<?xml version="1.0" encoding="ISO-8859-1"?><article xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink" xmlns:xsi="http://www.w3.org/2001/XMLSchema-instance">
<front>
<journal-meta>
<journal-id>0256-9574</journal-id>
<journal-title><![CDATA[SAMJ: South African Medical Journal]]></journal-title>
<abbrev-journal-title><![CDATA[SAMJ, S. Afr. med. j.]]></abbrev-journal-title>
<issn>0256-9574</issn>
<publisher>
<publisher-name><![CDATA[Health and Medical Publishing Group]]></publisher-name>
</publisher>
</journal-meta>
<article-meta>
<article-id>S0256-95742012000800030</article-id>
<title-group>
<article-title xml:lang=""><![CDATA[]]></article-title>
</title-group>
<aff id="A">
<institution><![CDATA[,  ]]></institution>
<addr-line><![CDATA[ ]]></addr-line>
</aff>
<pub-date pub-type="pub">
<day>00</day>
<month>08</month>
<year>2012</year>
</pub-date>
<pub-date pub-type="epub">
<day>00</day>
<month>08</month>
<year>2012</year>
</pub-date>
<volume>102</volume>
<numero>8</numero>
<fpage>703</fpage>
<lpage>714</lpage>
<copyright-statement/>
<copyright-year/>
<self-uri xlink:href="http://www.scielo.org.za/scielo.php?script=sci_arttext&amp;pid=S0256-95742012000800030&amp;lng=en&amp;nrm=iso&amp;tlng=en"></self-uri><self-uri xlink:href="http://www.scielo.org.za/scielo.php?script=sci_abstract&amp;pid=S0256-95742012000800030&amp;lng=en&amp;nrm=iso&amp;tlng=en"></self-uri><self-uri xlink:href="http://www.scielo.org.za/scielo.php?script=sci_pdf&amp;pid=S0256-95742012000800030&amp;lng=en&amp;nrm=iso&amp;tlng=en"></self-uri></article-meta>
</front><body><![CDATA[ <p align="right"><font face="Verdana, Arial, Helvetica, sans-serif" size="2"><b>ABSTRACTS</b></font></p>     <p>&nbsp;</p>     <p><font face="Verdana, Arial, Helvetica, sans-serif" size="4"><b>Audit of IV    iron sucrose (Venofer<sup>TM</sup>) usage in Groote Schuur Hospital's Gastrointestinal    Unit</b></font></p>     <p>&nbsp;</p>     <p>&nbsp;</p>     <p><font face="Verdana, Arial, Helvetica, sans-serif" size="2"><b>Y Awuku; C A    Viljoen; M Setshedi; G Watermeyer; S R Thomson</b></font></p>     <p><font face="Verdana, Arial, Helvetica, sans-serif" size="2">Gastrointestal    Unit, Division of Gastroenterology, Groote Schuur Hospital and University of    Cape Town</font></p>     <p>&nbsp;</p>     <p>&nbsp;</p> <hr noshade size="1">     <p><font face="Verdana, Arial, Helvetica, sans-serif" size="2"><B>INTRODUCTION:</b>    Intravenous (IV) iron sucrose (Venofer<sup>TM</sup>) therapy has merit in the    management of iron deficiency anaemia in 2 categories of gastrointestinal (GI)    patients: those with inflammatory bowel disease (IBD) and those with obscure    or recurrent GI bleeds. We wished to audit the effectiveness of this practice    in our GI patients.    ]]></body>
<body><![CDATA[<br>   <B>METHODS AND MATERIALS.</b> This was a retrospective descriptive study of    67 iron-deficiency anaemia patients seen at the GI clinic who received IV iron    therapy as day cases. Over a 6-month period results were obtained of their full    blood counts and iron studies (n=49) at baseline and post IV iron (Venofer<sup>TM</sup>).    Those that received a blood transfusion were excluded.    <br>   <B>RESULTS:</b> Participants were predominantly female (69.4%) and 32.6% had    IBD. Mean age was 56.2&plusmn;19.6 years and mean quantity of IV iron given    was 885 mg (&plusmn;355 mg) (approximately 3 infusions). No adverse effects    were documented. Mean baseline haemoglobin (Hb) was 10.0 g/dl, with no differences    between the IBD and non-IBD groups. However, post-IV infusion, the mean Hb increased    by 1.2 g/dl. Mean baseline iron was lower in IBD patients (5.89 mmol/l) than    in non-IBD patients (7.8 mmol/l). Post-infusion mean iron for the entire cohort    was higher by 6.3 units (11.5 in IBD and 14.4 in non-IBD patients).    <br>   <B>CONCLUSION:</b> IV iron therapy is safe and shows good response in Hb and    iron levels in IBD and non-IBD patients, but has a more pronounced effect on    iron levels in the latter. These data provide a benchmark for comparison with    other therapies in these categories of our GI patient population.</font></p>     <p><font face="Verdana, Arial, Helvetica, sans-serif" size="2"><b>Key words:</b>    Anaemia; iron deficiency anaemia; inflammatory bowel disease (IBD); intravenous    iron.</font></p> <hr noshade size="1">     <p>&nbsp;</p>     <p><font face="Verdana, Arial, Helvetica, sans-serif" size="3"><b>Successful use    of endoscopic SEMS to palliate malignant gastric outlet obstruction</b></font></p>     <p>&nbsp;</p>     <p>&nbsp;</p>     <p><font face="Verdana, Arial, Helvetica, sans-serif" size="2"><b>S Burmeister;    J E J Krige; J M Shaw; S R Thomson; G Chinnery; M Bernon; P C Bornman</b></font></p>     <p><font face="Verdana, Arial, Helvetica, sans-serif" size="2">Surgical Gastroenterology    Unit, Groote Schuur Hospital, and Department of Surgery, Faculty of Health Sciences,    University of Cape Town</font></p>     ]]></body>
<body><![CDATA[<p>&nbsp;</p>     <p>&nbsp;</p> <hr noshade size="1">     <p><font face="Verdana, Arial, Helvetica, sans-serif" size="2"><B>BACKGROUND:</b>    Enteral stenting has evolved over the past decade as an alternative to surgical    bypass in the palliation of malignant gastric outlet obstruction. In particular,    it offers a less invasive option in the management of patients who are often    at significant risk for peri-procedural morbidity and mortality. This single-centreprospective    study evaluated the success of enteral stenting for the relief of advanced malignant    gastroduodenal obstruction.    <br>   <B>METHODS:</b> Between January 2006 and April 2012, 127 patients (74 men, 53    women; mean age 60.1 years) with clinical, radiological and endoscopic gastric    outlet obstruction as a result of irresectable malignancy due to local extent,    regional or distant metastatic disease or patient choice, underwent endoscopic    placement of a self-expanding metal stent (SEMS) to relieve symptoms.    <br>   <B>RESULTS:</b> The technical success rate of endoscopic SEMS placement was    96%, allowing patients to be discharged at a mean of 4 days (range 1 - 23) post    stent placement. Fourteen patients (11%) required placement of a second stent    due to either a long distal stricture or early technical failure <i>(n</i>=8),    or as a result of delayed re-obstruction from tumour ingrowth <i>(n</i>=6).    Complications included bleeding <i>(n</i>=2), perforation <i>(n</i>=1), early    blockage from a food bolus <i>(n</i>=1), stent migration <i>(n</i>=1) and re-obstruction    from stent shortening/ deformation <i>(n</i>=1). One patient died following    failed stent placement with tumour perforation.    <br>   <B>CONCLUSION:</b> Endoscopic SEMS placement can be safely used to relieve non-resectable    malignant gastroduodenal obstruction, minimising hospital stay and peri-procedural    morbidity. Technical factors influence its early success rate while secondary    stenting is effective in managing subsequent tumour ingrowth.</font></p> <hr noshade size="1">     <p>&nbsp;</p>     <p><font face="Verdana, Arial, Helvetica, sans-serif" size="3"><b>Temporary placement    of covered SEMS for benign gastroduodenal pathology</b></font></p>     <p>&nbsp;</p>     <p>&nbsp;</p>     ]]></body>
<body><![CDATA[<p><font face="Verdana, Arial, Helvetica, sans-serif" size="2"> <b>G E Chinnery;    J E J Krige; M M Bernon; S Al-Harethi; M E C van Wyk; P C Bornman</b></font></p>     <p><font face="Verdana, Arial, Helvetica, sans-serif" size="2">Groote Schuur Hospital,    Cape Town</font></p>     <p>&nbsp;</p>     <p>&nbsp;</p> <hr noshade size="1">     <p><font face="Verdana, Arial, Helvetica, sans-serif" size="2"><b>BACKGROUND:</b>    Palliation of malignant gastroduodenal obstruction by self-expanding metal stent    (SEMS) placement is the accepted treatment of choice. While indications for    SEMS placement have expanded, the application for benign gastroduodenal pathology    remains controversial. We evaluated SEMS efficacy in patients with benign gastric    outlet obstruction (GOO) and duodenal fistulation.    <br>   <B>METHODS:</b> We analysed a prospective database documenting patients with    temporary placement of retrievable covered SEMS for benign gastroduodenal pathology.    Technical and clinical successes, as well as short- and long-term complications    were evaluated.    <br>   <B>RESULTS:</b> Eleven patients (6 men, 5 women) of median age 56.6 years (range    35 - 76), had endoscopic placement of a covered retrievable duodenal SEMS for    GOO due to peptic ulcer disease <i>(n</i>=6), extrinsic compression <i>(n</i>=3),    pyloric stricture following repair of a perforated peptic ulcer (n=1) and high-output    traumatic duodenal fistula (n=1). Indications included significant patient comorbidity,    recent surgery and patient preference. Ten stents (91%) were placed successfully;    1 misplaced stent was replaced within 48 hours. Hemoclips were applied to prevent    migration. Early proximal stent migration in 1 patient required replacement    after 4 days. Planned retrieval of stents was at 6 weeks. Late migration occurred    in 2 patients. One stent was not removable. Four patients (36%) required surgical    intervention for recurrent GOO within 4 - 6 weeks. Seven patients (63%) remain    symptom-free at 2 - 18 months after stent removal.    <br>   <B>CONCLUSION:</b> While surgery remains the standard of care for benign GOO,    temporary duodenal stenting may avoid surgery or improve nutritional status    and comorbidity before surgical intervention in selected patients.</font></p> <hr noshade size="1">     <p>&nbsp;</p>     <p><font face="Verdana, Arial, Helvetica, sans-serif" size="3"><b>Are current    ERCP training requirements adequate to ensure competency in interventional endoscopy?</b></font></p>     ]]></body>
<body><![CDATA[<p>&nbsp;</p>     <p>&nbsp;</p>     <p><font face="Verdana, Arial, Helvetica, sans-serif" size="2"><b>G E Chinnery;    J E J Krige; P C Bornman; M M Bernon; H P Essel; S Al-Harethi; E Deetlefs; N    Rajabally; S Burmeister; S R Thomson</b></font></p>     <p><font face="Verdana, Arial, Helvetica, sans-serif" size="2">Groote Schuur Hospital</font></p>     <p>&nbsp;</p>     <p>&nbsp;</p> <hr noshade size="1">     <p><font face="Verdana, Arial, Helvetica, sans-serif" size="2"><b>BACKGROUND:</b>    Endoscopic retrograde cholangiopancreatography (ERCP) is technically the most    difficult endoscopic procedure to master, with higher failure and complication    rates than other endoscopic interventions. Recommendations for the number of    procedures to be performed before achieving competence range from 35 to 200.    Traditionally, an 80% bile duct cannulation rate on completion of ERCP training    is the desired aim. In an era where the majority of ERCPs performed are therapeutic,    is this still an adequate form of assessment?    <br>   <B>METHODS:</b> An ERCP database was retrospectively reviewed to identify all    procedures performed by trainees. Eight surgical and medical gastroenterology    trainees completed their 2-year subspecialty training between 2006 and 2012    at Groote Schuur Hospital. No trainee had prior ERCP experience. Each trainee's    initial 120 supervised ERCPs were retrospectively reviewed, with division into    sequential chronological groups of 30. Comparisons between assisted and unassisted    completed ERCPs within the trainees' subgroups were made. Completing an ERCP    entailed both deep biliary cannulation and completion of the required therapeutic    intervention of sphincterotomy, stenting or stone extraction unaided.    <br>   <B>RESULTS:</b> Six trainees showed a definitive trend towards improvement over    the 4 quarters, but only 3 had an ERCP completion rate &gt;75% by the final    quarter. The average trainee unaided completion rate after 120 ERCPs was 67%.    <br>   <b>CONCLUSION:</b> Due to the complexity of interventional endoscopy, accreditation    in performing ERCPs should include steady progress and successful completion    rates, rather than an evaluation of biliary cannulation only. A prospective    study is needed to redefine the criteria of competency in interventional ERCPs.</font></p> <hr noshade size="1">     ]]></body>
