Diagnosis and surgical management of intractable constipation.

Seventy-four patients with intractable constipation, of whom thirty-three had slow and forty-one normal intestinal transit, were investigated to determine the aetiology of their disorder and plan treatment. Patients with slow transit had a greater incidence of abdominal pain and distension (P less than 0.001) and only 9 per cent had a normal call to stool compared with 71 per cent of those with normal transit (P less than 0.001). Internal and sphincter function as assessed by sphincter pressures, length and the recto-anal inhibitory reflex did not reveal any difference between the groups and normal controls; similarly anal sensation and rectal compliance were normal. However, those with normal transit had a higher threshold of rectal sensation than controls (P less than 0.05). Slow transit patients failed to show a postprandial increase in rectosigmoid motility compared with controls (P less than 0.05). Whilst the majority failed to inhibit the external sphincter on bearing down, half of those with normal transit produced either partial or complete inhibition. Both groups were able to increase the anorectal angle on straining. Twenty-two normal transit patients had abnormal perineal descent compared with controls (P less than 0.0005). Patients with perineal descent exhibited abnormal rectal morphology. Rectal intussusception was observed in 13 of 35 evacuation proctograms. On the basis of the data presented, we could not justify internal sphincterotomy of puborectalis division. Our policy in severe slow transit constipation was to offer colectomy and ileorectal anastomosis. In five out of seven to date, a successful result has been achieved. Eight patients with rectal intussusception have undergone an abdominal rectopexy with significant improvement in three. In our hands, the evacuation proctogram and transit studies were the most useful preoperative investigations.