Incidence and causes of rectal bleeding in general practice as detected by colonoscopy

Abstract

Background: Rectal bleeding is common, but it is still unclear which patients require investigation to exclude serious pathology, although it is known that colectoral cancer is very rare under the age of 40 years. Few studies have examined all patients presenting to their primary health physician rather than screening whole populations.

Aim: The aim of this study was to investigate the view that all patients over the age of 40 who present to their general practitioner with rectal bleeding should undergo investigation by colonoscopy to rule out serious pathology, regardless of symptomatology.

Method: A prospective study was carried out of 99 consecutive patients over 40 years presenting with rectal bleeding to 17 general practices in Newcastle upon Tyne.

Results: Serious pathology was detected by colonoscopy in 44.4% of patients. The diagnoses were: colorectal carcinoma, eight cases (two Dukes' grade A, two Dukes' grade B, four Dukes' grade C); one or more polyps, 25 cases (in 17 cases at least one polyp was 5 mm or greater in diameter); inflammatory bowel disease, 11 cases. In the remaining 55 patients, bleeding was associated with diverticular disease (16 cases) and haemorrhoids (28 cases). No cause was found in 11 patients. This high rate of pathology may be partly caused by selection of cases for referral by the general practitioner, despite efforts to minimize this. Three symptoms as elicited by the colonoscopist were found to be significantly associated with serious disease: blood mixed with stool (P < 0.001); change in bowel habit (P < 0.005); and the presence of abdominal pain (P < 0.025). However, symptoms elicited on primary presentation were less helpful and symptoms changed significantly between consultation with the general practitioner and colonoscopy.

Conclusion: All patients over the age of 40 years presenting with rectal bleeding should be referred for flexible sigmoidoscopy or colonoscopy. Symptoms are unhelpful in deciding who requires investigation.