Findings Murky on Extending Interval Between Sigmoidoscopies

CHICAGO – The recommended 5-year interval between screening sigmoidoscopies probably cannot be safely extended, according to an analysis of data on more than 70,000 enrollees of Kaiser Permanente of Northern California.

"We found that rates of colorectal cancer are low for at least 10 years after a negative sigmoidoscopy," reported Dr. Vincent P. Doria-Rose of Kaiser Permanente in Oakland. "But there may be consequences to waiting 10 years between endoscopies," he said at the annual Digestive Disease Week.

Current screening guidelines recommend a 10-year interval for colonoscopy and a 5-year interval for sigmoidoscopy. Prior investigations have suggested that there is a long duration of low risk for colorectal cancer (CRC) after endoscopy – up to 20 years in some series – but these studies have been limited. For example, many rely on patient self-reports, and others are case-control studies, he said.

Dr. Doria-Rose and his colleagues analyzed data from the colorectal cancer prevention program at Kaiser Permanente, through which individuals with an average risk for CRC underwent screening sigmoidoscopy once every 10 years starting at age 50. The study evaluated CRC incidence after 10.5 years of follow-up in persons screened between 1994 and 1996.

The evaluable population included 72,483 individuals with negative results on screening and excluded persons with a personal history of colorectal polyps, CRC or inflammatory bowel disease, or a high-risk family history of CRC. Of the study population, 50% were female, 51% were at least 60 years old, and 65% were white.

Within 10.5 years of index sigmoidoscopy, 103 cases of distal CRC (rectum and sigmoid) were diagnosed, as were 333 cases of proximal CRCs. The incidence rates per 100,000 person-years of both proximal and distal CRC increased over time, but did so especially during the second half of the interval, the study found.

The mean incidence rate of distal CRC during years 6-10 was approximately twice as high as that in years 1-5. The mean incidence rates were 28/100,000 person-years for the second half of the screening interval, compared with a rate of 13 during the first 5 years, for a difference of 15 cases per 100,000 person-years, Dr. Doria-Rose reported.

The rate ratio (RR) was 2.1 for all tumors diagnosed after 6 years compared with years 1-5. By age, the RR was 1.1 for those younger than 60 and 2.5 for those aged 60 and older.

The standardized incidence ratio (SIR) in the distal colon increased gradually from 0.40/100,000 in the first year post screening to 0.49 in the 10th year. For proximal cancers, increasing SIRs were due mostly to the aging of the population, and there was no clear increase over time, with a mean SIR of 0.75 for the entire follow-up period.

By subgroup analyses, the results of the current study were consistent, except for the age group that was younger than age 60 years at baseline, who experienced no increase in risk over the second half of follow-up, he said.

"The incidence rates of distal CRC increased considerably in years 6 and up as compared to that in years 1 through 5," he noted, "but remained low for the entire follow-up period."

He explained that the incidence rate in the current study is lower than has been observed in the SEER (Surveillance, Epidemiology, and End Results) population of the San Francisco Bay area, a finding that could be due to patient selection. The current study involved an insured population that presented for screening and whose lifestyle and behaviors would reduce their inherent risk, he suggested.

"The rate is low for 10 years of follow-up, compared with the SEER rate, but there was still a doubling in the second half," Dr. Doria-Rose emphasized.

"These data do not suggest that there is one obvious, clear-cut interval for screening sigmoidoscopy," he concluded, "and they provide no information regarding the duration of [a low-risk period] following the endoscopic examination of the right colon."

He added that to evaluate the relative benefits of sigmoidoscopy every 5 years, versus colonoscopy every 10 years, similar analyses in large cohorts of individuals screened by colonoscopy will be required.

Dr. Doria-Rose reported no relevant financial disclosures.

CHICAGO – The recommended 5-year interval between screening sigmoidoscopies probably cannot be safely extended, according to an analysis of data on more than 70,000 enrollees of Kaiser Permanente of Northern California.

"We found that rates of colorectal cancer are low for at least 10 years after a negative sigmoidoscopy," reported Dr. Vincent P. Doria-Rose of Kaiser Permanente in Oakland. "But there may be consequences to waiting 10 years between endoscopies," he said at the annual Digestive Disease Week.

Current screening guidelines recommend a 10-year interval for colonoscopy and a 5-year interval for sigmoidoscopy. Prior investigations have suggested that there is a long duration of low risk for colorectal cancer (CRC) after endoscopy – up to 20 years in some series – but these studies have been limited. For example, many rely on patient self-reports, and others are case-control studies, he said.

Dr. Doria-Rose and his colleagues analyzed data from the colorectal cancer prevention program at Kaiser Permanente, through which individuals with an average risk for CRC underwent screening sigmoidoscopy once every 10 years starting at age 50. The study evaluated CRC incidence after 10.5 years of follow-up in persons screened between 1994 and 1996.

The evaluable population included 72,483 individuals with negative results on screening and excluded persons with a personal history of colorectal polyps, CRC or inflammatory bowel disease, or a high-risk family history of CRC. Of the study population, 50% were female, 51% were at least 60 years old, and 65% were white.

Within 10.5 years of index sigmoidoscopy, 103 cases of distal CRC (rectum and sigmoid) were diagnosed, as were 333 cases of proximal CRCs. The incidence rates per 100,000 person-years of both proximal and distal CRC increased over time, but did so especially during the second half of the interval, the study found.

