The ArtAssist Device in chronic lower limb ischemia. A pilot study

BACKGROUND

Patients with chronic critical limb ischemia following a failed bypass graft or with non-reconstructable distal disease diagnosed angiographically, have a very poor prognosis. This is a prospective pilot study to assess the influence of the ArtAssist Device on pedal blood flow and amputation rate.

METHODS

Thirty-three legs in 25 patients were evaluated. Ten legs presented with rest pain, and 23 legs with tissue loss. Nine legs had previously undergone bypass surgery.

RESULTS

At a mean follow-up of 3 months, 14 (42%) legs were amputated, and 19 (58%) were saved. Eleven of the amputated legs were in patients with chronic renal failure, a known risk factor. The amputation rate, excluding this group, was 13.6% (3/22). Toe pressures measured initially and after 3 months on the pump showed a significant improvement (p=0.03). Forty percent of patients presenting with rest pain improved, while 26% of foot ulcers healed on the pump. Mortality rate was 12%.

CONCLUSIONS

The results from this prospective study are encouraging but need to be validated in a larger prospective randomized study.

Gunshot injuries of the extrahepatic biliary ducts

OBJECTIVE

To assess our management of gunshot injuries of the extrahepatic biliary ducts.

DESIGN

Retrospective study.

SETTING

Urban teaching hospital, South Africa.

SUBJECTS

17 patients who were found to have gunshot injuries of the extrahepatic bile ducts, January 1993-June 1998.

INTERVENTIONS

Packing and damage control, staged repair, or definitive repair.

MAIN OUTCOME MEASURES

Morbidity and mortality: Three of the 17 died intraoperatively or postoperatively from causes unrelated to the biliary injury. Of the 14 survivors, 3 were managed initially with packing and damage control, 7 by staged repair, and 4 by definitive biliary repair. Eleven patients had a Roux-en-Y biliary jejunostomy, of whom 1 developed a late biliary stricture related to a postoperative anastomotic leak. End-to-end anastomosis of the bile duct was done for 2 patients and they both developed late biliary strictures. Ligation of the injured duct and cholecystojejunostomy was done for I patient.

CONCLUSION

Roux-en-Y biliary-jejunal anastomosis is appropriate in the treatment of gunshot injuries of the extrahepatic biliary ducts as there is always a degree of tissue loss and some debridement is required, making it difficult to do a tension-free anastomosis.

Débridement of gunshot wounds: semantics and surgery

Débridement is a well established modality for management of gunshot wounds. The word "débridement" is originally French. It was used for the first time during the eighteenth century in the surgical context and meant "wound incision." For French surgeons, it has retained to this day its original meaning. In medical English, though, the use of the term has been marred by persisting confusion about its definition. Two quite different surgical procedures still compete for the definition of débridement: wound incision and wound excision. These procedures are also at the center of a modern controversy about the management of gunshot wounds. The orthodox doctrine, inherited from military surgeons, consists of aggressive tissue excision around the bullet track. This radical policy is being challenged by advocates of a more conservative approach. Minimal tissue excision is sufficient and safe in many cases provided careful monitoring of the wound is instituted. Wound incision alone to relieve tension and allow drainage is possible in certain cases. The tug-of-war between excision and incision is outlined herein with reference to the semantic tribulations of the word "débridement" and the implications for patient care.

Transmediastinal gunshot injuries

BACKGROUND

Transmediastinal gunshot injuries are a rarely reported injury. Until recently, assessment of the thoracic aorta by angiography preceded the investigation of the esophagus. This order has been recently debated.

METHODS

There were 118 patients with potential transmediastinal injuries included in this retrospective study. Unstable patients who were unresponsive to resuscitation were taken to the operating room without previous investigation. Stable patients were routinely investigated initially for injury of the aorta and then for injury of the esophagus.

