Reach of sigmoidoscopies inadequate in many older women

The number of sigmoidoscopy examinations that fail to attain an adequate depth of insertion increases progressively along with advancing age in men and women, according to a new study by a researcher at the San Francisco VA Medical Center (SFVAMC). In addition, the study found, women are up to twice as likely as men to have inadequate exams. The findings, which are based on reviews of thousands of records of sigmoidoscopies, suggest that a patient’s age and sex are important factors to consider in choosing among various colorectal cancer screening methods.

While the study does not pinpoint reasons for the climb in inadequate exams, the study’s author speculates that several factors that accumulate with age may be to blame, including medical conditions that can complicate the procedure.

Flexible sigmoidoscopy, one of the most common screening tools for colorectal cancer, is a fairly simple procedure that is commonly performed in a doctor’s examination room. A sigmoidoscope — a 60 centimeter-long flexible tube about the thickness of a finger — is threaded into the patient’s rectum and lower third of the colon. A tiny video camera in the sigmoidoscope allows the doctor to examine the wall of the colon for polyps, cancers or other abnormalities. In practice, the sensitivity of this procedure as a screening tool depends on how much of the colon can be viewed. A reach of 50 to 60 centimeters into the colon from the anus is considered adequate. Yet, until now, the frequency with which sigmoidoscopies attain adequate reach during routine screening has been unknown. Nor has data been available to compare performance of sigmoidoscopies between men and women or people in different age groups.

To determine the adequacy of the procedure among various age groups, Louise Walter, MD, a staff physician in geriatrics at SFVAMC and assistant professor of medicine at University of California, San Francisco, reviewed records of more than 15,000 sigmoidoscopies performed for routine colorectal cancer screening between 1997 and 2001. She found that the percentage of exams that failed to reach 50 centimeters into the colon increased from 10 percent in men aged 50 to 59, to 22 percent in those 80 years or older. The corresponding percentages of inadequate examinations in women climbed from 19 percent in the younger age group to 32 percent in women 80 years and older.

“I had been finding that a lot of my older patients were not getting adequate exams,” Walter says, “and I wanted to know whether this was a widespread problem. I was initially approaching this as an age issue. But then the gender differences popped up. The most disturbing thing this study shows is that women are twice as likely as men to have an inadequate exam.” The study is published in the Feb.1 issue of the American Journal of Medicine.

Colorectal cancer is the second most common cause of cancer-related death in the United States and almost 80 percent of colorectal cancer deaths occur in people older than 65 years. Yet there is disagreement about the best way to screen for these cancers. Some professional and governmental organizations recommend colonoscopy — a procedure that examines the entire colon and is performed on patients under sedation — while others recommend screening with flexible sigmoidoscopy every 5 years or annual fecal occult blood screening, or both.

“It was always a dilemma for me when a patient asked which test I recommended,” Walter says. “Other studies have never given a clear answer. Many guidelines say that it depends on the patient’s preferences. It does depend on the patient’s preferences, but it also depends on which test is most likely to succeed. My study didn’t look at the benefits or harms [potential complications of the procedure] of this test. It looked at adequacy, and for these particular populations [older people, and particularly older women] the study shows that a doctor might be better off recommending the fecal occult blood test, or going for the whole colonoscopy.”

An additional factor that must be taken under consideration in making a recommendation, Walter says, is the fact that older age and female sex are both associated with an increased likelihood that colon polyps and cancers will occur higher up in the colon. Cancer registries report that more than 50 percent of all colorectal cancers in women older than 70 are located beyond the sigmoid colon.

“I see the results of this study as further evidence that argues against the use of sigmoidoscopies to screen older women,” Walter says. “But even an inadequate exam will provide some benefit. It’s certainly better than no screening at all.”

The records that Walter examined did not identify reasons for the failure of sigmoidoscopies to attain adequate reach. But she cites two factors that may contribute to the differences she found between the sexes: the literature shows that women have more pain than men with sigmoidoscopies, which could prevent completion of the procedure, and women have a more pronounced curve in the sigmoid colon, which can obstruct the sigmoidoscope. One reason she did find for declining success rates of sigmoidoscopies with advancing age was that older people were less successful in attaining adequate bowel preparation, the tedious process of cleansing the bowels prior to the procedure. But there must be additional contributing factors, Walter says, because bowel preparation accounted for only a portion of the unsuccessful procedures.

The study is based on records from all 71 non-VA clinics and hospitals that participate in the Clinical Outcomes Research Initiative, a nationwide effort, supported by NIH and AstraZeneca Pharmaceuticals LP, to evaluate the outcomes of endoscopic procedures that are performed in clinical settings. Walter examined all 15,406 records of sigmoidoscopies for colorectal cancer screening on non-symptomatic people age 50 and older that were performed between April 1, 1997 and October 1, 2001.

Other researchers who participated in the study were Kenneth Covinsky, MD, MPH, staff physician at SFVAMC and UCSF assistant professor of medicine, and Pat de Garmo, ANP, MUP, Oregon Health Sciences University, Portland, OR.

The study was supported by a Veterans Affairs Career Development Award in Health Services Research and Development granted to Dr. Walter.