Background: A period of starvation after colorectal resections to allow for resolution of the clinical evidence of ileus has been an unchallenged surgical doctrine until recent times. A prospective randomized trial comparing early feeding to traditional management in patients undergoing open elective colorectal resections is reported. Methods: Patients undergoing elective intraperitoneal colorectal resections without stoma formation were randomized to either an early feeding or control group. The early feeding group were allowed free fluids from 4 h postoperatively progressing to a solid diet from the first postoperative day as they tolerated it. The control group remained nil orally until passage of flatus or bowel motion and were then commenced on fluids progressing to solids over 24-48 h. Results: There were 40 patients in each group well matched for age, sex, type and duration of operation, method of analgesia and mobilization. Thirty-two patients (80%) in the early feeding group tolerated a diet within 48 h. There was no significant difference in the rate of vomiting, nasogastric reinsertion or complications. The early feeding group tolerated a diet, passed flatus, used their bowels, and were discharged from hospital significantly earlier than the control group. Conclusion: Early feeding after elective open colorectal resections is successfully tolerated by the majority of patients, leading to earlier resolution of ileus and hospital discharge.

In this series laparoscopically assisted colorectal surgery was safe and was associated with a low incidence of complications, short hospitalization and a rapid return to preoperative activity levels when compared with open colorectal resections in this age group.

From 1983 to 1990.110 patients with abdominal trauma required laparotomy in the Albury-Wodonga region. Splenic and liver injuries occurred in 50% of cases, and bowel trauma in 20%. The mortality rate was 8.2% (nine deaths) and major postoperative complications occurred in 18 patients (16%). Delay in therapeutic intervention of greater than 4 h from hospital admission led to a statistically significant increase in the complication rate (P < 0.01) despite a lower injury severity score in this group. Delayed repair of bowel injuries in particular led to an 80% major complication rate in survivors. A high index of clinical suspicion and the regular use of diagnostic peritoneal lavage is suggested to avoid such delays in diagnosis and subsequent surgery.

Laparoscopic-assisted colorectal surgery is not associated with a higher incidence of perioperative hypothermia than open colorectal surgery using a standard warming regimen for both groups. On the basis of these results, standard temperature conservation is adequate, even for long, complex laparoscopic procedures.

Estimates of the efficacy of flexible sigmoidoscopy as a colorectal cancer screening modality should take into account the relatively high rate of incomplete studies, particularly in women.

Trauma to the right hepatic artery during biliary surgery can lead to false aneurysm formation. Subsequent rupture into the biliary system, which may occur after a considerable delay, will then result in major haemobilia. This report details five cases referred to the Royal Melbourne Hospital over a 12 month period, four of which followed initial laparoscopic procedures, and emphasizes important management procedures to prevent and treat this previously rare complication.

Post‐traumatic cerebrospinal fluid (CSF) rhinorrhoea is a well recognized complication of closed head injury. Most cases occur soon after injury and a delay in presentation of more than 1 month is unusual. A case is reported of CSF rhinorrhoea presenting 15 years after initial trauma which was complicated by meningitis after 12 months. The management of this condition is reviewed.

Patients with anastomoses at the anorectal ring, with or without anastomotic dehiscence, may develop large presacral collections. Such collections often drain poorly through the anastomosis, leading to chronic sepsis. A novel method of widely draining such collections by "marsupialization" into the bowel lumen with use of an endoscopic stapler inserted transanally is described.