<body><![CDATA[<p>&nbsp;</p>     <p><font face="Verdana, Arial, Helvetica, sans-serif" size="3"><b>Aetiological    discrimination of the causes of biliary obstruction in the presence of gallbladder    stones</b></font></p>     <p>&nbsp;</p>     <p>&nbsp;</p>     <p><font face="Verdana, Arial, Helvetica, sans-serif" size="2"> <b>L Ferndale;    S R Thomson</b></font></p>     <p>&nbsp;</p>     <p>&nbsp;</p> <hr noshade size="1">     <p><font face="Verdana, Arial, Helvetica, sans-serif" size="2"><B>INTRODUCTION:</b>    Malignant bile duct obstruction and choledocholithiasis are the most common    aetiologies of biliary obstruction. Trans-abdominal ultrasound can identify    gallstones with a high sensitivity but it is highly operator-dependent and does    not accurately detect bile duct stones. We assessed the utility of other clinical    and biochemical markers in predicting the actual cause of the biliary obstruction    in patients with gallbladder stones on trans-abdominal ultrasound.    <br>   <b>METHODS:</b> We performed a retrospective analysis of prospectively collected    (ERCP) data. Data on patients undergoing ERCP for suspected choledocholithiasis    over a 2-year period were reviewed. Data collected included demographics, presence    of comorbidities and laboratory investigations such as liver and renal function    and haematology tests.    <br>   <b>RESULTS:</b> A total of 162 patients were identified; 40 were excluded because    of failed cannulation or a normal cholangiogram. Of the 122 patients studied,    82 had confirmed bile duct stones (group 1) and 40 had alternative diagnoses    (group 2). Mean patient age in group 1 was 50.7 years v. 58.2 years in group    2 (p&lt;0.05). The female to male ratio was higher in group 1 than in group    2. Higher serum bilirubin and liver enzymes and lower albumin levels also significantly    predicted the finding of alternative diagnoses in these patients.    ]]></body>
<body><![CDATA[<br>   <b>CONCLUSION:</b> Male gender, higher serum bilirubin and liver enzymes, and    lower albumin levels are predictors of an alternative diagnosis to choledocholithiasis    in patients with gallbladder stones. Accurate identification of the one-third    of patients with no ductal stones would allow appropriate further imaging prior    to an appraisal of ERCP need.</font></p> <hr noshade size="1">     <p>&nbsp;</p>     <p><font face="Verdana, Arial, Helvetica, sans-serif" size="3"><b>SEMS insertion:    Fluoroscopy v. pure endoscopy - a cost comparison</b></font></p>     <p>&nbsp;</p>     <p>&nbsp;</p>     <p><font face="Verdana, Arial, Helvetica, sans-serif" size="2"><b>M Govender;    D Clarke</b></font></p>     <p><font face="Verdana, Arial, Helvetica, sans-serif" size="2">Greys Hospital    and Department of Health, KwaZulu-Natal</font></p>     <p>&nbsp;</p>     <p>&nbsp;</p> <hr noshade size="1">     <p><font face="Verdana, Arial, Helvetica, sans-serif" size="2"><b>BACKGROUND:</b>    Self-expanding metal stenting (SEMS) is used for palliation of dysphagia in    oesophageal cancer. In most centres it is performed under fluoroscopic guidance.    Limited access to fluoroscopy at our institution led us to develop a pure endoscopic    technique. We undertook a cost comparison of both approaches.    ]]></body>
<body><![CDATA[<br>   <B>PATIENTS AND METHODS:</b> We performed a prospective analysis of patients    stented at Greys Hospital utilising endoscopy alone compared with patients stented    at IALCH in Durban where fluoroscopy is routinely used. We observed and documented    20 procedures at each centre. Average costing was estimated using protocols    from the Revenue department and private practitioners. Individual cost drivers    include use of the fluoroscopy suite, use of contrast, screening time and overall    procedure time.    <br>   <b>RESULTS:</b> The average additional cost of utilising fluoroscopic guidance    for SEMS insertion is approximately R2 065.00 per patient. The average total    procedure time of the pure endoscopic technique was 5 min (range 4 - 11 min)    v. 17.5 min (range 5 - 24 min) for routine fluoroscopy. The use of screening    also exposed the operator to an estimated 1.05 mGy of radiation per procedure.    <br>   <b>CONCLUSION:</b> In a resource-limited setting with a high burden of inoperable    oesophageal cancer, time and cost are 2 significant variables. This simple analysis    confirms that the pure endoscopic technique of SEMS insertion is more cost-effective    and time-efficient than routine fluoroscopy.</font></p> <hr noshade size="1">     <p>&nbsp;</p>     <p><font face="Verdana, Arial, Helvetica, sans-serif" size="3"><b>Outcome in decompensated    alcoholic cirrhosis patients with acute variceal bleeding</b></font></p>     <p>&nbsp;</p>     <p>&nbsp;</p>     <p><font face="Verdana, Arial, Helvetica, sans-serif" size="2"><b>M I Hampton;    J E J Krige; U K Kotze; S Burmeister; M M Bernon; G E Chinnery; S Alherthi;    S Hofmeyr</b></font></p>     <p><font face="Verdana, Arial, Helvetica, sans-serif" size="2">Department of Health,    Western Cape</font></p>     <p>&nbsp;</p>     ]]></body>
<body><![CDATA[<p>&nbsp;</p> <hr noshade size="1">     <p><font face="Verdana, Arial, Helvetica, sans-serif" size="2"><B>OBJECTIVES:</b>    Variceal bleeding (VB) is the leading cause of death in cirrhotic patients with    portal hypertension and oesophageal varices. This prospective single-centre    study evaluated the efficacy of emergency endoscopic intervention in the control    of acute VB and the prevention of rebleeding and death during the index hospital    admission in a large cohort of consecutively treated alcoholic cirrhotic patients    after the first variceal bleed.    <br>   <b>METHODS:</b> From January 1984 to August 2011, 448 alcoholic cirrhotic patients    (349 men, 99 women; median age 50 years) with VB underwent 805 endoscopic treatments    (556 emergency, 249 elective) during their index hospital admission. Injection    sclerotherapy was used to control bleeding until 1990 and subsequently variceal    banding was used. Child-Pugh (C-P) class and score, endoscopic control of initial    bleeding, variceal rebleeding and survival after the first hospital admission    were recorded.    <br>   <b>RESULTS:</b> Endoscopic intervention alone controlled VB in 394 patients    (87.9%), while 54 patients also required balloon tamponade. Fifteen patients    rebled within 24 hours and 61 rebled after 24 hours (N=76; 17%). Ninety-three    patients (20.8%) died in hospital. No C-P class A patients died, while 16 (8.9%)    C-P class B and 77 (35.8%) C-P class C patients died. Mortality increased exponentially    as the C-P score increased, reaching 80% with a C-P score &gt;13.    <br>   <b>CONCLUSION:</b> Despite initial endoscopic control of variceal haemorrhage,    17% of patients (1/6) rebled during the first hospital admission. The survival    rate of 79.2% was influenced by the severity of liver failure, with most deaths    occurring in C-P grade C patients in this study.</font></p> <hr noshade size="1">     <p>&nbsp;</p>     <p><font face="Verdana, Arial, Helvetica, sans-serif" size="3"><b>Repairing major    laparoscopic bile duct injuries: What does it cost?</b></font></p>     <p>&nbsp;</p>     <p>&nbsp;</p>     <p><font face="Verdana, Arial, Helvetica, sans-serif" size="2"><b>S Hofmeyr; J    E J Krige; P C Bornman</b></font></p>     ]]></body>
<body><![CDATA[<p><font face="Verdana, Arial, Helvetica, sans-serif" size="2">Surgical Gastroenterology    Unit, Groote Schuur and UCTPrivate Academic Hospital and Department of Surgery,    Faculty of Health Sciences, University of Cape Town</font></p>     <p>&nbsp;</p>     <p>&nbsp;</p> <hr noshade size="1">     <p><font face="Verdana, Arial, Helvetica, sans-serif" size="2"><b>BACKGROUND:</b>    A major bile duct injury is an infrequent but potentially life-threatening complication    after laparoscopic cholecystectomy. Few data exist about the financial implications    of duct repair. We aimed to calculate the costs of operative repair in a cohort    of patients who underwent bile duct reconstruction after major ductal injury.    <br>   <b>METHODS:</b> A prospective database was reviewed to identify all patients    referred to the University of Cape Town Private Academic Hospital between 2002    and 2011 for assessment and repair of major laparoscopic bile duct injuries.    The detailed clinical records and billing information were evaluated to determine    all costs from admission to discharge. Total costs for each patient were adjusted    for inflation between year of repair and 2012.    <br>   <b>RESULTS:</b> Thirty-four patients, including 25 female and 9 male of median    age 49 years (range 32 - 71) with a major bile duct injury, were referred for    management after a median of 21 days (range 1 -280) following initial surgery.    Patients were admitted to hospital for a median of 13.5 days (range 6 - 52 days).    The mean cost of repair was R182 400 (range R70 112 - R395 515). Contributors    to cost were hospital bed costs (27.4%), theatre costs (27.2%), radiology (16.1%),    specialist fees (9.9%), consumables (8.1%), pharmacy (5.2%), endoscopy (3.3%)    and laboratory costs (2.9%).    <br>   <B>CONCLUSIONS;</b> The cost of repair of a major laparoscopic bile duct injury    is substantial due to prolonged admission to hospital, complex surgical intervention    and intensive imaging requirements.</font></p> <hr noshade size="1">     <p>&nbsp;</p>     <p><font face="Verdana, Arial, Helvetica, sans-serif" size="3"><b>Efficacy of    single-agent injection therapy in achieving haemostasis in patients with acutely    bleeding peptic ulcers</b></font></p>     <p>&nbsp;</p>     ]]></body>
<body><![CDATA[<p>&nbsp;</p>     <p><font face="Verdana, Arial, Helvetica, sans-serif" size="2"><b>F Jacobsohn;    J E J Krige; D A Levin; J M Shaw; D Epstein; S R Thomson; P C Bornman</b></font></p>     <p><font face="Verdana, Arial, Helvetica, sans-serif" size="2">Surgical and Medical    Gastroenterology Units, Groote Schuur Hospital, Departments of Surgery and Medicine,    University of Cape Town Health Sciences Faculty, Cape Town</font></p>     <p>&nbsp;</p>     <p>&nbsp;</p> <hr noshade size="1">     <p><font face="Verdana, Arial, Helvetica, sans-serif" size="2"><b>BACKGROUND:</b>    Endoscopic therapy is the standard of care in patients with bleeding peptic    ulcer disease. This study was undertaken to determine the efficacy of injection    therapy, using adrenaline/saline solution (ASS) alone, with regards to the rate    of rebleeding, need for surgery and death.    <br>   <b>METHODS:</b> Over a 21-month period, data were prospectively collected regarding    patients presenting to Groote Schuur Hospital with acute peptic ulcer bleeding    and treated with endoscopic therapy. The rate of rebleeding, surgical or other    intervention and mortality were recorded and analysed. In addition, demographic    data, comorbidity, non-steroidal and aspirin use, length of hospital stay, transfusion    requirements, endoscopic findings, Rockall score and aetiology of bleeding,    were recorded.    <br>   <b>RESULTS:</b> Eighty patients (median age 54 years, range 22 - 91) were included;    51/80 (63.8%) were male; 61/80 (76.3%) had successful primary haemostasis; 19    (23.7%) patients failed primary haemostasis and required either repeat endoscopy    or surgical intervention; 17/80 (21.3%) underwent second endoscopic therapy.    Overall, 6/80 (7.5%) required surgical intervention after failure of primary    or secondary endotherapy. One patient underwent embolisation of the gastroduodenal    artery. The mortality rate was 11.4% (9/80 patients); 8 died due to associated    comorbidity (malignancy, IHD, CRF, CVA); 1 death was due to bleeding and miliary    TB.    <br>   <b>CONCLUSION:</b> Single-agent endoscopic intervention controlled bleeding    in 76% of patients with acute peptic ulcer bleeding. In those who rebled, second    endoscopic management controlled bleeding in 70%. Surgery or embolisation was    required in 8.8% of patients. Although overall mortality was 11%, medical comorbidity,    rather than acute bleeding, accounted for 89% of deaths.</font></p> <hr noshade size="1">     <p>&nbsp;</p>     ]]></body>
<body><![CDATA[<p><font face="Verdana, Arial, Helvetica, sans-serif" size="3"><b>Correlation    between HFE gene mutations and ALT levels may increase cardiovascular disease    risk in patients with non-alcoholic steatohepatitis NASH</b> </font></p>     <p>&nbsp;</p>     <p>&nbsp;</p>     <p><font face="Verdana, Arial, Helvetica, sans-serif" size="2"><b>C Kruger; M    Kotze</b></font></p>     <p>&nbsp;</p>     <p>&nbsp;</p> <hr noshade size="1">     <p><font face="Verdana, Arial, Helvetica, sans-serif" size="2"><b>BACKGROUND:</b>    Both non-alcoholic fatty liver disease (NAFLD) and alanine transaminase (ALT)    are independent risk factors for cardiovascular mortality. Non-alcoholic steatohepatitis    (NASH), the histologically advanced form of NAFLD, is an independent risk factor    for cardiovascular disease (CVD) morbidity and mortality.    <br>   <B>AIM:</b> Key genes involved in metabolic pathways relating to atherogenic    dyslipidaemia, chronic inflammation, hypercoagulation and iron dysregulation    implicated in insulin resistance as a common feature of NAFLD, were studied    to identify a genetic subgroup at increased risk of cardiovascular mortality    compared with controls.    <br>   <b>METHODS:</b> A total of 178 patients diagnosed with NAFLD and 75 controls    were studied using direct sequencing and real-time polymerase chain reaction    (PCR) for mutation detection. The analyses included eight deleterious low-penetrance    mutations in 5 genes: APOE2, APOE4, F2-20210, FV-Leiden, HFE-C282Y, HFE-H63D,    MTHFR-677 and MTHFR-1298. Relevant biochemical determinations, including ALT    level, were performed for all subjects and compared using appropriate statistical    analyses.    <br>   <b>RESULTS:</b> There was no statistically significant difference in genotype    distribution for individual mutations between the NAFLD v. control subjects.    However, in a sub-analysis a significant increase (p=0.04) in ALT levels was    detected in NASH patients found to be heterozygous or homozygous for the HFE-C282Y    and HFE-H63D mutations <i>(n</i>=10), compared with mutation-negative patients    <i>(n</i>=34).    ]]></body>