RESULTS

There were 51 patients who underwent urgent thoracotomy/sternotomy. In 27, the hemorrhage was of mediastinal origin; 17 of these patients died of intraoperative bleeding. Eight of the patients had aortic injury, and only one of this group survived. There were 57 stable patients who were investigated initially for injury of the aorta by angiography. It was positive in only one patient who underwent an operation with good results. An investigation of the esophagus followed and revealed esophageal injury in 17 patients. All of them were treated operatively, 15 of them with satisfactory outcome.

CONCLUSIONS

Angiography should at present precede esophageal investigations. There is a need for shortening the time between admission and operation. Other modalities that could expedite the investigation of the thoracic aorta and the esophagus should be prospectively evaluated in multi-center studies.

Non-operative treatment of abdominal gunshot injuries

BACKGROUND

Non-operative treatment is a management option that challenges the traditional mandatory laparotomy for abdominal gunshot injuries.

METHODS

All published relevant clinical reports were retrieved by searching through the Medline database and manually. The theoretical arguments in favour of non-operative management as well as the results of the reviewed reports are analysed and evaluated.

RESULTS AND CONCLUSION

Patients with proven non-penetration of the abdominal cavity can be offered conservative treatment with a satisfactory outcome. Greater caution should be exercised in the presence of a documented visceral injury until the safety of this option has been established by further clinical trials.

Controversies in management of penetrating injuries of the pancreas

The management of penetrating pancreatic injuries is still beset by controversy, especially with regard to the role of pancreatography, the most appropriate operation, the type of drainage and the adjunctive role of ocreotide. The recommendations of various experts are outlined and evaluated in the light of the authors' personal experience. Consensus is easily achieved for very minor and very severe injuries. Most management discussions center around those injuries resulting in parenchymal disruption of a degree of severity sufficient to raise doubts about the integrity of the main pancreatic duct.

Multiple organ failure. How valid is the "two hit" model?

Inflammatory "one hit" and "two hit" models have recently been proposed to account for the development of multiple organ failure (MOF) in trauma and critically ill surgical patients when no source of infection can be found. In the "one hit" model, the initial insult is so massive that a systemic inflammatory response syndrome is triggered and leads rapidly to MOF. In the "two hit" scenario, initially less severely injured patients eventually develop MOF as a result of a reactivation of their inflammatory response caused by an adverse and often minor intercurrent event. At first sight, the theory is attractive because it seems to fit commonly observed clinical patterns. Indeed, injured patients often respond to initial resuscitation but, after an insult of some sort, develop organ dysfunction and die. The "two hit" model is furthermore mirrored at the cellular level. Inflammatory cells are indeed susceptible of being primed by an initial stimulus and reactivated subsequently by a relatively innocuous insult. However, in the absence of clinical and biological corroboration based on cytokine secretion patterns, these models should not be accepted uncritically.

Organisation of hospital responses for the trauma epidemic

The caseload of the Department of Surgery (Baragwanath Hospital, Johannesburg, South Africa) is characterised by a singularly heavy trauma component. Penetrating injuries account for the majority of cases. The management of penetrating trauma is not as demanding as that of blunt trauma, yet the sheer number of cases, over the past four decades, has imposed a clinical burden that has never been met by commensurate resources. The organisation of the hospital and departmental responses has manifested itself on two different levels. The first one is the structural and functional deployment of insufficient staff, facilities and equipment to cope as flexibly as possible with the trauma epidemic. The second one is the gradual adaptation of the clinical management philosophy to ensure that the best possible quality of care is provided to the majority of trauma victims, with the full knowledge that better resources would sometimes have elicited a different clinical approach.

Good clinical research practice: what is it and is it possible in the intensive care unit?

With the growing quest for answers to vexing dilemmas in critically ill patients, more intensive care units are embarking on clinical research. This places increasing importance on Good Clinical Research Practice (GCRP), a set of guidelines drawn up by the Pharmaceutical Industry to assist investigators in conducting ethical, reliable scientific studies. GCRP is a combination of good basic management skills allied to ethical principles for scientific research. Based on the principles of the Declaration of Helsinki, it consists of three tenets: patient protection (ethics), credible data (science) and data control. This article describes GCRP specifically relating it to the intensive care situation, illustrating some of the concepts with practical examples. With a minimum of extra time and effort these basic principles can be integrated as routine into all research projects.

