An Accurate and Automated Method for Adenoma Detection Rate and Report Card Generation Utilizing Common Electronic Health Records.

GOALS: To develop an automated method for Adenoma Detection Rate (ADR) calculation and report card generation using common electronic health records (EHRs). BACKGROUND: ADR is the most widely accepted colonoscopy quality indicator and is inversely associated with interval colorectal cancer incidence and mortality. However, ADR is difficult to efficiently measure and disseminate, due to need for data integration from distinct electronic databases. METHODS: We migrated data from an endoscopy reporting software (Endosoft) to Epic Reporting Servers where it was combined with anatomic pathology data (Beaker Lab Information System, EPIC Systems). A natural language processing expression was developed to search Beaker pathology reports for accurate identification of adenomatous polyps. A blinded physician manually validated a final cohort of 200 random procedures. ADR report cards were automatically generated utilizing the Crystal Reports feature within EPIC. RESULTS: Validation of the natural language processing algorithm for ADR showed a sensitivity, specificity, and accuracy of 100%. ADR was automatically calculated for 12 endoscopists over a calendar year. Two thousand two hundred seventy-six screening colonoscopies were performed with 775 procedures having a least one adenoma detected, for a total ADR of 34%. Report cards were successfully generated within the EPIC EHR and distributed to endoscopists by secure e-mail. CONCLUSION: We describe an accurate, automated and scalable process for ADR calculation and reporting utilizing commonly adopted EHRs and data integration methods. By integrating the process of ADR collection and streamlining dissemination of reports, this methodology is poised to enhance colonoscopy quality care across health care networks that use it.

'CROSS'-ing into the 'Real World': a retrospective cohort study of patients receiving trimodality and bimodality therapy for esophageal cancer.

Background: A standard of care for nonmetastatic esophageal cancer is trimodality therapy consisting of neoadjuvant chemoradiation and esophagectomy, with evidence for improved overall survival versus surgery alone in the ChemoRadiotherapy for Oesophageal cancer followed by Surgery Study (CROSS) trial. Patients who receive treatment with curative intent but are poor candidates for or decline surgery receive definitive bimodality therapy. Literature characterizing patients who receive bimodality therapy compared to trimodality therapy, and their relative outcomes, is sparse, especially among patients who are too old or too frail to qualify for clinical trials. In this study, we assess a single-institution real-world dataset of patients receiving bimodality and trimodality management. Methods: Patients treated for clinically resectable, nonmetastatic esophageal cancer between 2009 and 2019 who received bimodality or trimodality therapy were reviewed, generating a dataset of 95 patients. Clinical variables and patient characteristics were assessed for association with modality on multivariable logistic regression. Overall, relapse-free, and disease-free survival were assessed with Kaplan-Meier analyses and Cox proportional modeling. For patients nonadherent to planned esophagectomy, reasons for nonadherence were recorded. Results: Bimodality therapy was associated with greater age-adjusted comorbidity index, worse performance status, higher N-stage, presenting symptom other than dysphagia, and held chemotherapy cycles on multivariable analysis. Compared to bimodality therapy, trimodality therapy was associated with higher overall (3-year: 62% vs. 18%, P<0.001), relapse-free (3-year: 71% vs. 18%, P<0.001), and disease-free (3-year: 58% vs. 12%, P<0.001) survival. Similar results were observed among patients who did not meet CROSS trial qualifying criteria. Only treatment modality was associated with overall survival after adjusting for covariates (HR 0.37, P<0.001, reference group: bimodality). Patient choice accounted for 40% of surgery nonadherence in our population. Conclusions: Patients receiving trimodality therapy were observed to have superior overall survival compared to bimodality therapy. Patient preference for organ-preserving therapies appears to impact resection rate; further characterization of patient decision-making may be helpful. Our results suggest patients who wish to prioritize overall survival should be encouraged to pursue trimodality therapy and obtain early consultation with surgery. Development of evidence-based interventions to physiologically prepare patients before and during neoadjuvant therapy as well as efforts to optimize the tolerability of the chemoradiation plan are warranted.