<body><![CDATA[<br>   <b>CONCLUSION:</b> We could not implicate the genes selected from known pathways    in increasing CVD risk in NAFLD patients. However, the close association between    known risk factors for increased CVD morbidity and mortality, namely raised    ALT, NASH and HFE mutations, was confirmed. Subjects with HFE gene mutations    have been found to have an increased CVD risk. The role of such mutations in    increasing CVD risk in NASH patients needs to be investigated.</font></p> <hr noshade size="1">     <p>&nbsp;</p>     <p><font face="Verdana, Arial, Helvetica, sans-serif" size="3"><b>A pooled analysis    of perineal hernia repair after abdominoperineal resection</b></font></p>     <p>&nbsp;</p>     <p>&nbsp;</p>     <p><font face="Verdana, Arial, Helvetica, sans-serif" size="2"><b>M Mjoli<sup>I</sup>;    D Sloothaak<sup>II</sup>; C Buskens<sup>II</sup>; W Bemelman<sup>II</sup>; P    Tanis<sup>II</sup></b></font></p>     <p><font face="Verdana, Arial, Helvetica, sans-serif" size="2"><sup>I</sup>Department    of Surgery, Pietermaritzburg Hospital Complex and University of KwaZulu-Natal,    Durban    <br>   <sup>II</sup>Department of Surgery, Academic Medical Centre, University of Amsterdam,    The Netherlands</font></p>     <p>&nbsp;</p>     <p>&nbsp;</p> <hr noshade size="1">     ]]></body>
<body><![CDATA[<p><font face="Verdana, Arial, Helvetica, sans-serif" size="2"><B>AIM:</b> The    purpose of this study was to determine the treatment characteristics and clinical    outcomes of patients with perineal hernia after abdominoperineal resection (APR).    <br>   <B>METHOD.</b> A systematic literature search revealed 40 individually documented    patients (published between 1944 and 2010). Three patients treated at the Academic    Medical Centre in the Netherlands were also included. Patient characteristics,    repair type and outcome were recorded and a pooled analysis of the 43 patients    was performed.    <br>   <b>RESULTS:</b> Pooled analysis revealed a median time interval of 8 months    between APR and surgical repair of perineal hernia. The surgical approaches    were perineal in 22 patients, open abdominal in II,&nbsp;open abdominoperineal    in 3, laparoscopic in 5 and laparoscopic-perineal in 2 patients. A primary recurrence    was documented in 13 patients and a second recurrence in 3. The recurrence rate    was 5/25 for synthetic or biological mesh, 6/12 for primary closure and 2/6    for the remaining techniques. Recurrent perineal hernia was repaired using a    synthetic or biological mesh (N=6), primary closure (N=5) or a muscle flap (gluteus    or gracilis) <i>(N</i>=4).    <br>   <b>CONCLUSION:</b> From these limited and biased data based on published case    descriptions, it appears that the recurrence rate of primary perineal hernia    repair after APR is lower with the use of a mesh or other assisted closure in    comparison with primary suture repair.    <br>   <b>NOTE:</b> This work was accepted for publication in <i>Colorectal Disease:</i>    Mjoli M, <i>et al.</i> Colorectal Dis 2012;14(7):e400-e406 (online 8 June 2012).</font></p> <hr noshade size="1">     <p>&nbsp;</p>     <p><font face="Verdana, Arial, Helvetica, sans-serif" size="3"><b>A review of    patients with chronic hepatitis C treated with pegylated interferon and ribavirin    at a central referral hospital in Durban</b> </font></p>     <p>&nbsp;</p>     <p>&nbsp;</p>     <p><font face="Verdana, Arial, Helvetica, sans-serif" size="2"><b>V Naidoo; K    A Newton</b></font></p>     ]]></body>
<body><![CDATA[<p><font face="Verdana, Arial, Helvetica, sans-serif" size="2">University of KwaZulu-Natal</font></p>     <p>&nbsp;</p>     <p>&nbsp;</p> <hr noshade size="1">     <p><font face="Verdana, Arial, Helvetica, sans-serif" size="2"><B>INTRODUCTION    AND OBJECTIVES:</b> Local experience in the treatment of hepatitis C virus (HCV)    is relatively limited, primarily due to fewer HCV patients and limited resources    compared with the developed world. Since 2006 the Gastroenterology Department    at Inkosi Albert Luthuli Central Hospital (IALCH) has had access to pegylated    interferon and ribavirin to treat suitable HCV patients. We describe our early    experience with HCV treatment in a low-prevalence region, with particular reference    to treatment outcomes.    <br>   <b>METHODS:</b> A retrospective audit was performed of HCV patients who received    treatment from the Gastroenterology Department between 2006 and 2011. Data were    obtained from standard data sheets used to aid patient management during treatment    and patient records. Demographics, genotype, early virological response (EVR)    and sustained virological response (SVR) were recorded. Simple descriptive statistics    were used to analyse data.    <br>   <b>RESULTS:</b> A total of 7 patients received treatment. One patient who had    genotype-1 infection, a high baseline viral load and type 2 diabetes mellitus,    did not achieve an EVR and treatment was stopped after 16 weeks. One patient    had genotype-1 infection, 3 had genotype-3, 1 had genotype-4 and 2 had genotype-5    infection. Five out of the 6 patients who completed a standard course of treatment    achieved SVR (83%). Both patients with genotype-5 infection were treated for    48 weeks and achieved SVR.    <br>   <b>CONCLUSION:</b> Despite the small number of patients treated, it is rewarding    to document that a significant number who completed treatment achieved SVR which    equates to a cure.</font></p> <hr noshade size="1">     <p>&nbsp;</p>     <p><font face="Verdana, Arial, Helvetica, sans-serif" size="3"><b>Re-activated    Wnt signalling is crucial in hepatocellular carcinoma pathogenesis in double-transgenic    mice that constitutively over-express HBx and IRS-1 in the liver</b> </font></p>     <p>&nbsp;</p>     ]]></body>
<body><![CDATA[<p>&nbsp;</p>     <p><font face="Verdana, Arial, Helvetica, sans-serif" size="2"><b>M Setshedi</b></font></p>     <p><font face="Verdana, Arial, Helvetica, sans-serif" size="2">University of Cape    Town</font></p>     <p>&nbsp;</p>     <p>&nbsp;</p> <hr noshade size="1">     <p><font face="Verdana, Arial, Helvetica, sans-serif" size="2"><b>BACKGROUND:</b>    Hepatocellular carcinoma (HCC) is associated Wnt/&acirc;-catenin with aberrant    growth factor signalling. Both the insulin/ insulin-like growth factor (IGF)    and Wnt/p-catenin have been implicated in the pathogenesis of HCC. To date there    are no therapeutic molecules targeting this pathway; therefore, identifying    the target genes of Wnt signalling remains critical for understanding &acirc;-catenin-mediated    carcinogenesis.    <br>   <B>AIM:</b> The purpose of this work was to characterise the gene expression    patterns of Wnt/p-catenin signalling in a double-transgenic mouse model of HCC.    <br>   <b>METHODS:</b> We analysed liver tissue from mice sacrificed at various time    points <i>(n</i>=12 per time point) by quantitative real-time polymerase chain    reaction (qRT-PCR) using a PCR array panel. Wildtype mice were used as controls.    <br>   <b>RESULTS:</b> Tumours developed only in male mice generally after 15 -18 months.    The majority of Wnt ligands, receptors and downstream targets were upregulated    in mice without tumours at early time points. However, some genes including    <i>Wnt-1, -2, -3, -4, -6, -7b, -11</i> and <i>Fzd-8, -2, -3</i> and <i>Fzd-b</i>    were also upregulated in HCC (occurring at later time points), suggesting re-activation    of Wnt signalling in HCC. Specifically <i>Wnt-5b</i> and <i>-7b</i> were significantly    upregulated in mice with tumours, suggesting they may be key regulators in hepatitis    B virus (HBV)-related HCC.    <br>   <B>CONCLUSIONS:</b> Our study identified several candidate genes of Wnt signalling,    particularly <i>Wnt5b</i> and <i>-7b</i> that are dysregulated and in combination    with 1 or more co-factors, i.e. HBx and/or insulin receptor substrate-1 (IRS-1)    are carcinogenic in mouse HCC. These genes may serve as useful potential therapeutic    targets for the treatment of HCC.</font></p> <hr noshade size="1">     ]]></body>
<body><![CDATA[<p>&nbsp;</p>     <p><font face="Verdana, Arial, Helvetica, sans-serif" size="3"><b><i>IL28B</i>    polymorphisms are not predictive in South African patients infected with hepatitis    C genotype 5</b> </font></p>     <p>&nbsp;</p>     <p>&nbsp;</p>     <p><font face="Verdana, Arial, Helvetica, sans-serif" size="2"><b>M Sonderup<sup>I</sup>;    W Abuelhassan<sup>II</sup>; C W Spearman<sup>I</sup></b></font></p>     <p><font face="Verdana, Arial, Helvetica, sans-serif" size="2"><sup>I</sup>Department    of Medicine and Division of Hepatology, Groote Schuur Hospital and University    of Cape Town    <br>   <sup>II</sup>Department of Gastroenterology, Chris Hani Baragwanath Academic    Hospital, Soweto</font></p>     <p>&nbsp;</p>     <p>&nbsp;</p> <hr noshade size="1">     <p><font face="Verdana, Arial, Helvetica, sans-serif" size="2"><b>BACKGROUND:</b>    Hepatitis C virus (HCV)-genotype 5 (G5), endemic in southern Africa, is often    neglected in major trials. While single nucleotide polymorphisms (SNPs) of the    interleukin-28B gene <i>(IL28B)</i> on chromosome 19 have been associated with    treatment response in genotypes 1 - 4, scant data and only in white patients,    suggest no predictive value of <i>IL28B</i> in HCV-G5. No such data exist for    South African patients or those of different ethnicities.    ]]></body>
<body><![CDATA[<br>   <b>METHODS:</b> Genomic DNA was obtained from peripheral blood mononuclear cells    (PBMCs) in a cohort of G5 patients and analysed for the rs12979860 SNP near    <i>IL28B.</i> Treated patients in the cohort had received peg-interferon-ci-2a    and weight-based ribavirin. Genotypes were analysed for their association with    treatment response.    <br>   <b>RESULTS:</b> In a cohort of 35 patients, 28 who had completed treatment were    evaluated. Of these, 57% (16) were male, mean age was 52.5&plusmn;12.1 years    in men and 54.8&plusmn;10.5 years in women. Median alanine transaminase (ALT)    was 66 IU/ml (range 14 - 215), fibrosis score (F) was 2 (n=22) and baseline    HCV ribonucleic acid (RNA) was 6.05 (3.6 - 7.4) log IU/ml. F&gt;3 was present    in 27% of patients. Median body weight was 80 kg (range 53 - 130). Overall,    the frequencies of the rs12979860-IL28B genotype were CC=25%, TC=43%, TT=32%    with the distribution, respectively; in blacks <i>(n</i>=14) being CC=14%, TC=36%,    TT=50% and in whites (n=11) being CC=36%, TC=54%, TT=9%. Genotypes were equally    distributed in those of mixed ancestry. Overall, a rapid viral response (RVR),    early viral response (EVR), end-of-treatment response (ETR) and sustained viral    response (SVR) rate was seen in 60% (17), 100%, 96% (27) and 84% (22) of patients,    respectively. The odds ratios (95% confidence interval) for an RVR and SVR for    genotype CC v. non-CC (1.9, range 0.3 - 11.9, p=0.5; 0.6, range 0.1 - 4.2, p=0.59),    TT v. non-TT (0.4, range 0.1 - 1.9, p=0.37; 0.9, range 0.1 - 6.4, p=0.94) and    TC v. non-TC (1.6, range 0.3 - 7.4, p=0.57; 1.7, 0.3 - 11.1, p=0.59) were not    predictive of response to treatment. In addition, ethnicity viz. black v. white,    did not influence the probability of achieving an RVR (p=0.88) or SVR (p=0.35).    <br>   <B>CONCLUSIONS:</b> In this albeit modest cohort of treated HCV-G5 patients    of different ethnicities, <i>IL28B</i> polymorphisms did not significantly predict    RVR or SVR. This may reflect that the viral kinetics of HCV-G5 more closely    resemble -G2 and -G3, although a larger study is required to confirm this finding.</font></p> <hr noshade size="1">     <p>&nbsp;</p>     <p><font face="Verdana, Arial, Helvetica, sans-serif" size="3"><b>Pneumatic balloon    dilatation v. laparoscopic Heller's myotomy for achalasia: A prospective evaluation    </b> </font></p>     <p>&nbsp;</p>     <p>&nbsp;</p>     <p><font face="Verdana, Arial, Helvetica, sans-serif" size="2"><b>P C Bornman;    R Baigree; S Botha; S R Thomson; G Watermeyer; D Levin</b></font></p>     <p><font face="Verdana, Arial, Helvetica, sans-serif" size="2">Gastrointestinal    Unit, Groote Schuur Hospital and University of Cape Town</font></p>     <p>&nbsp;</p>     ]]></body>
