Background: The high prevalence of Human Immunodeficiency Virus (HIV) has added a new dimension to the management and outcomes of many general surgical conditions in South Africa. However, there is a paucity of data describing the impact of HIV status on surgical outcomes in our setting. Appendicitis is the most common gastrointestinal emergency, and its surgical outcomes in areas of high HIV prevalence are poorly described in the literature. Thus, the aim of this study is to describe and compare the outcomes of appendectomy between HIV-infected (HIV+) and HIV-negative (HIV-) patients. Methods: This is a retrospective study of patients undergoing appendectomy at a large regional hospital over a 12-month period. Demographic data, duration of pre-hospital symptoms, HIV status, surgical approach, operative findings, histopathology reports, hospital stay and complications were recorded. Data for the HIV+ and HIV-patient cohorts were then described, analysed and compared. Statistical analysis was performed using the Chi-Squared or Fisher's exact test for non-continuous variables, and non- parametric ANOVA and Wilcoxon ranked sum test for continuous variables. A P-value less than 0·05 was considered statistically significant. Results: The study group comprised 134 patients; 18 (13.4 %) tested positive for HIV. HIV+ patients were significantly older (mean age of 29.3 vs. 20.3 years, P= 0.002) and had longer duration of pre-hospital symptoms (mean of 3.94 vs. 2.57 days, P= 0.03). Postoperative complications (44.4 % vs. 17.2 %, P= 0.03) and lengthier hospital stays (7.28 days vs. 5.95 days, P= 0.004) were also more frequently seen in the HIV+ patients. There were no differences in appendiceal rupture rates, histopathological findings and mortality. Conclusion: HIV infection is common in patients admitted with clinical features of acute appendicitis in South Africa. Presentation in HIV+ patients was delayed, and surgery was associated with significant postoperative morbidity and longer hospital stay.

Background: There is an urgent clinical need for an alternative vascular graft, especially for smaller artery applications such as in below-knee and coronary artery bypass. Currently available synthetic grafts have unacceptably low patency rates, while autologous saphenous vein grafts are not feasible in one third of patients. Decellularised vascular grafts have been investigated as alternative conduits, but this chemical treatment results in degradation of the extracellular matrix. Chemical stabilization of elastin with penta-galloyl glucose (PGG) combined with collagen stabilisation during covalent heparinisation was previously investigated by our group in a small animal model and shown to be effective and safe. The current study describes their evaluation in a large animal (ovine) model. Methods: Porcine mammary arteries were harvested, decellularised according to an established protocol involving rinsing with sodium hydroxide, alcohol (ETOH), treatment with DNAse/RNAse enzymes, immersion in PGG and subsequently surface modified with covalently bound heparin. Samples of the grafts were also tested for radial and suture retention strength. The prepared grafts were implanted as interposition grafts into the carotid arteries of 6 sheep, using industry standard 6mm expanded polytetrafluoroethylene (ePTFE) on the contralateral side of each animal as control. In-situ patency was determined by ultrasound and angiography at two months, following which the grafts were explanted for macro- and microscopic analysis. Results: In-vitro evaluation: Grafts showed significant levels of bound heparin (14.56 mg/g vs 0.69mg/g in untreated tissue) and demonstrated similar mechanical properties to those of human carotid arteries. Survival: Five out of six sheep survived the full 2-month implant period, while the remaining animal developed sepsis shortly after implantation and was euthanized on day 4. Patency: None of the decellularised grafts were patent at explant, as assessed by ultrasound, angiography and macroscopic examination. Two of the five control (ePTFE) grafts were patent. Microscopic analysis: An inflammatory cell infiltrate with vascularised granulation tissue was found encasing the decellularised xenografts with little or no sign of endothelial cell infiltration. Signs of early occlusion, likely due to technical factors, was noted at the sites of anastomosis. Conclusion: Although demonstrating similar mechanical properties to human carotid arteries, and promising results in the small animal model, the stabilised decellularised vascular grafts failed to achieve endothelialisation or patency in this sheep carotid model. Significant calibre mismatch between the test graft and the native artery is thought to be the primary factor in the failure of these grafts, highlighting the potential difficulty in acquiring grafts of an appropriate size from animal sources.

