SAN DIEGO — Colonoscopies performed first thing in the morning yielded significantly more polyps and more histologically confirmed polyps than did those performed later in the day, according to a study presented at the annual Digestive Disease Week.
“In medicine it's well known that errors accumulate, particularly in anesthesia and surgery, as the day progresses,” said Dr. Brennan M. Spiegel of the University of California, Los Angeles. “Any surgeon will tell you that he'd rather be the first case of the day if he has to go under the knife.” The study he presented appears to extend these results to both surveillance and screening colonoscopy.
Dr. Spiegel and his colleagues performed a retrospective analysis of 500 consecutive patients seen at the West Los Angeles Veterans Administration Medical Center between 2006 and 2007. At that institution, colonoscopy cases begin at 7:45 a.m. and typically end at 1 p.m. The investigators divided at that time into five segments that they analyzed separately.
Colonoscopists found a mean of 2.6 polyps per patient seen before 8:30 a.m., 2.1 polyps between 10 a.m. and 11:30 a.m., and 1.2 polyps after 1 p.m. On average, the first colonoscopy of the day found 20% more polyps than did those performed later in the day, a statistically significant difference. The trend line was also statistically significant.
The investigators noticed a similar pattern when they restricted their analysis to histologically confirmed polyps. The colonoscopists found a mean of 2.1 hyperplastic polyps during the first case of the day, 1.6 in cases between 10 a.m. and 11:30 a.m., and 1.1 in cases after 1 p.m. That trend line also was statistically significant, but there was no significant difference between the time periods in the ratio of hyperplastic to adenomatous polyps.
“We were finding fewer of all polyps, not just hyperplastic or adenomatous,” Dr. Spiegel said.
The first case of the day remained a significant independent predictor of polyp yield even after the investigators controlled for a host of potential confounders in a multivariate analysis (P = .004). They controlled for patient-level factors such as age and body mass index, provider-level factors such as which individual colonoscopist performed the procedure and whether he or she was a fellow, and procedure-level factors such as the quality of the bowel prep and the withdrawal time.
Of those factors, the only other independent predictor of polyp yield was whether or not a fellow was participating in the procedure. Fellow participation was a strong predictor of higher yield (P = .00001). Dr. Spiegel suggested that one reason the presence of fellows may have improved yield is that there were, “two [sets of] eyes on the screen instead of one.”
The study generated some critical comment during the question-and-answer period. During his talk, Dr. Spiegel said that patients were assigned randomly to time slots, but by that he meant that no one intentionally assigned a specific type of case to a specific time period. But one audience member said that there might be some undetected bias in appointment times, with a certain type of patient choosing earlier or later appointments.
Another physician noted that some colonoscopy centers perform many more procedures per day than does the West Los Angeles VA Medical Center, and suggested that the result might have been different if the study had been conducted elsewhere.
Another audience member commented, “I worry that when the New York Times or the Wall Street Journal gets ahold of this paper and publishes it widely, we're going to begin to have great difficulties scheduling patients in the afternoon.”
Dr. Spiegel agreed that the study should be repeated elsewhere before anyone takes it too seriously. “And whether it's the New York Times or anyone else, we have to emphasize that we have no idea that this has an impact on advanced adenomatous cancer,” in terms of survival, he said.
But if the results are generalizable and colonoscopists are simply more vigilant earlier in the day, Dr. Spiegel wondered what could be done about this, “short of putting an electrical shock in the handle of the colonoscope.” He suggested that clinicians look to other industries, such as air-traffic control and long-distance trucking, that depend on constant vigilance. In those industries, strategies such as split-shift scheduling, visible prompts, and frequent reminders to be vigilant have proven to be helpful.
Dr. Spiegel acknowledged receiving consulting fees, research support, and/or other financial benefits from AstraZeneca, Ethicon, TAP Pharmaceutical, Novartis, and Procter & Gamble.