Surgery for high-output small bowel enterocutaneous fistula: a 30-year experience.

The outcome of surgery undertaken to repair high-output small bowel enterocutaneous fistula (SBECF) using the same technique was evaluated. Of a total of 282 patients with a high-output SBECF (daily fistula output > 500 ml), 183 patients were managed conservatively; in this group, a spontaneous closure rate of 81.4% (n = 149) and a mortality rate of 18.6% (n = 34) was noted. Ninety-nine patients (35.1%) underwent definitive surgical treatment. In all patients, after resection of the fistula, the entire small bowel was stented. Six patients (6%) died in the postoperative period. No complications developed as a result of intestinal stenting. Including the patients treated successfully for postoperative persistence of fistula, surgical repair was successful in 93 patients (93.9%). No patients returned with refistulization or small bowel obstruction within 6 months of surgery. We believe that routine intraluminal stenting should be an integral component of the surgical technique for the repair of SBECF.

Why we use the donor left kidney in live related transplantation.

UNLABELLED: In addition to the superior graft survival afforded by live related transplantation, this option has assumed an important role in the management of endstage renal failure in centres confronted with a scarcity of cadaveric kidneys. In pursuing this option, it is imperative that the donor has minimal morbidity. An ongoing dilemma is which side the kidney should be harvested from. This study reviews the anatomical basis for selecting the left kidney and the impact on outcome for patient and donor. A database comprising cadaveric and clinical subsets was analysed. The total sample size analysed was 1 244 kidney pairs (305 cadaveric; 939 clinical (61 live related left kidney transplants harvested by the extraperitoneal approach)). RESULTS: Additional renal arteries (ARAs): Right first, second = 18.6%, 4.7%; left first, second = 27.6%, 4.4%. Additional renal veins (ARV): Right first, second = 26%, 3.3%; left first only = 2.6%. Length of renal vein (cm): Right 2.4 +/- 0.7, left 5.9 +/- 1.5. Other venous variations encountered were on the left side only (renal collar 0.3%, retro-aortic vein 0.5%). In the live related transplant series 24.6% ARAs were encountered (first 19.7%, second 4.9%). The postoperative course and outcome of both donor and recipient were not associated with increased morbidity. While greater length of the left renal vein (LRV) afforded easier technical manipulation, it is interesting to note that its length is shorter than that reported in the literature. ARVs are infrequent on the left and when encountered the smaller calibre vessel may be ligated with impunity due to rich intrarenal anastomosis. In selecting the donor kidney the surgeon has to balance the prospect of fewer ARAs on the right against the benefit of a longer LRV. The solution to this dilemma can only arise from a randomised clinical study. In our practice, consistent use of the left kidney has not affected clinical outcome.

Cholecystectomy in a predominantly African population before and after the advent of the laparoscopic technique.

OBJECTIVE: There is a paucity of information on gallbladder disease in an African population. We, therefore, conducted a study to compare the immediate pre-laparoscopic era with the laparoscopic period in the predominantly African population at the King Edward VIII Hospital. MATERIAL AND METHODS: Data from a retrospective analysis of 144 patients undergoing open cholecystectomy (OC) between January 1990 and December 1992 were compared with a prospective analysis of 156 patients who underwent laparoscopic cholecystectomy (LC) between February 1992 and December 1994. Demographic data, presentation, operative management and outcome were the main factors analysed. RESULTS: Eighty-two per cent were Black African and the rest of Indian origin. Endoscopic retrograde cholangiopancreatography (ERCP) confirmed ductal stones in 11 patients in the OC and nine patients in the LC group. Endoscopic duct clearance was achieved in three and nine patients, respectively. Non biliary complications were rare. There were two major duct injuries in the OC group and one cystic duct leak in the LC group. The high conversion rate of 17.9% attests to the severity of their chronic disease making safe dissection in Calot's triangle problematic. The mortality in patients undergoing OC was 1 (0.07%) and 0% for LC. In South Africa, the hospital prevalence of calculous disease in African patients is increasing. However, cholecystectomy may be safely performed. CONCLUSION: The absence of any mortality and any major duct injury in the LC group allude to the safety of this procedure when appropriately applied to this population group.