Background: Breast cancer burden is on the increase in the developing world. Breast-conserving therapy (BCT) is prescribed for early breast cancer. It is the wide local excision of the tumour usually followed by radiation treatment to the breast. It is the mainstay treatment for carefully selected patients with early breast cancer presenting to the Groote Schuur Hospital's Oncology and Endocrine Surgical unit, Cape Town South Africa. There has not been a formal audit to review the outcomes of BCT in the unit. Objectives: The objective of this study is to determine and analyse the excision margins for all the wide local excisions and the re-excision and local recurrence rates during the study period. Methods: This is a histopathological and oncology records review of the patients that have undergone BCT in the unit from the 1st of January 2006 until the 31st of December 2010. The University of Cape Town's Faculty of Health Sciences Human Research Ethics Committee granted approval. Data points accrued included patient age, pathological tumour size and nodal status, histological tumour type, oestrogen receptor status, presence of lymphovascular invasion, volume of specimen excised, margin status, management of involved or close margins, completeness of radiotherapy, ipsilateral breast recurrence rate and total duration of follow up. Results: A total of 192 patients had BCT during the study period. The mean age is 53 years (range 25 to 84 years). A median of 229.5 cm3 volume of specimen was excised (range 4 cm3 to 10530 cm3). Infiltrating ductal carcinoma was the commonest histological type at 79.1%. 42.7% were pT1 tumours, 49.0% pT2 tumours and 2.6 % pT3. The resection margin status are: positive margins rate of 15.1%, 8.3 % close margin (≤ 1 mm), 35.9% 1 – 5 mm, 23.4% 6 – 10 mm and > 10 mm 17.2%. An overall of 26 (13.5%) patients underwent a repeat surgical procedure. 16 (8.3%) had re-excision and 10 (5.2%) had a mastectomy. Residual tumour was present in 50% of the re-excisions and 63.6% of mastectomies. As per category of the resection margins, 68.9% of patients with positive margins had repeat surgery (48.3% re-excision and 20.6% mastectomy). 31.1% of patients with positive margins did not have repeat surgery despite the indication due to advanced age, loss to follow up or residual tumour on the deep chest wall margin. 80.8% patients completed radiotherapy treatment post breast-conserving surgery. At a median follow up of 60 months (range 1 to 108 months), a total of 11 (6.8%) patients had ipsilateral breast local recurrence. Median time to recurrence is 39 months (range 12 to 106 months). Conclusion: Positive and close margin re-excision and local recurrence rates in our unit are acceptable and comparable to other units in South Africa and internationally.

Background: Various inguinal hernia repair techniques exist, without one ‘single best’ option. Hernia society guidelines recommend laparoscopic repair as one of its mainstays, provided surgeons are adequately trained. The current practice for hernia repair in South Africa as well as the surgical resident exposure to laparoscopic repair training is unknown. Aim: To quantify the current practice of inguinal hernia surgery in hospitals affiliated to the University of Cape Town (UCT) and to assess trainee exposure to laparoscopic repair. Methods: All adult patients who underwent inguinal hernia repair during the 12-month study period, at the four UCT affiliated hospitals (Groote Schuur, Mitchell’s Plain, Victoria and New Somerset) were included. Collected data parameters included age, gender, primary or recurrent hernia, uni- or bilaterality, primary surgeon consultant or non-consultant, operative time, and open or laparoscopic technique used. Results: 380 patients were included. Eighty-eight (23.2%) repairs were performed laparoscopically, of which 5 (5.7%) were converted to open. Non-consultants were present at 70/88 (79.5%) cases performed laparoscopically and were primary surgeon at 15 (17%). Laparoscopic repair was performed for 63.6% of bilateral versus 19.3% of unilateral hernias, 39.3% of recurrent hernias and 45% of hernias in females. Conclusion: Inguinal hernias in our setting are predominantly repaired by open surgery. The likelihood of laparoscopic repair varies significantly depending on which hospital the patient is referred to. Non-consultants have limited exposure to performing laparoscopic hernia repairs as the primary surgeon.

