The present study was undertaken in order to investigate temporal integration at threshold and suprathreshold levels in normal-hearing and cochlea-impaired subjects, and also to examine the effect of frequency on the amount of integration for each group. Thresholds for 500-and 20-msec tone pulses were established in a Bekesy tracking procedure by 20 subjects with normal hearing acuity and 20 subjects with noise-induced hearing loss. The amount of threshold shift between the long-and short-duration tones, or temporal integration, was examined as a function of subject group and signal frequency (500 Hz and 4000 Hz) both in quiet and in the presence of a white noise masker. Results of this study showed that the two groups could be differentiated by the mean threshold shift between the two tone durations, although overlap between groups was marked. The amount of integration observed for the group with normal hearing was frequency-dependent in both quiet and masked conditions The group of subjects with cochlear impairment exhibited normal integration values at 500 Hz, where no hearing loss was present and significantly smaller than normal values at 4000 Hz, where they demonstrated cochlear pathology. The masker altered the amount of temporal integration of the hearing-impaired group at both signal frequencies, while the normal subjects were unaffected by the presence of the masker at 4000 Hz. Critical ratios which were calculated from the masked thresholds showed larger values for the hearing-impaired group than the normal-hearing group. The present results support the use of brief-tone audiometry in the clinical assessment of cochlear impairment. However, the group overlap in performance which was demonstrated in this study indicates that an individual's performance using this assessment must be interpreted cautiously. That is, a subject cannot be reliably assigned to either group in this experiment when the amount of integration falls within the region of overlap. / Medicine, Faculty of / Obstetrics and Gynaecology, Department of / Graduate

Includes bibliographical references / Poor weight gain, recurrent vomiting and fussiness, chronic cough and recurrent chest infections are among the wide variety of signs that are often attributed to gastro-oesophageal reflux disease (GORD). The difficulty lies in distinguishing between physiological gastro-oesophageal reflux (GOR) and GORD and none of the tests available can, alone, give conclusive evidence for the latter. Clinicians are often at a loss which investigation to request in order to assess for GOR and assist in a diagnosis of GORD. Our hypothesis was that GORD investigations at Red Cross War Memorial Children's Hospital (RCWMCH) are requested without considering the appropriate modality required and without clear indications for suspecting GORD. This was supported by practical experience and a short preliminary review of request forms. In South Africa no specific guidelines exist regarding the diagnosis of GORD and there is a poor understanding of available tests and their role in aiding the diagnosis. Thus many unnecessary tests are requested. To review how appropriate the requests for GORD investigations were we analysed all requests made to the departments of nuclear medicine and radiology at RCWMCH for the purpose of GORD investigation between January and April 2011. This analysis was based on a review of the folders and the data of tests performed on all included patients. The two examination modalities involved were gastro-oesophageal radionuclide scintigrams (commonly known as milk scans) and contrast swallows. The specific points assessed were reasonability of the request, appropriate timing of the investigation, use of the correct modality for the question investigated and lastly evaluation of prior treatment with antacids. We found that most of the studies performed were requested on appropriate grounds and that the timing of the majority of the investigations was reasonable.

Background. There is an increasing trend in the use of long-term oral anticoagulation therapy in children. Monitoring the international normalised ratio (INR) is an integral part in management of these patients, but standard laboratory testing of the INR presents challenges in this age group. Point-of-care INR monitors such as the Mission® PT/INR monitor provide advantages in efficiency and accessibility but have not been evaluated for accuracy in the South African paediatric setting. Objectives. This is a feasibility study with the aim to evaluate the accuracy of the Mission® PT/INR Monitor in comparison to standard laboratory INR measurement, in children presenting for INR testing. Methods. We compared the accuracy of the Mission® PT/INR monitor to the Sysmex Cs2100i laboratory analyser in 37 children aged between 1 year and 17 years, who presented for INR testing. The sample size was limited due to time constraints. 40 paired POC INR and laboratory INR values were obtained. Results. The majority of participants in the study were outpatients (62%) and required INR testing as part of screening in non-cardiac disease (81%) - the majority had chronic liver disease, and a minority were on warfarin therapy (13.5%). The mean INR value on the Mission® PT/INR was 1.49 (standard deviation (SD) 0.73) and was comparable to the Sysmex Cs-2100i (mean INR value 1.39 with SD 0.69). The Bland-Altman difference plot revealed good agreement. Bias between the two methods was 0.13 (SD 0.23). In total, 92.5% of POC INR values were within 0.5 units of laboratory INR value. Conclusion. The Mission® PT/INR point-of-care monitor has a clinically acceptable level of accuracy in children when compared with laboratory INR measurement, but larger studies are needed in the paediatric setting to evaluate patient safety and clinical outcomes. There is a need for implementing POC INR monitoring in outpatient settings but this practice will require robust assessment of infrastructure and quality control before application.

