Introduction We have noticed in our endoscopy unit that abnormality on abdominal CT is becoming a frequent indication for colonoscopy.

Aims/Background The primary aim of this retrospective study was to compare the gold standard of colonoscopy with CT findings.

Method Using the Unisoft database of our endoscopy unit we retrospectively reviewed all colonoscopies done over a consecutive 6 months period from July to December. We analysed this data looking at colonoscopy results of those whose indication was abnormal CT scan and the CT scan results of those patients with abnormal colonoscopy. Statistical analysis was performed using Chi square and Fishers exact test on SPSS. A P- value of 0.05 was taken as significant.

Results 856 colonoscopies were performed within the study time period, of these 88 (9.62%) also had CT scans performed within the same month. 46 (52%) CT scans were performed before the colonoscopy. 33 CTs had been reported as showing colitis and 46 as showing tumour. Of 33 with colitis on CT scan only 16 (57%) had colitis confirmed on colonoscopy and 14(48%) were normal (P=0.0003 Chi square, P=0.0005 Fishers exact test) . Of 46 patients with colonic tumour reported on CT scan, 32(70%) had tumour confirmed on colonoscopy, 12 (26%) were normal (P=0.0034 on chi square, P=0.0053 Fishers exact test).While 9 CT scans were reported as normal, on colonoscopy, 6 of these had large colonic tumour and 3 had colitis.

Conclusion This study suggest that CT scan over diagnose colitis but more importantly may under diagnose colonic tumours.