<body><![CDATA[<p>&nbsp;</p> <hr noshade size="1">     <p><font face="Verdana, Arial, Helvetica, sans-serif" size="2"><b>BACKGROUND:</b>    Laparoscopic Heller's myotomy (LHM) and pneumatic balloon dilatation (PBD) are    widely used in the management of achalasia. Consensus among experts on the superior    treatment modality is lacking.    <br>   <b>METHODS:</b> Achalasia patients who presented between 1999 and 2007 were    assessed for entry into a randomised controlled trial between LHM and PBD. Clinical    data were collected prospectively, including weight loss, dysphagia, chest pain,    regurgitation, heartburn and dysphagia score. Due to patient preference, only    one-third of patients were randomised to either therapy. Both randomised and    non-randomised patients were evaluated. The primary outcome of success was defined    as relief of dysphagia without the need for an alternative intervention. Complication    rate was evaluated as a secondary outcome.    <br>   <b>RESULTS:</b> Twenty-nine of the 45 patients (64%) had PBD and 16 (36%) had    LHM. There was no difference between the 2 treatment groups with regards to    baseline characteristics. The median follow-up period was 30 months (range 1    - 96 months). The success rate in LHM patients was 81% (13/16) v. 69% (20/29)    for PBD. This trend was not significant. None of the clinical parameters evaluated    were predictive of failure risk. The complication rate was confined to the LHM    11% (5/45) and included 2 perforations in the LHM arm, recognised and treated    during the initial procedure and a peptic stricture.    <br>   <b>CONCLUSION:</b> Although there is a bias in favour of LHM, the 2 treatment    modalities are comparable in efficacy. Interventions were safe with no mortality.</font></p> <hr noshade size="1">     <p>&nbsp;</p>     <p><font face="Verdana, Arial, Helvetica, sans-serif" size="3"><b>Management of    hepatitis C virus infection at Chris Hani Baragwanath Academic Hospital, Johannesburg    </b> </font></p>     <p>&nbsp;</p>     <p>&nbsp;</p>     <p><font face="Verdana, Arial, Helvetica, sans-serif" size="2"><b>W Abuelhassan;    R Ally</b></font></p>     ]]></body>
<body><![CDATA[<p><font face="Verdana, Arial, Helvetica, sans-serif" size="2">Gastroenterology    Department, Chris Hani Baragwanath Academic Hospital, Soweto</font></p>     <p>&nbsp;</p>     <p>&nbsp;</p> <hr noshade size="1">     <p><font face="Verdana, Arial, Helvetica, sans-serif" size="2"><b>INTRODUCTION:</b>    Hepatitis C virus (HCV) is a leading cause of chronic liver disease, cirrhosis    and hepatocellular carcinoma. South Africa (SA) has a low prevalence of HCV    infection (0.16 - 1.8%).    <br>   <B>OBJECTIVES:</b> To report our experience in the management of HCV infection    at Chris Hani Baragwanath Academic Hospital (CHBAH).    <br>   <b>METHODS:</b> Consecutive patients attending CHBAH hepatitis C clinic between    2007 and April 2012 were considered for enrolment. Demographics, clinical data,    HCV genotype and treatment response were analysed. Patients were treated with    peginterferon-c-2a and ribavirin according to SA guidelines.    <br>   <b>RESULTS:</b> Fifty-eight patients (30 male, 28 female; mean age 52.4 years)    were enrolled. The majority were blacks (46; 79.3%), followed by Asians (7;    12.1%), and whites (5; 8.6%). The predominant HCV genotype was 5a (26; 44.8%),    all occurring in black patients. Among the remaining 32 patients, 11 had genotype    1; 9 had genotype 3 (mostly Asians); 6 had genotype 4 (3 from the DRC); and    3 patients had 2 genotypes (1b and 4a; 1b and 5a; 1 and 4). Genotyping was not    performed in 3 patients. Thirty-six patients (62.1%) received treatment: 17    achieved sustained viral response (SVR), 3 were nonresponders (all genotype    1 with compensated cirrhosis), treatment is ongoing in 13, 2 relapsed and treatment    was ceased in 1 due to adverse events. Nineteen patients did not qualify for    treatment (decompensated cirrhosis in 14) and 3 were lost to follow-up. Twenty-one    of the 36 patients developed side-effects.    <br>   <b>CONCLUSION:</b> HCV genotype 5 occurred exclusively in black patients. They    responded well to therapy but, if untreated, followed the same natural history.    Treatment was generally well tolerated with the most common side-effect being    bone marrow suppression, which responded to dose adjustments and supportive    therapy.</font></p> <hr noshade size="1">     <p>&nbsp;</p>     <p><font face="Verdana, Arial, Helvetica, sans-serif" size="3"><b>An unusual cause    of upper gastrointestinal bleeding</b> </font></p>     ]]></body>
<body><![CDATA[<p>&nbsp;</p>     <p>&nbsp;</p>     <p><font face="Verdana, Arial, Helvetica, sans-serif" size="2"><b>W Abuelhassan;    R Ally</b></font></p>     <p><font face="Verdana, Arial, Helvetica, sans-serif" size="2">Gastroenterology    Department, Chris Hani Baragwanath Academic Hospital</font></p>     <p>&nbsp;</p>     <p>&nbsp;</p> <hr noshade size="1">     <p><font face="Verdana, Arial, Helvetica, sans-serif" size="2"><b>INTRODUCTION:</b>    Most of the causes of brisk upper gastrointestinal bleeding can be diagnosed    during oesophagogastroduodenoscopy. However, the diagnostic dilemma ensues when    no cause for the bleeding is readily obvious.    <br>   <B>CASE PRESENTATION:</b> A 45-year-old female patient with HIV infection (CD4    count of 445 cells/&igrave;!) presented to the emergency department with acute    haematemesis. She was pale, haemodynamically stable with 8 g/dl haemoglobin,    had a normal coagulation profile and cholestasis on liver function tests (LFTs).    Urgent endoscopy was performed, which was normal. During admission, the patient    developed malaena and her haemoglobin dropped to 3.8 g/dl. Repeat endoscopy    showed a normal oesophagus, stomach and duodenal cap. Upon touching the ampulla    of Vater, a jet of blood spurted through the ampulla (haemobilia). Urgent angiography    of the coeliac axis was performed and revealed multiple aneurysms of the hepatic,    left gastric and left gastroduodenal arteries, which were subsequently embolised.    A diagnosis of HIV vasculopathy was made. The patient remained well with no    further bleeding in the ensuing period and was subsequently discharged with    a referral for initiation of antiretroviral therapy.    <br>   <B>DISCUSSION:</b> Haemobilia is a relatively common under-diagnosed cause of    upper gastrointestinal bleeding with considerable morbidity and mortality. Angiography    can aid in depicting the source of active bleeding. Aneurysms associated with    HIV vasculopathy have been described mainly in the cerebral and popliteal circulations,    with few reports of mesenteric and portal circulation vasculopathy. Clues to    diagnosis include the presence of multiple aneurysms on angiography in an HIV    patient, and are often associated with deranged LFTs.</font></p> <hr noshade size="1">     <p>&nbsp;</p>     ]]></body>
<body><![CDATA[<p><font face="Verdana, Arial, Helvetica, sans-serif" size="3"><b>Single institution    experience with Zollinger-Ellison syndrome: causes of death and survival pattern    </b> </font></p>     <p>&nbsp;</p>     <p>&nbsp;</p>     <p><font face="Verdana, Arial, Helvetica, sans-serif" size="2"><b>S Alharethi;    S R Thomson; P C Bornman</b></font></p>     <p><font face="Verdana, Arial, Helvetica, sans-serif" size="2">Groote Schuur Hospital,    Cape Town</font></p>     <p>&nbsp;</p>     <p>&nbsp;</p> <hr noshade size="1">     <p><font face="Verdana, Arial, Helvetica, sans-serif" size="2"><b>BACKGROUND:</b>    Mortality in Zollinger-Ellison syndrome (ZES) is not clearly established. We    report a mortality analysis in a tertiary institution cohort.    <br>   <i>Patients and METHODS:</i> Forty-eight consecutive ZES patients were managed    at Groote Schuur Hospital between 1978 and 2012. Thirty-five males (73%) and    13 females (27%) were diagnosed at a mean age of 40 years. Forty (83%) patients    had sporadic disease and 8 (17%) had multiple endocrine neoplasia type-1 (MEN-1).    <br>   <b>RESULTS:</b> Nineteen patients with a mean follow-up of 10.4 years died during    the study period; 4 had MEN-1-associated gastrinoma. Nine patients are still    attending follow-up (mean 19.7 years). Twenty patients had variable follow-up    with a mean of 8.5 years. Five deaths were related to ZES, 2 patients died from    duodenal ulcer haemorrhage, 1 from post-operative septic complications following    repeated surgery, 1 from tumour progression, and 14 deaths were unrelated to    ZES.    ]]></body>
<body><![CDATA[<br>   <b>CONCLUSION:</b> ZES is compatible with long-term survival. The majority of    deaths are unrelated to ZES. Death from tumour progression is rare. Patients    undergoing recurrent surgery are at increased risk of complications-related    death.</font></p> <hr noshade size="1">     <p>&nbsp;</p>     <p><font face="Verdana, Arial, Helvetica, sans-serif" size="3"><b>A single-institution    experience of type-1 multiple endocrine neoplasia (MEN-l)-associated gastrinoma    </b> </font></p>     <p>&nbsp;</p>     <p>&nbsp;</p>     <p><font face="Verdana, Arial, Helvetica, sans-serif" size="2"><b>S Alharethi;    S R Thomson; P C Bornman</b></font></p>     <p><font face="Verdana, Arial, Helvetica, sans-serif" size="2">Groote Schuur Hospital,    Cape Town</font></p>     <p>&nbsp;</p>     <p>&nbsp;</p> <hr noshade size="1">     <p><font face="Verdana, Arial, Helvetica, sans-serif" size="2"><b>BACKGROUND:</b>    Multiple endocrine neoplasia type-1 (MEN-1) is a rare autosomal dominant disease    affecting parathyroids, pancreas and pituitary glands.    ]]></body>
<body><![CDATA[<br>   <B>PATIENTS AND METHODS:</b> We report a single-institution experience over    a 32-year period of 8 patients with MEN-1 identified from 48 patients with Zollinger-Ellison    syndrome. The mean age at diagnosis was 38 years.    <br>   <b>RESULTS:</b> Seven patients presented with ZES or its complications, while    1 patient presented with hypercalcaemia. The average delay from onset of symptoms    to diagnosis of ZES/MEN-1 was 6.5 years. Prior to diagnosis, 8 patients had    surgery for peptic ulcer complications and 2 had parathyroidectomy. Prolactin    was raised in 8 patients, abnormal pituitary/cella turcica was seen on computed    tomography (CT) in 6 patients and 1 patient had acromegaly. Five patients underwent    post-diagnosis parathyroidectomy with immediate normalisation of calcium in    4 and repeat surgery required in 1 patient. After diagnosis, 3 patients had    no further surgical intervention for ZES. One patient had total gastrectomy.    One had distal pancreatectomy and total gastrectomy. The remaining 3 patients    each had 2 debulking procedures for a variety of pancreatic neuroendocrine tumours.    No clinical or biochemical cure was achieved in this cohort. The median survival    time was 18 years. There were 4 patient deaths, and 3 patients were lost to    follow-up. One patient is alive and well.    <br>   <b>CONCLUSION:</b> This study confirmed that surgery does not cure MEN-1-associated    ZES. Screening for MEN-1 is required in all ZES patients even in the absence    of family history. Long-term survival is the rule; hence, follow-up to detect    metachronous neuroendocrine tumours is important.</font></p> <hr noshade size="1">     <p>&nbsp;</p>     <p><font face="Verdana, Arial, Helvetica, sans-serif" size="3"><b>HLA-DQ genotype    distribution in type 1 diabetes mellitus patients with concomitant coeliac disease    </b> </font></p>     <p>&nbsp;</p>     <p>&nbsp;</p>     <p><font face="Verdana, Arial, Helvetica, sans-serif" size="2"><b>S F Bakker;    M E Tushuizen; J B A Crusius; S Simsek; C J J Mulder; B M E von Blomberg</b></font></p>     <p><font face="Verdana, Arial, Helvetica, sans-serif" size="2">VU Medical Center,    the Netherlands</font></p>     <p>&nbsp;</p>     ]]></body>