Sporadic hyperplastic polyp associated with above-average risk of developing metachronous colorectal cancer.

Post-colonoscopy surveillance interval for colorectal polyps depends on the size, number, and pathological classification of removed polyps. The risk of sporadic hyperplastic polyps (HPs) for developing colorectal adenocarcinoma remains debatable due to limited data. We aimed to evaluate the risk of metachronous colorectal cancer (CRC) in patients with sporadic HPs. A total of 249 patients with historical HP(s) diagnosed in 2003 were included as the disease group, and 393 patients without any polyp as the control group. All historical HPs were reclassified into SSA or true HP based on the recent 2010 and 2019 World Health Organization (WHO) criteria. Polyp size was measured under light microscope. Patients developed CRC were identified from the Tumor Registry database. Each tumor was tested for DNA mismatch repair proteins (MMR) by immunohistochemistry. Results showed that 21 (8%) and 48 (19%) historical HPs were reclassified as SSAs based on the 2010 and 2019 WHO criteria, respectively. The mean polyp size of SSAs (6.7 mm) was significantly larger than HPs (3.3 mm) (P<0.0001). For polyp size ≥5 mm, the sensitivity, specificity, positive predictive value (PPV), and negative predictive value (NPV) for diagnosing SSA was 90%, 90%, 46%, and 99%, respectively. Left-sided polyps with size <5 mm were 100% of HPs. Five of 249 (2%) patients developed metachronous CRC during the 14-year follow-up from 2003 to 2017, including 2 of 21 (9.5%) patients with SSA diagnosed at intervals of 2.5 and 7 years, and 3 of 228 (1.3%) patients with HP(s) at 7, 10.3, and 11.9 years. Two of 5 cancers showed MMR deficiency with concurrent loss of MLH1/PMS2. Based on the 2019 WHO criteria, the rate of developing metachronous CRC in patients with SSA (P=0.0116) and HP (P=0.0384) was significantly higher than the control group, and no significant difference was observed between patients with SSA and with HP (P=0.241) in this cohort. Patients with either SSA or HP also had higher risk of CRC than average-risk US population (P=0.0002 and 0.0001, respectively). Our data add a new line of evidence that patients with sporadic HP are associated with above-average risk of developing metachronous CRC. Post-polypectomy surveillance for sporadic HP may be adjusted in future practice given the low but increased risk of developing CRC.

Patient and Endoscopic Characteristics of Postcolonoscopy Colon Cancer-A Case-control Study.

Background and Aims: Colonoscopy is imperfect for colorectal cancer (CRC) prevention. Postcolonoscopy CRC (PCCRC) is defined as CRC detected after a screening or surveillance colonoscopy. PCCRCs can be divided into noninterval CRC and interval CRC. We performed a case-control study to identify risk factors for PCCRCs and to compare risks between noninterval and interval PCCRCs. Methods: We designed a retrospective case-control study. Using a Vermont tumor registry data set, we identified all PCCRCs diagnosed at our medical center from January 2012 to September 2017. Cases were matched 1:3 with controls of the same age, sex, and index colonoscopy date. Results: Fifty-four PCCRCs were matched with 162 controls and divided into noninterval (N = 27) and interval (N = 27) subsets. Overall PCCRC risk and noninterval PCCRC risk were significantly associated with history of polyps (odds ratio [OR] PCCRC = 2.71, OR noninterval = 4.41), sessile serrated polyps (OR PCCRC = 3.94, OR noninterval = 5.79), and high-risk adenoma (HRA) (OR PCCRC = 6.58, OR noninterval = 16.46) and with the index colonoscopy having a large polyp (OR PCCRC = 4.45, OR noninterval = 10.46) or having an HRA (OR PCCRC = 3.68, OR noninterval = 8.04). PCCRC risk and interval PCCRC risk were significantly associated with follow-up recommendations that did not correlate with American Gastroenterological Association surveillance guidelines (OR PCCRC = 3.30, OR interval = 4.85). Approximately 30% of PCCRCs could be attributed to endoscopic quality. Conclusion: Overall PCCRC risk and noninterval PCCRC risk were significantly associated with traditional CRC risk factors including precancerous polyps and HRA on the index colonoscopy. Interval PCCRC was not associated with these risk factors. Many PCCRCs can be attributed to endoscopic quality, and nonadherence to CRC surveillance guidelines may be a novel risk factor.

