The involvement of gastroenterology fellows in the administration of colonoscopy procedures may improve adenoma detection rate, according to study results published in Gastrointestinal Endoscopy. Compared with colonoscopies performed by an attending physician alone, colonoscopies performed by fellows under staff endoscopist supervision resulted in significantly greater rates of adenoma detection.
Investigators conducted a randomized clinical trial between April and December of 2019. Patients at 4 tertiary hospitals were randomized to undergo colonoscopy performed by either a fellow under endoscopist supervision (n=424) or by an attending physician alone (n=419). The control arm was led by 8 board-certified gastroenterologists with the combined experience of over 5000 colonoscopies. In the investigation arm, colonoscopies were performed by 13 gastroenterology fellows under the supervision of the aforementioned attending physicians. Supervising staff were present in the examination room for the entirety of procedures and provided verbal coaching. All patients received polyethylene glycol-based bowel preparation prior to colonoscopy. The same sedatives and endoscopy scopes were used at all study sites. Detected lesions were endoscopically removed and sent to pathology labs for histological analysis. The primary endpoint was adenoma detection rate. Secondary endpoints were advanced adenoma detection rate, sessile serrated adenoma detection rate, mean adenoma per colonoscopy, and mean polyp per colonoscopy. Logistic regression models were used to investigate the effect of treatment condition on adenoma detection rates.
Of 843 patients in the 2 study groups, 812 were included in analyses (406 in each arm). No significant demographic differences were observed between study groups. Median age was 59 years (range, 49-66) in the intervention group and 61 years (range, 52-68) in the control group (P =.78). The majority of patients in the intervention and control groups were men (55.9% and 53.2%, respectively; P =.43). For all patients, the most common indication for colonoscopy was colorectal cancer screening. The adenoma detection rate with fellow involvement was 44.8%, compared with 37.1% in the control group (P =.02). The mean number of adenomas per colonoscopy was significantly greater in the intervention arm compared with the control arm (0.65 ± 0.3 vs 0.53 ± 0.2; P <.001). Polyp detection rate was 69.7% in the fellow involvement arm and 62.5% in the control arm (P =.03). Mean number of polyps per colonoscopy was greater in the intervention group (1.01 ± 0.5 vs 0.94 ± 0.3; P <.001). Advanced and sessile serrated adenoma detection rates were not significantly different between trial arms. Fellows with ≥150 procedures worth of experience had higher adenoma detection rates than fellows with <150 procedures worth of experience (49.5% vs 39.7%; P =.04). However, detection rates were not significantly different between less experienced fellows and attending physicians alone (P =.53). In multivariate analyses, bowel preparation (OR, 2.32; 95% CI, 1.56-3.18; P =.001) and fellow involvement (OR, 1.55; 95% CI, 1.21-2.89; P =.01) emerged as significant predictors of adenoma detection rate.
These data suggest that fellow involvement in endoscopy procedures may be beneficial in increasing rates of adenoma detection. The beneficial effect was enhanced when the involved fellow had greater experience (≥150 procedures).
Investigators noted that participants could not be blinded to their study condition, which could potentially limit the generalizability of the findings. In addition, indications for colonoscopy were numerous among patients, which may have introduced significant heterogeneity among the study population. Even so, these data provide prospective validation of the benefits of fellow involvement in colonoscopy. “Our findings support the growing body of data suggesting that addition of a second observer might increase the rates of colon adenoma and polyp detection,” investigators wrote.
Facciorusso A, Buccino VR, Tonti P, et al. Impact of fellow participation on colon adenoma detection rates: a multicenter randomized trial [published online May 17, 2020]. Gastrointest Endosc. doi: 10.1016/j.gie.2020.05.015