BACKGROUND AND AIMS There has been a significant proximal shift in the distribution of colorectal cancer (CRC) in Northern Ireland over recent decades. The aim of this study was to investigate the potential implications of this proximal shift in CRC distribution on the efficacy of flexible sigmoidoscopy (FS) as a screening tool.

PATIENTS AND METHODS The site distribution of 5153 CRCs was available from the Northern Ireland Colorectal Cancer Register for the period 1990–1997. Similar data on 1241 CRCs between 1976 and 1978 were available from a previous study. Data on the site reached by FS were obtained from a prospectively collected endoscopy database at one of Northern Ireland's main teaching hospitals for the period 1993–1998.

RESULTS There was a significant proximal shift in CRC distribution between the two periods (23.5% proximal to the splenic flexure between 1976 and 1978v 36.7% between 1990 and 1997; p<0.001). The descending colon was visualised during 74.4% of FS examinations. By combining the observed extent of FS examination with CRC site distribution, it was calculated that FS could have visualised 68.8% of CRCs between 1976 and 1978 but only 56.0% between 1990 and 1997. Extrapolating these data to a Northern Ireland screening programme involving FS and faecal occult blood testing suggests that significantly more CRCs could have been detected between 1976 and 1978 than between 1990 and 1997 (51.7% v 48.2%, respectively; p=0.03).

CONCLUSIONS This study confirms the previously documented left to right shift in CRC distribution in Northern Ireland and demonstrates that if this shift continues, FS will become less successful as a screening tool than is currently predicted.