Knowledge, perceptions and behaviours of endoscopists towards the use of artificial intelligence-aided colonoscopy.

BACKGROUND: Recent developments in artificial intelligence (AI) systems have enabled advancements in endoscopy. Deep learning systems, using convolutional neural networks, have allowed for real-time AI-aided detection of polyps with higher sensitivity than the average endoscopist. However, not all endoscopists welcome the advent of AI systems. METHODS: We conducted a survey on the knowledge of AI, perceptions of AI in medicine, and behaviours regarding use of AI-aided colonoscopy, in a single centre 2 months after the implementation of Medtronic's GI Genius in colonoscopy. We obtained a response rate of 66.7% (16/24) amongst consultant-grade endoscopists. Fisher's exact test was used to calculate the significance of correlations. RESULTS: Knowledge of AI varied widely amongst endoscopists. Most endoscopists were optimistic about AI's capabilities in performing objective administrative and clinical tasks, but reserved about AI providing personalised, empathetic care. 68.8% (n = 11) of endoscopists agreed or strongly agreed that GI Genius should be used as an adjunct in colonoscopy. In analysing the 31.3% (n = 5) of endoscopists who disagreed or were ambivalent about its use, there was no significant correlation with their knowledge or perceptions of AI, but a significant number did not enjoy using the programme (p-value = 0.0128) and did not think it improved the quality of colonoscopy (p-value = 0.033). CONCLUSIONS: Acceptance of AI-aided colonoscopy systems is more related to the endoscopist's experience with using the programme, rather than general knowledge or perceptions towards AI. Uptake of such systems will rely greatly on how the device is delivered to the end user.

The imitation game: a review of the use of artificial intelligence in colonoscopy, and endoscopists' perceptions thereof.

The development of deep learning systems in artificial intelligence (AI) has enabled advances in endoscopy, and AI-aided colonoscopy has recently been ushered into clinical practice as a clinical decision-support tool. This has enabled real-time AI-aided detection of polyps with a higher sensitivity than the average endoscopist, and evidence to support its use has been promising thus far. This review article provides a summary of currently published data relating to AI-aided colonoscopy, discusses current clinical applications, and introduces ongoing research directions. We also explore endoscopists' perceptions and attitudes toward the use of this technology, and discuss factors influencing its uptake in clinical practice.

One-year review of real-time artificial intelligence (AI)-aided endoscopy performance.

BACKGROUND: Colonoscopies have long been the gold standard for detection of pre-malignant neoplastic lesions of the colon. Our previous study tried real-time artificial intelligence (AI)-aided colonoscopy over a three-month period and found significant improvements in collective and individual endoscopist's adenoma detection rates compared to baseline. As an expansion, this study evaluates the 1-year performance of AI-aided colonoscopy in the same institution. METHODS: A prospective cohort study was conducted in a single institution in Singapore. The AI software used was GI Genius™ Intelligent Endoscopy Module, US-DG-2000309 © 2021 Medtronic. Between July 2021 and June 2022, polypectomy rates in non-AI-aided colonoscopies and AI-aided colonoscopies were calculated and compared. Some of the AI-aided colonoscopies were recorded and video reviewed. A "hit" was defined as a sustained detection of an area by the AI. If a polypectomy was performed for a "hit," its histology was reviewed. Additional calculations for polyp detection rate (PDR), adenoma detection rate (ADR), and adenoma detection per colonoscopy (ADPC) were performed. Cost analysis was performed to determine cost effectiveness of subscription to the AI program. RESULTS: 2433 AI-aided colonoscopies were performed between July 2021 and June 2022 and compared against 1770 non-AI-aided colonoscopies. AI-aided colonoscopies yielded significantly higher rates of polypectomies (33.6%) as compared with non-AI-aided colonoscopies (28.4%) (p < 0.001). Among the AI-aided colonoscopies, 1050 were reviewed and a final 843 were included for additional analysis. The polypectomy to "hit" ratio was 57.4%, PDR = 45.6%, ADR = 32.4%, and ADPC = 2.08. Histological review showed that 25 polyps (3.13%) were sessile-serrated adenomas. Cost analysis found that the increased polypectomy rates in AI-aided colonoscopes led to an increase in revenue, which covered the subscription cost with an excess of USD 20,000. CONCLUSION: AI-aided colonoscopy is a cost effective means of improving colonoscopy quality and may help advance colorectal cancer screening in Singapore.

