Introduction Foley catheter balloon tamponade (FCBT) for bleeding penetrating neck injuries (PNIs) is an effective, readily available and easy-to-use technique. This study aims to audit the technique and highlight current investigative and management strategies. Methods All adult patients (18 years and older) with PNIs requiring FCBT presenting to Groote Schuur Hospital (GSH) within a 22-month study period were included. Data was captured from an approved electronic registry and analysed. Analysed parameters included demographics, major injuries, imaging, management and outcomes. Results Over the study period a total of 628 patients with PNI were managed at GSH, in which 95 patients (15.2%) FCBT was utilised. The majority were men (98%) with an average age was 27.9 years. Most injuries were caused by stab wounds (90.5%). The majority of catheters (81.1%) were inserted prior to arrival at GSH (1.1% prehospital, 45.3% at clinic level and 34.7% at district hospital level). Computerised tomography (CT) angiography was used in 92.6% of patients, while 8 patients (8.4%) required formal angiography. Of these, 2 were purely diagnostic and 6 were performed for definitive endovascular management. A total of 34 arterial injuries (19 major and 15 minor) were identified in 29 patients. Ongoing bleeding was noted in three patients, equating to a 97% success rate at haemorrhage control. Thirteen (13.7%) patients requried open neck surgery. Seventy-two (75.8%) patients without major arterial injury had removal of the catheter at 48-72 hours post injury. Only two of these had bleeding on catheter removal. Fifteen patients required ICU admission. A total of 36 separate morbidities were documented in 28 patients (29.5%). There were 4 deaths (4.2% mortality rate), with only one of these attributable to uncontrolled haemorrhage from the neck wound. Conclusion This large series shows the current use of FCBT for PNI. It highlights ease of use, high rates of success at haemorrhage control (97%) and good outcomes with the technique. Venous injuries and minor arterial injuries can be managed with this technique definitively.