<body><![CDATA[<p>&nbsp;</p> <hr noshade size="1">     <p><font face="Verdana, Arial, Helvetica, sans-serif" size="2"><B>BACKGROUND AND    AIMS:</b> The estimated prevalence of coeliac disease (CD) in patients with    type 1 diabetes mellitus (T1DM) is 5%. CD and T1DM, both multifactorial diseases,    are strongly clustered in families. In both, human leukocyte antigen (HLA) class    II molecules HLA-DQ2.5 (DQB1*02-DQA1-05) and DQ8 (DQB1*0302 - DQA1*0301) are    key genetic risk factors. We aimed to investigate HLA-DQ distributions in patients    diagnosed with both T1DM and CD. Associations were examined between HLA-DQ and    age of clinical onset and autoimmune comorbidity.    <br>   <B>MATERIAL AND METHODS:</b> Patients with T1DM and concomitant CD were recruited    from 33 hospitals in the Netherlands. We retrospectively collected data at T1DM    and CD diagnosis, and regarding comorbidity of autoimmune diseases. T1DM diagnosis    was defined as an absolute requirement of insulin, while CD diagnosis was based    on international criteria (European Society for Paediatric Gastroenterology,    Hepatology and Nutrition). Genomic DNA obtained from peripheral blood was used    for typing of HLA-DQA1* and DQB1* alleles, performed with a combined single-stranded    conformation polymorphism (SSCP)/ heteroduplex method by semi-automated electrophoresis    and gel-staining. Patients were divided into two groups, childhood-onset T1DM    (before age 20 years) and adult-onset T1DM, because childhood-onset is strongly    associated with HLA haplotypes.    <br>   <b>RESULTS:</b> The total cohort consisted of 61 patients diagnosed with T1DM    and CD (67.2% female; mean age 39.8&plusmn;19.8 years; T1DM and CD duration    of 22.6&plusmn;16.8 years and 8.3&plusmn;10.4 years, respectively). All patients    were unrelated and self-reported Dutch whites. Patients carried HLA-DQ2.5 in    80.3% (50.8% heterozygous and 29% homozygous). Only 6/61 (9.8%) patients were    diagnosed with CD before T1D; 50% of them were HLA-DQ2.5 homozygous. In the    childhood-onset T1DM group (n=38), mean age of T1DM onset was significantly    lower in HLA DQ-8 heterozygotes v. other genotypes (4.9 v. 8.0 years; <i>p</i>=0.04).    <br>   <B>CONCLUSIONS:</b> In patients with T1DM and CD, an 80.3% prevalence of carriers    of HLA-DQ2.5 was found. Interestingly, in the childhood-onset T1DM group, a    younger age of T1DM onset is associated with heterozygous HLA-DQ8. No associations    were found between HLA-DQ type and the prevalence of autoimmune comorbidity    or CD onset.</font></p> <hr noshade size="1">     <p>&nbsp;</p>     <p><font face="Verdana, Arial, Helvetica, sans-serif" size="3"><b>Surgery for    giant-giant gastric ulcers - the bedsore of the stomach: A report of 6 recent    cases</b> </font></p>     <p>&nbsp;</p>     <p>&nbsp;</p>     <p><font face="Verdana, Arial, Helvetica, sans-serif" size="2"><b>H Becker</b></font></p>     ]]></body>
<body><![CDATA[<p><font face="Verdana, Arial, Helvetica, sans-serif" size="2">Department of Surgery,    School of Medicine, Faculty of Health Sciences, University of Pretoria</font></p>     <p>&nbsp;</p>     <p>&nbsp;</p> <hr noshade size="1">     <p><font face="Verdana, Arial, Helvetica, sans-serif" size="2"><b>INTRODUCTION:</b>    Patients with giant gastric ulcers (&gt;3 cm) are significantly older and have    more aggressive disease, reflected by a higher incidence of bleeding, anorexia,    weight loss and emergency admission (Raju GS <i>et al., Am J Gastroenterol</i>    1999;94:3478-3486). These patients need prolonged aggressive treatment, with    excellent compliance. The author wishes to propose a further subgroup of giants    ulcers that are &gt;3 cm but also have a deep penetrating base into adjacent    organs (crater), where the base of the crater is an organ, usually the pancreas,    liver or porta hepatic.    <br>   <B>POSTULATE:</b> These giant-giant (bedsore) ulcers will not heal, but become    chronic and complicated with treatment. A low threshold for early referral to    surgery is necessary. If little improvement is seen after 3 months of adequate    medical treatment, surgery is indicated.    <br>   <B>MATERIAL AND METHODS:</b> Six patients underwent emergency operations for    uncontrollable haemorrhage from large penetrating gastric ulcers between December    2011 and May 2012. All patients needed definitive gastric surgery to control    haemorrhaging (subtotal gastrectomy/antrectomy, including the ulcer, with or    without truncal vagotomy).    <br>   <b>CONCLUSION:</b> Patients with giant penetrating gastric ulcers need early    referral for definitive surgery. A low threshold for referral is imperative.</font></p> <hr noshade size="1">     <p>&nbsp;</p>     <p><font face="Verdana, Arial, Helvetica, sans-serif" size="3"><b>Malignant biliary    obstruction: Plastic v. metal stents for palliation of symptomatic jaundice    </b> </font></p>     <p>&nbsp;</p>     ]]></body>
<body><![CDATA[<p>&nbsp;</p>     <p><font face="Verdana, Arial, Helvetica, sans-serif" size="2"><b>M Bernon; J    M Shaw; J E J Krige; S Burmeister; G E Chinnery; P C Bornman</b></font></p>     <p><font face="Verdana, Arial, Helvetica, sans-serif" size="2">Surgical Gastroenterology    Unit, Groote Schuur Hospital and Department of Surgery, Faculty of Health Sciences,    University of Cape Town</font></p>     <p>&nbsp;</p>     <p>&nbsp;</p> <hr noshade size="1">     <p><font face="Verdana, Arial, Helvetica, sans-serif" size="2"><b>BACKGROUND:</b>    Both plastic and self-expanding metal stents (SEMS) have been used to relieve    jaundice in patients with advanced malignant biliary obstruction. This study    compared the clinical efficacy of plastic v. metal biliary stents.    <br>   <B>MATERIALS AND METHODS:</b> In a prospective randomised controlled trial 37    patients with malignant common bile duct obstruction not amenable to curative    resection were offered palliative stenting from November 2008 to March 2012    and were followed-up until death. We compared patient survival and stent patency    rates.    <br>   <b>RESULTS:</b> Seventeen patients received 10 Fr. plastic stents and 20 patients    received SEMS. Mean duration of hospital stay after stenting was 2 days (range    1 - 2). One patient in each group remained jaundiced despite adequate biliary    drainage. Plastic stents blocked more frequently than SEMS (47.5% v. 10.0%;    p=0.015). In the SEMS group, 3 patients required re-admission to hospital (total    31 days) and median survival was 116 days, compared with 5 re-admissions (total    54 days) and a median survival of 105 days in the plastic stent group. Preliminary    cost analysis showed similar costs per patient in both groups.    <br>   <b>CONCLUSION:</b> Plastic 10 Fr. biliary stents block more frequently than    SEMS, which have a better patency rate and are associated with fewer hospital    re-admissions. Metal stents are cost-effective in palliating malignant biliary    obstruction in a public sector hospital.</font></p> <hr noshade size="1">     <p>&nbsp;</p>     ]]></body>
<body><![CDATA[<p><font face="Verdana, Arial, Helvetica, sans-serif" size="3"><b>Autoimmune hepatitis    at Chris Hani Baragwanath Academic Hospital</b> </font></p>     <p>&nbsp;</p>     <p>&nbsp;</p>     <p><font face="Verdana, Arial, Helvetica, sans-serif" size="2"><b>B Bobat; R Ally</b></font></p>     <p><font face="Verdana, Arial, Helvetica, sans-serif" size="2">Chris Hani Baragwanath    Academic Hospital, Johannesburg</font></p>     <p>&nbsp;</p>     <p>&nbsp;</p> <hr noshade size="1">     <p><font face="Verdana, Arial, Helvetica, sans-serif" size="2"><B>BACKGROUND:</b>    Autoimmune hepatitis (AIH) is a chronic inflammatory liver disease of any age,    evidenced by variable and fluctuating clinical features and serum auto-antibodies.    <br>   <b>AIM:</b> To report experience with AIH at Chris Hani Baragwanath Academic    Hospital (CHBAH).    <br>   <b>METHODS:</b> An audit was performed of 15 AIH cases identified after a review    of 243 patients attending the liver clinic at CHBAH between 2009 and 2012. Demographics,    clinical, laboratory and histology data, auto-antibodies, derived clinical scores    and therapy were reviewed.    ]]></body>
<body><![CDATA[<br>   <b>RESULTS:</b> Mean age of the 15 patients (14 females, 1 male) was 33.3 years    (range 14 - 68 years); 50% of the cohort had significant disease, initial mean    model for end-stage liver disease (MELD) score of 14.8 (6 - 26). Mean initial    bilirubin was 92.38, improving to 16.35 after treatment. Mean alanine transaminase    (ALT) improved from 345.23 to 45.85 IU/ml. Ten cases tested positive for antinuclear    antibodies (ANA). Anti-smooth-muscle antibody (ASMA) testing was positive in    a single case. No cases tested positive for antibodies to liver/kidney microsomes    (ALKM). Eight liver biopsies were performed; 6 were typical. Treatment improved    the Child-Pugh score to a mean of 6.2 (range 5 - 10). Comorbid diseases were    observed, including type 1 diabetes in 2 patients and single cases of hyperthyroidism,    interstitial lung disease, non-Hodgkin's lymphoma and ulcerative colitis with    primary sclerosing colangitis. Initial treatment included prednisone followed    by azathioprine. Currently, 10 patients are still receiving low-dose prednisone.    <br>   <B>CONCLUSIONS:</b> AIH at CHBAH has a female predominance and mainly affects    young adults; 50% are clinically ill at presentation. ANA testing was a good    indicator of disease, while ASMA and ALKM testing were not; anti-liver cytosol    (ALC) may be a better option. Liver biopsies proved to be typical. Most patients    responded very well to immunosuppressive therapy.</font></p> <hr noshade size="1">     <p>&nbsp;</p>     <p><font face="Verdana, Arial, Helvetica, sans-serif" size="3"><b>Endoscopic management    of bile leaks after laparoscopic cholecystectomy</b></font></p>     <p>&nbsp;</p>     <p>&nbsp;</p>     <p><font face="Verdana, Arial, Helvetica, sans-serif" size="2"><b>G E Chinnery;    J E J Krige; P C Bornman; M M Bernon; S Alharethi;S Hofmeyr; S Burmeister; S    Cullis; S R Thomson</b></font></p>     <p><font face="Verdana, Arial, Helvetica, sans-serif" size="2">Groote Schuur Hospital,    Cape Town</font></p>     <p>&nbsp;</p>     <p>&nbsp;</p> <hr noshade size="1">     ]]></body>
<body><![CDATA[<p><font face="Verdana, Arial, Helvetica, sans-serif" size="2"><b>BACKGROUND:</b>    A bile leak is an infrequent but potentially serious complication after biliary    surgery. The aim of this study was to assess the effectiveness of endoscopic    management.    <br>   <b>METHODS:</b> An endoscopic retrograde cholangiopancreatography (ERCP) database    was retrospectively reviewed to identify all patients with bile leaks after    laparoscopic cholecystectomy.    <br>   <b>RESULTS:</b> One hundred and thirteen patients (92 women, 21 men) of median    age 47 years (range 22 - 82) with a bile leak were referred at a median of 18.1    days (range 1 - 226) after surgery. Symptoms included pain (13; 11.5%), abnormal    liver function tests (LFTs) (22; 19.5%), bile leak (25; 22.1%), intra-abdominal    collections (45; 39.8%) and sepsis (8; 7%). Twenty-nine patients (25.7%) were    found to have a major bile duct injury without duct continuity, requiring surgery.    Forty-four patients had a cystic duct (CD) leak, 26 had a CD leak and common    bile duct (CBD) stones, and 14 patients had a CBD injury amenable to endoscopic    stenting. In the 70 patients with CD leaks, 24 underwent a sphincterotomy (including    8 stone extractions), 43 had a sphincterotomy and stenting (including 18 stone    extractions), 1 patient had stenting only, while 2 with previous sphincterotomies    required no further intervention. Of the 14 patients with CBD injuries treated    endoscopically, 7 had a class D injury, 3 had a class E5 injury, 3 had a class    B injury and 1 had a biliary stricture. The 113 patients underwent a total of    269 ERCPs (mean 2.4; range 1 - 7).    <br>   <B>CONCLUSIONS:</b> Bile leaks after laparoscopic cholecystectomy in 75% of    patients were due to CD leaks (with or without retained stones) or lesser bile    duct injuries; these were amenable to definitive endoscopic therapy.</font></p> <hr noshade size="1">     <p>&nbsp;</p>     <p><font face="Verdana, Arial, Helvetica, sans-serif" size="3"><b>Inflammatory    bowel disease (IBD) in Soweto</b> </font></p>     <p>&nbsp;</p>     <p>&nbsp;</p>     <p><font face="Verdana, Arial, Helvetica, sans-serif" size="2"><b>N A I Chopdat;    R Ally</b></font></p>     <p><font face="Verdana, Arial, Helvetica, sans-serif" size="2">Department of Health,    Chris Hani Baragwanath Academic Hospital, Johannesburg</font></p>     ]]></body>