Real-time artificial intelligence (AI)-aided endoscopy improves adenoma detection rates even in experienced endoscopists: a cohort study in Singapore.

BACKGROUND: Colonoscopy is a mainstay to detect premalignant neoplastic lesions in the colon. Real-time Artificial Intelligence (AI)-aided colonoscopy purportedly improves the polyp detection rate, especially for small flat lesions. The aim of this study is to evaluate the performance of real-time AI-aided colonoscopy in the detection of colonic polyps. METHODS: A prospective single institution cohort study was conducted in Singapore. All real-time AI-aided colonoscopies, regardless of indication, performed by specialist-grade endoscopists were anonymously recorded from July to September 2021 and reviewed by 2 independent authors (FHK, JL). Sustained detection of an area by the program was regarded as a "hit". Histology for the polypectomies were reviewed to determine adenoma detection rate (ADR). Individual endoscopist's performance with AI were compared against their baseline performance without AI endoscopy. RESULTS: A total of 24 (82.8%) endoscopists participated with 18 (62.1%) performing ≥ 5 AI-aided colonoscopies. Of the 18, 72.2% (n = 13) were general surgeons. During that 3-months period, 487 "hits" encountered in 298 colonoscopies. Polypectomies were performed for 51.3% and 68.4% of these polypectomies were adenomas on histology. The post-intervention median ADR was 30.4% was higher than the median baseline polypectomy rate of 24.3% (p = 0.02). Of the adenomas excised, 14 (5.6%) were sessile serrated adenomas. Of those who performed ≥ 5 AI-aided colonoscopies, 13 (72.2%) had an improvement of ADR compared to their polypectomy rate before the introduction of AI, of which 2 of them had significant improvement. CONCLUSIONS: Real-time AI-aided colonoscopy have the potential to improved ADR even for experienced endoscopists and would therefore, improve the quality of colonoscopy.

Effects of water stably-enriched with oxygen as a novel method of tissue oxygenation on mitochondrial function, and as adjuvant therapy for type 2 diabetes in a randomized placebo-controlled trial.

BACKGROUND: Diabetes mellitus is associated with inadequate delivery of oxygen to tissues. Cellular hypoxia is associated with mitochondrial dysfunction which increases oxidative stress and hyperglycaemia. Hyperbaric oxygenation therapy, which was shown to improve insulin sensitivity, is impractical for regular use. We evaluated the effects of water which is stably-enriched with oxygen (ELO water) to increase arterial blood oxygen levels, on mitochondrial function in the presence of normal- or high-glucose environments, and as glucose-lowering therapy in humans. METHODS: We compared arterial blood oxygen levels in Sprague-Dawley rats after 7 days of ad libitum ELO or tap water consumption. Mitochondrial stress testing, and flow cytometry analysis of mitochondrial mass and membrane potential, were performed on human HepG2 cells cultured in four Dulbecco's Modified Eagle Medium media, made with ELO water or regular (control) water, at normal (5.5 mM) or high (25 mM) glucose concentrations. We also randomized 150 adults with type 2 diabetes (mean age 53 years, glycated haemoglobin HbA1c 8.9% [74 mmol/mol], average duration of diabetes 12 years) to drink 1.5 litres daily of bottled ELO water or drinking water. RESULTS: ELO water raised arterial oxygen tension pO2 significantly (335 ± 26 vs. 188 ± 18 mmHg, p = 0.006) compared with tap water. In cells cultured in control water, mitochondrial mass and membrane potential were both significantly lower at 25 mM glucose compared with 5.5 mM glucose; in contrast, mitochondrial mass and membrane potential did not differ significantly at normal or high glucose concentrations in cells cultured in ELO water. The high-glucose environment induced a greater mitochondrial proton leak in cells cultured in ELO water compared to cells cultured in control medium at similar glucose concentration. In type 2 diabetic adults, HbA1c decreased significantly (p = 0.002) by 0.3 ± 0.7% (4 ± 8 mmol/mol), with ELO water after 12 weeks of treatment but was unchanged with placebo. CONCLUSIONS: ELO water raises arterial blood oxygen levels, appears to have a protective effect on hyperglycaemia-induced reduction in mitochondrial mass and mitochondrial dysfunction, and may be effective adjuvant therapy for type 2 diabetes.