Asymptomatic hyperparathyroidism caused by a giant parathyroid adenoma.

On routine investigation a 57-year-old woman was found to have primary hyperparathyroidism caused by a giant parathyroid gland. The gland was removed successfully and histological examination proved it to be a parathyroid adenoma.

Renal trauma secondary to stab, blunt and firearm injuries--a 5-year study.

OBJECTIVE: To review our experience of renal trauma at King Edward VIII Hospital, Durban. DESIGN: Retrospective study over 5 years. SETTING: Tertiary referral hospital in an urban area. PATIENTS: One hundred and seventy-four patients presenting with renal injuries treated at King Edward VIII Hospital. INTERVENTIONS: One hundred and thirty-six patients underwent laparotomy, while 38 were managed conservatively. RESULTS: There were 174 patients with renal injuries (148 male, median age 26 years). Eighty-seven injuries were from firearms, 43 from stabs and 44 from blunt trauma. Of 136 patients who underwent surgery, 60 were managed conservatively at laparotomy, 26 underwent renorrhaphy, 40 nephrectomy and 10 heminephrectomy. There was a 33% complication rate following surgery. The mortality rate was 20% (24%, 21% and 12% for firearm, stab and blunt injuries, respectively). There was no statistically significant difference in mortality for firearm, stab and blunt injuries. Furthermore, there was no difference in mortality and morbidity between those with associated organ injuries and those with isolated injuries (29% v. 21% and 30% v. 16%, respectively), but the difference in mortality became significant when patients with isolated injuries who did not undergo surgery were included (P = 0.0083). CONCLUSION: Renal injuries either in isolation or in association with other injuries carry a high mortality and morbidity rate. Firearm injury patients have a large number of associated injuries and have the highest complication rate. Operative strategy should be conservative unless the kidney is unsalvageable.

Additional renal arteries: incidence and morphometry.

Advances in surgical and uro-radiological techniques dictate a reappraisal and definition of renal arterial variations. This retrospective study aimed at establishing the incidence of additional renal arteries. Two subsets were analysed viz.: a) Clinical series--130 renal angiograms performed on renal transplant donors, 32 cadaver kidneys used in renal transplantation b) Cadaveric series--74 en-bloc morphologically normal kidney pairs. The sex and race distribution was: males 140, females 96; African 84, Indian 91, White 43 and "Coloured" 18, respectively. Incidence of first and second additional arteries were respectively, 23.2% (R: 18.6%; L: 27.6%) and 4.5% (R: 4.7%; L: 4.4%). Additional arteries occurred more frequently on the left (L: 32.0%; R: 23.3%). The incidence bilaterally was 10.2% (first additional arteries, only). The sex and race incidence (first and second additional) was: males, 28.0%, 5.1%; females, 16.4%, 3.8% and African 31.1%, 5.4%; Indian 13.5%, 4.5%; White 30.9%, 4.4% and "Coloured" 18.5%, 0%; respectively. Significant differences in the incidence of first additional arteries were noted between sex and race. The morphometry of additional renal arteries were lengths (cm) of first and second additional renal arteries: 4.5 and 3.8 (right), 4.9 and 3.7 (left); diameters: 0.4 and 0.3 (right), 0.3 and 0.3 (left). Detailed morphometry of sex and race were also recorded. No statistically significant differences were noted. Our results of the incidence of additional renal arteries of 27.7% compared favourably to that reported in the literature (weighted mean 28.1%). The study is unique in recording detailed morphometry of these vessels. Careful techniques in the identification of this anatomical variation is important since it impacts on renal transplantation surgery, vascular operations for renal artery stenosis, reno-vascular hypertension, Takayasu's disease, renal trauma and uro-radiological procedures.

Radio-contrast imaging of the rectum prior to colostomy closure for rectal trauma--is routine use still justified?