Acinetobacter baumannii (A. baumannii) is now increasingly recognised as an important cause of nosocomial infections in paediatric intensive care unit (PICU) patients, particularly in developing countries, where it contributes significantly to morbidity and mortality. Furthermore, it has been documented that emerging antimicrobial resistance patterns complicate antibiotic choice in these patients. At present, more paediatric data is needed regarding these infections. This is a retrospective case-control study that aims to document the demographic data and relevant clinical details of patients in whom A. baumannii was cultured, either from blood or respiratory specimens (thus including both infections and colonisation), in the PICU at Red Cross War Memorial Children's Hospital (RCWMCH) during 2010. Secondary objectives include comparing these patients with those in whom A. baumannii was not cultured and determining which isolates were causing infection and which were colonisers. In addition; of the isolates regarded as infections, documenting the antimicrobial sensitivities and resistance of the organisms cultured, determining whether infections were late or early onset and determining whether specific bed numbers were consistently involved.

Includes bibliographical references. / Methamphetamine, part of the amphetamine group of drugs, was first discovered in Japan in 1919. It has been clandestinely manufactured in the United States since the 1960s, and is still legally produced there as a nasal inhalant, as treatment for Attention Deficit Disorder and exogenous obesity, as well as off-label treatment for narcolepsy. (1) It is a cheap (about R15- 30 per 'straw'), easily obtainable, odourless, white powder, which has a bitter, taste, but dissolves easily in water or alcohol. Known as, amongst others, ' speed', 'ice', 'crystal', 'chalk', 'glass', 'crank', and locally, 'tik', it can be smoked, snorted, orally ingested, injected intravenously or even administered anally. In South Africa the preferred method consists of placing the powder or crystal in a light bulb (from which the metal threading has been removed) and inhaling the fumes produced while heating the bulb from below with a lighter.(2). The use of methamphetamine has risen sharply globally over the last decade, used by 26 million people worldwide by 2007, more than heroin and cocaine combined, according to the United Nations Office on Drugs and Crime. This has been ascribed to many interlocking reasons: it is cheap, easily obtainable, easy to use without the need for needles or other special 'equipment', and it produces in the user a characteristic 'rush'. This feeling of confidence, power and heightened sexual levels, of feeling 'on top of the world' has made it especially popular amongst teenagers and young adults. (3)

Includes bibliographical references. / Haemophagocytic Lymphohistiocytosis (HLH) is a rare haematological disorder in children. However, this is probably an underestimation due to the difficulty in diagnosing the disease. HLH is characterized clinically by persistent fevers, organomegaly, cytopaenias and typical biochemical derangements viz. hypertriglyceridaemia, hyperferritinaemia and hypofibrinogenaemia. Other associated findings include decreased natural killer cell (NKC) function and raised soluble CD 25. The exact pathophysiology of HLH is not completely understood but involves a trigger (often an infection) which sets off an uncontrolled inflammatory cascade, characterized by an increase in hyperactivated macrophages and T lymphocytes which leads to increased production of cytokines, alongside reduced cellular cytotoxicity as a result of reduced or absent NKC function.

Includes bibliographical references / Introduction Childhood anaemia is a major public health problem, iron deficiency being most common. WHO estimates anaemia to occur in 24.1% of pre-school South African children. Our study describes prevalence and management of anaemia in children aged 6 - 36 months presenting to a children's hospital. Methods In a retrospective cross-sectional study, laboratory data were used to estimate prevalence of anaemia in children aged 6 - 36 month presenting to medical emergency or ambulatory services of Red Cross Children's Hospital in 2012. A random sample of 50% of anaemic children was sampled for detailed review. Results 2661 subjects were included. Anaemia (H b < 10.5) was found in 40.8 % (1088/2661. Children presenting to medical emergency had a higher prevalence of anaemia compared to those presenting to ambulatory services ( 42.7% vs. 34.9 % ; p=0.001 ). Anaemia prevalence increased with decreasing age with RR 1.25 (95% CI 1.10 - 1.43) and RR 1.15 (95% CI 1.02 - 1.31) in children aged 6 - 11 months and 12 - 23 months respectively compared to children aged 24 - 36 months. Microcytosis was found in 51.3% (558/1088) of anaemic children and in 19.3% (n=303/1573) of children without anaemia ; p<0.001 Folders were reviewed i n 502 children with anaemia , 36.1% had mild anaemia (Hb 10 - 10.5g/dl), while moderate ( Hb 8 - 10 g/dl ) and severe ( Hb < 8 g/dl ) anaemia was found in 52.5% and 11.4% respectively. Breastfeeding for longer than six months was associated with higher risk of microcytic anaemia [RR 1.26 (95%CI 1.08 - 1.47)]. Only 12.2 % (31 /254) of children with microcytic anaemia received adequate iron therapy, 50.0 % (127/254) received no iron therapy. Conclusions Prevalence of anaemia in children presenting to hospital is higher than predicted for well children in South Africa. The risk is higher in younger and acutely sick children. Prolonged breasting is associated with increased risk of microcytosis. Most children with suspected iron deficiency anaemia did not receive appropriate treatment.

