Comparing Colonoscopy Quality Indicators between Surgeons and Gastroenterologists in A Rural Healthcare System

Adjusting Detection Measures for Colonoscopy: How Far Should We Go?

Uptake of Colorectal Cancer Screening in 45 to 49 Year Olds: An Early-Inning View from the Endoscopy Suite

Colonoscopy

Cuff-assisted colonoscopy: Should you be riding the mechanical enhancement wave?

INTRODUCTION: Adenoma detection rate (ADR) and its surrogate marker, mean withdrawal time (WT) of ≥ 6 min, is the current golden metric for colonoscopy quality. Serrated polyp detection rate (SPDR) is also an important marker for CRC. Longer WTs are linked with higher ADRs. We aim to assess if there is a correlation between SPDR and ADR, and if the same link exists between WTs and SPDR, as it exists for WT on ADR. METHODS: Screening colonoscopies between 2012 and 2017 were evaluated. Endoscopists with < 40 colonoscopies and procedures with polyps >6 mm, and/or WTs >30 min were excluded. Data was grouped into fellow (FG) and attending (AG) groups. The bivariate association of SPDR and ADR was calculated by correlation coefficients. Statistical analysis was performed using chi-squared tests, two-sample t-tests, and ANOVA t-testing. RESULTS: A total of 5951 procedures performed by 51 fellows and 13 attendings were assessed. The MWT was 15.4 min, SD 4.72, average total procedure time (TT) was 28.3 min, and average ADR was 40%. WT had a strong positive correlation with ADR and TT. The average SPDR was 0.1, SD 0.056. SPDR had a positive correlation with WT and ADR. There were significant differences between FG and AG in ADR, WT, TT and in regression models between the two for TT. There was no difference in the mean SPDR between groups. SPDR had a strong positive correlation with ADR and WT in the AG. There was no significant difference between bowel prep vs. SPDR between groups. CONCLUSION: Our study demonstrates a strong linear relationship between WT, ADR and TT. While FG have a higher ADR compared with AG at the cost of a longer WT, how much ADR changes as WT increases is not significant. AG and FG had similar SPDRs, however, only for AG were SPDR positively correlated with longer WTs and higher ADRs. We postulate that given the increased technical difficulty in detecting sessile polyps, AG are better equipped to improve SPDRs with increased WT. Bowel prep scores had no effect on SPDR, explained by exclusion criteria including only fully completed colonoscopies, and therefore subpar prep patients were likely pre-excluded. Our study highlights the importance of expertise and training in the identification of sessile polyps, the association between high ADR and SPDR, and the positive effect of WT on SPDR. Kahi et al., identified variability in SPDR ranging from 1% to 18%, compared with much less variability seen in ADRs. As a result, our findings are worthy of validation in larger multiple center studies.

Introduction: Measuring withdrawal time (WT) is a vital component of colonoscopy quality improvement but is limited in practice. Machine learning (ML) may automate quality metrics measurement and facilitate directed feedback. We aimed to 1) develop an algorithm to measure WT (ML-WT) and 2) assess the correlation of ML-WT with manual WT and adenoma detection rate (ADR). Methods: Endoscopy procedures at a single academic center were recorded beginning in 3/2018 using a cloud-based video recording system. We developed an algorithm to calculate ML-WT via identification of cecal landmarks (appendiceal orifice, ileum, ileocecal valve, and cecal base). We also calculated “high-quality” ML-WT, defined as frames during withdrawal where the lumen was clearly visible (excluding blurry or “red out” frames). Image processing was handled by an 18-layer Resnet convolutional neural network. The model was trained on predictive tasks for detection on 9,498 images. To ensure a realistic distribution, approximately 30% of the training frames were poor quality. ADR was calculated using a 2-year historical mean. We only included colonoscopists with ≥25 recorded normal (i.e., no biopsy or polypectomy) screening or surveillance colonoscopies. Manual WT was extracted for each corresponding procedure from the electronic health record. We used the Pearson correlation coefficient to assess the relationships between normal colonoscopy WT, ML-WT and ADR. Results: A total of 16 colonoscopists met inclusion criteria (median historical ADR 39.2%; Interquartile Range [IQR] 35.0%, 44.8%). ML-WT was calculated on a total of 1,823 normal screening and surveillance colonoscopies. Median ML-WT was 12.7 min (IQR 10.9, 14.5). Colonoscopist mean ML-WT very strongly correlated with their historical manually calculated mean WT (r=0.97; Figure 1a). The absolute difference between mean physician ML-WT and WT was 0.86 min (SD 0.7). Overall, 79% of the ML-WT was “high-quality” (frames interpretable for polyp detection). ML-WT moderately correlated with endoscopist historical ADR (r=0.66; Figure 1b). Similarly, high-quality ML-WT moderately correlated with ADR (r=0.69; Figure 1c). Conclusion: We report the development of an accurate automated ML assessment of WT that correlates with ADR in a large cohort of screening colonoscopists. We propose utilizing ML-WT to measure and improve colonoscopy quality in settings where abstraction of quality metrics is infeasible.Figure 1:: A. Machine learning withdrawal time (ML-WT) strongly correlates with manually calculated withdrawal time (WT) B. ML-WT moderately correlates with adenoma detection rate C. High-quality ML-WT (time spent during withdrawal where the colon mucosa can be clearly evaluated) moderately correlates with adenoma detection rate.