This retrospective study was undertaken to assess the yield of radio-contrast imaging of the rectum before closure of colostomy following extraperitoneal rectal trauma. Sixty-nine patients (63 males) underwent colostomy closure in 36 months. All radio-contrast studies (colograms) performed before closure of colostomy were normal, and there were no deaths following closure. This study demonstrated that the yield from pre-closure radio-contrast imaging of the rectum after rectal trauma was negligible and did not influence colostomy closure. We conclude that while it may be appealing to suggest abandonment of its routine use, this investigation needs to be further evaluated prospectively with special attention given to injury to associated structures such as bone, bladder and vagina.

Congenital absence of the gall bladder.

Congenital absence of the gall bladder is an extremely rare embryological aberration with a reported incidence ranging between 0.013 and 0.075%. This report, the first from South Africa, discusses 2 cases of gall bladder agenesis, bringing to 413 the number of cases reported in the literature. In confirming the diagnosis of an agenesis of the gall bladder, it is necessary to exclude the abnormal locations which are intrahepatic, retrohepatic, on the left side, or within the lesser omentum or falciform ligament and retroperitoneal. Patients with gall bladder agenesis are classified into 3 categories: i) Multiple foetal anomaly (12.9%), ii) Asymptomatic (31.6%) and iii) Symptomatic (55.6%). Notwithstanding current diagnostic modalities, this rare condition may still present a dilemma to the abdominal surgeon. Agenesis of the gall bladder is a well-recognised but uncommon congenital abnormality. With the advent of minimal access surgery laparotomy may be avoided as the condition, when suspected, may be confirmed by ERCP and CT scan.

Left renal vein variations.

The highly complex embryological development of the left renal vein compared to its right counterpart results in greater variations which are clinically significant. The study aimed to identify these variations and to document its incidence. Cadaveric study: 153 kidney pairs were harvested en bloc, dissected, 100 resin casts prepared and 53 plastinated; renal venography performed on further 58 adults and 20 foetal cadavers. Clinical study: (retrospective analysis): a) radiological study, 104 renal venograms; b) live related renal transplantation, 148 donor left kidneys; c) abdominal aortic aneurysm surgery, 525 patients. Total sample size: 1008. Renal collars observed in 0.3%; retro-aortic vein 0.5%; additional veins 0.4%; posterior primary tributary 23.2%, (16.7% Type IB; 6.5% Type IIB, cadaveric series, only). Our results differ significantly in incidence to that reported in the literature: renal collar 0.2-30%; retro-aortic vein 0.8-7.1%; additional renal vein 0.8-6%. Variations are clinically silent and remain unnoticed until discovered during venography, operation or autopsy. To a transplant surgeon, morphology acquires special significance, since variations influence technical feasibility of operation. Prior knowledge of circum-aortic vein is important when blood samples from suprarenal or renal veins are collected. Collar may provide developed collateral pathway immediately after surgery if renal interruption planned without awareness of its presence. Variations restrict availability of vein for mobilisation procedures. In aortic aneurysm repair, retro-aortic vein is important. During retroperitoneal surgery, the surgeon may visualise a pre-aortic vein but be unaware of an additional retroaortic component or a posterior primary tributary, and may avulse it while mobilising the kidney or clamping the aorta.

Causes and outcome of bladder injuries in Durban.

OBJECTIVE: To document our experience with the management of bladder injuries. DESIGN: A retrospective study. SETTING: King Edward VIII Hospital in Durban, South Africa. PATIENTS: One hundred and twenty patients with urinary bladder injuries. INTERVENTIONS: Patients with intraperitoneal bladder injuries underwent laparotomy whereas those with extraperitoneal bladder injuries were managed non-operatively with a suprapubic catheter. RESULTS: The patients' median age was 28.5 years and the male to female ratio was 5:1. Sixty injuries were due to firearms, seven to stabs and fifty-three were due to blunt trauma. There were other associated injuries in sixty-six patients while fifty patients had isolated bladder injuries. Ninety-four intraperitoneal injuries were repaired while 26 extraperitoneal injuries were managed conservatively. The mortality rate was fourteen per cent (twenty-four per cent for patients with other associated injuries and two per cent for patients with isolated injuries). The mean hospital stay was 18.65 +/- 23.35 days (13.3 +/- 13.5 for firearms 30.56 +/- 33.39 for blunt injuries and 11.00 +/- 2.55 for stabs). CONCLUSION: The majority of bladder injuries were penetrating and the major cause was firearm wounds. Blunt injuries were associated with prolonged hospital stay. Isolated bladder injuries carried a low mortality rate even in patients with delayed diagnosis. Associated injuries were responsible for the high mortality. Suprapubic cystostomy for extraperitoneal bladder injuries led to no complications in our hands.