Prospective development and validation of a volumetric laser endomicroscopy computer algorithm for detection of Barrett's neoplasia.

BACKGROUND AND AIMS: Volumetric laser endomicroscopy (VLE) is an advanced imaging modality used to detect Barrett's esophagus (BE) dysplasia. However, real-time interpretation of VLE scans is complex and time-consuming. Computer-aided detection (CAD) may help in the process of VLE image interpretation. Our aim was to train and validate a CAD algorithm for VLE-based detection of BE neoplasia. METHODS: The multicenter, VLE PREDICT study, prospectively enrolled 47 patients with BE. In total, 229 nondysplastic BE and 89 neoplastic (high-grade dysplasia/esophageal adenocarcinoma) targets were laser marked under VLE guidance and subsequently underwent a biopsy for histologic diagnosis. Deep convolutional neural networks were used to construct a CAD algorithm for differentiation between nondysplastic and neoplastic BE tissue. The CAD algorithm was trained on a set consisting of the first 22 patients (134 nondysplastic BE and 38 neoplastic targets) and validated on a separate test set from patients 23 to 47 (95 nondysplastic BE and 51 neoplastic targets). The performance of the algorithm was benchmarked against the performance of 10 VLE experts. RESULTS: Using the training set to construct the algorithm resulted in an accuracy of 92%, sensitivity of 95%, and specificity of 92%. When performance was assessed on the test set, accuracy, sensitivity, and specificity were 85%, 91%, and 82%, respectively. The algorithm outperformed all 10 VLE experts, who demonstrated an overall accuracy of 77%, sensitivity of 70%, and specificity of 81%. CONCLUSIONS: We developed, validated, and benchmarked a VLE CAD algorithm for detection of BE neoplasia using prospectively collected and biopsy-correlated VLE targets. The algorithm detected neoplasia with high accuracy and outperformed 10 VLE experts. (The Netherlands National Trials Registry (NTR) number: NTR 6728.).

Expert assessment on volumetric laser endomicroscopy full scans in Barrett's esophagus patients with or without high grade dysplasia or early cancer.

BACKGROUND: Volumetric laser endomicroscopy (VLE) allows for near-microscopic imaging of the superficial esophageal wall and may improve detection of early neoplasia in Barrett's esophagus (BE). Interpretation of a 6-cm long, circumferential VLE "full scan" may however be challenging for endoscopists. We aimed to evaluate the accuracy of VLE experts in correctly diagnosing VLE full scans of early neoplasia and non-dysplastic BE (NDBE). METHODS: 29 VLE full scan videos (15 neoplastic and 14 NDBE) were randomly evaluated by 12 VLE experts using a web-based module. Experts were blinded to the endoscopic BE images and histology. The 15 neoplastic cases contained a subtle endoscopically visible lesion, which on endoscopic resection showed high grade dysplasia or cancer. NDBE cases had no visible lesions and an absence of dysplasia in all biopsies. VLE videos were first scored as "neoplastic" or "NDBE." If neoplastic, assessors located the area most suspicious for neoplasia. Primary outcome was the performance of VLE experts in differentiating between non-dysplastic and neoplastic full scan videos, calculated by accuracy, sensitivity, and specificity. Secondary outcomes included correct location of neoplasia, interobserver agreement, and level of confidence. RESULTS: VLE experts correctly labelled 73 % (95 % confidence interval [CI] 67 % - 79 %) of neoplastic VLE videos. In 54 % (range 27 % - 66 %) both neoplastic diagnosis and lesion location were correct. NDBE videos were consistent with endoscopic biopsies in 52 % (95 %CI 46 % - 57 %). Interobserver agreement was fair (kappa 0.28). High level of confidence was associated with a higher rate of correct neoplastic diagnosis (81 %) and lesion location (73 %). CONCLUSIONS: Identification of subtle neoplastic lesions in VLE full scans by experts was disappointing. Future studies should focus on improving methodologies for reviewing full scans, development of refined VLE criteria for neoplasia, and computer-aided diagnosis of VLE scans.

