Rectal bleeding is a relatively common presenting complaint, with an estimated 40-50 per cent of the population experiencing it at some point.
The mean consultation incidence rate is around 7 per 1,000 patients. In a study of patients presenting with rectal bleeding, only 3.6 per cent were diagnosed with colorectal cancer.1
The majority had relatively minor underlying causes, such as haemorrhoids, anal fissure, gastroenteritis, trauma or anticoagulant/NSAID therapy.
These complaints can be remedied with advice and simple pharmacological therapies. Trauma may need further assessment.
Of the remainder, excluding colorectal cancer, the following should be considered as differential diagnoses: diverticular disease, IBD, villous adenoma, colonic carcinoma or coeliac disease.
|Red flag symtptoms|
It should be remembered that 80 per cent of rectal tumours are within fingertip range, so a rectal examination is essential in patients with rectal bleeding.
If blood is seen on the surface of toilet paper, or on the top of the motion, the cause of the bleeding is likely to be able to be seen via proctoscopy.
Blood mixed with the stool is likely to need a further referral for investigation. The presence of blood mixed with stool in young or middle-aged adults suggests gastroenteritis (e.g. campylobacter) or colitis.
Useful investigations in primary care include abdominal examination and digital rectal examination, FBC, ESR and CRP. Stool studies, such as faecal occult blood, and microscopy, culture and sensitivity, and ova, cysts and parasites, are necessary along with coeliac studies and a clotting screen.
Other investigations may be necessary if a more sinister cause is suspected. Consider proctoscopy, sigmoidoscopy and colonoscopy. A barium enema or CT bowel studies may also be required. Refer for fast-track/formal surgical assessment for palpable tumours if necessary.
- Dr Tinsley is a salaried GP in Bradford, West Yorkshire
1. Robertson R, et al. Predicting colorectal cancer risk in patients with rectal bleeding. Br J Gen Pract 2006; 56: 763-7.