Includes bibliographical references.

There is sparse literature about HIV transmission in preterm infants. Eighty-two HIV-exposed preterm infants received birth polymerase chain reactions (PCRs). Five (6.1%) were HIV positive with all 5 mothers receiving inadequate antiretrovirals. Of the PCRnegative infants, 9 died and 87% of the survivors received further PCR testing which remained negative. With correct care, intrapartum transmission of HIV can virtually be eliminated.

Changing body position is commonly used in the management of individuals with respiratory diseases and those receiving mechanical ventilation, in order to optimise ventilation and oxygenation. In acute respiratory distress syndrome (ARDS), prone positioning is reported to improve oxygenation by recruiting collapsed dorsal lung regions, although this has not been confirmed in children. Ventilation distribution is well established in adults as being gravity dependent. Clinical practice in the paediatric population has been guided by the notion that all children, irrespective of the presence or absence of disease and age, consistently demonstrate the opposite ventilation distribution pattern to adults and this pattern is said to occur until the second decade of life. Studies in the paediatric population are limited to a few reported from the 1980's, on very heterogeneous populations. With advances in technology, new methods of examining regional ventilation, such as electrical impedance tomography (EIT), have become available. Recent neonatal studies using EIT have reported a dissimilar ventilation distribution to the conventional paediatric pattern. Despite a growing number of studies examining the effects of various interventions on ventilation distribution, very few exist in infants and children older than 6 months of age. Furthermore, differing methodologies and the manner in which ventilation distribution is described and analysed makes pooling the available data in the paediatric population extremely difficult. An understanding of how ventilation is distributed under normal conditions is imperative when examining the effects of different interventions and medical conditions on ventilation distribution. This thesis aimed to describe the effects of body position, head position, age, and respiratory muscle activity on ventilation distribution in children between six months and nine years of age under normal conditions, with respiratory disease, neuromuscular disease, and during mechanical ventilation. Furthermore, the effect on ventilation distribution of prone positioning in children with ARDS was evaluated. Regional ventilation distribution was measured using thoracic EIT and respiratory muscle activity was measured using surface electromyography (sEMG) using standardised methodology. Results of a series of sub-studies indicate that ventilation distribution is more complex and variable than previously thought, with no standard "paediatric pattern" of ventilation. Overall, greater ventilation occurred in the right and dorsal lungs, respectively, in different positons. Head position did not affect regional ventilation in the children studied. Age had a variable effect on ventilation distribution, with healthy children under 12 months of age more likely to follow the paediatric pattern, particularly in side lying positions; however the response was not uniform. The presence of mechanical ventilation, disease state and respiratory muscle activity did not affect ventilation distribution with these children also showing variable patterns of regional ventilation distribution. Data suggests that turning children with ARDS into the prone position does not result in recruitment of the dorsal lung regions, but rather more homogenous ventilation throughout the lungs. Furthermore, results suggest that children with greater ventilation inhomogeneity at baseline are more likely to respond positively (improvement in oxygenation index) to prone positioning. This research provides novel insights into ventilation distribution and respiratory muscle activity in infants and children older than six months of age under a number of different conditions. These results contribute to a better understanding of the factors influencing the distribution of regional ventilation and the mechanisms by which prone positioning in ARDS may improve oxygenation in this population. These findings have potentially important clinical implications, as well as providing baseline data for future clinical studies. Given the variability observed, these studies highlight the potential clinical utility of EIT to monitor different interventions and outcomes. An important strength of the studies presented in this thesis, is that they were performed in a standardised manner, using relatively homogenous individual populations and validated measures of describing ventilation distribution. This methodology could provide a template for future studies in the paediatric population, to allow for comparison between studies.