Studies suggest 10-year colonoscopy intervals to age 80

May 22, 2006

Two population-based studies in the latest Journal of the American Medical Association address critical topics in colorectal cancer screening with optical colonoscopy. Cobbled together, the results suggest that once-a-decade screening is sufficient for average-risk patients, but that the potential of screening to extend lives diminishes by age 80.

How the implementation of virtual colonoscopy screening might alter the results in terms of safety or sensitivity was not directly addressed.

In the first study, Dr. Harminder Singh and a team from the University of Manitoba in Canada found a low risk of colorectal cancer for as long as 10 years after screening with optical colonoscopy, suggesting that a decade between exams may not be too long (JAMA, May 24/31, 2006, Vol. 295:20, pp. 2366-2373).

In the second study, Dr. Otto Lin and colleagues from three institutions found that gains in life expectancy from colonoscopy were far shorter among patients over 80 years compared with younger screening subjects. So even though the prevalence of colorectal cancer continues to increase with age, the costs and risks of colonoscopy may outweigh its potential benefits among the very elderly (JAMA, May 24/31, 2006, Vol. 295:20, pp. 2357-2364).

How often to screen

"A screening interval of 10 years after normal colonoscopy has been adopted based on the estimate of the time it takes for an adenomatous polyp to transform into carcinoma," wrote Singh; Donna Turner, Ph.D.; Linda Xue; and colleagues from Manitoba. "However, the duration over which the risk of CRC (colorectal cancer) remains decreased following the performance of a normal colonoscopy remains unknown."

The study, based on the results of 35,975 colonoscopy exams billed in the province between April 1989 and December 2003, sought to determine the risk of developing colorectal cancer after optical colonoscopy.

Standard incidence ratios (SIRs) were used to compare colorectal cancer incidence in the study cohort with that of the provincial populations. "Stratified analysis was performed to determine the duration of the reduced risk," the group wrote.

"Because the goal of the study was to evaluate the risk relative to the general population of developing CRC in average-risk individuals after a negative colonoscopy result," patients with a history of colorectal cancer, inflammatory bowel disease, resective colorectal surgery, or endoscopy within five years previously were excluded from the study, wrote the authors.

However, an editorial in the same issue by Timothy Church, Ph.D., complained that this decision ensured the applicability of study results only to individuals with the lowest pretest potential of developing colorectal cancer (JAMA, May 24/31, 2006, Vol. 295:20, pp. 2411-2412).

Singh et al found that negative colonoscopy was associated with SIRs of 0.69 (95% CI, 0.59-0.81) at six months, 0.66 (95% CI, 0.56-0.78) at one year, 0.59 (95% CI, 0.48-.0.72) at two years, 0.55 (95% CI, 0.41-0.73) at five years, and 0.28 (95% CI, 0.09-0.65) at 10 years.

"The proportion of colorectal cancer located in the right side of the colon was significantly higher in the colonoscopy cohort than the rate in the general population (76 [47%] of 163 versus 2,884 [28%] of 10,197; p < 0.001)," the researchers wrote.

"There is a need to improve the early detection of right-sided colorectal neoplasia in the usual clinical practice," the authors wrote. "Failure of endoscopists to reach the cecum, even when they believe they have reached the cecum, or the likelihood that inadequate bowel preparations affect the right side more than the rest of the colon may explain this higher rate of right-sided cancers...."

CRC was more likely to be diagnosed in the two years following index colonoscopy (33 [56%] of 59) compared with those diagnosed more than five years following the initial colonoscopy (19 [38%] of 50), though the difference did not reach statistical significance, they added.

"This study demonstrates that following a negative colonoscopy ... the risk of developing CRC is at most 60% to 70% of the risk of developing CRC in the general population and the duration of the interval of decreased CRC risk persists for more than 10 years.... Our findings suggest that screening colonoscopies do not need to be performed at intervals shorter than 10 years."

Benefit diminishes for very elderly

The study by Lin and colleagues, from the Virginia Mason Medical Center in Seattle, the Changhua Christian Medical Center in Taiwan, and the University of Toronto in Canada, sought to compare estimated life years saved with screening colonoscopy in very elderly versus younger individuals.

The team studied 1,244 asymptomatic subjects who underwent screening colonoscopy at a teaching hospital in three age groups, including 50-54 years (n = 1,034), 75-79 years (n = 107), and 80 years or older (n = 63).

The prevalence of neoplasia continued to rise with advancing age. It was 13.8% among the 50- to 54-year-old group, 26.5% in the 75- to 79-year-old group, and 28.6% among those 80 years and older.