An estimated 36.7 million people were living with HIV in 2015, with 2.1 million newly infected people with HIV in 2015 worldwide. The highest prevalence of HIV in 2015 was in the Eastern and Southern African regions. This highlights the importance for research in this field to further prevent the number of new HIV cases. Mother-to-child transmission of HIV is due to either cell-associated or cell-free virus present in the breast milk of an HIV positive mother. Most often, HIV positive mothers choose to breastfeed their infants due to the nutritional and immunological benefits outweighing that of the risk of HIV transmission. Importantly, approximately 85% of infants do not acquire HIV through daily exposure to breast milk from their HIV positive mothers who are not on ART suggesting that human breast milk has antiviral properties. Previously in our laboratory, MUC1 and MUC4 has been implicated in the inhibition of HIV-1 in an in vitro assay. Furthermore, crude breast milk was tested in this assay showing strong HIV-1 neutralisation. Pasteurisation (80°C for 10 minutes) of both HIV positive and negative breast milk indicated good neutralisation of HIV-1 in our laboratory. Another breast milk protein, tenascin-C (TNC), was recently shown to strongly neutralise HIV-1 in a study performed by another group. Therefore, with this knowledge, this study was employed to firstly compare the antiviral properties of MUC1, MUC4 and TNC. Furthermore, the HIV-1 neutralisation ability of crude breast milk was sought to be investigated along with the investigation of two different pasteurisation methods including 80°C for 10 minutes and 62.5°C for 30 minutes (Holder pasteurisation). Human breast milk was separated into milk fat and skim milk using caesium chloride density gradient ultracentrifugation. MUC1 was purified from the milk fat using gel extraction from a 4-20% sodium dodecyl sulphate polyacrylamide gel. The skim milk was chromatographed on a Sepharose 2B-CL column from which the void volume was collected to purify TNC using gel extraction from a 4-20% sodium dodecyl sulphate polyacrylamide gel. During this purification, a band consisting of MUC1 which adhered to TNC was used to co-purify the MUC1/TNC glycoprotein using gel extraction. MUC1 and TNC were individually purified using gel extraction. MUC4 was not successfully purified and from ELISA data it was concluded that the concentration of MUC4 was below the detectable limit of the ELISA kit. The average concentration of MUC1 was determined to be 307.85 ng/ml, while the concentration of TNC could not be determined due to the majority of absorbance values (450 nm) lying above the upper limit of the curve. The HIV neutralisation of each of the samples was tested in an in vitro HIV-1 assay. This assay utilises a luciferase reporter gene in modified TZM-bl/JC cells using Du422.1 virus derived from clad C of HIV-1. These assays are being performed to assess the antiviral properties of crude and heat treated breast milk and purified MUC1 and TNC separately as well as co-eluted and co-purified MUC1/TNC. The two pasteurisation methods increased the HIV-1 neutralisation when compared to crude breast milk. The HIV-1 neutralisation of these groups were compared with a Kruskal Wallis test and a statistically significant difference was detected among the crude and 62.5°C heat treated breast milk cohorts (Mann-Whitney U, p-value = 0.0021). Furthermore, a statistically significant difference in the HIV-1 neutralization was detected among the 80°C and 62.5°C heat treated breast milk cohorts (Mann-Whitney, p-value = 0.0033). From the data, and the range of IC₅₀ values (50% inhibitory concentration), the HIV-1 potency was deemed the strongest in the 62.5°C heat treated breast milk. This pasteurisation method could potentially be promoted in lower resource settings to decrease mother-to-child transmission of HIV-1. Purified MUC1 and TNC, as well as co-eluted and co-purified MUC1/TNC, was tested in the same neutralization assay in order to compare the HIV-1 potency of these glycoproteins. The difference in HIV-1 neutralisation was not statistically significant among all three groups (Kruskal Wallis, p-value = 0.13). From the range of IC₅₀ values, it was suggested that TNC has a stronger HIV-1 potency when compared to MUC1. Overall, the co-eluted and co-purified MUC1/TNC showed a lower HIV-1 potency when compared to the single, purified glycoprotein. MUC1 and TNC could be purified and cloned to aid in the protection against contracting HIV-1, especially in mother-to-child transmission of HIV-1. Histochemistry and immunohistochemistry was performed on breast tissue sections to investigate the morphology of the cells and the presence of MUC1, MUC4 and TNC respectively. The lactating breast tissue was confirmed to have wide, dilated lumina and vacuolated cytoplasm with neutral and sialomucins. The presence of MUC1, MUC4 (β subunit) and TNC were confirmed in the lactating breast tissue using immunohistochemistry.

Treatment of renal failure while awaiting transplant requires vascular access, which comes with both complications and failure rates. In order to improve this, information about the AVF or AVG itself, as well as the haemodynamics is required. This data will then be used for computer modelling techniques and computational flow dynamics. Previously, the required imaging was provided by contrasted MRI, contraindicated in renal failure. Haemodynamic data was prvided by, amongst other things, duplex Doppler. New MRI software that provides imaging data as well as haemodynamic information without using contrast could be used to provide new high-quality data for modelling. Methods: This was a prospective pilot study. Six control cases (with no history of vascular illness or surgery of any kind to the right upper arm), as well as three grafts and five fistulae underwent phase contrast MR angiography of the right upper arm with a Siemens Magnetom Symphony 1.5T MRI Scanner. Images were then processed using Supertool in Matlab, and flow velocities at predetermined points on the brachial artery and cephalic vein, graft and fistula were calculated. Results: Velocities ranged from 5.8 cm/sec in a volunteer's brachial artery to 85.5 cm/sec in an arteriovenous fistula patient's brachial artery. Flow volumes in the cephalic vein or access varied from 6.9 ml/min. in a volunteer and up to 4398.1 ml/min. in an arteriovenous fistula. Graphical representations show marked haemodynamic changes throughout the imaged vessels. Conclusion: This technique provides good imaging and quantitative data about small vessel haemodynamics.

