South African Journal of Surgery
On-line version ISSN 2078-5151
Print version ISSN 0038-2361
BACKGROUND: Pancreaticoduodenal injuries are uncommon owing to the protected position of the pancreas and duodenum in the retroperitoneum. Management depends on the extent of injury. This study was undertaken to document outcome of pancreaticoduodenal injuries and to re-evaluate our approach. PATIENTS AND METHODS: A prospective study of all patients treated for pancreaticoduodenal trauma in one surgical ward at King Edward VIII hospital over a 7-year period (1998 -2004). Demographic data, clinical presentation, findings at laparotomy and outcome were documented. Prophylactic antibiotics were given at induction of anaesthesia. RESULTS: A total of 488 patients underwent laparotomy over this period, 43 (9%) of whom (all males) had pancreatic and duodenal injuries. Injury mechanisms were gunshot (30), stabbing (10) and blunt trauma (3). Their mean age was 30.1+9.6 years. Delay before laparotomy was 12.8+29.1 hours. Seven were admitted in shock. Mean Injury Severity Score (ISS) was 14+8.6. Management of 20 duodenal injuries was primary repair (14), repair and pyloric exclusion (3) and conservative (3). Management of 15 pancreatic injuries was drainage alone (13), conservative management of pseudocyst (1) and distal pancreatectomy (1). Management of 8 combined pancreati-coduodenal injuries was primary duodenal repair and pancreatic drainage (5) and repair with pyloric exclusion of duodenal injury and pancreatic drainage (3). Twenty-one patients (49%) developed complications, and 28 required ICU admission with a median ICU stay of 4 days. Ten patients died (23%). Mean hospital stay was 18.3+24.4 days. CONCLUSIONS: The overall mortality was comparable with that in the world literature. We still recommend adequate exploration of the pancreas and duodenum and conservative operative management where possible.