Laparoscopic inguinal hernia repair by an intraperitoneal onlay mesh technique using expanded PTFE: a prospective study

This prospective trial was initiated to assess laparoscopic inguinal hernia repairs by an intraabdominal intraperitoneal onlay mesh (IPOM) technique. An IPOM method utilizing 10 x 7.5-cm expanded polytef (PTFE) patches was used to repair 52 inguinal hernias in 50 patients. There were two patients with bilateral defects. Fourteen were direct and 38 were indirect hernias. There was one patient with bladder perforation, one with strangulation of the small bowel that entered the sac of a recurrent defect, and five patients with other minor complications. There were two recurrences. The follow-up period ranged from 7 to 31 months. The operating time ranged from 35 to 180 min and was 35-60 min in 40 patients. The postoperative pain was minimal in 40 patients. Our conclusions are as follows: This IPOM method was less time consuming in theater time than other laparoscopic methods. There was minimal postoperative pain in the majority of cases. The hospital stay was shorter in comparison with the standardized stay for conventional methods. The case with recurrence and strangulation raises some questions as to the safety of this method. Hospital costs of conventional herniorrhaphies were lower (63%) than those of this laparoscopic method.

Pelvic hemangio-pericyptoma: A case report

A large pelvic hemangiopericyptoma causing intestinal obstruction is reported.

Management of combined penetrating cardiac and abdominal trauma

This is a retrospective review of patients with combined penetrating cardiac and abdominal trauma. Clinical presentation patterns are described and a management strategy is outlined. The series comprises 25 patients. On the basis of the mechanism of injury and the prognosis, the patients were divided into two groups: 'low risk' (single high epigastric stab wound) and 'high risk' (multiple stabs, single or multiple gunshot wounds). There were six patients in the low-risk group. Their intra-abdominal injuries were moderately severe. None of this group died. There were 19 patients in the high-risk group. Three underwent emergency-room thoracotomy and died. Of the remaining patients, four underwent a thoracotomy first for cardiac tamponade or massive haemothorax and 12 underwent a laparotomy first because of massive haemoperitoneum. The mortality in this group was 63 per cent. It is essential to recognize the cardiac injury in low-risk patients; the cardiorrhaphy must be performed before the laparotomy. In high-risk patients, the sequence of operations depends on the clinical presentation. Obvious cardiac tamponade or massive haemothorax mandate a thoracic approach first, while severe hypovolaemic shock with a massive haemoperitoneum justifies the performance of a laparotomy first; a transdiaphragmatic pericardiotomy is useful, in these cases, before proceeding to median sternotomy.

Management of penetrating injuries of the cervical trachea

This is a report of a 2-year experience with the management of penetrating injury of the cervical trachea. There were 29 cases. The respiratory status of the patient on admission dominated the initial management: 12 patients required emergency intubation and were immediately taken to operation, while 17 patients were more stable and could be subjected to the preoperative assessment of the oesophagus. Associated injuries were significant and dominated the postoperative morbidity and mortality. Primary repair of the trachea, without tracheostomy, was successful in a relatively high proportion of patients (55%).

Gunshot injuries of the abdominal aorta: a continuing challenge

This is a retrospective study of 57 patients with gunshot injuries of the abdominal aorta. The aortic repair was achieved by various means: lateral aortorraphy, patch aortoplasty and graft insertion. There was an 85 per cent mortality rate from bleeding or secondary coagulopathy directly related to the aortic injuries. The need for resuscitative thoracotomy, shock, lack of response to fluid resuscitation and intraperitoneal bleeding were directly related to mortality. We feel that shortening of prehospital transfer time will increase the absolute number of patients surviving this grave injury.