Aim: To study the effects of muscle damage and fatigue on neuromuscular preactivation and performance during submaximal and maximal running. Setting: University of Cape Town, Sports Science Institute of South Africa. Methods: 12 male distance runners (19 - 39 years of age) with a minimum weekly training distance of 40 kilometers per week were randomly assigned to either control (n = 6) or experimental (n = 6) groups. Subjects’ visited the laboratory over an 11 day period during which testing included a submaximal and maximal run (5 km time trial) on the first and last day of testing. Neuromuscular preactivation, rating of perceived exertion, heart rate and performance times were recorded during the performance trials. The intervention between performance trials included two 40 minute bouts on a treadmill at 70 % peak treadmill running speed at –10 º elevation (experimental) or horizontal (control). Results: Running performance in the 5 km time trial (5K) improved in the experimental group alone by an average of 40 seconds over 5 km (P < 0.04) in the presence of muscle damage and without altered neuromuscular preactivation. There was no evidence any interaction between altered neuromuscular activity with regard to fatigue and muscle damage during submaximal and maximal running. Evidence of muscle damage in the experimental group was supported by a significant group versus time interaction effect in subjective pain score for daily living and increased plasma creatinine kinase levels in the experimental group (P<0.03). A significant decrease in rating of perceived exertion (RPE) was observed in both groups during both the submaximal (P<0.04) and 5 km time trial (P<0.03) post intervention. There was an interaction effect for group versus pre-post 5K (P<0.06), with the post 5K RPE in the experimental group showing an average decreased RPE score of 2.6 for each kilometer and the control group an average decrease in RPE score of only 0.03. Conclusion: The research design of this study was appropriate to study the interaction between fatigue and muscle damage during submaximal and maximal running. This study suggests that there is no neuromuscular compensation after muscle damage and that EMG is regulated similarly for both fatigue and muscle damage during submaximal and maximal running. Improvement in running performance and decreased rating of perceived exertion after muscle damage is due to some unknown variable.

This work was based on the hypothesis that optimization of growth factor delivery rate and duration, combined with a biomaterial scaffold, could lead to an improved strategy for therapeutic neovascularization. To test this hypothesis, a novel in vivo model system that allows for characterization of stability and mural cell investment of newly created vessels was designed.

Introduction: Management of Gustilo-Anderson grade 3b tibia fractures are challenging due to the high rate of complications which includes infection, nonunion and possible amputation. Due to limited soft tissue coverage of the tibia antero-medially, open fractures remain a treatment challenge. Despite many advances, the ideal time delay to definitive soft tissue cover remains controversial. Aim: We aimed to investigate the management strategy and the outcome of soft tissue reconstruction of Gustilo-Anderson grade 3b tibia fractures at a tertiary hospital in Cape Town, South Africa. Methods: A retrospective study was conducted on 22 patients who underwent soft tissue reconstruction for grade 3b tibia fractures from January 2014 to July 2017. Patient demographics, comorbidities, injury characteristics and management practices such as time to debridement, relook time, Negative Pressure Wound Therapy (NPWT), soft tissue coverage and complications were recorded. Results: Most patients were males (n=18; 81.8%) with an average age of 39.3 years. Pedestrian vehicle accidents accounted for 45.4%(n=10), motor-vehicle accidents (n=6; 27.3%) and gunshot wounds (n=2; 9.1%). The commonest site of injury was the middle third of the tibia (n=13; 59.1%), distal third (n=7; 31.8%) and proximal third (n=2; 9.1%). Most patients (n=18; 81.8%) were debrided within 24 hours. The mean times for NPWT prior to cover was 12.5 days and for soft tissue cover 13.7 days (range 2-35), respectively. Fasciocutaneous flaps (n=11; 50%) were predominantly used as cover, then pedicled muscle flaps (n=8; 36.4%), free flaps (n=2) and skin graft(n=1). Most patients (n=13; 59.1%) received satisfactory outcomes. Seven (31.8%) required soft tissue revisions. Three patients (13.6%) suffered complications namely, complete flap loss resulting in amputation, partial skin graft loss and soft tissue infection, respectively. Patients who underwent debridement after 24 hours reported the least complications and there appeared to be better outcomes in the relooks beyond 48 hours. Conclusion: Despite achieving outcomes which concur with other published studies, the BOAST 4 guidelines were not fully reflected in our management strategy. We will require larger numbers in future studies to formulate a standardized management protocol going forward.