Clinical implications of prompt ascitic drain removal in cirrhosis with refractory ascites.

INTRODUCTION: Large-volume paracentesis (LVP) is the first-line treatment for decompensated cirrhosis with refractory ascites. While ascitic drain removal (ADR) within 72 hours of the procedure was once considered safe, it was uncertain whether ADR within 24 hours could further reduce the risk of ascitic drain-related bacterial peritonitis (AdBP). This study aimed to investigate the association between the timing of ADR and the presence of AdBP. METHODS: All patients with cirrhosis with refractory ascites who underwent LVP in our institution from 2014 to 2017 were studied. AdBP was diagnosed based on an ascitic fluid neutrophil count ≥ 250 cells/mm3 or positive ascitic fluid culture following recent paracentesis within two weeks. RESULTS: A total of 131 patients who underwent LVP were followed up for 1,806 patient-months. Their mean age was 68.3 ± 11.6 years, and 65.6% were male. Their mean Model for End-Stage Liver Disease score was 15.2. The overall incidence of AdBP was 5.3%. ADR beyond 24 hours was significantly associated with a longer median length of stay (five days vs. three days, p < 0.001), higher risk of AdBP (0% vs. 8.9%, p = 0.042) and acute kidney injury (AKI) following LVP (odds ratio 20.0, 95% confidence interval 2.4-164.2, p = 0.021). The overall survival was similar in patients who underwent ADR within and beyond 24 hours of LVP. CONCLUSION: ADR within 24 hours of LVP is associated with a reduced risk of AdBP and AKI. As AdBP is associated with resistant organisms and AKI, we recommend prompt ADR within 24 hours, especially in patients who have Child-Pugh class C alcoholic cirrhosis.

Novel albumin, bilirubin and platelet criteria for the exclusion of high-risk varices in compensated advanced chronic liver disease: A validation study.

BACKGROUND AND AIMS: Availability of transient elastography (TE) limits the application of Baveno-VI criteria. In a derivation study, the ABP criteria (Albumin >40 g/l, Bilirubin <22 µmol/l and Platelet >114,000/µl) had been shown to perform well in identifying compensated advanced chronic liver disease (cACLD) patients without high-risk varices (HRV). We aim to externally validate this novel ABP criteria for the exclusion of HRVs among cACLD patients. METHODS: Data was retrospectively collected from consecutive cACLD patients with paired TE and esophagogastroduodenoscopy (EGD) performed between 2011 and 2017 in Changi General Hospital, Singapore. We estimate the discriminative ability of ABP criteria in validation cohort using AUROC and calibration-in-the-large. We subsequently compare the performance between ABP and Baveno-VI criteria in the validation cohort. RESULTS: Among 314 patients included in our validation cohort, 32 (10.2%) had HRV on screening EGD. Application of ABP criteria within this validation cohort has increased discriminative ability than the derivation cohort. The AUROC of validation and derivation cohort were 0.68 (0.60-0.76) and 0.66 (0.60-0.76), respectively. The mean and standard error for calibration-in-the-large and calibration slope were -0.08 (0.22) and 0.93 (0.26) respectively. The ABP criteria had excellent performance in excluding HRV and will spare more screening EGDs than the Baveno-VI criteria (39.2% vs 27.4%, p < 0.001), without missing more HRVs. CONCLUSION: We validated the performance of ABP criteria for the exclusion of HRVs in cACLD patients. ABP criteria is superior to Baveno-VI criteria by sparing more screening EGD without the need of TE.

The impact of unrestricted access to direct-acting antiviral among incarcerated hepatitis C virus-infected patients.

