Diss. Linköping : Univ., 2001.

Objective. Over the last two decades there has been increasing evidence that radiosurgery and radiotherapy management of skull-base paragangliomas is as effective as microsurgical resection and carries less morbidity. This 10 year retrospective review of 24 patients in a single institution, treated over 10 years assesses tumour control rates and morbidity associated with radiosurgery and radiotherapy treatment. Method. Patients with a radiological diagnosis of skull-base paragangliomas were treated with different techniques of stereotactic and image-guided radiotherapy delivering hypo fractionated irradiation. Techniques used included conventional radiotherapy or intensity modulated radiotherapy (IMRT), dynamic arc (DA) and volumetric modulated arc therapy (VMAT). Analysis of local tumour control was performed using RECIST criteria and the KaplanMeier method. 69% of patients received 14-16gy in 1-3 fractions while 31% received 48- 50gy in 25 fractions. Radiation-associated toxicity was graded according to the commonly used Radiation therapy Oncology group (RTOG) toxicity criteria. Results. 24 patients with skull-base paragangliomas were treated with a median follow up of 43 months. One patient lost to follow up and was excluded. Tumour control was achieved in 96% of patients. 76% of patients treated reported no radiation associated toxicity. 24% of patients had some radiation associated toxicity: the conventional group 12%, stereotactic radiosurgery 8% and stereotactic radiotherapy 4%. 43% of patients in the conventional group had progression of hearing loss in the affected ear. One patient in the radiosurgery group developed osteonecrosis of the temporal bone at 5 year follow up. Conclusion. Radiosurgery and radiotherapy are efficacious in achieving tumour control with minimal morbidity. Tumour control rates in the study are similar to control rates in literature. Radiation associated toxicities are mainly minor. Study is limited by the retrospective nature and limited duration of follow up.

Includes abstract. Includes bibliographical references.

INTRODUCTION: The incidence of intraventricular hemorrhage (IVH) in very low birth weight infants is between 25 and 50%. Approximately 13-60% of these patients will develop progressive post hemorrhagic hydrocephalus (PHH) and of these 22- 70% will require CSF diversion. The most common therapeutic intervention is insertion of a ventriculoperitoneal shunt but there is considerable controversy surrounding the timing of the operation. Most authors promote early surgery to prevent secondary injury from hydrocephalus but it was our impression that this was associated with a higher incidence of shunt complications. METHOD: The incidence of shunt complications in 36 patients shunted for PHH were retrospectively reviewed. Patients were treated at Red Cross Children's Hospital over an 8 year period. RESULTS: Nine (25%) of the 36 patients required shunt revision for obstruction, seven required revision during the initial admission. Shunt infection occurred in 4 patients (11 %) all during the initial hospital admission. Four patients died, one from a shunt related complication. There was a clear relationship between the timing of surgery and the incidence of complications (chi square test p,0.01 ). Nineteen patients underwent surgery before 5 weeks of age and 9 developed early shunt complications. Of those shunted after 5 weeks none had an early complication. Groups were matched for weight and grade of IVH. DISCUSSION: A possible explanation for these results is that shunt complications are related to the quantity of blood present in the CSF at the time of shunting. A short delay before intervention is recommended in an effort to reduce the morbidity of shunt complications.

This dissertation describes the development and clinical use of a novel stereotactic neurosurgical system, the Cape Town Stereotactic Pointer (CTSP). This system has four main components; a halo containing three fiducials also serves as the platform for a tripod pointing device which is set with the aid of a 3D phantom or a printed setting diagram, and software which enables transformation of imaging space into patient space. Laboratory tests indicated an application accuracy of 1.9 +/- 0.6mm using the 3D phantom to set the tripod. From the first clinical application, the system underwent a series of iterations which could broadly be divided into four successive phases of refinement. This took place over a six year period, encompassing one hundred patients who underwent 115 stereotactic procedures. Indications for surgery included biopsy (62.6%), aspiration (15.7%) and cannulation (21.7%) and the surgical objective was realized in 101/109 cases (92.7%). Given the fact that six of the eight failures represented errors of surgical judgment that could not be ascribed to the device, and each of two system errors resulted in a significant modification to the system, the CTSP demonstrated a satisfactory level of accuracy in the clinical setting. This was accomplished at an acceptable complication rate, with one death five days after surgery attributable to a stereotactic procedure (mortality 0.9%) and major morbidity in two cases (1.7%); thirteen patients experienced minor complications, all of which proved to be transient (11.3%). A simple protocol for use of the CTSP evolved over the course of this study, making it easier for neurosurgeons from varying backgrounds to introduce stereotaxis into their practice with the help of this system. In addition to satisfactory levels of clinical reliability and safety, the system was versatile and also well tolerated by patients. It is hoped that the CTSP provides a costeffective alternative for neurosurgeons working in under-resourced settings. Sixty units of the production version of the CTSP have been sold and the system is now in use in ten countries.

