Summary Seventy-one consecutive patients with acute anal fissure were randomly allocated to treatment by lateral subcutaneous sphincterotomy under either local anaesthesia (n=34) or general anaesthesia (n=37). Four months after treatment there were 18 patients with a recurrent or persistent anal fissure, 17 had had local anaesthesia (50 per cent) and only one had a recurrent fissure after general anaesthesia (3 per cent). These results indicate that if lateral subcutaneous sphincterotomy is used for treatment of anal fissure the operation should be performed under a general anaesthetic.

Summary A prospective randomized trial has compared rubber band ligation with phenol injection in 82 patients with symptomatic haemorrhoids. The symptomatic results in all patients 12 months after treatment were indistinguishable, 64 per cent being improved after rubber band ligation, compared with 70 per cent after injection. Complications of treatment were recorded in 5 patients after rubber band ligation, 3 of whom required hospital admission compared with one minor complication after injection. However, repeated treatment was necessary in 13 patients after phenol injection compared with only 4 following rubber band ligation (P<0·05). Additional treatment was necessary in 6 patients from each group. Two-thirds of patients with third-degree piles required additional therapy. It is concluded that rubber band ligation and injection provide similar results, but rubber band ligation is more liable to complications with inexperienced operators.

In a prospective randomized trial of 72 patients undergoing elective colorectal surgery for malignant disease the incidence of post-operative infection after cefoxitin prophylaxis was 27 per cent compared with 31 per cent after metronidazole and gentamicin prophylaxis. Sepsis from Bacteroides species occurred in five patients after cefoxitin (one septicaemia) compared with only one after metronidazole and gentamicin. Sepsis with aerobic bacteria was more common in patients who had received pre-operative oral mannitol for bowel preparation. Oral mannitol bowel preparation was associated with higher counts of Escherichia coli in the colon than other methods of bowel preparation. Five cases of antibiotic associated colitis occurred, all in patients receiving cefoxitin.

A prospective randomized trial compared anal stretch with or without continued dilatation for three months in 89 consecutive patients with haemorrhoids. Complications of therapy occurred in only 4 patients, 3 of whom had transient incontinence of flatus. When patients were reviewed four months after treatment, only 4 of 44 patients (9%) who used a dilator had not been improved and required additional therapy, compared with 15 of 42 patients (36%) who had anal stretch alone (P <0.02). It is concluded that the use of an anal dilator improves the results of anal stretch.