Article Text

  1. K. Whitt,
  2. M. Ismail
  1. University of Tennessee at Memphis, Memphis, TN


Background and Aims Colonoscopy has limitations in diagnosing colorectal neoplasia (CRC) and 4-12% of lesions can be missed. CRC miss rate has been studied in various settings, but no data are available for CRC miss rate among US veterans undergoing colonoscopy.

Aims To evaluate miss colorectal cancer rate among veterans undergoing colonoscopy at VA Medical Center at Memphis (VAMC) and to identify factors related to CRC miss rate.

Methods Cases of primary colorectal adenocarcinoma at VAMC between January 1, 2000 and April 30, 2005 were identified from a local tumor registry. These cases were cross-referenced with pathology databases to ensure accuracy of report. During this period, patients with two or more colonoscopies at our institution were identified from the endoscopic database. Retrospective chart review was done to obtain data related to missed cancer. A "real-life " estimate of missed cancer diagnosis was then determined.

Results 685 veterans underwent two or more colonoscopies during January 1, 2000 to April 30, 2005. Five cases of missed colorectal cancer were found among this group (2% of all the CRC diagnosis at VAMC, Memphis during this time interval). All patients were male, and 4/5 were white. Mean age at index colonoscopy was 73 years (range 63-82 years). Gastroenterology fellows or surgical residents performed 2/5 index colonoscopies under the supervision of a staff physician. Cancer diagnosis was made at a mean interval of 33.2 months after index colonoscopy (range 17-47 months); 4/5 lesions were proximal to the splenic flexure. Colon preparation was described as fair or good in all patients. Indications for the follow-up exam were surveillance colonoscopy (2 patients), persistent anemia with weight loss or abdominal pain (2 patients), and large bowel obstruction (1 patient). In our patient group, the primary factor identified for missing a diagnosis of CRC was failure to complete a follow-up exam in the recommended interval (1-yr surveillance in 2 patients and 3-yr surveillance in 1 patient). Eighty percent of the neoplasms were stage III or higher and 3/5 patients had expired by the end of follow-up. Missed colorectal cancer diagnosis among veterans population undergoing repeat colonoscopy occurred in only 0.73% of cases.

Conclusions Missed colorectal cancer diagnosis among veterans at our institution is rare. Failure to diagnose colorectal cancer in this population relates to a discrepancy in physician-recommended surveillance intervals and actual patient follow-up. Patients with delayed diagnosis of colon cancer present with advanced tumor stages, but it is unclear how survival is impacted.

Statistics from

If you wish to reuse any or all of this article please use the link below which will take you to the Copyright Clearance Center’s RightsLink service. You will be able to get a quick price and instant permission to reuse the content in many different ways.