Penetrating injuries of the abdominal inferior vena cava

This is a retrospective study of 74 patients with penetrating injuries of the abdominal inferior vena cava; the cause of injury was gunshot in 91% and stabbing in 9%. Of the patients, 77% underwent lateral venorrhaphy, 5% underwent infrarenal ligation of the inferior vena cava (IVC), and 18% died perioperatively before any caval repair could be carried out. There was an overall perioperative mortality of 39%. Persistent shock, the site of the venous injury, particularly in the retrohepatic position, and the number of associated vascular injuries were directly related to mortality. Irrespective of the improvements in resuscitation and the various operative methods available, penetrating trauma of the abdominal IVC remains a life-threatening injury.

Selective surgical management of zone II gunshot injuries of the neck: a prospective study

BACKGROUND

The management of penetrating neck trauma remains controversial, with many studies supporting either mandatory exploration or selective conservatism.

METHODS AND RESULTS

This is a prospective study of 75 patients with gunshot injuries to zone II of the neck. Forty patients (53.3%) underwent immediate exploration because of clinical indications or positive initial investigations. A 7.5% incidence of unnecessary explorations, a 5% mortality rate, and average hospital stay of 10.5 days were noted. Thirty-five patients with negative clinical or investigational findings underwent observation with constant monitoring. A 5.7% incidence of missed injuries, no mortality, and an average hospital stay of 3.5 days were noted for these patients.

CONCLUSIONS

We suggest that conservative management in gunshot injuries confined to zone II of the neck selectively supplemented by appropriate investigations is a viable proposition in this type of injury. Further contemporary studies reporting specifically on this injury will enable us to reach statistically significant conclusions.

Gunshot arterial injuries to the groin: comparison of iliac and femoral injuries

In a prospective study of 69 patients with gunshot arterial injuries to the groin, there were 18 patients with iliac and 51 with femoral artery injuries. In the former group, the majority of patients were admitted with gross physiological derangement that did not respond to pre-operative resuscitation, so were taken directly to theatre. There was a 39 per cent peri-operative mortality. A prompt operation is mandatory to improve the chances of survival of patients with this injury. In the latter group all patients responded to pre-operative resuscitation and in the great majority, the diagnosis was established by the presence of 'hard' signs of arterial injury alone. There was a 4 per cent amputation rate and no mortality.

Perforation of the gastro-intestinal tract by a foreign body. A case report

Most ingested foreign bodies pass through the gastro-intestinal tract without giving rise to complications. If they become lodged in a narrow segment of the gastro-intestinal tract, perforation may occur. The resulting morbidity depends on the further route of the penetrating object and whether septic sequelae ensue. This article describes an unusual case of foreign body perforation with a protracted clinical course. It emphasises that there are no insurmountable barriers in the pathway of foreign bodies.

Chondroid syringoma of the abdominal wall. A case report and review of the literature

A case of an unusually large chondroid syringoma of the abdominal wall is described. Chondroid syringomas are rare neoplasms believed to be of sweat gland origin. They are often slow-growing and benign. Local recurrence and distal spread characterise the malignant form.

Volvulus of the sigmoid colon at Baragwanath Hospital

Fifty-six patients treated for sigmoid volvulus over a period of 5 years are reviewed. The purpose of this study is to evaluate our management policy and to compare it with other studies, particularly from other parts of Africa. The patients were evaluated by retrospective allocation into one of three treatment modalities. Patients with clinical evidence of ischaemia underwent a Hartmann's procedure. There was a 33% mortality rate in this group. In those patients who had sigmoidoscopic reduction, a second-stage sigmoidectomy and primary anastomosis were performed. The results in this group were excellent. In the group of patients with failed sigmoidoscopic resection the majority underwent a Hartmann's procedure with a mortality rate of 15%.

Penetrating injuries of the iliac arteries: a South African experience

BACKGROUND

We did a retrospective study of 62 patients with penetrating injuries of the iliac arteries.

