Includes bibliographical references. / The aims of the study were to compare the structure, process and outcomes of care of patients with acute stroke (first ever stroke) who were and who were not admitted to hospital; to compare the demographic and clinical features of patients who were and who were not admitted to hospital; to identify demographic and clinical features associated with hospital admission following acute stroke.

This doctoral work aimed to describe the epidemiology of TB in a high HIV and TB prevalent community from 1997 to 2008, and to assess the effect of HIV and a highly active antiretroviral therapy (HAART) programme on the TB epidemic in this setting.

Tuberculous meningitis (TBM) is a common cause of meningitis in adults in South Africa (1-3), second only to cryptococcal meningitis in studies of microbiologically confirmed meningitis, and accounting for 28% of cases in one (1). Conventional diagnostic tests for TBM are, however, relatively insensitive, and the true incidence is likely to be underreported. When both microbiological and clinical diagnostic criteria (4) are used in the same setting, the incidence of TBM rose to 57% (3), emerging as the most common cause of meningitis in adults in a district level hospital in the Western Cape. In the setting of high human immunodeficiency virus (HIV) and tuberculosis (TB) prevalence, approximately 88% of patients with definite or probable TBM are co-infected with HIV (3, 5) and six-month mortality in this group approaches 50% (3). Survivors may be left with long-term disability secondary to hydrocephalus, cranial neuropathies, seizures and strokes (6).

Bibliography: p. 131-170. / This study examines the effects of various protoporphyrinogen oxidase mutations responsible for variegate porphyria, the role of the arginine-59 residue, and the glycines in the conserved flavin binding site, in catalysis and/or cofactor binding. Wild type recombinant human protoporphyrinogen oxidase and a selection of both naturally occurring and self-designed mutants were generated, expresses and purified. The self designed mutants included a conservative and two non-conservative arginine-59 replacements, and substitution of glycine residues at positions 9, 11, and 14 by alanine. The expression and purification for all protoporphyrinogen oxidases was optimised, enabling their purification to homogeneity by single step metal affinity chromatography.

Bibliography: leaves 109-137. / Cardiac preconditioning describes a cell survival program whereby a trigger renders the heart partially resistant to subsequent ischaemia/reperfusion induced cell death.

Includes bibliographical references.

Includes abstract. Includes bibliographical references (p. 233-318).

Bibliography: leaves 163-203. / This work involved the characterisation of protoporphyrinogen oxidase (PPO), the penultimate enzyme in haem biosynthesis, from Bacillus subtilis, Myxococcus xanthus, and human. A defect in human PPO causes variegate porphyria, an autosomal dominant disorder characterised by skin photosensitivity and propensity towards acute neurovisceral crises. At the beginning of this project little information was available on the kinetic and biophysical properties of isolated PPOs due largely to difficulties associated with their purification from natural sources.

Background Randomised control trials (RCTs) are capital-intensive projects and demand substantial human and capital resources. Therefore, proper planning, precise research questions and adequate thoughts are required in areas such as acceptability of the intervention, participant recruitment, and selection of measurable outcomes. Ensuring all these are possible before delving into the main work can be forecasted through pilot trials. They help in determining the feasibility of the intended critical endpoints and ensure the applicability of the result findings. However, no matter how noble and vital the results are, improper reporting can make them unusable. The thesis brings to the fore the importance of pilot trials in low- and medium-income countries and how they can help make a case for more extensive definitive trials. It then focuses on how subgroup analysis can be used as an essential statistical tool for fully understanding clinical trial results and can be used to unearth non-apparent results in RCT. In the thesis, we highlight the need for accurate, systematic and complete reporting of pilot trials, by critically appraising the literature on abstract reporting in heart failure. The thesis discusses several aspects of pilot trial processes to understand better its unique role in helping refine the components of RCT, to make the running smooth and findings affirmative. Leveraging on the experience of working as a clinical research fellow in the second Investigation for Management of Pericarditis in Africa (IMPI-2) trial, the lessons learnt in planning, designing, implementation, recruitment and reporting of the IMPI-2 pilot forms the nucleus of this thesis. The experience acquired in the process and how they can help in planning future definitive studies are discussed in different sections of the thesis. Methodology The thesis uses the experience gained in critical appraisal of the literature, participation in preliminary planning and active participation in a multicentre randomised control trial to understand the importance of some issues during an RCT. These areas include the need for specific objective setting, identification of research participants and collaborators, the acceptability of research intervention, proper identification of possible outcome measures, retention of participants and quality reporting of research findings. It begins with an overview of pilot trials, subgroup analysis and tuberculous pericarditis which is the primary disease focus of the IMPI project. Each subsequent chapter of the thesis is presented either as a published manuscript or prepared for submission as a manuscript. The quality of reporting of pilot trials is then examined by systematically surveying the reporting of abstracts of pilot trials in heart failure using the checklist of the Consolidated standard for reporting of trials (CONSORT) extension for pilot trials. A subgroup analysis of IMPI-1 trial planned a priori on the modification of the effect of prednisolone by baseline pericardiocentesis status of trial participants is used to highlight the role subgroup analysis can play in unmasking the group effect in the randomised control trial. The thesis then goes on to present the preliminary report of the IMPI-2 pilot study, highlighting the lessons learned and aspirations in need of refining. Retention of study participants is essential to achieve success in clinical trials, one way of ensuring this is by letting the study participants understand the objectives and processes of the research and gaining their confidence. Thus, in chapter six, we piloted the use of the University of California San Diego Brief Assessment of Capacity to Consent (UBACC), a screening tool for evaluation of informed consent (IC) comprehension as a training tool for iterative learning and evaluation of consent comprehension among IMPI-2 pilot trial participants. Results and Conclusion Enormous resources expended in clinical research can yield good returns before the main work commences, a well-planned micro trial run in the form of a pilot study is undertaken. Our systematic survey of abstracts of pilot trials in heart failure showed that reporting of abstracts of pilot trials is currently suboptimal. Deciding ahead of time on what to report by systematically identifying the different sections needed to inform the audience can improve the quality adequately. Planning subgroup analysis during the design of main studies can help reveal unsuspected findings. The subgroup analysis result showed that pericardiocentesis, despite its essential use among patients with pericardial effusion, did not significantly influence the effect of prednisolone on the primary critical outcomes among IMPI-1 participants. The preliminary report of IMPI-2 trial was designed as a two-phase study; phase 1 results showed that at 50mg, intrapericardial alteplase was safe in facilitating complete pericardiocentesis, while phase two showed that it was feasible to recruit, randomised and follow up patients in line with the study protocol. However, we identified participant retention as a considerable challenge. The result of the pilot revealed that more effort should be expended on participants’ education on the clinical condition, the reason for the trial and the need for follow-up adherence. There is also a need to make adequate provision for the use of field workers for contact tracing to reduce the dropout rate. In the main trial protocol, there may also be a need to reconsider the patient's selection and use of fibrinolysis in malignant effusion, judging from the high rate of 3 months mortality in this group of patients. The results of the informed consent study showed that an improved level of comprehension followed the use of iterative learning, a higher level of education and non-use of interpreters during informed consent delivery. These finding led us to conclude that every effort should be made to ensure that research participants entirely buy into the research they are asked to be part of through thorough information delivery. Doing so can help improve participants adherence to the trial follow-up.

Includes abstract. Includes bibliographical references.