The mean incidence rate of distal CRC during years 6-10 was approximately twice as high as that in years 1-5. The mean incidence rates were 28/100,000 person-years for the second half of the screening interval, compared with a rate of 13 during the first 5 years, for a difference of 15 cases per 100,000 person-years, Dr. Doria-Rose reported.

The rate ratio (RR) was 2.1 for all tumors diagnosed after 6 years compared with years 1-5. By age, the RR was 1.1 for those younger than 60 and 2.5 for those aged 60 and older.

The standardized incidence ratio (SIR) in the distal colon increased gradually from 0.40/100,000 in the first year post screening to 0.49 in the 10th year. For proximal cancers, increasing SIRs were due mostly to the aging of the population, and there was no clear increase over time, with a mean SIR of 0.75 for the entire follow-up period.

By subgroup analyses, the results of the current study were consistent, except for the age group that was younger than age 60 years at baseline, who experienced no increase in risk over the second half of follow-up, he said.

"The incidence rates of distal CRC increased considerably in years 6 and up as compared to that in years 1 through 5," he noted, "but remained low for the entire follow-up period."

He explained that the incidence rate in the current study is lower than has been observed in the SEER (Surveillance, Epidemiology, and End Results) population of the San Francisco Bay area, a finding that could be due to patient selection. The current study involved an insured population that presented for screening and whose lifestyle and behaviors would reduce their inherent risk, he suggested.

"The rate is low for 10 years of follow-up, compared with the SEER rate, but there was still a doubling in the second half," Dr. Doria-Rose emphasized.

"These data do not suggest that there is one obvious, clear-cut interval for screening sigmoidoscopy," he concluded, "and they provide no information regarding the duration of [a low-risk period] following the endoscopic examination of the right colon."

He added that to evaluate the relative benefits of sigmoidoscopy every 5 years, versus colonoscopy every 10 years, similar analyses in large cohorts of individuals screened by colonoscopy will be required.

Dr. Doria-Rose reported no relevant financial disclosures.

CHICAGO – The recommended 5-year interval between screening sigmoidoscopies probably cannot be safely extended, according to an analysis of data on more than 70,000 enrollees of Kaiser Permanente of Northern California.

"We found that rates of colorectal cancer are low for at least 10 years after a negative sigmoidoscopy," reported Dr. Vincent P. Doria-Rose of Kaiser Permanente in Oakland. "But there may be consequences to waiting 10 years between endoscopies," he said at the annual Digestive Disease Week.

Current screening guidelines recommend a 10-year interval for colonoscopy and a 5-year interval for sigmoidoscopy. Prior investigations have suggested that there is a long duration of low risk for colorectal cancer (CRC) after endoscopy – up to 20 years in some series – but these studies have been limited. For example, many rely on patient self-reports, and others are case-control studies, he said.

Dr. Doria-Rose and his colleagues analyzed data from the colorectal cancer prevention program at Kaiser Permanente, through which individuals with an average risk for CRC underwent screening sigmoidoscopy once every 10 years starting at age 50. The study evaluated CRC incidence after 10.5 years of follow-up in persons screened between 1994 and 1996.

The evaluable population included 72,483 individuals with negative results on screening and excluded persons with a personal history of colorectal polyps, CRC or inflammatory bowel disease, or a high-risk family history of CRC. Of the study population, 50% were female, 51% were at least 60 years old, and 65% were white.

Within 10.5 years of index sigmoidoscopy, 103 cases of distal CRC (rectum and sigmoid) were diagnosed, as were 333 cases of proximal CRCs. The incidence rates per 100,000 person-years of both proximal and distal CRC increased over time, but did so especially during the second half of the interval, the study found.

The mean incidence rate of distal CRC during years 6-10 was approximately twice as high as that in years 1-5. The mean incidence rates were 28/100,000 person-years for the second half of the screening interval, compared with a rate of 13 during the first 5 years, for a difference of 15 cases per 100,000 person-years, Dr. Doria-Rose reported.

The rate ratio (RR) was 2.1 for all tumors diagnosed after 6 years compared with years 1-5. By age, the RR was 1.1 for those younger than 60 and 2.5 for those aged 60 and older.

The standardized incidence ratio (SIR) in the distal colon increased gradually from 0.40/100,000 in the first year post screening to 0.49 in the 10th year. For proximal cancers, increasing SIRs were due mostly to the aging of the population, and there was no clear increase over time, with a mean SIR of 0.75 for the entire follow-up period.

By subgroup analyses, the results of the current study were consistent, except for the age group that was younger than age 60 years at baseline, who experienced no increase in risk over the second half of follow-up, he said.

"The incidence rates of distal CRC increased considerably in years 6 and up as compared to that in years 1 through 5," he noted, "but remained low for the entire follow-up period."

He explained that the incidence rate in the current study is lower than has been observed in the SEER (Surveillance, Epidemiology, and End Results) population of the San Francisco Bay area, a finding that could be due to patient selection. The current study involved an insured population that presented for screening and whose lifestyle and behaviors would reduce their inherent risk, he suggested.

"The rate is low for 10 years of follow-up, compared with the SEER rate, but there was still a doubling in the second half," Dr. Doria-Rose emphasized.

"These data do not suggest that there is one obvious, clear-cut interval for screening sigmoidoscopy," he concluded, "and they provide no information regarding the duration of [a low-risk period] following the endoscopic examination of the right colon."

He added that to evaluate the relative benefits of sigmoidoscopy every 5 years, versus colonoscopy every 10 years, similar analyses in large cohorts of individuals screened by colonoscopy will be required.

Dr. Doria-Rose reported no relevant financial disclosures.