METHODS

The cause of injury was gunshot wound in 85.5% and stabbing in 14.5%. The arterial repair was achieved by various means: lateral arteriorrhaphy, end-to-end anastomosis, and polytetrafluoroethylene interposition grafts.

RESULTS

There was a 42% mortality rate from exsanguination or secondary coagulopathy directly related to the arterial injury. Persistent shock, resuscitative thoracotomy, free intraperitoneal hemorrhage, and the number of vascular injuries were directly related to mortality.

CONCLUSIONS

A high index of suspicion, aggressive resuscitation, and prompt surgery are necessary to improve the chances of surviving this ominous injury.

Gunshot injuries of the head of the pancreas: conservative approach

This study is comprised of 48 patients with gunshot injuries of the head of the pancreas, many of which were high velocity injuries. The purpose of this study was to evaluate our management policy for these injuries based on our recent wide experience. Patients with grade II and III injuries underwent conservative surgery, with 0% and 21% postoperative mortality, respectively, directly related to the pancreatic injury. For patients in whom the duodenum was involved, pyloric exclusion was applied depending on the grade of the duodenal injury. We concluded that moderate gunshot injuries of the head of the pancreas (grade II) can be safely treated by débridement and suture repair, with or without drainage. Severe (grade IV) injuries warrant a pancreaticoduodenectomy. Most grade III injuries can be treated by débridement and drainage unless an associated severe duodenal injury is present, in which case resection may be indicated.

Diaphragmatic herniation after penetrating trauma

A study was made of 45 patients with diaphragmatic herniation after penetrating trauma. In 29 the diagnosis was established during the first admission (early presentation) and in 16 during a subsequent admission (delayed presentation). The mortality rate in the early presentation group was 3 per cent compared with 25 per cent in the delayed presentation group. The presence of gangrenous or perforated abdominal viscus in the chest cavity was the single most common and severe aggravating factor. The need for diagnosis of diaphragmatic herniation during the initial admission is emphasized. As isolated diaphragmatic injuries provide few helpful clinical features to aid diagnosis, appropriate investigations and good follow-up are of paramount importance in preventing late herniation of intra-abdominal viscera through a penetrating diaphragmatic injury.

Early closure of colostomies in trauma patients--a prospective randomized trial

BACKGROUND

Most traumatic colon injuries can be repaired primarily, but a colostomy may still be required for severe colonic or rectal injury. The current trend is to reverse the colostomy early, rather than to wait the traditional 3 months before closure.

METHODS

Forty-nine patients with colostomies after abdominal trauma were entered into the study. All patients had undergone a contrast enema in the second postoperative week to assess distal colon healing. Patients were excluded from early closure for nonhealing of the bowel injury, unresolving wound sepsis, or an unstable condition. We then compared the outcome of the remaining 38 (77.6%) patients allocated to either an early or a late colostomy group in a controlled, prospective, randomized trial.

RESULTS

We found no significant difference in morbidity between the two groups, with an overall complication rate of 26.3%. Technically the early closure of colostomies was far easier than late closure and required significantly less operating time (p = 0.036) and with less intraoperative blood loss (p = 0.020). The closure of end colostomies was more time consuming, both early (p < 0.001) and late (p < 0.001) and caused more bleeding (p < 0.001 and p < 0.001, respectively). Total hospitalization was marginally shorter overall for early closure, but late closure of end colostomies resulted in prolonged hospitalization (p = 0.023).

CONCLUSIONS

The early closure of colostomies and the use of loop colostomies whenever possible are recommended as both safe and beneficial for patients with colonic injury after trauma. Contraindications for early closure include nonhealing distal bowel, persistent wound sepsis, or persistent postoperative instability.