Volumetric laser endomicroscopy interpretation and feature analysis in dysplastic Barrett's esophagus.

BACKGROUND AND AIM: Volumetric laser endomicroscopy (VLE) is used to identify Barrett's esophagus (BE) dysplasia. Selection of a dysplastic region of interest (ROI) can be challenging due to feature variability across a large amount of data. The degree of agreement among VLE users in selecting a ROI has not been studied. METHODS: High-definition videos that divided a VLE scan from 18 patients with biopsy-proven BE dysplasia into 1-cm segments were reviewed using a four-quadrant grid superimposed for systematic interpretation. VLE scans were selected based on image quality and appropriate visualization of BE epithelium. Four experienced VLE users rated each quadrant as dysplastic or non-dysplastic. For quadrants rated as dysplastic, reviewers selected a single timeframe with representative features. A high-degree of agreement among reviewers was defined as ≥75% agreement on the quadrant diagnosis and ≥50% agreement on selected timeframe (±2 s). RESULTS: Thirty-one videos, each 32 s in length, comprising 124 quadrants were reviewed. There was high-agreement among reviewers in 99 (80%) quadrants, of which 68 (69%) were rated as dysplastic. Compared with quadrants rated as non-dysplastic, ROIs of quadrants rated as dysplastic contained a higher number of epithelial glands (12.7 vs 1.2, P < 0.001) with atypical architecture (54 vs 1, P < 0.001). A statistically significant difference was observed between the signal intensity profiles of quadrants rated as dysplastic and quadrants rated as non-dysplastic (P = 0.004). CONCLUSION: This study highlights that experienced VLE users can identify ROIs with high-degree of agreement. Selected ROIs contained VLE features associated with BE dysplasia.

Celiac disease detection in hypothyroid patients requiring elevated thyroid supplementation: A prospective cohort study.

BACKGROUND: Celiac disease (CD) is associated with hypothyroidism, but the disease prevalence is not thought to be great enough to warrant testing all hypothyroid patients. We hypothesized that hypothyroid patients with concomitant CD would require elevated doses of levothyroxine, and there is a threshold daily dose, above which, hypothyroid patients should be tested for CD. METHODS: Hypothyroid patients presenting to the endoscopy or endocrinology clinics at the University of Vermont Medical Center were included. Patients were categorized by whether or not they required ≥125mcg/day of levothyroxine. A serum tissue transglutaminase (tTG) was performed on enrolled patients. Patients with an elevated serum tTG underwent endoscopy with duodenal biopsies. Symptoms were assessed by the Gastrointestinal Symptom Rating Scale. RESULTS: Overall, 500 patients were enrolled and 29% (144 patients) required ≥125mcg/day of levothyroxine. CD was detected in 9 patients. The prevalence of CD ranged from 1.8% in our entire cohort to 12.5% in patients requiring ≥200mcg/day of levothyroxine. Eight patients with CD (89%) required ≥125mcg/day of levothyroxine. Patients who required ≥125mcg/day of levothyroxine had a significantly increased risk of CD (p<0.001). CD was detected in 5.6% of patients requiring ≥125mcg/day of levothyroxine. CONCLUSIONS: Hypothyroid patients requiring elevated daily doses of levothyroxine are more likely to have CD. Hypothyroid patients requiring ≥125mcg/day of levothyroxine should undergo serologic testing for CD.

