Flexible Sigmoidoscopy Misses Interval Colorectal Cancers

Caroline Helwick

June 06, 2011

June 6, 2011 (Chicago, Illinois) — Flexible sigmoidoscopy as an effective screening tool for colorectal cancer (CRC) was dealt a blow here at Digestive Disease Week (DDW) 2011 when researchers reported finding 4 times the number of lesions in the left side of the colon with colonoscopy than with flexible sigmoidoscopy, when used in older persons with a history of CRC.

It has been unclear how flexible sigmoidoscopy compares with colonoscopy in the rate of interval CRCs. This study's objective was to determine whether the approaches differ with regard to the incidence of interval, or "missed," cancers that were diagnosed 6 to 36 months after the index endoscopy.

"We concluded that despite its imperfections, colonoscopy remains the gold standard for detecting or preventing [CRC]," said lead investigator Yize Richard Wang, MD, PhD, a gastroenterology fellow at the Mayo Clinic in Jacksonville, Florida. "Older persons should undergo colonoscopy instead of flexible sigmoidoscopy."

Dr. Wang and his colleagues reviewed the national Surveillance, Epidemiology and End Results Medicare-linked database for patients aged 67 years and older who were diagnosed with CRC in the left colon (distal to the splenic flexure) between 1998 and 2005. Of the 52,236 patients in the study, 52,412 had colonoscopies and 3523 had flexible sigmoidoscopies.

There were 25,541 older patients with CRC diagnosed in the left colon. Compared with patients undergoing colonoscopy, persons having sigmoidoscopy tended to be slightly older, to be women, and to live in Zip codes with higher income and educational levels. Nongastroenterologists performed two thirds of the sigmoidoscopies and one third of the colonoscopies.

The rate of new or missed left-sided CRCs diagnosed after flexible sigmoidoscopy was 11.6% (n = 251), the rate being highest in the descending colon. In contrast, the rate of interval CRCs after colonoscopy was 2.6% (n = 598), which was a highly significant difference (P <.001), Dr. Wang reported.

With flexible sigmoidoscopy, therefore, the risk for new or missed CRCs in the distal colon quadrupled compared with colonoscopy (odds ratio, 4.00; 95% confidence interval [CI], 3.51 - 4.55), he noted.

Women had also had a 15% increased risk for interval cancers (95% CI, 1.02 - 1.30), whereas persons having an endoscopy performed as inpatients had a 47% reduced risk for interval cancers (95 CI, 0.45 - 0.62).

Insignificant variables were age, non-white race, Zip code area, metropolitan residence, and physician specialty.

By location, the interval cancers found after sigmoidoscopy vs colonoscopy were, respectively:

  • 18.7% vs 3.3% in the descending colon,

  • 12.5% vs 2.7% in the rectum,

  • 11.7% vs 2.6% in the left colon combined,

  • 11.3% vs 2.4% in the sigmoid colon, and

  • 8.0% vs 2.2% at the rectosigmoid junction.

The retrospective study could not show why sigmoidoscopies proved far less sensitive in identifying CRCs, Dr. Wang acknowledged, but he suggested: "We think that differences in bowel preparation, lack of sedation, and uncertainty of the depth that was reached with sigmoidoscopy are possible explanations."

He acknowledged several study limitations: it is a retrospective study of older patients whose indication for the lower endoscopy procedure is not known, and the results do not reflect recent advancements in endoscopy equipment and techniques. Dr. Wang also cautioned that the findings are not applicable to patients undergoing CRC screening with no past history of CRC.

At a press briefing, other panelists expressed little surprise at the findings and contributed to the discussion. When asked whether flexible sigmoidoscopy should be reserved for selected populations, moderator Jerome Waye, MD, clinical professor of medicine at Mt. Sinai Medical Center, New York City, responded, "I think this is a reasonable conclusion."

"We don't know how far the sigmoidoscope went. That's a difficult number to assess in a retrospective study. That is, for patients who developed left-sided cancers, we don't know where these lesions were, in relation to where the scope actually reached," he noted. "A lot of sigmoidoscopy patients are poorly prepped and unsedated, and the scope may not go up very far. You can miss lesions just around the bend."

Brooks D. Cash, MD, chief of medicine at the National Naval Medical Center and the Walter Reed Army Medical Center, Washington, DC, added, "These are very important findings. They are a logical follow-up to data [from] large studies, including our CONCERN trial, which found a significantly increased yield with colonoscopy and helped establish colonoscopy as the gold standard endoscopic approach. This shows the natural history with regard to interval cancers with these approaches."

Dr. Wang, Dr. Waye, and Dr. Cash have disclosed no relevant financial relationships.

Digestive Disease Week (DDW) 2011: Abstract 906. Presented May 10, 2011.


Comments on Medscape are moderated and should be professional in tone and on topic. You must declare any conflicts of interest related to your comments and responses. Please see our Commenting Guide for further information. We reserve the right to remove posts at our sole discretion.
Post as: