Colonoscopy in prevention of colorectal carcinoma

Continuing from my previous post on colorectal carcinoma, I just came across a study by Hermann Brenner et al. showing the effectiveness of colonoscopy in preventing colorectal carcinoma. They had conducted a population based case-control study in Germany to quantify the effectiveness of colonoscopy. 1688 cases and 1932 controls aged over 50 years participated in the study. The selection criteria for the cases were - a first diagnosis of invasive colorectal carcinoma (>30 years), physically and mentally fit to participate and ability to communicate in German. The study used randomly selected population based controls without history of colorectal carcinoma and were age, sex and country matched. The use of population based control prevents selection bias in recruiting controls and Berkson's bias, so this can be counted as a strength of the study.

Colon and rectum. Modified screenshot from Google body browser.

The cases were interviewed by trained interviewers in hospital setting while the controls were examined at their home. It would have been better if the interviewers were blinded to remove interviewer bias. The data on endoscopies during the last 10 years were corroborated with medical records which were available in 84% of the cases; hence there isn't much chance of recall bias. The authors excluded from statistical analysis cases less than 50 years old as screening colonoscopy is not recommended in them, cases with history of inflammatory bowel disease as they undergo surveillance colonoscopy due increased risk of colorectal carcinoma, missing information regarding colonoscopy, had undergone any other endoscopic procedure or if the last colonoscopy was done less than 1 year or more than 10 years ago.

Flow diagram of the study.

The confounding factors accounted for were age, sex, and education level, family history of colorectal carcinoma, smoking, body mass index, non-steroidal anti-inflammatory drug use, hormone replacement therapy use and participation in health screening examinations. The risk of colorectal cancer was estimated by calculating Odds ratio (OR).

The findings of the study clearly show the benefit of colonoscopy in preventing colorectal carcinoma. It should be noted that colonoscopy by itself is not preventive, colonoscopy identifies the early adenomas which are removed. As development of colon cancer follows adenoma-carcinoma sequence, colonoscopy becomes protective by early detection and removal of adenomas. The study found that colonoscopy reduces the overall risk of colorectal carcinoma by 77%. This is substantial. The greatest risk reduction was seen in case of cancers located in the sigmoid colon (adjusted OR=0.14). The least risk reduction was for cancers located in the ascending colon. In general the risk reduction was more for left colon cancers than for right colon cancers.

Adjusted odds ratio according to the site of colorectal cancer (less OR means more protection by colonoscopy).

The beneficial effect of colonoscopy did not vary with sex or with family history of colorectal carcinoma. The risk reducing effect of colonoscopy increased with increasing age in case of right sided colon cancer but such an effect was not observed in case of left sided colon cancers.

The study has few limitations like unaccounted confounders, possible differential rate of participation in the study. Also the authors did not account for the dietary habits of the participants. Increased intake of dietary fiber has a protective effect in development of colorectal cancer while intake of red meat and animal fat has a deleterious effect. It can be argued that BMI is a measure of the diet but I don’t think it can replace dietary history. This is one major point that I think the authors should have addressed.

Brenner H, Chang-Claude J, Seiler CM, Rickert A, & Hoffmeister M (2011). Protection from colorectal cancer after colonoscopy: a population-based, case-control study. Annals of internal medicine, 154 (1), 22-30 PMID: 21200035