Take Risk Into Account Before Repeat Surveillance Colonoscopy, Study Says

A majority of older adults undergoing surveillance colonoscopy were invited to return for a subsequent colonoscopy, even when no significant polyps were detected and they had a limited life expectancy, a study using Medicare claims data found.

Of nearly 10,000 adults 65 and up undergoing surveillance colonoscopy due to a prior polyp, colonoscopy turned up colorectal cancer (CRC) in 0.2%, advanced polyps in 7.8%, and at least three adenomas or sessile serrated polyps in 8.7%, according to researchers led by Audrey Calderwood, MD, MS, of the Dartmouth-Hitchcock Medical Center in Lebanon, New Hampshire.

In the subset of people with an available recommendation to stop or return for future surveillance colonoscopies, the vast majority with no polyps or only small hyperplastic polyps were recommended to return, including 58.1% of those with an estimated life expectancy of under 5 years, the group reported in JAMA Internal Medicine.

"These data may help refine decision-making about pursuing or stopping surveillance colonoscopy in older adults with a history of polyps," the study authors concluded.

Most colon polyps are not harmful, though some can transform into cancers, typically over a period of 10 to 15 years.

"Given this long biological timeline, the identification and removal of polyps in those with limited life expectancy (e.g., <10 years) may not provide benefit in CRC prevention within their lifetime," wrote Calderwood and co-authors.

"Studies of screening flexible sigmoidoscopy and colonoscopy have shown that the reduced CRC incidence among individuals exposed to lower endoscopy compared with no exposure started to appear after 6 to 8 years of screening," they continued. "Thus, it is possible that surveillance that includes polypectomy could lead to a CRC incidence reduction in less than 10 years, but this is unlikely within 5 years."

Significant colonoscopy findings (CRC, advanced polyps, or at least three adenomas or sessile serrated polyps) were identified in 14.9% of those with a life expectancy of at least 10 years and 22.9% of those with a life expectancy under 5 years (P<0.001 for trend).

Another factor worth considering in light of the "rare finding of CRC during surveillance colonoscopy," Calderwood's group said, is the growing risks for complications with colonoscopy as individuals age.

According to a systematic review and meta-analysis they cited, roughly 26 per 1,000 individuals age 65 and up experience an adverse gastrointestinal event (perforation, bleeding, or cardiovascular or pulmonary complications) and one of every 1,000 die. And the risk for an adverse gastrointestinal event increases to 35 per 1,000 for people in their 80s.

In the current study, 10% had conditions associated with a higher risk for complications, and inadequate bowel preparation -- also linked with adverse events -- was more common among those with a limited life expectancy (15.0% for those with a life expectancy of less than 5 years), Calderwood's group reported.

"While considering colonoscopy findings in the context of life expectancy is helpful, it is worth acknowledging that predicted life expectancy is not typically readily available in current practice," the study authors noted, adding that "primary care practitioners and patients have expressed concern about the use of life expectancy in the context of preventive care."

Their study included 9,831 adults age 65 and older who underwent a surveillance colonoscopy from 2009 to 2018, with the individuals identified using the New Hampshire Colonoscopy Registry, which was linked with Medicare claims data. Patients had a mean age of 73 years, and 54% were men. Most of the patients (83.5%) were white and non-Hispanic (80.2%).

A tenth of the cohort had one comorbidity associated with an increased risk with colonoscopy -- chronic lung disease, kidney failure, peripheral vascular disease, heart failure, liver disease, etc. -- while 2% had at least two of these comorbidities, and 0.7% had three or more.

Life expectancy was 10 or more years in 5,649 of the patients (57.5%), while 3,443 (35%) had a life expectancy of 5 years to less than 10 years, and 739 (7.5%) had a life expectancy of less than 5 years.

A little more than half of the patients (53.7%) had available recommendations to stop or continue a colonoscopy, with 86.9% of those individuals invited to return. Regardless of clinical findings, people with a life expectancy of 5 years or more were more likely to be recommended to return for surveillance colonoscopy when compared with those with a life expectancy of less than 5 years:

• Life expectancy ≥10 years: adjusted odds ratio (aOR) 21.5 (95% CI 15.2-30.3)
• Life expectancy 5 to <10 years: aOR 2.7 (95% CI 2.0-3.6)

One limitation of the analysis is a Medicare restriction that did not allow for separating out significant findings (CRC, advanced polyps, or at least three adenomas or sessile serrated polyps) for each life expectancy group. Calderwood and co-authors also cited the possibility for selection bias and that other important decision-making factors leading to discontinuation of surveillance could not be accounted for -- like frailty in older adults.

Comment