Background: Outcomes of patients subjected to damage control laparotomy (DCL) for abdominal gunshot wounds (GSWs) remains relatively unknown. There is limited evidence as to which variables may reliably predict morbidity and mortality. The aim of this study was to evaluate the impact of DCL on long term morbidity and survival, to determine clinical characteristics associated with increased mortality, and to evaluate the indications for DCL in patients with abdominal GSWs. Methods: A retrospective study of patients who underwent a damage control laparotomy for abdominal GSWs at Groote Schuur Hospital (GSH) was conducted. Data was collected on 50 consecutive trauma patients over a 4.5 years period (between August 1st, 2004 and September 30th, 2009). Patients were stratified by, age, preoperative and intraoperative physiological parameters, trauma indices, numbers and locations of abdominal GSWs, extra abdominal involvement, intensive care unit and hospital length of stay, morbidity and mortality. Unadjusted and adjusted estimates of the association between these factors and the odds of survival were computed with univariate and multivariate logistic regression. Results: Most of the patients were male (96%) with a mean age 29.7 year. Most patients had a single abdominal gunshot wound (60%). Liver injuries were the most common injury (58%) followed by small bowel (44%), 20 majors venous (40%), and colonic injury (38%) injuries. The overall mortality was 54%. The mean of length stay in the intensive care unit was 7 days with overall mean hospital length of stay of 13 days. Factor an associated with a decreased odd of survival included Penetrating abdominal trauma index(PATI) >25, pre-operative infusion of less than two litres of crystalloids, intra-operative blood lactate level >8mmol/L, massive transfusion >10 units PRBCs. Conclusion: The overall mortality of patients requiring DCL for abdominal GSWs was 54%. In this limited study, there is significant evidence that after controlling for confounding PATI score of >25 is associated with a decreased odds of survival (OR:0.20, p-value 0.04).

Includes bibliographical references / Introduction: ERAS programmes employed in elective colorectal, vascular, urologic and orthopaedic surgery has provided strong evidence for decreased lengths of hospital stay without increase in postoperative complications. Aim: The aim of this study was to explore the role and benefits, if any, of ERAS / ERP (early recovery programmes) implemented in patients undergoing emergency laparotomy for trauma at a level 1 trauma centre. Methods: Institutional UCT-HREC approved study. A prospective cohort of 38 consecutive patients with isolated penetrating abdominal trauma undergoing emergency laparotomy were included in the study. The ERP included: early feeding, early urinary catheter removal, early mobilisation/physiotherapy, early intravenous line removal and early optimal oral analgesia. This group was compared to a historical control group of 40 consecutive patients undergoing emergency laparotomy for penetrating abdominal trauma, prior to introduction of ERP. Demographics, mechanism of injury and injury severity scores (ISS and PATI) were determined for both groups. The primary end-points were the length of hospital stay and incidence of complications (Clavien-Dindo classification) in the 2 groups. The difference in means was tested using the t-test assuming unequal variances. Statistical significance was defined as p < 0.05. Results: The two groups were comparable with regards to age, gender, mechanism of injury and ISS and PATI scores. The mean time to solid diet, urinary catheter removal and NGT removal was (non ERAS) 3.6 and (ERAS) 2.8 days [p < 0.035], (non ERAS) 3.3 and (ERAS) 1.9 days [p < 0.00003], (non ERAS) 2.1 and (ERAS) 1.2 days [p < 0.0042], respectively. There was no difference in time from admission to time of laparotomy [(non ERAS) 313 vs. (ERAS) 358] min (p < 0.07). There were 11 and 12 complications in the control and study group, respectively. When graded as per the Clavien-Dindo classification there was no significant difference in the 2 groups (p < 0.59). Hospital stay was significantly shorter in the ERAS group: 5.5 (SD 1.8) days vs. 8.4 (SD 4.2) days [p < 0.00021]. Conclusion: This small pilot study shows that ERPs can be successfully implemented with significant shorter hospital stays without any increase in postoperative complications in trauma patients undergoing laparotomy for penetrating abdominal trauma. Furthermore, the study shows that ERP can also be applied to patients undergoing emergency surgery.

