Gastroenterology

Gastroenterology

Volume 132, Issue 1, January 2007, Pages 96-102
Gastroenterology

Clinical–alimentary tract
Rates of New or Missed Colorectal Cancers After Colonoscopy and Their Risk Factors: A Population-Based Analysis

https://doi.org/10.1053/j.gastro.2006.10.027Get rights and content

Background & Aims: The rate of new or missed colorectal cancer (CRC) after colonoscopy and their risk factors in usual practice are unknown. Our objective was to evaluate the rate and risk factors in a population-based study. Methods: We analyzed data from the Canadian Institute for Health Information, the Ontario Health Insurance Program, and Ontario Cancer Registry for all patients (≥20 years of age) with a new diagnosis of right-sided, transverse, splenic flexure/descending, rectal or sigmoid CRC in Ontario from April 1, 1997 to March 31, 2002, who had a colonoscopy within the 3 years before their diagnosis. Patients with new or missed cancers were those whose most recent colonoscopy was 6 to 36 months before diagnosis. We examined characteristics that might be risk factors for new or missed CRC. Results: We identified a diagnosis of CRC in 3288 (right sided), 777 (transverse), 710 (splenic flexure/descending), and 7712 (rectal or sigmoid) patients. The rates of new or missed cancers were 5.9%, 5.5%, 2.1%, and 2.3%, respectively. Independent risk factors for these cancers in men and women were older age; diverticular disease; right-sided or transverse CRC; colonoscopy by an internist or family physician; and colonoscopy in an office. Conclusions: Because having an office colonoscopy and certain patient, procedure, and physician characteristics are independent risk factors for new or missed CRC, physicians must inform patients of the small risk (2% to 6%) of these cancers after colonoscopy. The influence of type of physician and setting on the accuracy of colonoscopy, potentially modifiable risk factors, warrants further study.

Section snippets

Materials and Methods

The Research Ethics Board of Sunnybrook and Women’s College Health Sciences Centre in Toronto, Ontario, Canada, approved this study.

Patient Characteristics

Between April 1, 1997, and March 31, 2002, we identified 31,074 patients with a diagnosis of CRC in the province of Ontario. The mean age was 69.3 ± 12.2 years; 14,356 (46.9%) were women. We excluded 4 patients <20 years old and 272 with inflammatory bowel disease. We excluded 4710 patients with CRC site classified as synchronous or “other.” The mean age of the “other” group was slightly greater at 70.7 years (standard deviation [SD] 12.3) compared with 69.0 years (SD 12.1) for the remaining

Discussion

In our study, the rate of new or missed CRC after colonoscopy in usual clinical practice was 2%–6%, depending on the site of the cancer. Given the widespread use of colonoscopy, an estimate of the accuracy of this procedure is essential. Approximately 150,000 people in the United States will be diagnosed with CRC in 2006.1 Based on our findings, approximately 5100 of these cancers would not be detected at the initial colonoscopy if all of these persons were to have the procedure. We found that

Cited by (545)

  • Effect of selection bias on Automatic Colonoscopy Polyp Detection

    2023, Biomedical Signal Processing and Control
  • Approach of artificial intelligence in colorectal cancer and in precision medicine

    2023, Computational Methods in Drug Discovery and Repurposing for Cancer Therapy
View all citing articles on Scopus
1

L.R. is a senior investigator with the Cancer Quality Council of Ontario.

View full text