Clinicopathological trends in colorectal carcinoma in a Black South African population

The clinico pathological characteristics of colorectal carcinoma in 127 Black South African patients were studied. The main presenting symptoms were altered bowel habits in 70% of the patients, weight loss in 64.3% and abdominal pain in 47.1%. Anaemia was present in 75.7%. There was a 31% incidence of mucinous carcinoma with a particular predilection for the younger age groups. Mucinous tumors were found more commonly in Duke stage C and D than in earlier stages. Tumors arising from a pre-existing adenoma constituted to 5.5% of the lot of patients with cancer colon.

Distal pancreatectomy for gunshot injuries of the distal pancreas

This study comprised 57 patients with gunshot injury of the distal pancreas. There were 16 grade II, 29 grade III and 12 grade IV pancreatic injuries. The intraoperative mortality rate was 12 per cent. Patients with the most severe grade II injuries and all those with grade III and IV injuries had distal pancreatectomy and splenectomy, with a 14 per cent fistula formation rate and 2.3 per cent postoperative mortality rate directly related to the pancreatic injury. The remaining patients with grade II injuries were managed by debridement and drainage; there were no fistulas or deaths. The method of closure of the pancreatic resection margin is unrelated to fistula formation, and identification of the duct for ligation is unnecessary. Liberal use of distal pancreatectomy with splenectomy for gunshot injuries of the distal pancreas is suggested.

Arterial gunshot injuries of the extremities: a South African experience

This is a retrospective study of 173 patients with gunshot injuries of the major arteries of the extremities. A selective policy for the use of angiography was followed. The arterial repair was achieved by various means: primary end-to-end anastomosis, vein interposition graft, and polytetrafluoroethylene (Teflon) interposition grafts. Overall, there were nine amputations in the lower limb. Ninety-eight percent of the patients had a palpable pulse on discharge. There were 3 preoperative deaths, 1 intraoperative death, and 5 postoperative deaths (overall perioperative mortality, 3.5%). We conclude that the results of vascular extremity gunshot injuries are satisfactory when standard methods of management are used. Morbidity and mortality can be further reduced by prompt admission to appropriate centers.

Outcome of a strict policy on emergency department thoracotomies

OBJECTIVE

To audit emergency department thoracotomies from January 1981 to May 1993.

DESIGN

Retrospective analysis of case records.

SETTING

A large (3000-bed) tertiary care academic hospital; the department of general surgery (including trauma) consists of 360 beds.

PATIENTS

All patients who underwent a thoracotomy in the emergency department during the above period.

INTERVENTION

An emergency department thoracotomy was performed on trauma patients with recordable vital signs and rapid deterioration and on patients with uncontrollable bleeding or profound hypotension not responsive to resuscitation. The procedure was performed either on the resuscitation trolley in the emergency department or in the adjacent operating room.

MAIN OUTCOME MEASURES

Survival and subsequent neurological function after thoracotomy.

RESULTS

There were 312 stab injuries, 358 gunshot injuries, and 176 blunt injuries. Survival occurred in 26 stab-wound cases (8.3%), in 16 gunshot cases (4.4%), and in one blunt injury case (0.6%). There was one patient with neurological impairment in each of the three injury groups. Those with penetrating chest injuries had the best survival rate (20%), and the survival rate for penetrating abdominal trauma was 6.8%.

CONCLUSIONS

Emergency department thoracotomies have a definite role in the management of trauma patients. The best results are obtained in patients with penetrating chest injuries.

Penetrating injuries of the axillary artery

This is a retrospective study of 32 patients with penetrating injury of the axillary artery. There was an overall mortality of 6% entirely accounted for by associated injuries. Pre-operative angiography was used in 12 of these patients either to confirm the presence of an injury or to define its location. Twelve patients underwent lateral arteriorrhaphy or an end-to-end anastomosis and 19 patients had an interposition graft. No immediate problems were experienced with polytetrafluoroethylene grafts compared with autogenous vein grafts. There were 14 patients with a concomitant venous injury; 13 were repaired and only transient arm oedema was experienced. Eleven patients had a brachial plexus injury and, of these, nine underwent a secondary nerve repair with a poor outcome. Axillary artery injury has a good prognosis with a morbidity related mainly to associated nerve injury and a mortality accounted for by injuries to other body systems.

