© Georg Thieme Verlag KG Stuttgart · New York
Colonoscopy, tumors, and inflammatory bowel disease
23 October 2009 (eFirst)
Association of colonoscopy and death from colorectal cancer (Baxter et al., Ann Intern Med 2009 )
Colonoscopy has been accepted as a modality for colorectal screening and surveillance for several years now. It has been shown to have a high sensitivity in detecting gastrointestinal lesions, and also allows complete examination of the colon and rectum and removal of polyps at the time of detection, making it a suitable screening tool. The demand of screening colonoscopies has increased over recent years for several reasons . Despite this increased use, few studies have looked at the impact of colonoscopy screening on the incidence and mortality of colorectal cancer in the general population.
This retrospective case-control study by Baxter et al. evaluated the association between colonoscopy and the number of deaths from colorectal cancer (CRC) by studying 10 292 case patients who died of CRC and, for each case patient, five control patients who did not. Case patients were derived from the Ontario Health Insurance Plan billing codes and matched to controls for factors known to influence colonoscopy rates and risk for CRC (sex, socioeconomic status, and age). Because these billing codes do not distinguish between screening and diagnostic colonoscopies, all colonoscopies that were carried out within 6 months before the referent date were excluded, in an effort to decrease the proportion of diagnostic colonoscopies.
The authors found an inverse association of death from left-sided CRC with colonoscopy (odds ratio [OR] 0.33; 95 % confidence interval [CI] 0.28 – 0.39). However, colonoscopy was not associated with fewer deaths from right-sided CRC (OR 0.99; 95 %CI 0.86 – 1.14). They concluded that colonoscopy is associated with fewer deaths from CRC developing in the left side of the colon but not in the right side.
Why would screening colonoscopy be less effective in preventing right-sided CRC? Possible explanations are suboptimal bowel preparation that tends to obscure the right colon to a greater degree than the left or macroscopic features of right-sided lesions that are different (e. g. sessile or flat) compared with those in the left side making them more difficult to visualize.
This study suffers from a few methodologic drawbacks that somewhat limit its conclusions. First, the quality aspects of the colonoscopies in this study are unknown. Colon preparation might have been poor or withdrawal time short, attributing to missed lesions. Also, colonoscopies with cecal intubation were identified from the billing codes, but with no direct evidence for actual cecal intubation. Furthermore, as the majority of colonoscopies were done by internists and surgeons rather than gastroenterologists, differences in experience might have contributed to a lower cecal intubation rate and a lower adenoma detection rate.
Secondly, screening tests among patients with cancer and control patients during the same period before the date of the patient’s diagnosis are not distributed in time in the same way. Almost all screening tests in patients with CRC will have been done close to the time of diagnosis. In case the screening had been done earlier in the group of case patients, the tumor would have been found then. The distribution of screening colonoscopy during the corresponding period in the control group is therefore more uniform than the one in the case patients.
Finally, screening and diagnostic colonoscopies in this study could not be distinguished. Patients that were diagnosed with cancer within 6 months before the colonoscopy were excluded, in order to decrease the proportion of diagnostic colonoscopies. Even though this seems like an appropriate amount of time, diagnostic colonoscopies cannot be completely excluded.
Despite these methodologic drawbacks, this interesting study raises clear questions on the effectiveness of currently practiced screening colonoscopy for right-sided colorectal cancer prevention. It warrants further research into the possible difference in biology of right- and left-colonic neoplasia. Furthermore, it stresses the importance of quality assessment in colonoscopy; achieving acceptable rates of cecal intubation, good preparation of patients, and an acceptable level of experience and technical skills of the endoscopist to adequately identify polyps and cancer, particularly on the right side.