Phillip Lum

Natural Language Processing As an Alternative to Manual Reporting of Colonoscopy Quality Metrics

BACKGROUND AND AIMS The adenoma detection rate (ADR) is a quality metric tied to interval colon cancer occurrence. However, manual extraction of data to calculate and track the ADR in clinical practice is labor-intensive. To overcome this difficulty, we developed a natural language processing (NLP) method to identify adenomas and sessile serrated adenomas (SSAs) in patients undergoing their first screening colonoscopy. We compared the NLP-generated results with that of manual data extraction to test the accuracy of NLP and report on colonoscopy quality metrics using NLP. METHODS Identification of screening colonoscopies using NLP was compared with that using the manual method for 12,748 patients who underwent colonoscopies from July 2010 to February 2013. Also, identification of adenomas and SSAs using NLP was compared with that using the manual method with 2259 matched patient records. Colonoscopy ADRs using these methods were generated for each physician. RESULTS NLP correctly identified 91.3% of the screening examinations, whereas the manual method identified 87.8% of them. Both the manual method and NLP correctly identified examinations of patients with adenomas and SSAs in the matched records almost perfectly. Both NLP and the manual method produced comparable values for ADRs for each endoscopist and for the group as a whole. CONCLUSIONS NLP can correctly identify screening colonoscopies, accurately identify adenomas and SSAs in a pathology database, and provide real-time quality metrics for colonoscopy.

Large Sessile Serrated Polyps Can Be Safely and Effectively Removed by Endoscopic Mucosal Resection.

BACKGROUND & AIMS As many as 50% of large sessile serrated adenomas/polyps (SSPs) are removed incompletely, which is significant because SSPs have been implicated in the development of interval cancers. It is unclear if endoscopic mucosal resection (EMR) is an optimal method for removal of SSPs. We assessed the efficacy and safety of removal of SSPs 10 mm and larger using a standardized inject-and-cut EMR technique. METHODS We performed a retrospective analysis of colonoscopy data, collected over 7 years (2007-2013) at 2 centers, from 199 patients with proximal colon SSPs 10 mm and larger (251 polyps) removed by EMR by 4 endoscopists. The primary outcome measure was local recurrence. The secondary outcome measure was safety. RESULTS At the index colonoscopy, patients had a median of 1 serrated lesion (range, 1-12) and 1 nonserrated neoplastic lesion (range, 0-15). The mean SSP size was 15.9 ± 5.3 mm; most were superficially elevated (84.5%) and located in the ascending colon (51%), and 3 SSPs (1.2%) had dysplasia. Surveillance colonoscopies were performed on 138 patients (69.3%) over a mean follow-up period of 25.5 ± 17.4 months. Of these patients, 5 had local recurrences (3.6%; 95% confidence interval, 0.5%-6.7%), detected after 17.8 ± 15.4 months, with a median size of 4 mm. No patients developed postprocedural bleeding, perforation, or advanced colon cancer, or had a death related to the index colorectal lesion during the study period. CONCLUSIONS Inject-and-cut EMR is a safe and effective technique for the resection of SSPs. Less than 5% of patients have a local recurrence, which is usually small and can be treated endoscopically.

Tu2027 Endoscopic Versus Percutaneous Drainage for Management of Hilar Cholangiocarcinoma

Background: The incidence of hepatocellular carcinoma (HCC) has increased significantly in United States over the last few decades in parallel with the epidemic of nonalcoholic fatty liver disease (NAFLD). Limited data suggest that HCC could arise in steatotic livers without the presence of cirrhosis. The aim of the current study was to characterize patients with NAFLD presenting with HCC in non-cirrhotic liver (NCL) compared to those with HCC in association with liver cirrhosis (LC). Methods: A retrospective analysis was performed on all patients with HCC and NAFLD diagnosis seen at our institution between 2003 and 2012. We characterized the patients with respect to demographic, clinical, histological and tumor features. Comparisons between the NCL and LC groups were done using t-test, Mann Whitney U test, and Chi square test as appropriate. P value 5 cm) (77.8% vs. 23 %), and receive hepatic resection as the modality for HCC treatment (66.7% vs. 17%); and were less likely to receive loco-regional therapy (16.7% vs. 51.1%) or liver transplant (0% vs. 72.3%), p value < 0.001 for all. Furthermore, 62% of patients without cirrhosis had HCC recurrence compared to only 12.5% in patients with cirrhosis (p < 0.001) and had worse survival with mortality rate of 41.7% at last follow up for the HCC-NCL group compared to 27.7% in the cirrhotic group (p = 0.008). Conclusion: Patients with HCC in the absence of liver cirrhosis are more likely to present at an older age with larger tumor and have higher rates of tumor recurrence. Studies to assess the cost-effectiveness of HCC surveillance in this group should be conducted.