<body><![CDATA[<p>&nbsp;</p>     <p>&nbsp;</p> <hr noshade size="1">     <p><font face="Verdana, Arial, Helvetica, sans-serif" size="2"><b>BACKGROUND:</b>    IBD is an integrated group of disorders characterised by recurrent, destructive    inflammation of the gastrointestinal tract. The most common forms include ulcerative    colitis (UC) and Crohn's disease (CD). Most studies arise from Western populations;    few, if any, describe IBD in the South African black population.    <br>   <B>AIM:</b> To establish a formal IBD clinic at Chris Hani Baragwanath Academic    Hospital (CHBAH), we assessed the demographics, disease spectrum and treatment    responses in a cohort of CHBAH patients.    <br>   <B>METHODS:</b> From January 2011 to March 2012 all patients with a confirmed    diagnosis of IBD on histology, irrespective of original date of diagnosis, were    recruited. Patient files were analysed and data collected.    <br>   <b>RESULTS:</b> Thirty-five patients, including 20 females, were recruited;    88.6% were black, 8.6% were white and 2.8% were Asian. IBD subtypes included    UC (91.4%) and CD (8.6%). In the latter, only 2 patients were black and 1 was    white. In the UC group, left-sided colitis comprised 47%, pancolitis 44% and    distal colitis 9%. The most common extra intestinal manifestation was primary    sclerosing cholangitis, followed by arthropathy. No patient had skin or eye    involvement. The mainstay of treatment included salazopyrin, corticosteroids    and azathioprine. No patient was on 6-mercaptopurine (6MP) or biologicals. Three    patients were defined as having steroid-dependent CD (too small for further    analysis).    <br>   <b>CONCLUSION:</b> IBD is common in our Soweto community. A high index of suspicion    is needed for the diagnosis of more cases. Specialised clinics and registries    are needed to improve patient outcome and quality of life.</font></p> <hr noshade size="1">     <p>&nbsp;</p>     <p><font face="Verdana, Arial, Helvetica, sans-serif" size="3"><b>Case summary:    the neuroendocrine tumour effect</b> </font></p>     <p>&nbsp;</p>     ]]></body>
<body><![CDATA[<p>&nbsp;</p>     <p><font face="Verdana, Arial, Helvetica, sans-serif" size="2"><b>N A I Chopdat;    R Ally</b></font></p>     <p><font face="Verdana, Arial, Helvetica, sans-serif" size="2">Department of Health,    Chris Hani Baragwanath Academic Hospital, Johannesburg</font></p>     <p>&nbsp;</p>     <p>&nbsp;</p> <hr noshade size="1">     <p><font face="Verdana, Arial, Helvetica, sans-serif" size="2"><B>HISTORY:</b>    We report a case of a 53-year-old male patient who presented with syncope and    haematemesis in a second presentation after almost 1 year. Gastroduodenoscopy    1 year preceding was reported as normal. No other previous background history    was notable.    <br>   <B>EXAMINATION AND FINDINGS:</b> The patient appeared well, apart from postural    hypotension and sinus tachycardia. His haemoglobin was 14.5 g/dl and coagulation    profile was normal. At 22h00 he complicated and had massive haematemesis necessitating    intensive care unit (ICU) admission. The patient was resuscitated and received    a proton pump inhibitor (PPI) infusion. Emergency endoscopy revealed blood in    the stomach, making visualisation difficult. Endoscopic haemostasis was achieved    with epinephrine injections blindly at the site thought to be bleeding. Repeat    endoscopy revealed a reddish mass lesion along the lesser curvature of the stomach,    with active bleeding, resulting in emergency surgery. At surgery, a well-delineated    mass was found with histology consistent with a well-differentiated neuroendocrine    tumour (NET) that extends into the muscularis. Gastrin and chromogranin A were    normal. A diagnosis of a type 3 NET was made. Octreoscan showed no evidence    of distant disease. He was referred to oncology and completed chemotherapy.    He has had an excellent outcome so far.    <br>   <b>DISCUSSION:</b> Although rare, NET must be considered as a cause of haematemesis.    Various diagnostic tools including specific urine and serum markers may help    to identify a specific tumour type Tumour localisation and metastasis is performed    via endoscopy, radiological imaging and somastatin receptor scintigraphy. Treatment    modalities include surgery, somastatin analogues, chemotherapy and radiotherapy.</font></p> <hr noshade size="1">     <p>&nbsp;</p>     <p><font face="Verdana, Arial, Helvetica, sans-serif" size="3"><b>HIV/AIDS cholangiopathy,    biochemical, histological and cholangiographic features in a South African cohort    </b> </font></p>     ]]></body>
<body><![CDATA[<p>&nbsp;</p>     <p>&nbsp;</p>     <p><font face="Verdana, Arial, Helvetica, sans-serif" size="2"><b>N A I Chopdat;    R Ally</b></font></p>     <p><font face="Verdana, Arial, Helvetica, sans-serif" size="2">Department of Health,    Chris Hani Baragwanath Academic Hospital, Johannesburg</font></p>     <p>&nbsp;</p>     <p>&nbsp;</p> <hr noshade size="1">     <p><font face="Verdana, Arial, Helvetica, sans-serif" size="2"><b>BACKGROUND:</b>    HIV-associated cholangiopathy is considered rare - a form of biliary inflammation    characterised by abnormal cholestatic liver enzymes and biliary duct irregularities.    The aetiology remains unclear. Opportunistic infections are implicated.    <br>   <B>OBJECTIVES.</b> To describe the biochemical, cholangiographic and histological    findings in HIV-positive patients with abnormal cholestatic liver enzymes.    <br>   <b>METHODS:</b> In a pilot study, data were collected from our Liver Clinic.    HIV patients with elevated cholestatic enzymes (&gt;2 times the upper limit    of normal for alkaline phosphatase (ALP) and gamma glutamyl transferase (GGT)),    but without viral, drug, autoimmune hepatitis or diabetes or high level of alcohol    consumption were recruited. All patients underwent an ultrasound to exclude    obstructive aetiologies. Endoscopic retrograde/magnetic resonance cholangiopancreatography    (ERCP)/(MRCP) and liver biopsy was performed in selected cases.    <br>   <B>RESULTS:</b> From December 2011 to March 2012, 19 patients (including 10    men) were recruited. Abdominal pain <i>(n</i>=12) was the most common manifestation;    the remaining patients were asymptomatic. Mean ALP and GGT were 720 U/l and    1 127 U/l, respectively. The mean CD4 lymphocyte count was 120 cells/mm<sup>3</sup>.    Only 3 patients were on highly active antiretroviral therapy (HAART) at presentation.    Ultrasonography was abnormal in only 3 patients (dilated common bile duct);    63% underwent ERCP, with the major finding being paucity of the peripheral bile    ducts (58%). The most common finding on liver histology was granulomatous hepatitis    (32%). Mycobacterium was the most common opportunistic infection.    ]]></body>
<body><![CDATA[<br>   <b>CONCLUSION:</b> HIV-associated cholangiopathy is not uncommon in the SA HIV    population. ERCP is an important diagnostic and therapeutic modality. Liver    histology is a powerful diagnostic tool for opportunistic infections.</font></p> <hr noshade size="1">     <p>&nbsp;</p>     <p><font face="Verdana, Arial, Helvetica, sans-serif" size="3"><b>Clinical, laboratory    and outcome data of hepatocellular carcinoma in Mozambique</b></font></p>     <p>&nbsp;</p>     <p>&nbsp;</p>     <p><font face="Verdana, Arial, Helvetica, sans-serif" size="2"><b>L Cunha; N B    Bhatt; P Modcoicar; F Lorenzo; F Mbofana; F Fernandes; E Samo Gudo; N Macie;    B Meggi; C Maueia; C Carrilho; I V Jani</b></font></p>     <p><font face="Verdana, Arial, Helvetica, sans-serif" size="2">Hospital Central    de Maputo, Mozambique</font></p>     <p>&nbsp;</p>     <p>&nbsp;</p> <hr noshade size="1">     <p><font face="Verdana, Arial, Helvetica, sans-serif" size="2"><b>BACKGROUND:</b>    Hepatocellular carcinoma (HCC) is a lethal cancer and represents the fifth most    common malignancy in the world. Data on clinical and laboratory characteristics    and prevalence of hepatitis virus infection among HCC patients in Mozambique    are scarce. We aimed to investigate these parameters in HCC patients attending    Maputo Central Hospital (Hospital Central de Maputo) (HCM).    ]]></body>
<body><![CDATA[<br>   <b>METHODS:</b> Between March 2011 and April 2012, 105 patients with HCC attending    HCM, were enrolled and screened for hepatitis B virus (HBV), hepatitis C virus    (HCV) and HIV. Alpha-fetoprotein levels, abdominal ultrasound and fine needle    aspiration (FNA) for cytology were performed routinely. Clinical and demographic    data were collected using a standard questionnaire.    <br>   <b>RESULTS:</b> HCC was most frequent in men (72.4%) and mean patient age was    49.7&plusmn;15.7 years. Lower educational level, history of smoking and alcohol    intake were found in 69%, 5.7% and 58% of patients, respectively. The prevalence    of HBV, HCV and HIV infection was 52.9%, 4.9% and 11.9%, respectively. HBV/HIV    co-infection was found in 9% of patients. No patient was co-infected with HBV/HCV    or HCV/HIV. Alpha-fetoprotein was &gt;400 UI in 62.7% of patients. Multinodular    presentation in the liver was detected by ultrasound in 80%. Mean survival was    66.5&plusmn;6.7 days.    <br>   <B>CONCLUSIONS:</b> Overall, our data demonstrated that chronic HBV infection    is highly prevalent in Mozambican HCC patients, suggesting that HBV represents    the major cause of HCC. The poor survival rate may reflect the advanced stage    of tumours at the time of diagnosis.</font></p> <hr noshade size="1">     <p>&nbsp;</p>     <p><font face="Verdana, Arial, Helvetica, sans-serif" size="3"><b>Colorectal cancer    in Ugandan patients: a morphological study</b></font></p>     <p>&nbsp;</p>     <p>&nbsp;</p>     <p><font face="Verdana, Arial, Helvetica, sans-serif" size="2"><b>D N Dijxhoorn;    A Boutall; C J Mulder; A Mall; R Ssebuufu; S Kalungi; C Baigrie; P A Goldberg</b></font></p>     <p>&nbsp;</p>     <p>&nbsp;</p> <hr noshade size="1">     ]]></body>
<body><![CDATA[<p><font face="Verdana, Arial, Helvetica, sans-serif" size="2"><b>INTRODUCTION:</b>    Colorectal cancer (CRC) is uncommon in Africa. Inherited cancers may account    for a greater portion of the disease burden in low- v. high-incidence areas.    CRC-related demographics and histological features and the incidence of hereditary    nonpolyposis colorectal cancer (HNPCC) in African populations are largely unknown.    <br>   <B>AIM:</b> To assess the demographic and morphological features of CRC patients    from hospitals in Mulago and Mbarara, Uganda.    <br>   <b>METHODS:</b> Histopathological specimens of 81 Ugandan CRC patients (2006    - 2010) were retrospectively reviewed. Demographic data were retrieved from    medical records. Tumours were examined to determine histological features suggestive    of mismatch repair gene mutations (according to the revised Bethesda guidelines    for microsatellite instability testing).    <br>   <b>RESULTS:</b> The median CRC patient age was 55 years (range 20 - 89) at diagnosis;    30.6% and 6.9% were aged &lt;50 and &lt;30 years, respectively. Tumours were    mostly (94.9%) located in the left side of the colon. Mucinous adenocarcinoma    was significantly more common in patients aged &lt;50 years. Based on the histological    and demographic features, 22/81 (27.3%) patients met at least 1 criterion of    the revised Bethesda Guidelines for MSI testing.    <br>   <B>CONCLUSIONS:</b> Histological and demographic features suggestive of HNPCC    were identified in 27.3% of patients. These features appear to be quite different    from published data from First-World countries. The lack of endoscopic equipment    in Uganda may account for the majority of cancers being left-sided.</font></p> <hr noshade size="1">     <p>&nbsp;</p>     <p><font face="Verdana, Arial, Helvetica, sans-serif" size="3"><b>MRCP and ERCP    demonstration of biliary tract disease</b> </font></p>     <p>&nbsp;</p>     <p>&nbsp;</p>     <p><font face="Verdana, Arial, Helvetica, sans-serif" size="2"><b>M Lagaud; Z    Ghoor; O Mzileni; A Elnagar</b></font></p>     ]]></body>
<body><![CDATA[<p><font face="Verdana, Arial, Helvetica, sans-serif" size="2">Dr George Mukhari    Hospital, Department of Health, Gauteng</font></p>     <p>&nbsp;</p>     <p>&nbsp;</p> <hr noshade size="1">     <p><font face="Verdana, Arial, Helvetica, sans-serif" size="2"><b>INTRODUCTION:</b>    Hepatobiliary diseases are often effectively diagnosed by abnormal liver biochemical    tests and ultrasonography. However, there are also unusual cases that may require    additional diagnostic procedures such as magnetic resonance cholangiopancreatography    (MRCP) and endoscopic retrograde cholangiopancreatography (ERCP). Clinical features    often do not present the true extent of the disease, especially in patients    with AIDS-related cholangiopathy. This presentation reports a case of biliary    disease diagnosed and managed by MRCP and ERCP.    <br>   <b>CASE DESCRIPTION:</b> A 23-year-old female who had used highly active antiretroviral    therapy (HAART) and anti-tuberculosis drugs, presented with jaundice, fever    and dry cough. Clinical examination and liver enzymes tests were suggestive    of hepatobiliary disease. The patient's alanine transaminase (ALT) was 48 IU/l,    aspartate transaminase (AST) was 559 U/l, AST-platelets-rate-index was 6.1,    alkaline phosphatase (ALP) was 209 U/l and gamma-glutamyl transpeptidase (GGT)    was 38 U/l. In addition, cytomegalovirus (CMV) DNA analysis by polymerase chain    reaction (PCR) revealed 584 000 copies/ml, with a CD4 cell count of 125 cells/mm<sup>3</sup>    and CMV pneumonitis. Ultrasonography showed only acalculous cholecystitis (ACC)    with pericholecystitis fluid; also seen with MRCP. ERCP revealed sclerosing    cholangitis in a setting of vanishing bile duct syndrome (VBDS), ulcerated stenotic    duodenal papillitis and ACC.    <br>   <b>DISCUSSION:</b> MRCP and ERCP performed equally well. MRCP should be used    to screen AIDS patients, irrespective of the pattern of the liver enzymes abnormalities.    MRCP also helps to select appropriate patients for ERCP, which will be required    to obtain tissue samples for microbiological studies and for sphincterotomy    (if needed).</font></p> <hr noshade size="1">     <p>&nbsp;</p>     <p><font face="Verdana, Arial, Helvetica, sans-serif" size="3"><b>Repeat endoscopic    rapid urease test as a surrogate marker for the success of <i>Helicobacter pylori</i>    eradication</b> </font></p>     <p>&nbsp;</p>     <p>&nbsp;</p>     ]]></body>
<body><![CDATA[<p><font face="Verdana, Arial, Helvetica, sans-serif" size="2"><b>D Levin; P Williams;    S R Thomson</b></font></p>     <p><font face="Verdana, Arial, Helvetica, sans-serif" size="2">PAWC and Groote    Schuur Hospital, Cape Town</font></p>     <p>&nbsp;</p>     <p>&nbsp;</p> <hr noshade size="1">     <p><font face="Verdana, Arial, Helvetica, sans-serif" size="2"><b>BACKGROUND:</b>    Amoxil and metronidazole are first-line antimicrobial eradication therapy for    <i>Helicobacter pylori</i> infection at Groote Schuur Hospital. Metronidazole    resistance is purported to be high. There are no local data assessing the efficacy    of this strategy.    <br>   <b>METHODS:</b> We evaluated the efficacy of antimicrobial therapy in patients    with <i>H. pylori</i> infection, in patients with a positive rapid urease test    (RUT) at initial endoscopy and who underwent a repeat endoscopy and RUT. Data    analysed included those from a prospective endoscopy database that were merged    with the antibiotic regimen dispensed from our pharmacy database. A positive    RUT at index and a negative RUT at repeat endoscopy were considered successful    eradication.    <br>   <b>RESULTS:</b> Over a 48-month period, 270 patients positive for <i>H. pylori</i>    infection at index endoscopy underwent a repeat endoscopy and repeat RUT; 132    patients had a positive RUT at index, but negative RUT at repeat endoscopy;    <i>H. pylori</i> was considered to be successfully eradicated in 132/270 (49%).    One hundred and thirty-eight patients were positive for <i>H. pylori</i> infection    at both index and repeat endoscopy. Seventy-four patients received antibiotic    therapy at the index endoscopy, from our pharmacy. Sixty-four of 138 (46.4%)    patients did not receive treatment from our pharmacy at index endoscopy. After    excluding those who did not receive therapy, the failure rate for <i>H. pylori</i>    eradication was 74/206 (36%).    <br>   <B>CONCLUSIONS:</b> This selected sample revealed an <i>H. pylori</i> eradication    rate of 36%, far below the standard efficacy rate of &lt;10%. The study also    identified a possible prescription-dispensing problem that may have contributed    to the high eradication failure rate.</font></p> <hr noshade size="1">     <p>&nbsp;</p>     <p><font face="Verdana, Arial, Helvetica, sans-serif" size="3"><b>A clinicopathological    spectrum of anal cancer in Kwazulu-Natal</b></font></p>     ]]></body>
<body><![CDATA[<p>&nbsp;</p>     <p>&nbsp;</p>     <p><font face="Verdana, Arial, Helvetica, sans-serif" size="2"><b>T E Madiba;    X H Ntombela</b></font></p>     <p><font face="Verdana, Arial, Helvetica, sans-serif" size="2">University of KwaZulu-Natal</font></p>     <p>&nbsp;</p>     <p>&nbsp;</p> <hr noshade size="1">     <p><font face="Verdana, Arial, Helvetica, sans-serif" size="2"><B>AIM:</b> To    document our experience with the management of anal cancer presenting to the    KwaZulu-Natal teaching hospitals.    <br>   <b>METHODS:</b> We analysed prospectively collected data on anal cancer patients    from 2004 to 2011, including demographics, clinical presentation, tumour site,    treatment, outcome and follow-up.    <br>   <b>RESULTS:</b> One hundred and thirty patients were included in the study (M:F    ratio, 1:1, mean age, 51&plusmn;14.0 years). Presenting symptoms were anal mass    (41), bleeding (27), ulcer (24), loss of weight and/ or appetite (19), anal    pain (18), peri-anal abscess/fistula (16), change in bowel habit (14), warts    (8), and incontinence (6). Mean duration of symptoms was 15.37&plusmn;19.41    months. Histology confirmed squamous carcinoma in 95, adenocarcinoma in 33,    melanoma in 1 and neuroendocrine tumour in 1. There were 104 anal margin tumours    and 26 anal canal tumours. Ten patients (8%) had distant metastases at diagnosis.    Ten patients (8%) were eligible for surgery. The remainder were managed non-operatively.    Seventy-nine patients were lost to follow-up and the rest were followed up for    1 - 69 months (mean 16&plusmn;17.0 months). Eleven patients have been confirmed    to have died so far.    <br>   <B>CONCLUSION:</b> Anal cancer affects all population groups with an equal incidence    in both sexes. Squamous carcinoma was 3 times as common as adenocarcinoma. Anal    margin tumours were 5 times as common as anal canal tumours. Delayed clinical    diagnosis leads to poor prognosis.</font></p> <hr noshade size="1">     ]]></body>
<body><![CDATA[<p>&nbsp;</p>     <p><font face="Verdana, Arial, Helvetica, sans-serif" size="3"><b>Metastatic colorectal    cancer in KwaZulu-Natal: a 12-year experience</b></font></p>     <p>&nbsp;</p>     <p>&nbsp;</p>     <p><font face="Verdana, Arial, Helvetica, sans-serif" size="2"><b>T E Madiba;    R Naidoo</b></font></p>     <p><font face="Verdana, Arial, Helvetica, sans-serif" size="2">University of KwaZulu-Natal</font></p>     <p>&nbsp;</p>     <p>&nbsp;</p> <hr noshade size="1">     <p><font face="Verdana, Arial, Helvetica, sans-serif" size="2"><b>BACKGROUND:</b>    The liver is the most common site of colorectal metastases, followed by the    lungs.    <br>   <B>AIM:</b> To evaluate our experience with the clinical presentation, management    and outcome of metastatic colorectal cancer (CRC).    ]]></body>
<body><![CDATA[<br>   <b>METHODS:</b> We analysed data from a prospectively compiled CRC database    (2000 - 2011), pertaining to demographics, clinical presentation, disease staging,    treatment, outcome and follow-up.    <br>   <b>RESULTS:</b> Of 1 424 CRC patients, 245 (17%) had metastatic CRC (mean age    57.5&plusmn;14.6 years; 1:1 male:female ratio). Primary sites were ascending    colon and caecum (31), hepatic flexure (7), transverse (9), splenic flexure    (4) descending (8), sigmoid colon (150), rectum (122). Sites of metastases were    liver (172; 70%), lung (47; 19%), peritoneum (21; 9%), omentum (12; 5%), ovaries    (8; 3%) and miscellaneous (9). Ninety-five patients (39%) underwent resection    of the primary tumour. Two hundred and twenty-five patients (92%) presented    for oncology treatment and follow-up, while 60 patients refused, absconded or    were too ill for oncological treatment. Excision of liver and lung metastasis    was performed in 1 patient each. Mean follow-up was 15&plusmn;17.3 months. Overall,    58 patients died (24%); 2 following resection and 56 due to disease progression.    <br>   <b>CONCLUSION:</b> Metastatic colorectal carcinoma accounts for 17% of CRC in    KwaZulu-Natal. The liver is the most common site. The surgical intervention    rate is extremely low in our setting. Patient follow-up remains a problem. Not    surprisingly, there is an appreciable mortality.</font></p> <hr noshade size="1">     <p>&nbsp;</p>     <p><font face="Verdana, Arial, Helvetica, sans-serif" size="3"><b>Predictors for    new bleeding in patients with higher digestive haemorrhage due to portal hypertension    in Maputo Central Hospital, Mozambique</b> </font></p>     <p>&nbsp;</p>     <p>&nbsp;</p>     <p><font face="Verdana, Arial, Helvetica, sans-serif" size="2"><b>P Modcoicar;    L Cunha; J Arteaga</b></font></p>     <p><font face="Verdana, Arial, Helvetica, sans-serif" size="2">Service of Gastroenterology,    Hospital Central de Maputo, Mozambique</font></p>     <p>&nbsp;</p>     ]]></body>
<body><![CDATA[<p>&nbsp;</p> <hr noshade size="1">     <p><font face="Verdana, Arial, Helvetica, sans-serif" size="2"><B>OBJECTIVE:</b>    Variceal rebleeding frequently develops after acute variceal bleeding and is    the most life-threatening complication in patients with cirrhosis. We aimed    to ascertain the predictive factors of new episodes of bleeding due to rupture    of oesophageal varices (EHV) and first episodes of bleeding after placement    of elastic bandages in patients suffering from portal hypertension (PHT).    <br>   <b>METHODS:</b> In 2011 we analysed 436 consecutive patients, presenting between    2003 and 2009, with endoscopically proved degree III and IV varices at the Endoscopy    Unit of Maputo Central Hospital (Hospital Central de Maputo) (HCM). Two-hundred    and thirteen patients were diagnosed with higher digestive haemorrhage at admission    (first arm) and 223 with PHT (second arm) underwent endoscopic variceal band    ligation.    <br>   <b>RESULTS:</b> In the first arm, 28 (13.1%) patients rebled; 19 (68.0%) rebled    after 6 weeks. In the second arm, 15 (6.7%) patients bled; in 12 (80.0%) this    occurred after 6 weeks. Using multivariate analysis, irregular follow-up (OR    4.0, 95% confidence interval (CI) 1.7 - 9.2, p&lt;0.001) and pancytopenia (2.99,    95% CI 1.33 - 6.73, p&lt;0.01) were, independently, predictors of rebleeding    in the first arm. In the second arm, irregular follow-up (4.2, 95% CI, 1,4 -    12.4, <i>p</i>&lt;0.004), pancytopenia (OR 4.2, 95% CI, 1.32 - 12.4, <i>p</i>&lt;0.009)    and anaemia at admission (OR 10.5, 95% CI, 1.6 - 68.5, <i>p</i>&lt;0.003) were,    independently, predictors of bleeding.    <br>   <B>CONCLUSIONS:</b> Irregular follow-up and pancytopenia in both arms, and anaemia    in the second arm, are predictors of new episodes of bleeding in our setting.</font></p> <hr noshade size="1">     <p>&nbsp;</p>     <p><font face="Verdana, Arial, Helvetica, sans-serif" size="3"><b>A case of collagenous    gastritis in an 18-year-old black female from Zambia</b> </font></p>     <p>&nbsp;</p>     <p>&nbsp;</p>     <p><font face="Verdana, Arial, Helvetica, sans-serif" size="2"><b>I Moola; A Mahomed</b></font></p>     ]]></body>
<body><![CDATA[<p><font face="Verdana, Arial, Helvetica, sans-serif" size="2">Charlotte Maxeke    Johannesburg Academic Hospital</font></p>     <p>&nbsp;</p>     <p>&nbsp;</p> <hr noshade size="1">     <p><font face="Verdana, Arial, Helvetica, sans-serif" size="2"><B>CASE REPORT:</b>    An 18-year-old black Zambian female residing in Johannesburg presented with    vomiting, fatigue and malaise. Aside from pallor, her physical examination was    normal. Her full blood count showed hypochromic microcytic anaemia. Her haemoglobin    (Hb) was 4.7 g/dl, MCV (mean corpuscular volume) was 68.8 fl, and white cell    count and platelet counts were normal. Further investigation showed severe iron    deficiency anaemia with 1.1 &igrave;mol/l (9.0 - 30.4) serum iron, 2 &igrave;g/l    (5 - 148) ferritin and 1% percentage saturation. Vitamin B<sub>12</sub> and    red cell folate levels were normal. Gastroscopy showed an extensive pangastritis    with nodular infiltrate most marked in the fundus and corpus of the stomach.    The nodules were 4 - 8 mm in diameter. No active bleeding, erosions or ulceration    was noted. The duodenum and oesophagus were normal macroscopically. Biopsy of    the affected lesions showed severe chronic gastritis with severe activity, with    evidence of erosions, focal ulceration and atrophy with focally prominent submucosal    fibrosis. No intestinal metaplasia, dysplasia or malignancy was noted. <i>Helicobacter    pylori</i> infection was not observed. The differential diagnosis was collagenous    gastritis and auto-immune gastritis. Other differentials were scleroderma and    post-radiation therapy changes. Extensive physical and biochemical investigation    revealed no other abnormalities. Follow-up colonoscopy showed a normal-looking    mucosa with no features of collagenous colitis. A diagnosis of collagenous gastritis    was made. Follow-up gastroscopy after 1 year of oral 20 mg omeprazole daily    showed similar endoscopic appearance. The iron deficiency was treated with intravenous    iron and the patient's full blood count normalised. She is currently asymptomatic.</font></p> <hr noshade size="1">     <p>&nbsp;</p>     <p><font face="Verdana, Arial, Helvetica, sans-serif" size="3"><b>Outcome of cricopharyngomyotomy    and diverticulopexy for Zenker's diverticulum</b> </font></p>     <p>&nbsp;</p>     <p>&nbsp;</p>     <p><font face="Verdana, Arial, Helvetica, sans-serif" size="2"><b>P Nel; H van    der Walt</b></font></p>     <p>&nbsp;</p>     ]]></body>