Background: Penetrating abdominal trauma (PAT) in South Africa represents a significant burden of disease. The current global trend has seen management shift towards selective conservatism. The purpose of this study is to describe the presentation, management and outcomes of PAT in a level I trauma unit, which routinely practices selective non-operative management (SNOM). Methods: This was a retrospective descriptive audit of prospectively collected data. The Setting was Groote Schuur Hospital Trauma Centre, Cape Town, South Africa over 24 months (1 May 2015 to 30 April 2017). All patients presenting to the centre with PAT during the study period were included. The data captured and analysed included: basic demographics; admission vital signs; blood investigations; number of traumatic insults; penetrating wound positions; radiological investigations and interventions; indication for laparotomy; operative or nonoperative management; laparotomy findings: negative, therapeutic or non-therapeutic; abdominal visceral injuries and associated injuries. The Revised Trauma Score (RTS); Injury Severity Score (ISS); Penetrating Abdominal Trauma Index (PATI); and Kampala Trauma Score (KTS) were then calculated. The descriptive end points included the following: Length of hospital stay (LOS); ICU admission time; relaparotomy; readmission; mortality; and in-hospital complications. Results: During the study period, 805 patients with penetrating abdominal trauma were managed. There were 502 (62.4%) and 303 (37.6%) patients with gunshot and stab wounds, respectively. The majority were young men (762 – 94.7%) with a mean age of 28.3 (95%CI: 27.7-28.9) years. The median trauma scores were as follows: RTS – 7.84 (IQR: 7.00-7.84); ISS: 13 (IQR: 9-22), PATI: 6 (IQR: 1-14); and KTS: 14 (IQR: 14-15). Abdominal penetration was thoracoabdominal in 332 (41.2%), abdominal in 694 (86.5%), and pelvic in 192 (23.9%) patients. Immediate laparotomy was performed in 446 (55.4%) patients for: haemodynamic instability – 42 (5.2%); peritonism – 296 (36.8%); evisceration - 27 (3.4%); unreliable clinical evaluation – 24 (3.0%); and positive radiological findings – 57 (7.1%). There were 406 (50.4%) therapeutic laparotomies; 18 (2.3%) negative laparotomies; and 22 (2.7%) nontherapeutic laparotomies in the immediately operated group. Initial SNOM was performed in 359 (44.5%) patients, of which 208 (68.7%) sustained stab wounds and 151 (30.1%) gunshot wounds. Thirty-five (4.3%) patients failed SNOM and underwent delayed laparotomy. Should a policy of mandatory laparotomy have been implemented in this series, 206 (68.0%) SW and 163 (32.5%) GSW patients would have underwent unnecessary exploration. Overall non-fatal complications were 179 (22.2%) which were then further classified according to the Clavien-Dindo grading system. The median hospital stay was 4.5 (IQR: 3-7) and 7 (IQR: 5-12) days for SW and GSW, respectively. Overall 114 (14.2%) patients required admission to critical care unit for a median stay of 3 (IQR: 2-5) days. Total mortality was 7.2% (n=58). Conclusion: Clinical evaluation (haemodynamic instability, peritonism and evisceration) was remarkably accurate in determining the need for early laparotomy. The unnecessary laparotomy rate of this group was 5.0% (negative: 2.3% and nontherapeutic: 2.7%) overall. Selective nonoperative management was performed in 44.5% of patients with a successful SNOM rate of 90.3%. The overall mortality was 7.2 %.

Purpose: Each year, a cohort in the Northern Cape undergo colonoscopies as part of a surveillance program for individuals who have a C1528T mutation in the hMLH1 gene that puts them at very high risk for the development of colon cancer (Lynch syndrome). A clean colon is essential as it allows a thorough evaluation and surveillance for small polyps or mucosal lesions mostly encountered in the ascending colon. This study evaluated both the subject acceptance and the effectiveness of a 2L PEG electrolyte lavage solution containing ascorbic acid and sodium ascorbate (Moviprep®) as a preparation solution. Methods: The screening program was divided into two stages. Stage 1, 71 subjects were counselled individually on the importance of bowel cleansing and the use of Moviprep® as their bowel cleansing agent. Preparation was either a) 2L the night prior to colonoscopy or b) 1L the night prior to and the second litre on the morning of the colonoscopy. Subjects were encouraged to drink at least 500ml clear fluid in addition to each litre of Moviprep®. Informed consent was obtained for participation in the study. Stage 2, approximately 6 weeks later, each subject completed a questionnaire, evaluating their experience with Moviprep® and also had their screening colonoscopy performed. Colonoscopies were performed at 4 medical facilities in the Northern Cape. All subjects were assessed for bowel cleanliness on arrival at the facility where colonoscopy was to be performed. If any of the subjects were found to be inadequately cleaned, extra oral preparation was given prior to colonoscopy. The Harefield cleansing scale was used to evaluate the quality of colonic cleansing during each colonoscopy. The colon was divided into 6 segments (rectum, sigmoid, descending-, transverse-, ascending colon and cecum). Preparation was scored as A = all colon segments clean; B = at least 1 segment with residual amounts of brown liquid or semisolid stool, which can easily be displaced or removed; C = at least 1 segment with only partially removable stool, preventing complete visualization; D = at least 1 segment which could not be examined due to solid stool). Grades A or B were considered successful cleansing and grades C or D were considered a failed colonic preparation. Results: A total of 46 subjects had colonoscopies performed. 41(89%) of them had successful and 5(11%) failed preparation. Three of those subjects that prepared successfully had previously undergone right hemicolectomies, leaving 38 with intact colons. 22/38 (58%) subjects achieved an A grade for caecal cleansing and 16/38 (42%) a B grading. 2438 (63%) subjects scored an A grade for the ascending colon and 14/38 (37%) a B grade. A total of 64 subjects completed the questionnaires of which 83% (53/64) had used other colon preparations previously. When asked if they would use Moviprep® again in the future, 89% (57/64) said yes and 11% (7/64) said no. 94% of subjects (60/64) would recommend Moviprep® to friend and family. Conclusion: Moviprep® provided adequate colonic cleansing in 89% of subjects. In addition, nearly 90% of subjects were satisfied with the product and would use it again.

