Using Computer Vision to Improve Endoscopic Disease Quantification in Therapeutic Clinical Trials of Ulcerative Colitis.

BACKGROUND & AIMS: Endoscopic assessment of ulcerative colitis (UC) typically reports only the maximum severity observed. Computer vision methods may better quantify mucosal injury detail, which varies among patients. METHODS: Endoscopic video from the UNIFI clinical trial (A Study to Evaluate the Safety and Efficacy of Ustekinumab Induction and Maintenance Therapy in Participants With Moderately to Severely Active Ulcerative Colitis) comparing ustekinumab and placebo for UC were processed in a computer vision analysis that spatially mapped Mayo Endoscopic Score (MES) to generate the Cumulative Disease Score (CDS). CDS was compared with the MES for differentiating ustekinumab vs placebo treatment response and agreement with symptomatic remission at week 44. Statistical power, effect, and estimated sample sizes for detecting endoscopic differences between treatments were calculated using both CDS and MES measures. Endoscopic video from a separate phase 2 clinical trial replication cohort was performed for validation of CDS performance. RESULTS: Among 748 induction and 348 maintenance patients, CDS was lower in ustekinumab vs placebo users at week 8 (141.9 vs 184.3; P < .0001) and week 44 (78.2 vs 151.5; P < .0001). CDS was correlated with the MES (P < .0001) and all clinical components of the partial Mayo score (P < .0001). Stratification by pretreatment CDS revealed ustekinumab was more effective than placebo (P < .0001) with increasing effect in severe vs mild disease (-85.0 vs -55.4; P < .0001). Compared with the MES, CDS was more sensitive to change, requiring 50% fewer participants to demonstrate endoscopic differences between ustekinumab and placebo (Hedges' g = 0.743 vs 0.460). CDS performance in the JAK-UC replication cohort was similar to UNIFI. CONCLUSIONS: As an automated and quantitative measure of global endoscopic disease severity, the CDS offers artificial intelligence enhancement of traditional MES capability to better evaluate UC in clinical trials and potentially practice.

Variation in colectomy rates for benign polyp and colorectal cancer.

BACKGROUND: Removal of pre-cancerous polyps on screening colonoscopy is a mainstay of colorectal cancer (CRC) prevention. Complex polyps may require surgical removal with colectomy, an operation with a 17% morbidity and 1.5% mortality rate. Recently, advanced endoscopic techniques have allowed some patients with complex polyps to avoid the morbidity of colectomy. However, the rate of colectomy for benign polyp in the United States is unclear, and variation in this rate across geographic regions has not been studied. We compared regional variation in colectomy rates for CRC versus benign polyp. METHODS: We performed a retrospective population-based study of Medicare beneficiaries undergoing colectomy for CRC or benign polyp, using the 100% Medicare Provider Analysis and Review files from 2010 to 2015. We used multivariable linear regression to obtain population-based colectomy rates for CRC and benign polyp at the hospital referral region (HRR) level, adjusted for age, sex, and race. RESULTS: Of 280,815 patients, 157,802 (65.8%) underwent colectomy for CRC compared to 81,937 (34.2%) for benign polyp. Across HRRs, colectomy rates varied 5.8-fold for cancer (0.32-1.84 per 1000 beneficiaries). However, there was a 69-fold variation for benign polyp (0.01-0.69). While the rate of colectomy for CRC was correlated with the rate of colectomy for benign polyp (slope = 0.61, 95% CI 0.48-0.75), HRRs with the lowest or highest rates of colectomy for CRC did not necessarily have similarly low or high rates for benign polyp. CONCLUSIONS: The use of colectomy for benign polyp is much more variable compared to CRC, suggesting overuse of colectomy for benign polyp in some regions. This variation may stem from provider-level differences, such as endoscopists' referral practice or skill or surgeons' decision to perform colectomy, or from limited access to advanced endoscopists. Interventions to increase endoscopic resection of benign polyps may spare some patients the morbidity and cost of surgery.

Specialist Endoscopists Are Associated with a Decreased Risk of Incomplete Polyp Resection During Endoscopic Mucosal Resection in the Colon.

BACKGROUND: Endoscopic experience is known to correlate with outcomes of endoscopic mucosal resection (EMR), particularly complete resection of the polyp tissue. Whether specialist endoscopists can protect against incomplete polypectomy in the setting of known risk factors for incomplete resection (IR) is unknown. AIMS: We aimed to characterize how specialist endoscopists may help to mitigate the risk of IR of large sessile polyps. METHODS: This is a retrospective cohort study of patients who underwent EMR at the University of Michigan from January 1, 2006, to November 15, 2015. The primary outcome was endoscopist-reported polyp tissue remaining at the end of the initial EMR attempt. Specialist endoscopists were defined as endoscopists who receive tertiary referrals for difficult colonoscopy cases and completed at least 20 EMR colonic polyp resections over the study period. RESULTS: A total of 257 patients with 269 polyps were included in the study. IR occurred in 40 (16%) cases. IR was associated with polyp size ≥ 40 mm [adjusted odds ratio (aOR) 3.31, 95% confidence interval (CI) 1.38-7.93], flat/laterally spreading polyps (aOR 2.61, 95% CI 1.24-5.48), and difficulty lifting the polyp (aOR 11.0, 95% CI 2.66-45.3). A specialist endoscopist performing the initial EMR was protective against IR, even in the setting of risk factors for IR (aOR 0.13, 95% CI 0.04-0.41). CONCLUSIONS: IR is associated with polyp size ≥ 40 mm, flat and/or laterally spreading polyps, and difficulty lifting the polyp. A specialist endoscopist initiating the EMR was protective of IR.

Gastric Intestinal Metaplasia in an Underserved Population in the USA: Prevalence, Epidemiologic and Clinical Features.

Gastric intestinal metaplasia is an important stage in the development of gastric cancer. Limited data is available regarding the prevalence of gastric intestinal metaplasia in the United States. We conducted a retrospective review of esophagogastroduodenoscopies performed in our endoscopy unit between the months of April and October 2010 to evaluate the prevalence and the epidemiologic and endoscopic features of gastric intestinal metaplasia in an underserved population in the United States.

Hepatocellular carcinoma in a noncirrhotic patient with hereditary hemochromatosis.

We report the case of a 66-year-old man with genetic hemochromatosis who was found to have hepatocellular carcinoma (HCC) in the absence of cirrhosis. This is a rare and life-threatening complication of noncirrhotic hemochromatosis that has been described only 10 times in the English literature. In addition to presenting our case, we also cite some other potential causes of HCC in noncirrhotic patients that should be clinically considered. Although the incidence of HCC in noncirrhotic hemochromatosis patients is not sufficiently high to warrant routine screening, physicians should be aware that this fatal complication may rarely occur.