Hospitalization for variceal hemorrhage in an era with more prevalent cirrhosis.

AIM: To examine hospitalization rates for variceal hemorrhage and relation to cause of cirrhosis during an era of increased cirrhosis prevalence. METHODS: We performed a retrospective review of patients with cirrhosis and gastroesophageal variceal hemorrhage who were admitted to a tertiary care referral center from 1998 to 2009. Subjects were classified according to the etiology of their liver disease: alcoholic cirrhosis and non-alcoholic cirrhosis. Rates of hospitalization for variceal bleeding were determined. Data were also collected on total hospital admissions per year and cirrhosis-related admissions per year over the same time period. These data were then compared and analyzed for trends in admission rates. RESULTS: Hospitalizations for cirrhosis significantly increased from 611 per 100000 admissions in 1998-2001 to 1232 per 100000 admissions in 2006-9 (P value for trend < 0.0001). This increase was seen in admissions for both alcoholic and non-alcoholic cirrhosis (P values for trend < 0.001 and < 0.0001 respectively). During the same time period, there were 243 admissions for gastroesophageal variceal bleeding (68% male, mean age 54.3 years, 62% alcoholic cirrhosis). Hospitalizations for gastroesophageal variceal bleeding significantly decreased from 96.6 per 100000 admissions for the time period 1998-2001 to 70.6 per 100000 admissions for the time period 2006-2009 (P value for trend = 0.01). There were significant reductions in variceal hemorrhage from non-alcoholic cirrhosis (41.6 per 100000 admissions in 1998-2001 to 19.7 per 100000 admissions in 2006-2009, P value for trend = 0.007). CONCLUSION: Hospitalizations for variceal hemorrhage have decreased, most notably in patients with non-alcoholic cirrhosis, and this may reflect broader use of strategies to prevent bleeding.

The rosemont criteria can predict the pain response to pancreatic enzyme supplementation in patients with suspected chronic pancreatitis undergoing endoscopic ultrasoun.

BACKGROUND/AIMS: The Rosemont classification system was designed to standardize the endosonographic assessment of chronic pancreatitis. To determine whether the Rosemont classification system can predict the response to pancreatic enzyme supplementation in patients undergoing endoscopic ultrasound (EUS) evaluation of suspected chronic pancreatitis. METHODS: Sixty-five patients were included with abdominal pain undergoing endosonography for suspected chronic pancreatitis were included. Patients completed a questionnaire for evaluation of their abdominal pain. Group 1 (n=13) had EUS findings consistent with or suggestive of chronic pancreatitis. Group 2 (n=45) had EUS findings that were normal or indeterminate in the Rosemont classification system. Patients were given pancreatic enzyme supplementation and then given a follow-up pain questionnaire for a mean of 37 days subsequent to EUS regarding the change in pain. RESULTS: Group 1 patients were more likely to have a response to pancreatic enzymes (62% vs 24%, p=0.012) and a decrease in their pain scale ratings (2.62 vs 0.29, p=0.01). Computed tomography findings of chronic pancreatitis and narcotic use did not predict the response to pancreatic enzyme supplementation. The individual Rosemont criteria of hyperechoic foci with shadowing (p=0.03), lobularity (p=0.02), and stranding (p=0.001) were associated with improvement of pain after treatment. CONCLUSIONS: The Rosemont classification system can identify patients who are more likely to have improvement in abdominal pain after treatment with pancreatic enzyme supplementation.

Screening for pancreatic cancer in a high-risk population with serum CA 19-9 and targeted EUS: a feasibility study.