Includes bibliographical references (leaves 48-59). / This dissertation represents the culmination of work and research on the management of pancreatic fluid collections (PFCs) undertaken at Groote Schuur Hospital. The aim of the study was to review the management outcome, via different treatment modalities, of pancreatic fluid collections according to a locally derived classification. Patients with symptomatic and unresolved PFCs treated during a nine year period were reviewed, assessing clinical features and anatomical location determined using CT scanning. The classification system applied distinguishes PFCs according to the associated pancreatitis, being acute or chronic. This was based on definitions established at the Atlanta Symposium in 1992. The chronic pancreatitis related pseudocysts were further selected anatomically into intra or extra pancreatic.

Introduction: The emergency centre forms the first formal interaction between the prehospital and inhospital phases of the patient care continuum. There are several variables that have the potential to affect handover efficacy. Poor handover has been associated with an increase in sentient events and a risk to patient safety. This thesis aimed to investigate the perceptions of the practice of patient handover between prehospital emergency care providers and the emergency centre. This information was used to generate a model that addresses identified aspects of the emergency centre handover, namely content, process, and communication. Methods: The methodology followed a sequential, explanatory, mixed-methods design. Data were collected from prehospital emergency care personnel (PECP) and emergency centre personnel (ECP) in the Johannesburg area of South Africa. Study One and Study Two formed the quantitative and qualitative data collection phases respectively. Study One formed the quantitative component of the study using a survey that utilised a crosssectional, convenience design. Questionnaires were compiled de novo using data sourced from a search of major databases and were pilot tested prior to distribution. Questionnaires contained a mix of Likert-type, forced binary and open-ended questions. Questionnaires were distributed using a purposive, convenience strategy where potential participants were approached at their place of work. Data were analysed descriptively and reported on. The responses to the open-ended questions were used to compile the interview schedule used in Study Two. Study Two formed the qualitative approach of the thesis and used a qualitative descriptive design. Questions for Study Two were compiled using the results of the coding, analysis and interpretation of the responses to the open-ended questions from the paper-based questionnaire. Data were collected from 15 PECP and 15 ECP using face-to-face, semi structured interviews. Participants were approached using a purposive strategy and, where consent was obtained, were interviewed in a location that was conveniently available and afforded an adequate amount of privacy. Interviews were transcribed and then analysed using Computer-Assisted Qualitative Data Analysis Software Atlas.ti. Data were read and reread, coded and analysed to identify categories and themes that were then reported. A code-recode strategy ensured trustworthiness. Results: Handover content variables were ranked according to the perceived level of importance by prehospital emergency care and ECP. Physiological variables dominated the ten most important variables for both PECP and ECP. Handover quality was perceived by both PECP and ECP as requiring improvement. Less than half of both PECP and ECP had been exposed to formal handover training. Mnemonic knowledge was generally poor, and the most familiar mnemonic used by PECP was unfamiliar to the ECP. The same was true for the mnemonic most familiar to the ECP. Process factors that had the potential to affect the efficacy of emergency centre handover included repetition of information and having to hand over multiple times. The busyness of the emergency centre and the noisy environment associated with it were linked to compromised patient privacy and a distractive environment in which to hand over. Understaffing and overworked staff were identified as barriers to an effective handover process and contributors to some of the identified issues related to poor emergency centre handover. There were several communication factors identified by both PECP and ECP that negatively affected handover efficacy. Verbal, non-verbal and paraverbal cues were identified as having he potential to act as facilitators of or barriers to effective emergency centre handover. Listening skills were identified as a barrier to effective handover by both PECP and ECP. Interprofessional communication and relationships were identified as important for effective emergency centre handover. Using the data, a novel model was developed using an iterative process. The model proposes solutions to some of the content, process and communication problems that were identified in this thesis. The model sees handover as comprising of five phases of information flow and unlike many previous models, recognises the bidirectional nature of communication within the handover process. Conclusion: Emergency centre handover between PECP and ECP needs improvement. The novel model proposed in this thesis divides handover into phases, each of which has identified factors that have the potential to act as facilitators of or barriers to effective handover. The model has potential to be implemented in emergency centre handover environments and may also have relevance in other patient handover environments.

