Rethinking Colonoscopy Intervals in Patients With Multiple Non-Advanced Adenomas

The increased numbers of small, precancerous non-advanced adenomas that are detected with newer, high-definition (HD) colonoscopes do not appear to pose a heightened risk of metachronous advanced neoplasia (MAN), Cleveland Clinic researchers have determined.

Their findings, presented at the 2019 annual meeting of the American College of Gastroenterology (ACG), could spur consideration of a longer colonoscopy surveillance interval for a subset of patients with adenomas who were previously considered to be at high risk for MAN, the researchers say. The current post-polypectomy surveillance guidelines are based on studies more than a decade old, prior to the introduction of HD colonoscopes and quality metrics like adenoma detection rates and split-dose bowel preparation.

An estimated 14 million colonoscopies are done in the United States every year and adenomatous polyps are a common finding. Removal of them may prevent many cancers and reduce mortality.

Studies using standard-definition colonoscopes in the early 2000s demonstrated that risk of MAN was increased in patients with more than three, versus one to two, small tubular adenomas. Since 2006, introduction of HD colonoscopes and attention to the adenoma detection rate (ADR) have enabled screening endoscopists to detect more patients with numerous and smaller adenomas than had been possible before.

The upward trend in identification and removal of multiple diminutive, non-advanced adenomas prompted researchers in Cleveland Clinic’s Center for Colon Polyp and Cancer Prevention to investigate whether patients who have them actually are at increased risk of developing MAN.

The U.S. Multi-Society Task Force (MSTF) on Colorectal Cancer, which represents the ACG, the American Gastroenterological Association and the American Society for Gastrointestinal Endoscopy, currently recommends a three-year post-polypectomy colonoscopy interval in patients considered at high risk for adenomas (HRA). Historically, in addition to multiplicity (≥3 adenomas), other HRA features associated with an increased risk of MAN include villous histology, high-grade dysplasia (HGD) and size ≥ 10 mm.

“The current MSTF guidelines were based on studies done in the late 1990s/early 2000s” says Carol Rouphael, MD, a Gastroenterology Fellow in Cleveland Clinic’s Digestive Disease & Surgery Institute and the study’s principal investigator. “In the contemporary era, with high-definition colonoscopes, quality metrics, and split-dose bowel preparation, we are seeing more patients with multiple non-advanced adenomas. Our hypothesis was that those individuals might not be at increased risk of MAN, as had previously been thought.”

Study criteria and results

The study analyzed the association between baseline adenoma features, particularly three and more non-advanced adenomas, and risk of MAN in patients undergoing HD colonoscopy. Data from electronic medical records at Cleveland Clinic were used to identify individuals aged 50 or older who had undergone their first colonoscopy in 2006 or later and who had a follow-up colonoscopy more than 24 months thereafter.

Mean age of the 3,383 patients in the study was 60.5 years and 54.6% were female. Patients were considered to have HRA if they met the MSTF criteria as previously defined. Low-risk adenoma (LRA) was defined as having one to two tubular adenomas measuring 10 mm or less, per the MSTF guidelines. Median time interval between first and follow-up colonoscopy in the HRA group was 41.6 months (range 36.3-52.6) versus 53.5 months (range 39.1 to 62.9) in the LRA group versus 60.7 months (range 45.0-68.4) in patients with no adenoma on their first examination.

Comparing adenoma features in the two groups, the investigators found that risk of MAN was 3.8% in patients with no adenomas at baseline; 4.6% with LRA; and 9.3% with HRA. However, when breaking down HRA features, in individuals with three or more non-advanced adenomas, risk of MAN was 6.3% versus 6.1% with three to four and 7.7% with five or more such adenomas.

“We expected to find that villous features, high-grade pathology, and large adenoma size would be significantly associated with MAN, and that was the case,” says Dr. Rouphael.

On multivariable analysis, the risk of MAN did not differ between the HRA and LRA groups based on the number of non-advanced adenomas. “The absolute risk of MAN with five or more non-advanced adenomas, however, was close to that associated with adenomas measuring at least 10 mm [7.7% vs. 8.3%] ,” says Dr. Rouphael.

The investigators’ conclusion, therefore, was that “longer intervals for surveillance colonoscopy likely are safe in patients who have three to four non-advanced adenomas, but more data are needed to determine whether an extended interval is appropriate for those with five or more such adenomas,” says Dr. Rouphael.

Potential clinical impact

The study findings have implications for health care resource utilization and the MSTF guidelines, according to Carol A. Burke, MD, Director of the Center for Colon Polyp and Cancer Prevention, who oversaw the research. “Only 65% of the population that needs colorectal cancer screening has undergone the procedure. Performing colonoscopy frequently in low-risk groups who do not gain an advantage from it takes away resources from patients who have never been screened.”

Regarding the MSTF recommendations, Dr. Burke, who is the ACG’s former president, concludes that “this is one area in which there has not been a lot of evidence on which to base guidelines related to resource utilization. We believe our research is likely going to have an impact on clinical practice very soon in terms of modifying the standard of care.”

Photo: Endoscopic view of a non-advanced adenoma.