BACKGROUND: Earlier detection of pancreatic adenocarcinoma is needed. OBJECTIVE: To determine whether early pancreatic neoplasia can be detected in a high-risk population by using CA 19-9 followed by targeted EUS. DESIGN: Prospective cohort study. SETTING: Two academic medical centers. PATIENTS: Eligible patients met age criteria and had at least 1 first-degree relative with pancreatic adenocarcinoma. INTERVENTIONS: A serum CA 19-9 was performed on all patients. EUS was performed if the CA 19-9 level was elevated. FNA of identified lesions was performed. Patients with pancreatic cancer detected by using this screening protocol were compared with patients presenting off-protocol for staging data. Medicare reimbursement rates were used to derive cost data. MAIN OUTCOME MEASUREMENTS: Detection of early pancreatic neoplasia. RESULTS: A total of 546 patients were enrolled. CA 19-9 was elevated in 27 patients (4.9%, 95% CI, 3.2%-7.1%). Neoplastic or malignant findings were detected in 5 patients (0.9%, 95% CI, 0.3%-2.1%), and pancreatic adenocarcinoma in 1 patient (0.2%, 95% CI, 0.005%-1.02%). The patient with pancreatic cancer detected as part of this protocol was 1 of 2 patients presenting to the University of Vermont with stage 1 cancer. The cost to detect 1 pancreatic neoplasia was $8431. The cost to detect 1 pancreatic adenocarcinoma was $41,133. LIMITATIONS: The sample size is adequate only to demonstrate the feasibility of this approach. CONCLUSIONS: Potentially curative pancreatic adenocarcinoma can be identified with this screening protocol. Stage 1 pancreatic cancer is more likely to be detected by using this screening protocol than by using standard means of detection.

Endoscopic occlusion of cystic duct using N-butyl cyanoacrylate for postoperative bile leakage.

Bile leak after cholecystectomy is well described, with the cystic duct remnant the site of the leak in the majority of cases. Endoscopic retrograde cholangiopancreatography (ERCP) with biliary stent placement has a high success rate in such cases. When ERCP fails, options include surgery, and percutaneous and endoscopic transcatheter occlusion of the site of bile leak. Here, we describe a case of endoscopic transcatheter occlusion of a persistent cystic duct bile leak after cholecystectomy using N-butyl cyanoacrylate glue. A 51-year-old man had persistent pain and bilious drainage following a laparoscopic cholecystectomy. The bile leak persisted after endoscopic placement of a biliary stent for a confirmed cystic duct leak. A repeat ERCP was carried out and the cystic duct was occluded with a combination of angiographic coils and N-butyl cyanoacrylate glue. The patient's pain and bilious drainage resolved. A follow-up cholangiogram confirmed complete resolution of the cystic duct leak and a patent common bile duct.

Mucosal serotonin signaling is altered in chronic constipation but not in opiate-induced constipation.

OBJECTIVES: Changes in mucosal serotonin (5-HT) signaling have been detected in a number of functional and inflammatory disorders of the gastrointestinal (GI) tract. This study was undertaken to determine whether chronic constipation (CC) is associated with disordered 5-HT signaling and to evaluate whether constipation caused by opiate use causes such changes. METHODS: Human rectal biopsy samples were obtained from healthy volunteers, individuals with idiopathic CC, and individuals taking opiate medication with or without occurrence of constipation. EC cells were identified by 5-HT immunohistochemistry. 5-HT content and release levels were determined by enzyme immunoassay, and mRNA levels for the synthetic enzyme tryptophan hydroxylase-1 (TpH-1) and serotonin-selective reuptake transporter (SERT) were assessed by quantitative real-time reverse transcription PCR. RESULTS: CC was associated with increases in TpH-1 transcript, 5-HT content, and 5-HT release under basal and stimulated conditions, whereas EC cell numbers and SERT transcript levels were not altered. No changes in these elements of 5-HT signaling were detected in opiate-induced constipation (OIC). CONCLUSIONS: These findings demonstrate that CC is associated with a pattern of altered 5-HT signaling that leads to increased 5-HT availability but does not involve a decrease in SERT expression. It is possible that increased 5-HT availability due to increased synthesis and release contributes to constipation due to receptor desensitization. Furthermore, the finding that elements of 5-HT signaling were not altered in the mucosa of individuals with OIC indicates that constipation as a condition does not lead to compensatory changes in 5-HT synthesis, release, or signal termination.