Background: The Trauma Unit at Groote Schuur Hospital is a mature, tertiary, high-volume trauma referral centre. The number of patients being treated in the unit at any given time can vary greatly. There is evidence to suggest that these fluctuations may be related to external and environmental factors, such as time of the day, week and month, local weather and significant home-team sport matches. Objective: The objective of this audit was to determine the relationship between volumes of patients in the unit, and environmental factors. Specifically, we aimed to determine whether the numbers of patients presenting after motor vehicle collisions and interpersonal violence was related in any way to temporal factors (time of day, week and month), weather variables (temperature and precipitation), and whether or not major or home-team soccer matches were being played. Methods: Trauma Unit admission records were examined retrospectively, and the numbers of patients presenting to the unit per shift for a total of 17 months was recorded. Patients were grouped according to their presenting complaints. Weather data, Premier Soccer League and Bafana Bafana match locations and results, and information regarding public holidays and long weekends were obtained for the relevant shifts. Average daily attendances for interpersonal violence (IPV)-related injuries and motor vehicle collisions (MVCs) were compared across the various external factors described. Poisson regression models were fitted using Stata 13 (StataCorp. 2013. Stata Statistical Software: Release 13. College Station, TX: StataCorp LP). and used to express the relative incidence of attendances. These results were expressed using incidence rate ratios (IRRs). Results: In total, 16 706 attendances were recorded over 1 074 shifts. Of these, 7 350 (44%) attendances were due to injuries sustained as a result of interpersonal violence (IPV), and 3 188 (19%) were due to MVCs. Predictors of increased attendances due to MVC-related injuries were week day shifts, and night shifts on long weekends, and on weekends that fell on the last day of the month. Weekend nights shifts were busier than week night shifts from this perspective. Public holiday shifts were shown to have less MVC-related attendances than an ordinary week day. The presence of precipitation was also shown to increase the number of MVC-related attendances. IPV-related attendances were always increased on night shifts compared to day shifts, except on public holidays, long weekends, and on weekends that fell on the last day of the month. All weekend shifts were busier than their corresponding week day shifts from an IPV-related perspective, and this effect was enhanced on weekends that fell on the last day of the month. Long weekends showed very similar trends to ordinary weekends, and public holidays showed similar trends to ordinary week days. Increasing temperatures are associated with increased attendances due to IPV. Soccer matches and their outcomes have no significant effect on attendances due to IPV. Conclusions: Temporal and weather factors can be used to predict which trauma unit shifts will be busiest.

The aim of this study is to identify factors associated with limb loss in patients with popliteal artery injuries. Retrospective chart review of a prospectively collected data base of all patients with popliteal artery injuries presenting to the Groote Schuur Hospital Trauma Centre from 01 January 1999 to 31 December 2008. Demographic data, mechanism of injury, hemodynamic status, limb status (viable, non-viable or ischaemic), special investigations, associated injuries, ischaemic time, surgical treatment and amputation rate were analysed.