"Participants aged 80 years and older had a significantly higher prevalence of advanced neoplasia than the 50- to 54-year-old group (14% versus 3.2%; p < 0.001)," the team reported. "Overall, men had a significantly higher prevalence of neoplasia than women" (18.5% versus 13%; p = 0.008).

Nevertheless the gain in life expectancy was far smaller for the oldest patients compared to the 50- to 54-year-old group: 0.13 versus 0.85 years.

Screening colonoscopy in subjects 80 and older offered only "15% of the expected gain in life expectancy in younger patients," based on the most favorable assumptions for polyp lag time and polyp progression probability. The least favorable assumptions yielded an expected 5.6% gain.

As a result, even though the prevalence of neoplasia increases with age, the very elderly are much likelier to die of "natural" causes than colorectal cancer even if they have it, the group wrote, and the potential of being harmed from the exam is greater.

"Previous studies of colonoscopy in symptomatic and asymptomatic elderly patients have consistently shown a high prevalence of colorectal neoplasia," the authors noted. "However, cecal intubation rates are lower, procedure times are longer, perforation risks are higher, and suboptimal bowel preparation is more likely in very elderly patients."

Colonoscopy in the very elderly should only occur after careful consideration of the potential benefits, risks, and patient preferences, they cautioned.

An optimal assessment of the impact of screening would require a 30- to 40-year follow-up period, Lin et al noted. Still, a previous modeling study using the DEALE technique concluded that colonoscopy screening would add 170 days to the life expectancy of 50- to 54-year-olds versus 41 days for those age 70-74 years. Another modeling study concluded that colonoscopy would offer no increase in life expectancy at all for men 85 and older or women 90 and older.

Critic's turn

"These two studies address two very important and unresolved questions in colorectal cancer screening by colonoscopy, and the answers are made more difficult because the efficacy of colonoscopic screening has never been examined in a randomized field," wrote Church in his editorial.

With regard to Singh et al's study on follow-up intervals, Church stated that the low reported SIRs have the effect of excluding individuals with colorectal cancer and polyps, guaranteeing a lower detection rate for these individuals as well as others with prior endoscopy, inflammatory bowel disease, and prior resection, who were also excluded.

"The absolute decrease will be sustained indefinitely unless the individuals with negative colonoscopy results somehow develop higher incidence rates to compensate for the removed and prevented cancers," he wrote. "However, the SIRs indicate a proportionately lower risk and are relevant to the selection effect of a negative colonoscopy."

By excluding those at higher risk, the paper recommends longer screening intervals for precisely the group that should have longer screening intervals, Church stated. And the study does not address the effect of polyp detection and removal.

In addition, while Singh et al suggested that inadequacies in colonoscopy practice were likely to blame for lower right-sided lesion detection, "a more likely explanation is that screening differentially affects left-sided lesions, missing more rapidly growing neoplasia in the right colon," Church countered. If the differences were due to colonoscopy practice, he stated, "it would affect the general population equally and therefore would keep the SIR the same."

As for the paper by Lin et al on the age of subjects at screening, Church noted its heavy reliance on the assumptions of the model applied to neoplastic findings regarding their potential progression to cancer, as did the authors, who reported potential mortality benefits in a wide range. Yet only the incidence-prevention benefit was modeled and tested specifically, Church wrote, while noting the diminished statistical power of the study for very elderly subjects due to their sharply lower representation in the study compared to the youngest group.

Also, the study authors suggested that the effect of colonoscopy screening might be similar to that of fecal occult blood testing (FOBT), but Church answered that the effect of colonoscopy screening would likely be more beneficial than FOBT, since the former catches more cancers earlier, thus potentially boosting life expectancy across all age levels.

Despite their shortcomings, both studies offer important new information that had not been previously addressed, Church noted, recommending more research.

"Given (the) enormous annual cost of universal colonoscopy and the modest cost of a single randomized trial relative to this sum, the critical issue is whether present methods are reliable enough to establish efficacy in either incidence or mortality, regardless of the age being considered," Church wrote. "It may well be time to address this in a randomized trial of the efficacy of colorectal cancer screening by colonoscopy in symptomatic men and women aged 50 years and older."

By Eric Barnes
AuntMinnie.com staff writer
May 23, 2006

VC would raise screening costs, study concludes, October 31, 2005

Fecal occult blood screening cuts colorectal cancer mortality, June 25, 2004

Lower cardiovascular risk makes virtual colonoscopy safer, study finds, December 31, 2003

Screening colonoscopy worthwhile into the eighth and ninth decades, September 1, 2003