Hyponatraemia--a prospective analysis of surgical patients.

A prospective study of 71 patients with hyponatraemia was undertaken over an 18-month period in one surgical unit at King Edward VIII Hospital, Durban, to study the incidence and pattern of hyponatraemia. Electrolytes and urea values were measured serially in all patients. Hyponatraemia was defined as a serum sodium level of < 130 mmol/l. The incidence of hyponatraemia was 2.2%, the most common type being normovolaemic hypotonic hyponatraemia. Hyponatraemia was either mild (sodium level 120-130 mmol/l) or moderate (111-120 mmol/l). No patient had severe hyponatraemia (< 110 mmol/l). Hyponatraemia was corrected within 1-6 days using normal saline; in 73% of patients it was corrected within 24 hours. No patient developed neurological symptoms. A mortality rate of 28% was attributed to underlying illness, and hyponatraemia per se was asymptomatic in this study. Aggressive sodium correction was therefore not indicated.

Resympathectomy for sympathetic regeneration.

Explanations for recurrent sympathetic activity after an apparently successful sympathectomy are varied and often tenuous. Among the theories given for recurrent sympathetic activity are the development of alternate neuroanatomic pathways, the possibility of an incomplete operation (failure to appreciate an alternative anatomic pathway at the time of surgery, i.e., nerve of Kuntz), and sympathetic regeneration. The latter, although long suspected, has never been conclusively demonstrated in humans. In this report, a case of recurrent sympathetic activity with conclusive evidence of sympathetic regeneration is described.

Does type of colostomy influence outcome of colostomy closure?

A retrospective analysis of 81 patients who had closure of colostomy over a 32-month period was carried out to establish factors affecting the outcome of this operation. Their ages averaged 27 years and there were 69 male patients. The sigmoid colon was the most common site and the loop colostomy was most frequently performed. The majority were closed 3 or more months after construction. Loop colostomy took significantly less time to close and patients were fed significantly earlier compared with the other types. Patients who underwent closure after Hartmann's procedure had the longest hospital stay. The complication rate was 12% and there was no mortality. Colostomy closures in this study had minimal complications and no mortality. The loop colostomy is as easy to close as it is to perform and results in shorter hospital stay.

Appendicitis among African patients at King Edward VIII Hospital, Durban, South Africa: a review.

This retrospective study of 645 black patients, carried out over a five year period, showed that appendicitis is twice as common in males as in females and that it occurs predominantly in young people (median age 20 years). The classical presentation of periumbilical pain (16%) was outnumbered by right iliac fossa pain (36%) and non-specific pain (27%). The majority perforated (43%) and appendiceal inflammation was second commonest (37%). The negative appendicectomy rate was 8.8% and there was a diagnostic error of 14%. Mortality was two per cent mainly from patients complicated by peritonitis. Hospital stay was 7 +/- 7 days, with the longest stay following peritonitis. A diagnosis of appendicitis should always be entertained in an African patient presenting with an acute abdomen and, where the diagnosis is in doubt, a laparotomy should be performed.

Ileal amoebiasis.

Whilst colonic amoebiasis and its complications are well-documented, ileal amoebiasis has not been described in the literature. We report the cases of two patients with ileal amoebiasis and outline their management and outcome. It is important that surgeons working in areas endemic for amoebiasis be mindful of this entity.