Penetrating injuries of the brachial artery

This is a retrospective study of 72 patients with non-iatrogenic penetrating injuries of the brachial artery treated between 1991 and 1993. The mechanism of injury was stabbing in 39 per cent, a bullet in 51 per cent, pellets in 4 per cent and a dog bite in 6 per cent. We believe that angiography is rarely indicated, as in all our patients the diagnosis was made on clinical grounds. The proximity of the injury to neurovascular bundles was a poor predictor of arterial injury. One month after discharge 95 per cent of the patients had palpable distal pulses. Long-term morbidity is mainly attributed to associated nerve injuries.

Gunshot injuries of the liver: the Baragwanath experience

BACKGROUND

This study comprised 304 patients with gunshot injuries of the liver, many of which from high-velocity firearms. The purpose of this study is to evaluate our management policy in gunshot injuries of the liver in light of our recent wider experience.

METHODS

All grade I and II injuries and most grade III injuries were managed by simple operative measures, without postoperative mortality directly related to the liver trauma.

RESULTS

Grade III, IV, and V injuries had 8.5%, 52%, and 16% resectional debridement rates and 8.5%, 38%, and 84% perihepatic packing rates, respectively. In the resectional debridement group the postoperative mortality rate was 15% (half the deaths were directly caused by the hepatic injury). The postoperative mortality rate in the perihepatic packing group was 31.5% of which 45% of deaths were due to ongoing bleeding, 27.5% to sepsis, and 27.5% to associated trauma. The septic complications were less common when packs were removed early.

CONCLUSIONS

We suggest that resectional debridement and perihepatic packing should be liberally applied in the most severe grade III, most grade IV, and grade V gunshot injuries of the liver and that perihepatic packing should be removed as early as the physiologic derangements are corrected. Our experience with grade VI injuries is very limited, and their management should be studied in larger series.

Penetrating injuries of the femoral artery

This is a retrospective study of 106 patients with penetrating injuries to the femoral artery. The cause of injury was gunshot in 82 per cent, stabbing in 13 per cent and pellets in 5 per cent. We believe that angiography is unnecessary in patients with 'hard' signs of vascular injury: 97 patients underwent immediate exploration with positive findings. The presence of 'soft' signs alone mandates angiography as five of 12 patients with suspected injury were saved a negative exploration. Proximity of injury was a poor predictor of arterial injury. There were four amputations. Ligation of the femoral vein is not reliably associated with amputation but it was an aggravating factor in the setting of severe arterial disruption and delay to surgery.

Leiomyosarcoma of the breast. A case report

A case of primary leiomyosarcoma of the breast in a 35-year-old woman is presented. The condition is rare, only 11 cases of primary leiomyosarcoma of the breast having been reported previously. The clinical and pathological features are reviewed.

Changing profiles in spinal cord injuries and risk factors influencing recovery after penetrating injuries

OBJECTIVE

The changing profiles of spinal cord injuries in South Africa are addressed in this study.

DESIGN

A retrospective analysis of 551 patients with spinal cord injury.

MATERIALS AND METHODS

The cause of injury was motor vehicle crashes in 30%, stab wounds in 26%, gunshot wounds in 35%, and miscellaneous causes 9%.

MEASUREMENTS AND MAIN RESULTS

There was a significant shift from stab wounds towards bullet wounds over the last five years. Bullet spinal cord injuries increased from 30 cases in 1988 to 55 cases in 1992, while stab spinal cord injuries decreased from 39 cases in 1988 to 20 cases in 1992. The incidence of spinal cord injuries following a motor vehicle crash showed a declining tendency after a transient increase (28 cases in 1988, 40 in 1990, 31 in 1992). Moreover, the problem of severe septic complications has been investigated and various risk factors for sepsis that might impair the rehabilitation process have been examined. The risk of developing septic complications was higher in gunshot spine injuries (21 cases out of 193) than in knife injuries (5 cases out of 143). The presence of a retained bullet did not seem to increase the chances for sepsis. In seven patients the sepsis was the direct consequence of the retained bullet while in 14 patients sepsis developed with no bullet in situ. Furthermore, the site of the injury (cervical, thoracic, lumbar spine) did not correlate with the abovementioned risks.

