. Aim/objective and methods The study investigated the discrepancy rate and discrepancy pattern of diagnosis between antemortem brain computed tomography (CT) scan findings and autopsy findings in deceased patients with traumatic intracranial haemorrhage (TIH). A quantitative retrospective descriptive study was conducted based on bodies presented with TIH at Pietersburg Hospital Forensic Pathology Department. A total of consecutive 85 cases with antemortem CT (ACT) scan findings were compared to autopsy findings using percentage agreement and Cohen’s kappa statistics. Results and conclusion There was a fair overall agreement (k=0.38) with overall discrepancy rate of 24.74%, ranging from 9.41% to 34.12% for individual TIH between ACT scan and autopsy findings. Subarachnoid haemorrhage had the lowest agreement between the ACT scan and autopsy findings for TIH. Patient and doctor factors associated with the discrepancies were assessed. Those associated factors, if addressed, may have a positive impact on patient outcome. As far as the debate on non-invasive autopsy procedure is concerned, as a result of existing discrepancy rate, we conclude that ACT should not be used alone in the determination of cause of death but may be used in conjunction with autopsy findings.

京都大学 / 新制・課程博士 / 博士(医学) / 甲第23380号 / 医博第4749号 / 京都大学大学院医学研究科医学専攻 / (主査)教授 山下 潤, 教授 木村 剛, 教授 YOUSSEFIAN Shohab / 学位規則第4条第1項該当 / Doctor of Medical Science / Kyoto University / DFAM

京都大学 / 新制・論文博士 / 博士(医学) / 乙第13436号 / 論医博第2235号 / 新制||医||1054(附属図書館) / (主査)教授 花川 隆, 教授 YOUSSEFIAN Shohab, 教授 髙橋 良輔 / 学位規則第4条第2項該当 / Doctor of Medical Science / Kyoto University / DFAM

This thesis deals with the detection of intracranial haemorrhages and their type classification in head CT images. The method of haemorrhages detection is based on a series of classifiers of the presence and type of haemorrhages in 2D CT slices in axial, sagittal and coronal plane, that may localise the bleedings and determine their types. The classifiers are based on the convolutional neural network architecture Inception-ResNet-v2. The head CT dataset CQ500 which is made available for public access, is used for the experiments. The thesis describes an additional manual annotation of the data, as the available annotations are insufficient for the purposes of the experiments. This thesis includes a theoretical basis of the essential medical knowledge, machine learning based classification and detection methods, and the detection algorithm proposal, realisation and testing. The algorithm performance is evaluated and discussed together with the potential implementation of the algorithm in computer-aided diagnosis systems.

No description available.

Cerebral malaria is a common encephalopathy in African children, but the cause of death and neurological sequelae are unknown. This dissertation examines the hypothesis that raised intracranial pressure (ICP) is a determinant of poor outcome in Kenyan children with cerebral malaria. The opening cerebrospinal fluid pressure was raised in all 26 children in whom it was measured on admission and 92% of 35 children in whom it was measured after admission. Brain stem signs, particularly an abnormal respiratory pattern, absent pupillary responses and a lack of spontaneous eye movement were associated with a death. In 33 children who died with cerebral malaria, at least 18-42% had clinical features of transtentorial herniation, according to the criteria used. Intracranial pressure monitoring was performed in 18 children with severe CM, of whom 14 had computerised tomography (CT) and in 10 the basal cranial arteries were monitored with transcranial Doppler (TCD) sonography. Three children with severe intracranial hypertension (maximum ICP > 60 mmHg and minimum cerebral perfusion pressure (CPP) < 40 mmHg) had a poor outcome despite aggressive therapy with mannitol. One child with a maximum ICP of 151 mmHg died with the signs of uncal and medullary stages of herniation. In the other 2 children, middle cerebral artery velocity and vascular resistance monitored with TCD sonography changed with ICP and CPP. Both of these children had diffuse brain swelling associated with generalised hypodensity on their acute CT scans. These children survived° with cerebral atrophy on their convalescent scans and severe neurological deficits. In the 8 children with intermediate intracranial hypertension (maximum ICP 20-60 mmHg and CPP < 50 mmHg) mannitol was effective in controlling the intracranial hypertension. TCD was not reliable in detecting changes in ICP or CPP. Two of these children had acute brain swelling, but the tomographic density was normal and the swelling had resolved when the repeat scans were performed 12-24 days later. All the children with intermediate intracranial hypertension survived without major neurological sequelae. In the remaining 7 children who had ICP monitoring, the maximum ICP was <20 mmHg and mannitol was not administered. None of the CT scans showed brain swelling and the children survived without severe sequelae. In a further 9 children with severe malaria (6 with CM) the agonal stages were monitored with TCD. Three children with CM had sonographic features of progressive intracranial hypertension associated with signs of herniation, whilst the other children (including 3 with CM) did not have these sonographic features, although one had evidence of brainstem compromise before dying. Thus raised ICP is a feature of CM in Kenyan children. Severe intracranial hypertension is associated with a poor outcome and could be responsible for at least a third of the children dying from CM. Mannitol reduces the ICP, but does not prevent nor control severe intracranial hypertension.

Traumatic brain injuries (TBIs) are a major public health problem worldwide where the symptoms can be anything from mild concussion to severe swelling of the brain tissue. As a result of TBI the intracranial pressure (ICP) can elevate to pathological levels with severe consequences such as hypoxia, ischemia and brain hemorrhage. TBI and the subsequent ICP increase could hence lead to disability or in worst cases death. Therefore to understand the severity of a head injury and the path regarding further treatments, monitoring of a patient's ICP is crucial in the intensive care units (ICU) environment. Invasive methods of ICP monitoring are at this present date the standard in ICU because of the accuracy when compared to non-invasive methods. All invasive ICP monitoring methods come with a risk to the patient and require the presence of a neurosurgeon. The thesis's objective was to develop a gradually increasing ICP testbed for a new non-invasive microwave based skin patch sensor. The aim with this project was to verify if a dependence in the resonance characteristic of the NASA SansEC microwave sensor with respect to ICP exists as suggested by previous works in a novel testbed and to provide a correlation model based on the testbed experiment. The developed testbed simulate increasing ICP by increasing volume of an artificial cerebro-spinal fluid (aCSF) liquid, a liquid emulating the CSF. The microwave sensor's resonance frequency is due to the permittivity changes caused by the change (increasing) in the fluid volume, which for this setup is directly correlated to the pressure change as well. Trials with different aCSF samples were made to ensure that the used aCSF in the testbed had the same dielectric properties as human CSF. The developed testbed had a simple structure made with several plastic containers of rectangular shape which were found to be well suited for the purpose of the experiment. For the microwave sensor trials an Fieldfox microwave analyzer was used and the sensor was evaluated around 1 - 4 GHz. The testbed pressure was increasing from 0 - 47 mmHg covering most useful ICP ranges. Larger pressures were also possible but limited by the height of the work room and the increase of complexity in the testbed design. The results from the trials showed a total resonance frequency shift of 76 MHz from 4 - 30 mmHg with an linear correlation of R2 = 0,91. The sensor measurements above 30 mmHg showed a saturation where the first principal frequencies were stable at 1,368 GHz. The linear relationship obtained for 4-30 mmHg is a reassurance that the Nasa SansEC sensor should be studied further. Future work should include new trials with modifications to the testbed setup and sensor design.