BACKGROUND/AIMS: Despite the disproportionally high prevalence rates of hepatitis C virus (HCV) amongst the incarcerated population, eradication remains challenging due to logistic and financial barriers. Although treatment prioritization based on disease severity is commonly practiced, the efficacy of such approach remained uncertain. We aimed to compare the impact of unrestricted access to direct-acting antiviral (DAA) among incarcerated HCV-infected patients in Singapore. METHODS: In this retrospective study, we reviewed all incarcerated HCV-infected patients treated in our hospital during the restricted DAA era (2013-2018) and unrestricted DAA access era (2019). Study outcomes included the rate of sustained virological response (SVR), treatment completion and treatment default. Subgroup analysis was performed based on the presence of liver cirrhosis, HCV genotype and HCV treatment types. RESULTS: A total of 1,001 HCV patients was followed-up for 1,489 person-year. They were predominantly male (93%) with genotype-3 HCV infection (71%), and 38% were cirrhotic. The overall SVR during the restricted DAA access era and unrestricted DAA access era were 92.1% and 99.1%, respectively. Unrestricted access to DAA exponentially improved the treatment access among HCV-infected patients by 460%, resulting in a higher SVR rate (99% vs. 92%, P=0.003), higher treatment completion rate (99% vs. 93%, P<0.001) and lower treatment default rate (1% vs. 9%, P<0.001). CONCLUSION: In this large cohort of incarcerated HCV-infected patients, we demonstrated that unrestricted access to DAA is an impactful strategy to allow rapid treatment up-scale in HCV micro-elimination.

Efficacy and Safety of Statin for Hepatocellular Carcinoma Prevention Among Chronic Liver Disease Patients: A Systematic Review and Meta-analysis.

INTRODUCTION AND AIM: Hepatocellular carcinoma (HCC) is a deadly complication among patients with chronic liver disease (CLD). Controversies on the efficacy and safety of statin to prevent HCC among patients with CLD remain despite the growing evidences. We aim to investigate the efficacy and safety of using statin for HCC prevention among adult with CLD. METHODS: We performed a systematic search of 4 electronic databases (PubMed/MEDLINE, EMBASE, Cochrane library, and ClinicalTrial.gov) up to April 15, 2020. We selected all types of studies evaluating the statin use and the risk of HCC among CLD patients, regardless of language, region, publication date, or status. The primary endpoint was the pooled risk of HCC. The secondary endpoint was the risk of statin-associated myopathy. RESULT: From 583 citations, we included a total of 13 studies (1,742,260 subjects, 7 types of statins), fulfilling the inclusion criteria, evaluating efficacy and safety of statin in CLD patients for HCC prevention. All studies were observational (2 nested case-control studies, 11 cohort studies), and no randomised trial was identified. We found that statin user has a lower pooled risk of HCC development (hazard ratio=0.57, 95% confidence interval: 0.52-0.62, I2=42%). HCC reduction was consistent among statin users in cirrhosis, hepatitis B virus, and hepatitis C virus infections. The risk of statin-associated myopathy was similar between statin user and nonuser (hazard ratio=1.07, 95% confidence interval=0.91-1.27). CONCLUSION: Statin use was safe and associated with a lower pooled risk of HCC development among adults with CLD. Given the bias with observation studies, prospective randomised trial is needed to confirm this finding.

Efficacy and safety of sofosbuvir/velpatasvir in a real-world chronic hepatitis C genotype 3 cohort.

BACKGROUND AND AIM: Real-world data on sofosbuvir/velpatasvir with and without ribavirin (SOF/VEL ± RBV), particularly among patients with genotype 3 (GT3) decompensated cirrhosis, prior treatment, coinfection, and hepatocellular carcinoma (HCC), are scarce. We aimed to assess the efficacy and safety of SOF/VEL ± RBV in a real-world setting that included both community and incarcerated GT3 hepatitis C virus (HCV) patients. METHODS: We included all GT3 HCV patients treated with SOF/VEL ± RBV in our institution. The primary outcome measure was the overall sustained virological response 12 weeks after treatment (SVR12), reported in both intention-to-treat (ITT) and per-protocol analyses. The secondary outcome measures were SVR12 stratified by the presence of decompensated cirrhosis, prior treatment, HCC, and HIV/hepatitis C virus coinfection and the occurrence rate of serious adverse events requiring treatment cessation or hospitalization. RESULTS: A total of 779 HCV patients were treated with 12 weeks of SOF/VEL ± RBV, of which 85% were treated during incarceration. Among the 530 GT3 HCV patients, 31% had liver cirrhosis, and 6% were treatment-experienced. The overall SVR12 for GT3 was 98.7% (95% confidence interval: 97.3%, 99.5%) and 99.2% (95% confidence interval: 98.1%, 99.8%) in ITT and per-protocol analyses, respectively. High SVR12 was also seen in ITT analysis among GT3 HCV patients with decompensated cirrhosis (88%), prior treatment (100%), HCC (100%), and HIV/hepatitis B virus coinfection (100%). Apart from one patient who developed myositis, no other serious adverse events were observed. CONCLUSION: The SOF/VEL ± RBV is a safe and efficacious treatment option for GT3 HCV patients in a real-world setting. SOF/VEL with RBV may be considered for decompensated GT3 HCV patients.

