番茄社区

Trainee involvement during colonoscopy offered mixed results when it came to exam quality and polyp detection, according to a population-based cohort study from Canada.

Having a gastroenterology or general surgery resident or fellow on board did not reduce the overall quality of the colonoscopy procedure, but their involvement did reduce the detection rate of sessile serrated polyps, reported Michael Sey, MD, MPH, of Western University in London, Ontario, and colleagues.

Looking at nearly 35,500 colonoscopies, univariate analysis showed significant differences in polyp detection rate when trainees participated versus when they did not (39.2% vs 42.0%, respectively) and in the sessile serrated polyp detection rate specifically (4.4% vs 5.2%).

But only the latter remained significant on multivariable analysis that adjusted for endoscopist, patient, and procedural confounders (risk ratio 0.79, 95% CI 0.64-0.98, P=0.03).

"Detection of sessile serrated polyps is highly important given that they have unique clinical and molecular profiles and account for up to one-third of colon cancers," the group wrote in . "Ultimately, sessile serrated polyps are difficult to detect endoscopically, and in light of the findings of our study, extra attention should be paid while withdrawing through the right colon when supervising trainees, such as a second look by the attending endoscopist."

"Unlike board-certified endoscopists, trainees have to balance the need for skill acquisition with the provision of patient care," Sey and colleagues added. "At times, these competing interests may affect clinical outcomes."

Concerns have been raised that trainee participation in colonoscopy can affect procedural quality, the authors noted, though prior studies have failed to provide evidence of a positive or negative impact. The group explained, however, that ADR is by far the most important quality assurance benchmark for colonoscopy "owing to its direct correlation with post-colonoscopy risk of colon cancer."

In the current study, no significant differences were seen on univariate analysis for adenoma detection rate (ADR) when trainees joined versus when they did not (26.4% vs 27.3%) or for cecal intubation rates (96.7% vs 97.2%) or perforation rates (0.05% vs 0.06%). And no differences on multivariable analysis for these key quality indicators were detected either.

"The results of this study are good news for those involved in training of gastroenterology fellows by showing that quality metrics are maintained during procedures where trainees are involved, but also suggests a caution that extra care should be exercised in examining the right colon when trainees are participating in the procedure," said David Greenwald, MD, of Mount Sinai Hospital in New York City, who was not involved in this study.

The findings should "allay concerns of patients regarding trainees, and demonstrates to clinicians what to be more cautious of," Allen Kamrava, MD, of Cedars-Sinai Medical Center in Los Angeles, told 番茄社区.

"Understandably, some patients are anxious about having residents in their procedures," said Kamrava, who was not involved in the research. "The study helps demonstrate that, other than [sessile serrated polyps detection rate], having a trainee in the room during endoscopy does not negatively impact the procedure."

Sey's group examined data on 35,499 consecutive colonoscopies for any indication performed by 71 board-certified endoscopists across 21 hospitals in Southwest Ontario from 2017 to 2018.

The vast majority of colonoscopies were performed to investigate symptoms (55%) or screening (42%), and nearly all were outpatient procedures. Good bowel preparation was reported in 83% of the cases.

Endoscopists had a mean practice time of 14 years and were split between general surgeons (54.5%) and gastroenterologists (45.5%). Endoscopists who did not perform at least 50 annual colonoscopies were excluded, as were those who did not supervise trainees.

Overall, 55 trainees were involved in 5,746 colonoscopies (16.7%). Colonoscopies that involved trainees were more likely to be diagnostic (58% vs 54%), inpatient procedures (13% vs 3%), and were more likely to use sedation (83% vs 67%). Trainee procedures were also more likely to have poor bowel preparation (5% vs 3%), more gastroenterologist supervision (61% vs 42%), and were more likely to be conducted at academic centers (88% vs 48%).

Study limitations included the fact that the authors did not directly examine colonoscopy withdrawal time and, despite trainee hands-on participation, the specific role or extent of trainee involvement during colonoscopy was not documented.

Comment