CONCLUSIONS

Gunshots carry a heavier prognosis. Only 32% of our gunshot cases underwent a significant recovery as opposed to 61% of stab cases and 44% of the motor vehicle crash victims.

Extraperitoneal rectal gunshot injuries

Extraperitoneal rectal gunshot injuries are rare, but may be encountered in civilian practice. We report on a series of 26 such cases. The aim of the study is to attempt to evolve a treatment policy of this injury. The principles of management include the repair of rectal wound in selected cases and the formation of a diverting colostomy. Distal rectal washout and presacral drainage, although advocated by some authors, do not seem to be indispensable adjuncts to the management of these injuries.

Selective surgical management in penetrating neck injuries

OBJECTIVE

To evaluate selective operative management in penetrating neck trauma.

DESIGN

A chart review.

SETTING

A university-affiliated hospital in Johannesburg, South Africa.

PATIENTS

All adults admitted to the hospital between January 1988 and June 1993 with a penetrating neck wound. Excluded were patients in whom there was no suspicion of an occult injury that might need further investigation. There were 755 patients in the study.

INTERVENTIONS

Immediate surgical exploration (group A, 613 patients) and observation with constant monitoring (group B, 142 patients).

MAIN OUTCOME MEASURES

Unnecessary explorations in group A and missed significant injuries in group B.

RESULTS

In group A there was a 3% incidence of unnecessary explorations, and 4.2% of the patients died. In group B there was a 9.1% incidence of missed injuries, and 2.8% of the patients died as a result of the delayed diagnosis. Overall the death rate was 4%.

CONCLUSION

Selective operative intervention for penetrating neck trauma results in fewer negative explorations and a death rate comparable to those of series that support mandatory neck exploration.

Stool weight and faecal pH in two South African populations with a dissimilar colon cancer risk

Bowel diseases (appendicitis, diverticulitis and colon cancer) in South African whites and populations of prosperous Western countries are similar. Among rural South African blacks with a traditional lifestyle these diseases are very uncommon or almost unknown. A study was undertaken to measure stool weight and faecal pH in South African urban blacks and whites. There was no difference in stool weight, but the faecal pH was significantly higher in whites (P = 0.002).

Penetrating cardiac injuries: clinical classification and management strategy

The management of penetrating cardiac injury is controversial. To facilitate decision making, a simple clinical classification of patients with such an injury is proposed. Five categories are considered: (1) lifeless, (2) critically unstable, (3) cardiac tamponade, (4) thoracoabdominal injury and (5) benign presentation. Investigation, if indicated, and the timing and setting of surgical intervention are discussed for each category.

Penetrating injuries of the popliteal artery: the Baragwanath experience

This study describes the management of 43 patients with penetrating injury of the popliteal artery. Of these patients, 33 (76.5%) had bullet wounds, four patients (9.5%) pellet wounds and 6 (14%) knife wounds. Patients with 'hard' signs of arterial injury underwent exploration without preoperative angiograms. There were no negative explorations. Patients with only 'soft' signs of arterial injury underwent preoperative angiograms. Of this group, 75% had positive angiograms and underwent exploration. There were no false-positive or false-negative preoperative angiograms in the group of patients with 'soft' signs in this study. Definitive orthopaedic management of associated fractures followed vascular reconstruction. There was no difference in the short-term patency of autologous saphenous vein graft as against PTFE grafts. Fasciotomy was performed on patients who had arterial and venous injury or presented late. Overall amputation rate was 14% and for bullet injuries 18%.