Modified AST to platelet ratio index improves APRI and better predicts advanced fibrosis and liver cirrhosis in patients with non-alcoholic fatty liver disease.

AIMS: Advanced fibrosis (AF) and liver cirrhosis (LC) are important milestones in non-alcoholic fatty liver disease (NAFLD). FIB-4, NFS and BARD are validated scores with good accuracy in detecting AF and LC. APRI does not have similar predictive accuracy. While a modification (m-APRI) improves its use in viral hepatitis, this has yet to be evaluated in NAFLD. This study compares diagnostic performance of aforementioned scores in predicting AF and LC in NAFLD. METHODS: Consecutive NAFLD patients undergoing Transient Elastography (TE) using Echosens® Fibroscan® for fibrosis staging were included. Cut-off liver stiffness measurements for AF and LC were 7.9 kPa and 11.5 kPa respectively. Anthropometric and laboratory tests done within 3 months were used. Diagnostic performances of scores were analyzed by standard statistical tests. RESULTS: 161 patients qualified for the study. Mean age was 60.2 ±â€¯14 years, BMI 26.8 ±â€¯4.6 kg/m2. M-probe was used in 113, XL in 48. Optimal cut-offs of m-APRI for AF and LC were 5.84 and 9 respectively. Area under receiver operator characteristic curves (AUROC) for prediction of AF at optimal cut-off points were m-APRI 0.84, APRI 0.80, FIB-4: 0.77, NFS 0.77 and BARD 0.65. For prediction of LC, AUROC were m-APRI: 0.83, APRI: 0.76, FIB-4: 0.81, NFS: 0.77 and BARD: 0.66. m-APRI was significantly superior to all scores compared in detecting AF (p < 0.05 for all) and superior to APRI (p = 0.008) and BARD (p = 0.007) in predicting LC. There was no significant difference between m-APRI and FIB-4 or NFS in prediction of LC. CONCLUSIONS: For prediction of AF in NAFLD, m-APRI outperforms BARD, APRI, NFS and FIB-4, while for the prediction of cirrhosis, m-APRI is superior to APRI and BARD but comparable to NFS and FIB-4.

A systematic review and meta-analysis of the COVID-19 associated liver injury.

INTRODUCTION AND OBJECTIVES: The novel coronavirus disease 2019 (COVID-19) has affected more than 5 million people globally. Data on the prevalence and degree of COVID-19 associated liver injury among patients with COVID-19 remain limited. We conducted a systematic review and meta-analysis to assess the prevalence and degree of liver injury between patients with severe and non-severe COVID-19. METHODS: We performed a systematic search of three electronic databases (PubMed/MEDLINE, EMBASE and Cochrane Library), from inception to 24th April 2020. We included all adult human studies (>20 subjects) regardless of language, region or publication date or status. We assessed the pooled odds ratio (OR), mean difference (MD) and 95% confidence interval (95%CI) using the random-effects model. RESULTS: Among 1543 citations, there were 24 studies (5961 subjects) which fulfilled our inclusion criteria. The pooled odds ratio for elevated ALT (OR = 2.5, 95%CI: 1.6-3.7, I2 = 57%), AST (OR = 3.4, 95%CI: 2.3-5.0, I2 = 56%), hyperbilirubinemia (OR = 1.7, 95%CI: 1.2-2.5, I2 = 0%) and hypoalbuminemia (OR = 7.1, 95%CI: 2.1-24.1, I2 = 71%) were higher subjects in critical COVID-19. CONCLUSION: COVID-19 associated liver injury is more common in severe COVID-19 than non-severe COVID-19. Physicians should be aware of possible progression to severe disease in subjects with COVID-19-associated liver injury.

Impact of enhanced personal protective equipment on colonoscopy performance during the COVID-19 pandemic.