Brain tumours in children are one of the most challenging diseases to treat, and so outcomes are variable and often lacking. There are currently no reliable data of presentation of disease, the spectrum of tumours treated, how these are treated, and what the outcomes are for children in South Africa, and certainly no molecular biology data. In this respect, this thesis investigated the two commonest types of childhood brain tumour, the highly malignant Medulloblastoma (MB) and the generally less aggressive Juvenile Pilocytic Astrocytoma (JPA) with relation to their molecular biology and their clinical correlates to begin to address this gap and build capacity for further molecular-based studies in an African context. The study design in this thesis takes a systematic approach and is structured into MB and JPA biochemical characterisation followed by 4 studies of their respective proteomic profiles. The study design involved creating appropriate patient cohorts and determining sample characteristics for interpretation of results. The statistical power achieved in this thesis showed a minimum of 2-fold difference for a power greater than 0.8 in each case. Proteomic clustering was used to validate or delineate any discrepancies in subtype assignments for MB. Molecular profiles together with proteomic data of MB and JPA cases in this thesis provide evidence for some novel molecular pathways, proteins and peptides associated with pathogenesis. This work therefore provides extensive data that is hypothesis generating for further studies that could build upon molecular understanding in a South African and larger African context.

Includes bibliographical references.

Includes abstract. / Includes bibliographical references (leaves 98-112). / Pituitary apoplexy is an uncommon, yet potentially fatal illness, usually the result of infarction, hemorrhage or a combination of both in a pituitary tumor. The management of pituitary apoplexy consists of replacement therapy and in the majority of patients, surgical decompression, although some cases may be treated conservatively. Up to now no study has attempted to separate the two histopathological types of pituitary apoplexy or to analyze their clinical and radiological significance on presentation and outcome.

Bibliography: leaves 248-285.

Includes bibliographical references / Background: Assessment of intracranial pressure (ICP) is an essential aspect in the management of most neurosurgical conditions in children. While invasive ICP monitoring is considered the criterion standard, the need for a reliable, non-invasive, easy-to-use and accurate method to detect and monitor raised ICP has inspired the development of many useful techniques. The present study examined the relationship between transorbital ultrasound measurement of the optic nerve sheath diameter (ONSD) and invasively measured ICP in children, as well as the influence of relevant physiological and demographic variables on this relationship. Methodology: ONSD measurement was performed using a high frequency, small footprint linear array probe, and prior to invasive ICP measurement. All patients were under general anaesthesia and being mechanically ventilated. Physiological variables including systolic blood pressure (SBP), diastolic blood pressure (DBP), mean arterial pressure (MAP), pulse rate, temperature, respiratory rate and end tidal carbon dioxide (ETCO2) level were recorded at the time of ONSD measurement. The ONSD measurements were analysed for repeatability, intra- and inter-observer variability as well for correlation between images acquired in different planes and those obtained from either eye. The diagnostic accuracy of ONSD measurement for detecting ICP at different thresholds of 20, 15, 10 and 5 mmHg was analysed. This analysis included evaluation of age-related thresholds for defining different ONSD cut-off values in children. Dynamic image acquisition was performed and analysed to evaluate the relevant pulsatile motion of the ONS as a marker of the sheath stiffness. Results: One hundred and seventy four children undergoing diagnostic or therapeutic surgical procedures were included in this study. ONSD measurement demonstrated good correlation with ICP across the entire patient cohort (r = 0.66, p < 0.001), but was better in children > 1 year or with a closed anterior fontanelle (AF) (r = 0.7, p < 0.001). Age above and below 1 year was found to be an appropriate age threshold for defining two different sets of ONSD cut-off values. The study however, supported using patency of the AF as a stronger clinical marker for describing different ONSD cut-off values in children. The second part of this work described a dynamic technique for analysing the pulsatile motion of the ONS. Analysis of the deformability index (DI) as an indirect marker of ONS stiffness, revealed a statistically significant relationship with ICP (sensitivity of 90%, specificity of 87% for detecting ICP ≥ 20 mmHg). Conclusion: Measurement of the ONSD is a sensitive surrogate marker of raised ICP, but demonstrated poorer specificity. This relationship was more reliable in older children, particularly when the AF was closed. Analysis of the dynamic characteristics of the ONS appeared to provide useful additional information as an independent marker, and may contribute to our overall understanding of ONSD measurement in raised ICP.