<body><![CDATA[<p>&nbsp;</p> <hr noshade size="1">     <p><font face="Verdana, Arial, Helvetica, sans-serif" size="2"><b>OBJECTIVES:</b>    To determine the outcomes after long-term follow-up of patients who underwent    Zenker's diverticulum repair using only cricopharyngomyotomy with diverticulopexy.    <br>   <B>METHODS:</b> This is an observational descriptive study. Only patients with    proven Zenker's diverticulum who underwent cricopharyngeal myotomy and diverticulopexy    were included in the study. Patients with recurrences were included.    <br>   <b>RESULTS:</b> Fifty-four cases were included; subject age ranged from 30 to    89 years with no sex predominance. Mean follow-up was 6 months. The main complaint    was dysphagia and regurgitation. All patients improved with a cricopharyngomyotomy    and diverticulopexy. There were 4 major complications and 1 recurrence.    <br>   <b>CONCLUSION:</b> Cricopharyngomyotomy and diverticulopexy for Zenker's diverticulum    is a safe procedure for all patients, with very good outcome and long-term RESULTS:</font></p> <hr noshade size="1">     <p>&nbsp;</p>     <p><font face="Verdana, Arial, Helvetica, sans-serif" size="3"><b>An unusual cause    of cholestasis: IgG4-associated sclerosing cholangiopathy</b></font></p>     <p>&nbsp;</p>     <p>&nbsp;</p>     <p><font face="Verdana, Arial, Helvetica, sans-serif" size="2"><b>N Parsoo; V    G Naidoo; K A Newton</b></font></p>     ]]></body>
<body><![CDATA[<p><font face="Verdana, Arial, Helvetica, sans-serif" size="2">Inkosi Albert Luthuli    Hospital and Department of Health, KwaZulu-Natal</font></p>     <p>&nbsp;</p>     <p>&nbsp;</p> <hr noshade size="1">     <p><font face="Verdana, Arial, Helvetica, sans-serif" size="2"><b>INTRODUCTION:</b>    Autoimmune pancreatitis and IgG4-associated cholangitis are recently recognised    pancreaticobiliary manifestations of IgG4-associated systemic disease. It may    mimic pancreatic cancer, primary sclerosing cholangitis or cholangiocarcinoma.    <br>   <B>CASE PRESENTATION:</b> A 68-year-old male presented with a 1-year history    of steatorrhoea, weight loss and a 2-week history of jaundice with pruritus.    The patient had no history of alcohol consumption. Clinical examination revealed    jaundice and a firm 3 cm hepatomegaly. Biochemical testing showed cholestasis.    An ultrasound demonstrated prominent intrahepatic ducts and dilated common bile    duct; however, no obstructing lesion could be visualised. A magnetic resonance    cholangiopancreatogram (MRCP) showed multiple strictures involving the intrahepatic    ducts. Primary sclerosing cholangitis (PSC) was the most likely consideration.    Colonoscopy was normal. Endoscopic ultrasound (EUS) revealed a thickened common    bile duct wall of 3.3 mm. Fine-needle aspiration of an indistinct hypoechoic    pancreatic head mass showed lymphocytes with no evidence of malignancy. The    IgG4 subset was elevated to 10 times the upper limit of normal. A diagnosis    of IgG4 sclerosing cholangiopathy was made and the patient was commenced on    prednisone therapy. There was remarkable improvement in the liver function tests    and pruritus after 2 weeks of steroids. Repeat endoscopic ultrasound showed    reduction in common bile duct wall thickness to 2.0 mm.    <br>   <b>CONCLUSION:</b> IgG4-associated sclerosing cholangiopathy should be considered    in suspected PSC as it responds well to steroids. The use of EUS to evaluate    bile duct wall thickness and for subsequent monitoring of therapeutic response    is not well described and may be of value.</font></p> <hr noshade size="1">     <p>&nbsp;</p>     <p><font face="Verdana, Arial, Helvetica, sans-serif" size="3"><b>Geographic distribution    of gastroscopy referrals to a tertiary unit: implications for the development    of an equitable endoscopy service</b> </font></p>     <p>&nbsp;</p>     <p>&nbsp;</p>     ]]></body>
<body><![CDATA[<p><font face="Verdana, Arial, Helvetica, sans-serif" size="2"><b>A Rajula; S    R Thomson; G Watermeyer</b></font></p>     <p><font face="Verdana, Arial, Helvetica, sans-serif" size="2">Groote Schuur Hospital    and the University of Cape Town</font></p>     <p>&nbsp;</p>     <p>&nbsp;</p> <hr noshade size="1">     <p><font face="Verdana, Arial, Helvetica, sans-serif" size="2"><b>BACKGROUND:</b>    Gastroscopy is not yet a tertiary hospital procedure. Groote Schuur Hospital    (GSH) provides 80% of the gastroscopy service for the Cape Metro West Health    Area (CMWHA), with the remainder provided by Somerset Hospital (SH), G F Jooste    Hospital (GFJH), and Victoria Hospital (VH). To plan a comprehensive equitable    gastroscopy service, the geographical distribution of referrals needs to be    mapped.    <br>   <b>METHODS:</b> From 1 September to 31 October all gastroscopy records entered    into a prospective database at the gastrointestinal unit of GSH added home location    to the dataset. Locations were allocated according to the catchment areas/proximity    of the 4 CMWHA hospitals - and whether patients' home locations fell within    the catchment areas or not.    <br>   <b>RESULTS:</b> Gastroscopies performed annually included 6 300 at GSH, 1 439    at VH, 1 141 at SH and 0 at GFJH. Of the 482 endoscopies performed during the    2-month study period, 80 had no recorded address. Sixteen per cent of the patients    were not from CMWHA, while 49%, 10% and 3% should have gone to GFJH, VH and    SH, respectively. Only 22% of the patients were located within GSH's true catchment    area.    <br>   <b>CONCLUSION:</b> The gross inadequacy of a secondary-level gastroscopy service    is highlighted. There is an urgent need to establish an endoscopy service at    GFJH and to ensure that the services in the adjacent districts are responsible    for their own patients.</font></p> <hr noshade size="1">     <p>&nbsp;</p>     <p><font face="Verdana, Arial, Helvetica, sans-serif" size="3"><b>Revisional laparoscopic    surgery for failed laparoscopic Nissen fundoplication: a private practice experience    </b> </font></p>     ]]></body>
<body><![CDATA[<p>&nbsp;</p>     <p>&nbsp;</p>     <p><font face="Verdana, Arial, Helvetica, sans-serif" size="2"><b>M Z Shaik; B    Govender; A D R Reddy; P Rajaruthnam</b></font></p>     <p>&nbsp;</p>     <p>&nbsp;</p> <hr noshade size="1">     <p><font face="Verdana, Arial, Helvetica, sans-serif" size="2"><b>BACKGROUND:</b>    The efficacy of revisional surgery in failed laparoscopic Nissen fundoplications    (LNFs) (failed due to recurrent reflux and/or new symptoms following a successful    period of antireflux surgery) is promising in current literature. This is an    audit of our practice.    <br>   <b>METHODS:</b> Nine cases of revision of failed LNFs were carried out over    a 3-year period at a private surgical practice and analysed by retrospective    chart review. Two cases were failed redo LNFs.    <br>   <b>RESULTS:</b> Symptoms before re-operation were dysphagia <i>(n</i>=5), heartburn    <i>(n</i>=4) and persistent epigastric pain <i>(n</i>=3). Endoscopic findings    included reflux oesophagitis (n=2), hiatal hernia (n=2), stenoses <i>(n</i>=1)    and no abnormality <i>(n</i>=4). Barium swallows showed reflux (n=3), wrap herniation    (n=1), contrast hold-up (n=1), 'cup and spill' deformity (n=1), oesophageal    dysmotility (n=1) and no abnormality <i>(n</i>=1). Mechanisms of failure at    re-operation were cruraplasty stenosis (n=5), tight wrap (n=1), wrap herniation    (n=1), wrap disruption (n=2), slipped wrap (n=1). Intra-operative oesophageal    and stomach injuries were limited to patients with previous redo operations    <i>(n</i>=2); these were repaired laparoscopically. Wraps and cruraplasties    were undone only <i>(n</i>=5) in tight wraps and stenosed cruraplasties. Redo    LNFs <i>(n</i>=4) were performed in slipped, herniated and disrupted wraps.    Follow-up at 1, 3, and 12 months revealed overall patient satisfaction and relief    of symptoms <i>(n</i>=7) with persistent dysphagia in 1 patient.    <br>   <b>CONCLUSION:</b> Revisional laparoscopic surgery is safe and offers effective    symptom relief for failed LNFs. The risk of iatrogenic injury increases with    subsequent redo fundoplications.</font></p> <hr noshade size="1">     <p>&nbsp;</p>     ]]></body>
<body><![CDATA[<p><font face="Verdana, Arial, Helvetica, sans-serif" size="3"><b>Diarrhoea in    Schnitzler syndrome</b></font></p>     <p>&nbsp;</p>     <p>&nbsp;</p>     <p><font face="Verdana, Arial, Helvetica, sans-serif" size="2"><b>W Simmonds;    O C Buchel; R C Flooks; V J Louw</b></font></p>     <p><font face="Verdana, Arial, Helvetica, sans-serif" size="2">Universitas Hospital    and Department of Health, Bloemfontein, Free State</font></p>     <p>&nbsp;</p>     <p>&nbsp;</p> <hr noshade size="1">     <p><font face="Verdana, Arial, Helvetica, sans-serif" size="2"><b>BACKGROUND:</b>    Schnitzler syndrome is a rare idiopathic condition characterised by recurrent    episodes of non-pruritic urticarial rash, fever, bone pain, arthralgia or arthritis,    and a monoclonal gammopathy (IgM or more rarely, IgG) in concentrations &lt;10    g/l. According to the literature, approximately 100 cases of Schnitzler syndrome    have been reported to date.    <br>   <B>CASE DESCRIPTION:</b> We report a case of Schnitzler syndrome affecting a    54-year-old male who, in addition, had diarrhoea with his episodes of urticaria.    No other cause for diarrhoea was found on further evaluation. We postulate that    the diarrhoea may be caused by excess interleukin-1 activity, thought to be    a primary mediator in Schnitzler syndrome. Ciprofloxacin and loratadine therapy    was initiated in the patient, with good effect.    <br>   <B>CONCLUSION:</b> Diarrhoea in Schnitzler syndrome has not previously been    documented. The probable role of interleukin-1 as the cause for the diarrhoea    requires confirmation.</font></p> <hr noshade size="1">     ]]></body>
<body><![CDATA[<p>&nbsp;</p>     <p><font face="Verdana, Arial, Helvetica, sans-serif" size="3"><b>Initial experience    with HROPT subtype classification of achalasia: a potential tool to predict    outcome</b> </font></p>     <p>&nbsp;</p>     <p>&nbsp;</p>     <p><font face="Verdana, Arial, Helvetica, sans-serif" size="2"><b>J E C Botha;    S Hlatshwayo; S R Thomson; P C Bornman; G Watermeyer</b></font></p>     <p><font face="Verdana, Arial, Helvetica, sans-serif" size="2">Gastrointestinal    Unit, Groote Schuur Hospital and the University of Cape Town</font></p>     <p>&nbsp;</p>     <p>&nbsp;</p> <hr noshade size="1">     <p><font face="Verdana, Arial, Helvetica, sans-serif" size="2"><b>BACKGROUND:</b>    High-resolution oesophageal pressure topography (HROPT) details pressure topography    of the oesophagus, and an integrated relaxation pressure (IRP) for the lower    oesophageal sphincter complex which allows subtyping. This study describes the    3 achalasia subtypes in our patient population and the early outcome of different    therapies.    <br>   <b>METHODS:</b> Over 14 months patients diagnosed with achalasia on clinical,    barium swallow and endoscopy features underwent HROPT and were categorised into    the 3 subtypes: I - classic achalasia, failed peristalsis; II - achalasia with    compression, pan-oesophageal pressurisation; III - achalasia with rapid propagated    pressurisation spasm. Treatment was at the individual clinician's discretion.    ]]></body>
<body><![CDATA[<br>   <B>RESULTS:</b> Eighteen patients had primary achalasia. Average age was 42    years (range 11 - 77). The frequency of the subtypes was as follows: I - 5;    II - 12; III - 1. The mean lower oesophageal sphincter pressure (LOSP) of 37    mmHg was highest in subtype II. Female patients were 3 times more common in    type II. IRP increased from subtype I to III. Nine patients had pneumatic balloon    dilatation (PBD), 4 had a Hellers myotomy, and 1 is awaiting surgery. Four patients    were treated elsewhere. Eight have had a good initial result. Two patients had    a repeat PBD. Two patients have not yet been evaluated post-operatively. One    patient has not returned for clinical assessment, and 1 patient died from a    PBD perforation.    <br>   <b>CONCLUSION:</b> Our early experience with HROTP clearly identifies subsets    of patients with achalasia, which are indistinguishable on standard clinical    and investigational grounds. There is a predominance of subtype II. Greater    accrual and long-term follow-up are required to evaluate if these subtypes predict    durable outcome.</font></p> <hr noshade size="1">      ]]></body>
<REFERENCES></REFERENCES
</article>