Background and study aims Using personal protective equipment (PPE) can reduce risk of disease transmission. During the COVID-19 pandemic, enhanced PPE (EPPE) is recommended when performing endoscopy. We aimed to evaluate the impact of EPPE on colonoscopy performance when compared to standard PPE (SPPE). Patients and methods A review of electronic medical records and endoscopy reports of consecutive patients who underwent colonoscopy during two similar one-month time periods (in 2019 and during the COVID-19 pandemic in 2020) was performed. SPPE was used in 2019 and EPPE was used in 2020. Patient clinical data and procedure-related information were captured and analyzed. The primary outcomes were time to cecum (TTC) and total procedure time. Secondary outcomes were adenoma detection rate (ADR), polyp detection rate (PDR) and cecal intubation rate (CIR). Statistical analysis was performed using STATA v16.1. Results Two hundred and forty-seven colonoscopy procedures were analyzed. Baseline demographics and indications for colonoscopy of patients in both groups were similar. There were no significant differences in median TTC (10.0 vs 10.0 min, P  = 0.524) or total procedure time (22.5 vs 23.0 min, P  = 0.946) between colonoscopy performed in SPPE and EPPE. The ADR, PDR and CIR were also similar. Conclusion Our findings suggest that use of EPPE does not affect colonoscopy performance.

Short- and long-term predictors of spontaneous bacterial peritonitis in Singapore.

INTRODUCTION: Spontaneous bacterial peritonitis (SBP) is the commonest complication of liver cirrhosis. Timely and appropriate treatment of SBP is crucial, particularly with the rising worldwide prevalence of multidrug-resistant organisms (MDROs). We aimed to investigate the clinical outcomes of SBP in Singapore. METHODS: All cirrhotic patients with SBP diagnosed between January 2014 and December 2017 were included. Nosocomial SBP (N-SBP) was defined as SBP diagnosed more than 48 hours after hospitalisation. Clinical outcomes were analysed as categorical outcomes using univariate and multivariate analysis. RESULTS: There were 33 patients with 39 episodes of SBP. Their mean age was 64.5 years and 69.7% were male. The commonest aetiology of cirrhosis was hepatitis B (27.3%). The Median Model for End-stage Liver Disease (MELD) score was 17; 33.3% had acute-on-chronic liver failure and 60.6% had septic shock at presentation. N-SBP occurred in 25.6% of SBP cases. N-SBP was more commonly associated with MDROs, previous antibiotic use in the past three months (p = 0.014) and longer length of stay (p = 0.011). The 30-day and 90-day mortality among SBP patients was 30.8% and 51.3%, respectively. MELD score > 20 was a predictor for 30-day mortality. N-SBP and MELD score > 20 were predictors for 90-day mortality. CONCLUSION: N-SBP was significantly associated with recent antibiotic use, longer hospitalisation, more resistant organisms and poorer survival among patients with SBP. N-SBP and MELD score predict higher mortality in SBP. Judicious use of antibiotics may reduce N-SBP and improve survival among cirrhotic patients.

A prospective randomized study of colonoscopy using blue laser imaging and white light imaging in detection and differentiation of colonic polyps.

Background and study aims Published data on blue laser imaging (BLI) for detection and differentiation of colonic polyps are limited compared to narrow band imaging (NBI). This study investigated whether BLI can increase the detection rate of colonic polyps and adenomas when compared to white light imaging (WLI), and examined use of NICE (NBI International Colorectal Endoscopic) and JNET (Japan NBI Expert Team) classifications with BLI. Patients and methods Patients aged 50 years and above referred for colonoscopy were randomized to BLI or WLI on withdrawal. Detected polyps were characterized using NICE and JNET classifications under BLI mode and correlated with histology. Primary outcome was adenoma detection rate. Secondary outcomes were utility of NICE and JNET classifications to predict histology using BLI. Results A total of 182 patients were randomized to BLI (92) or WLI (90). Comparing BLI with WLI, the polyp detection rate was 59.8 % vs 40.0 %, P  = 0.008, and the adenoma detection rate was 46.2 % vs 27.8 %, P  = 0.010. NICE 1 and JNET 1 diagnosed hyperplastic polyps with sensitivity of 87.18 % and specificity of 84.35 %. NICE 2 diagnosed low- (LGD) or high-grade dysplasia (HGD) with sensitivity of 92.31 % and specificity of 77.45 %. JNET 2A diagnosed LGD with sensitivity of 91.95 %, and specificity of 74.53 %. Four cases of focal HGD all had JNET 2A morphology. Conclusion BLI increased adenoma detection rate compared to WLI. NICE and JNET classifications can be applied when using BLI for endoscopic diagnosis of HP and LGD but histological confirmation remains crucial.

