Colonoscopic findings in patients with lower gastrointestinal bleeding send to a hospital for their study. Value of clinical data in predicting normal or pathological findings.

BACKGROUND

Lower gastrointestinal bleeding is a highly frequent clinical problem that may reflect serious pathology in the colon. Colonoscopy is generally accepted as the diagnostic procedure of choice. Decisions as to whether to carry out colonoscopy or not, are not well defined.

METHODS

536 colonoscopies, made to discover the cause of lower gastrointestinal bleeding were analyzed and a final 457 included in the study. All of these patients came to the hospital because they had presented at least in one occasion, one episode of rectal bleeding, and were send by the specialist of the zone, in order to achieve a correct diagnosis of its process. In all cases the following associated symptoms were analyzed: the presence of diarrhea, constipation, abdominal pain and rectal mass on examination. The characteristics of lower gastrointestinal bleeding were analyzed in a subset of 150 consecutive patients.

RESULTS

Mean age was 59 +/- 16.9 years. 54.5% were male and 45.5% female. The exploration was normal until the cecum in 146 patients (32%). In the remaining 311, the findings were: polyps (25.1%), diverticular disease (24%), neoplasia (12.6%), inflammatory bowel disease (9.4%), unspecific proctitis (2.4%), ischemic colitis (2.4%), angiodysplasia (1.9%), infectious colitis (1.1%), and miscellaneous (0.7%). An age of less than 40 years and the existence of anal pathology were significantly more frequent among patients with a normal examination (p < 0.001), but with a sensitivity of only 66%. No differences were found among patients with disordered bowel frequency or abdominal pain in relation to the colonoscopic findings. Previous presence of a rectal mass when the examination proved abnormal (p=0.06). Intermittent bleeding and the presence of blood in the stools were more frequent in patients with normal examination (p= 0.07 and p< 0.05, respectively). No significant differences in relation to colour, duration of bleeding, or to whether toilet paper was stained with blood were found.

CONCLUSIONS

1. The more frequent endoscopic findings were polyps and diverticular disease. 2) Clinical data are of little value in predicting a normal examination. 3) Total colonoscopy appears to be the first procedure of choice in all patients with lower gastrointestinal bleeding, irrespective of the clinical data and the presence of anal pathology.