SEATTLEHaving at least one follow-up colonoscopy within 5 years of a
diagnosis of nonmetastatic colorectal cancer decreased the mortality risk by
40%, according to a study presented at the 67th Annual Scientific Meeting of
the American College of Gastroenterology (president’s plenary session
Lead author Deborah Fisher, MD, an associate in the Department of
Medicine, Duke University, said that the study supports the current practice
of performing endoscopic surveillance in this group of patients and is the
first to show a mortality benefit. In collaboration with colleagues from Duke
University and the Durham VA Medical Center, Dr. Fisher utilized three
national VA databasesthe inpatient treatment file, the outpatient clinic
file, and the beneficiary identification and records location system (BIRLS)to
find a suitable study cohort.
While the optimal study design would be a randomized controlled trial, Dr.
Fisher said, the researchers decided on a retrospective cohort design because
of the expense and large enrollment needed for a randomized trial, along with
the fact that randomizing patients to no colonoscopy follow-up would
contradict current guidelines.
A total of 3,546 patients were included in the study. They were 98% male
and 81% white, with a mean age of 68 years. Patients were included if they
had a diagnosis of colon or rectal cancer in fiscal years 1995 to 1996, and
were excluded if there was any history of inflammatory bowel disease,
metastatic disease at initial diagnosis, death within 1 year of initial
diagnosis, and race other than black or white.
Approximately half of the patients had no major comorbidities, 26% had
one, and 19% had two or more. A total of 28% received chemotherapy, and 16%
radiation therapy. The 5-year survival rate was 67%, and 64% had undergone at
least one follow-up colonoscopy. Patients in both groups (with or without
follow-up colonoscopy) averaged 123 outpatient visits for any reason.
The researchers found that demographics, such as age and race, and
distribution of comorbidities were similar between patients who underwent a
follow-up colonoscopy and those who did not. "There were, however,
significantly more patients in the follow-up group who received chemotherapy
and radiation therapy," Dr. Fisher said.
Using the log rank test, the 5-year survival rate was found to be
significantly higher for the follow-up colonoscopy group (P < .0001).
Analysis by the Cox proportional hazard method also showed that colonoscopy
was associated with a significant decrease in risk of death at any point
during the 5-year follow-up (risk ratio 0.6).