Endoscopic submucosal dissection of colorectal neoplasms: an audit of its safety and efficacy in a single tertiary centre in Singapore.

INTRODUCTION: Endoscopic submucosal dissection (ESD) provides a higher en bloc and R0 resection rate than endoscopic mucosal resection. Colorectal ESD is not widely used because of its technical difficulty and risk of complications, especially for right-sided colonic lesions. We audited the clinical outcomes of our initial experience with colorectal ESD. METHODS: We reviewed data collected from a prospective registry of patients who underwent colorectal ESD from December 2014 to March 2018. Therapeutic outcomes and procedure-related complications were analysed. RESULTS: A total of 41 patients (mean age 67.4 years, 21 male) underwent colorectal ESD. The distribution of the lesions was as follows: rectum (n = 9), sigmoid colon (n = 8), descending colon (n = 6), splenic flexure (n = 1), transverse colon (n = 5), ascending colon (n = 8) and caecum (n = 4). The mean size was 23 (range 12-50) mm. En bloc resection was achieved in 35 (85.4%) out of 41 patients, and R0 resection or clear resection margins was achieved in 33 (94.3%) of the en bloc resection patients. The lesion was upstaged in 14 (34.1%) patients after ESD. Colonic perforation occurred in 3 (7.3%) patients during ESD and was successfully treated with endoscopic clips. There was no procedure-related bleeding. No patient required surgery for management of complications. The median duration of hospitalisation was 1 (range 0-7) day. Four patients with lesions that were upstaged after ESD, from high-grade dysplasia to intramucosal carcinoma with deep submucosal invasion, were referred for colectomy. CONCLUSION: Our early outcome data was comparable to that from large published series. ESD is an effective and feasible treatment for colorectal lesions.

Novel non-invasive score to predict cirrhosis in the era of hepatitis C elimination: A population study of ex-substance users in Singapore.

BACKGROUND: Chronic hepatitis C infection is common among people with history of substance use. Liver fibrosis assessment is a barrier to linkage to care, particularly among those with history of substance users. The use of non-invasive scores can be helpful in predicting liver cirrhosis in the era of HCV elimination, especially in countries where transient elastography (TE) is not available. We compared the commonly used non-invasive scores with a novel non-invasive score in predicting liver cirrhosis in this population. METHODS: HCV patients with history of substance use between 2011 and 2016 were analyzed. All patients had TE for liver fibrosis assessment. Clinical performance of established non-invasive scores for fibrosis assessment and novel score were compared. Youden's index was used to determine optimal cut-off of the novel score. RESULTS: A total of 579 patients were included. In multivariate logistic regression, cirrhosis on TE was associated with age (P = 0.002), aspartate aminotransferase (AST) (P = 0.004), and platelet count (P < 0.001), but not alanine aminotransferase (ALT) (P = 0.896). These form the components of modified AST-to-platelet ratio index (APRI) score. Modified APRI was superior to APRI in predicting cirrhosis (AUROC, 0.796 vs. 0.770, P = 0.007), but not fibrosis-4 score (FIB-4) (P = 1.00). Modified APRI at cut-off of 4 has sensitivity, specificity and negative predictive value (NPV) of 94.4%, 26.9% and 92.6%, respectively, and at 19, has sensitivity, specificity and positive predictive value (PPV) of 33.3%, 96.2% and 77.1%, respectively. FIB-4 has a NPV and PPV of 88.6%, 41.8% and 78.5%, 77.6%, at cut-off of 1.45 and 3.25, respectively. Using the cut-off of 4 and 14 for modified APRI, 32.5% of patients can be correctly classified and misses out only 5.6% of cirrhosis patients. CONCLUSIONS: Modified APRI score is superior in predicting cirrhosis in HCV population, with 32.5% of the population being correctly classified using cut-off of 4 and 14. Further studies are required to validate the findings.