Colorectal Cancers Soon After Colonoscopy A Pooled Multicohort Analysis Gut. 2014;63(6):949-956.
Douglas J Robertson, David A Lieberman, Sidney J Winawer, Dennis J Ahnen, John A Baron, Arthur Schatzkin, Amanda J Cross, Ann G Zauber, Timothy R Church, Peter Lance, E Robert Greenberg, María Elena Martínez

1 Department of Veterans Affairs Medical Center, White River Junction, VT and Dartmouth Medical School & The Dartmouth Institute, Hanover, New Hampshire, USA
2 Department of Veterans Affairs Medical Center, Portland, Oregon, USA
3 Memorial Sloan-Kettering Cancer Center, New York, New York, USA

Abstract

Objective. Some individuals are diagnosed with colorectal cancer (CRC) despite recent colonoscopy. We examined individuals under colonoscopic surveillance for colonic adenomas to assess possible reasons for diagnosing cancer after a recent colonoscopy with complete removal of any identified polyps.

Design. Primary data were pooled from eight large (>800 patients) North American studies in which participants with adenoma(s) had a baseline colonoscopy (with intent to remove all visualised lesions) and were followed with subsequent colonoscopy. We used an algorithm based on the time from previous colonoscopy and the presence, size and histology of adenomas detected at prior exam to assign interval cancers as likely being new, missed, incompletely resected (while previously an adenoma) or due to failed biopsy detection.

Results. 9167 participants (mean age 62) were included in the analyses, with a median follow-up of 47.2 months. Invasive cancer was diagnosed in 58 patients (0.6%) during follow-up (1.71 per 1000 person-years follow-up). Most cancers (78%) were early stage (I or II); however, 9 (16%) resulted in death from CRC. We classified 30 cancers (52%) as probable missed lesions, 11 (19%) as possibly related to incomplete resection of an earlier, non-invasive lesion and 14 (24%) as probable new lesions. The cancer diagnosis may have been delayed in three cases (5%) because of failed biopsy detection.

Conclusions. Despite recent colonoscopy with intent to remove all neoplasia, CRC will occasionally be diagnosed. These cancers primarily seem to represent lesions that were missed or incompletely removed at the prior colonoscopy and might be avoided by increased emphasis on identifying and completely removing all neoplastic lesions at colonoscopy.

NOTE In the last one year there is evidence that using the whole cleansing prep the evening before will not clean the mucosa for adequate examination, as the overnight secretions stick to the colonic wall and is difficult to clean even with water jet during the procedure. The standard of practice now is to use split prep, i.e. half the dose of the laxative is given the evening before and the other half is given 4-5 hours before the procedure. In my personal experience this makes a huge difference.