Audit of sepsis in operations for inflammatory bowel disease

We report the results of a prospective audit of the rates of postoperative infection in patients having operations for inflammatory bowel disease. Apart from a single prospective controlled trial, all other groups have been studied sequentially using the original placebo control group for comparison. The rate of abdominal wound sepsis when no antibiotic was used was 37 per cent. This was reduced to 23.3 per cent with 24-hour cover using metronidazole and gentamicin. However, only after prolonged use of metronidazole and gentamicin for five days was there a significant reduction in abdominal wound infections to 13.3 per cent. Prophylaxis, using 24-hour cover with metronidazole combined with five-day therapy with mezlocillin, achieved an abdominal wound infection rate of 15.6 per cent. The most recent group of patients studied have received 24-hour cover with metronidazole and five-day exposure to latamoxef. In the last group the rate of abdominal wound infections was only 13.5 per cent and serious postoperative bleeding was recorded in eight patients (15 per cent) compared with serious bleeding is only three of the remaining 129 patients (2.3 per cent). The only patients in whom streptococcal isolates were eliminated were those receiving metronidazole and mezlocillin. The majority of infections was due to antibiotic-sensitive strains of Escherichia coli, Proteus, and Staphylococcus species.