Fredy Nehme, MD1, Daniel Merrill, MD1, Harris Zamir, MD1, Ahmed ElKafrawy, MD1, Laith Al Momani, MD1, Hasan Bader, MD1, Tejas Joshi, MD2, Mohammad Alomari, MD3, Wendell Clarkston, MD1; 1University of Missouri, Kansas City, MO; 2Louisiana State University School of Medicine, New Orleans, LA; 3East Tennessee State University, Johnson City, TN
Introduction: Accurate estimation of polyp size during screening colonoscopy is an important factor in determining future surveillance intervals for colorectal cancer screening. However, the determination of polyp size is based on a subjective assessment by the endoscopist during the procedure. Studies have reported substantial discordance between endoscopic and pathology-based estimates of polyp size. Trainee participation in screening colonoscopy may lead to improvement in some quality metrics such as adenoma detection rate. It remains unclear whether the participation of GI fellows affects other procedural factors. We aim to investigate whether the participation of GI fellows at different stages of training affect accurate sizing of polyps compared to screening colonoscopies performed by attendings without fellows. Methods: We performed a retrospective review of all patients who underwent a screening colonoscopy at our institution between January 2018 and June 2019. We included polyps resected in total and having both endoscopic estimation and pathology measurements documented. Polyps reported as fragmented on pathology review were excluded. Categorical variables were compared using Chi-Square test with a 95% confidence interval. Results: A total of 895 colonoscopies were reviewed. From 950 polyps resected during the study period, 514 met the inclusion criteria and were included in the analysis, of which 43% were performed with a GI fellow. Polyp size was accurately estimated in 38.2% of diminutive polyps as compared to 18.3% of polyps ≥ 1 cm (p< 0.001). Polyp size accuracy was comparable with and without participation of a GI fellow (p=0.56). The polyp was overestimated in 28.1% of cases with a GI fellow compared to 22.1% without a GI fellow (p=0.127). 58% of polyps estimated ≥ 1 cm on endoscopy were < 1 cm on pathology and this was not affected by fellow participation. In addition, there was no significant difference in polyp size accuracy between fellows at different levels of training. Discussion: Inaccuracy in the appraisal of polyps is high, especially for polyps ≥ 1 cm and is unaffected by fellow participation regardless of training stage. While technical colonoscopy skills are emphasized and continue to improve during fellowship training, GI fellows could benefit from additional training in polyp size estimation. Using of biopsy forceps or snare to estimate polyp size may be helpful tools when learning this important aspect of polyp assessment.
Disclosures: Fredy Nehme indicated no relevant financial relationships. Daniel Merrill indicated no relevant financial relationships. Harris Zamir indicated no relevant financial relationships. Ahmed ElKafrawy indicated no relevant financial relationships. Laith Al Momani indicated no relevant financial relationships. Hasan Bader indicated no relevant financial relationships. Tejas Joshi indicated no relevant financial relationships. Mohammad Alomari indicated no relevant financial relationships. Wendell Clarkston indicated no relevant financial relationships.