Adhering to quality metrics in colonoscopy: we can do better

Differences in Detection Rates of Adenomas and Serrated Polyps in Screening Versus Surveillance Colonoscopies, Based on the New Hampshire Colonoscopy Registry

Improving measurement of the adenoma detection rate and adenoma per colonoscopy quality metric: the Indiana University experience

Impact of sedation type on adenoma detection rate by colonoscopy

Introduction: Sessile serrated polyps are important precursors of colon cancer, but are more difficult to detect at colonoscopy. While serrated polyp detection rate (SDR) may be an important measure of colonoscopy quality, an accurate diagnosis of a serrated polyp is dependent upon pathologist expertise. It is unclear whether established quality metrics of withdrawal time (WT) and adenoma detection rate (ADR) sufficiently correlate with SDR. The primary aim is to determine the correlation of SDR with WT and ADR. The secondary aim is to determine the SDR in high ADR endoscopists. Methods: This is a retrospective cohort study of all providers at a single academic medical center performing ≥200 eligible screening colonoscopies over a 26-month period (4/2014-5/2016). An eligible screening colonoscopy was a colonoscopy in a patient age 50-75 without history of neoplastic polyps. SDR and ADR are the proportion of screening colonoscopies in which a serrated polyp (SDR) or adenoma (ADR) was biopsied or removed. WT is the time spent on colonoscope withdrawal in screening colonoscopies in which no biopsy/polypectomy occurred. Endoscopists were stratified into tertiles by ADR (High/Intermediate/Low ADR Endoscopists). All data was abstracted from our institutional data warehouse which integrates pathology, demographics, and endoscopy reports. Results: The 24 endoscopists who met inclusion criteria performed a total of 15,168 screening colonoscopies (median volume 486; range 200-1126). All endoscopists had a mean WT of ≥6 minutes (range 6-20 min). The mean institutional ADR was 39% (range 21-59%) and SDR was 8% (range 0.5-19%). There was a strong positive correlation between WT and both ADR (r=0.72, p < 0.0001) and SDR (r=0.71, p < 0.0001; Fig 1). There was a very strong positive correlation between ADR and SDR (r=0.86, p < 0.0001; Fig 2). The pooled SDR among High ADR endoscopists was 13%, which was significantly higher than Intermediate (SDR 7%, p < 0.01) and Low (SDR 4%, p < 0.001) ADR endoscopists.Figure 1Figure 2Conclusion: Screening colonoscopy SDR varies widely between endoscopists, identifying important variations in screening colonoscopy quality. However, there is a very strong correlation between endoscopist ADR and SDR. As SDR is more dependent on pathologist expertise than ADR, there appears to be no additional benefit in calculating endoscopist SDR to measure screening colonoscopy quality. Improvements in ADR may lead to simultaneous improvements in SDR, though this requires further study.

Impact of withdrawal time on adenoma detection rate: results from a prospective multicenter trial

In Search of Quality Colonoscopy

Impact of artificial intelligence-assisted colonoscopy on gastroenterology fellow performance: a pragmatic randomized controlled trial.

Background/AimsThe adenoma detection rate (ADR) is used as a quality indicator for screening and surveillance colonoscopy. The study aimed to determine if moderate versus deep sedation affects the outcomes of the ADR and other quality metrics in the veteran population.MethodsA retrospective review of colonoscopies performed at Memphis Veterans Affairs Medical Center over a one-year period was conducted. A total of 900 colonoscopy reports were reviewed. After exclusion criteria, a total of 229 index, average-risk screening colonoscopies were identified. Data were collected to determine the impact of moderate (benzodiazepine plus opioids) versus deep (propofol) sedation on the ADR, polyp detection rate (PDR), and withdrawal time.ResultsAmong 229 screening colonoscopies, 103 (44.9%) used moderate sedation while 126 (55%) were done under deep sedation. The ADR and PDR were not significantly different between moderate versus deep sedation at 35.9% vs. 37.3% (p=0.82) and 58.2% vs. 48.4% (p=0.13), respectively. Similarly, there was no significant difference in withdrawal time between moderate and deep sedation (13.4 min vs. 14 min, p=0.56) during screening colonoscopies.ConclusionsIn veterans undergoing index, average-risk screening colonoscopies, the quality metrics of the ADR, PDR, and withdrawal time are not influenced by deep sedation compared with moderate sedation.