Background and aim: An audit of the Fellowship of the College of Surgeons FCS (SA) Final Examination results has not been previously performed. The purpose of this study was to review and determine any predictors of outcome. Methods: The results of the FCS (SA) Final Examinations from October 2005, to and including, October 2014, were retrieved from the College of Medicine of South Africa database. The current format of the examinations consists of: two written essay question papers, an OSCE, two clinical cases and two vivas. These were retrospectively reviewed and analyzed. Predictors of failure or success were determined. Analysis was performed using IPython for scientific computing. Assumptions for the normal distribution of numerical values were made based on the Kolmogorov-Smirnov test and quantile-quantile plots. Normally distributed variables were analyzed by parametric tests. In all other cases nonparametric tests were employed. An alpha value of 0.05 was chosen to indicate statistical significance, using a confidence level of 95%. Results: During the 10-year study period, 472 candidates attempted the examinations. A total of 388 (82,2%) candidates were successful in the written component of the examination and were subsequently invited to participate in the oral/clinical component of the examinations. 9 Overall, 296 (62,7%) of candidates passed and 176 (37,3%) failed. A total of 19 candidates achieved less than 50% for both papers, yet still managed an average of more than 45%. A total of 15 (79%) of these candidates went on to fail the examination. There were 51 candidates who were invited to the oral examinations despite an average of less than 50% in the two papers, and 34 (67%) failed the overall examination. Similarly, 126 candidates were invited having failed one of the two papers of which 81 (64.3%) ultimately failed. A total of 49 candidates failed the OSCE, 82% of these candidates failed overall. There was strong correlation between paper one and paper two (r = 0.56, p-value < 0.01), oral one and oral two (r = 0.41, p-value < 0.01) and case one and case two (r = 0.38, p-value < 0.01). Similar correlations were seen between the averages of the papers versus the orals (r = 0.52, p-value < 0.01), the papers versus the cases (r = 0.5, p-value < 0.01) and the papers versus the OSCE (r = 0.54, p-vale < 0.01). Conclusion: The written papers are the main determinant of invitation to the second part of the examination. Candidates with marginal scores in the written component had an overall failure rate of 67%. Failing one paper and passing the other, resulted in an overall failure rate 64,3%. Failing the OSCE resulted in an overall 82% failure rate. With the high failure rate of candidates with marginal scores and with the inter-examination variability of the papers, it might be prudent to revisit both the process of invitation selection and the decision to continue with the long-form for the written component.

Includes bibliographical references. / Distal rectal washout and presacral drainage appear to have little or no influence on the morbidity and mortality in patients with low-energy trauma to the rectum. The ever-increasing popularity and obvious advantages of minimal access surgery have prompted surgeons to apply its use to a variety of surgical diseases, including trauma-related conditions. This study retrospectively reviews and examines the safety and efficacy of laparoscopy and the formation of a diverting sigmoid loop colostomy through an abdominal wall trephine, in a limited number of carefully selected patients with isolated extra-peritoneal rectal injuries. The patient is thus spared a major laparotomy wound. The value of distal rectal washout and presacral drainage in such injuries is also examined.