Background: An association between environmental pollutants and alcohol-related liver disease (ALD) has not been determined until now. The objectives of this study were to examine the association of the pollutants with ALD, and whether the pollutants together increased the risk of ALD. Methods: Data were extracted from the Korea National Health and Nutrition Examination Survey (2010–2013 and 2016–2017; n = 11,993). Blood levels of lead, cadmium, and mercury were measured. ALD was defined by a combination of excessive alcohol consumption and ALDon-alcoholic fatty liver disease index > 0. The aspartate aminotransferase-to-platelet ratio index and fibrosis (FIB)-4 score were used to evaluate ALD FIB. Results: The odds ratios (ORs) of ALD for the highest versus the lowest quartiles of exposure were for lead, 7.39 (95% confidence interval [CI], 5.51–9.91); cadmium, 1.68 (95% CI, 1.32–2.14); and mercury, 5.03 (95% CI, 3.88–6.53). Adjusting for age, gender, smoking, occupation, education, and personal income attenuated the associations but indicated significant positive trends (all P trend < 0.001). A positive additive interaction between cadmium and lead was observed. The relative excess OR due to the interaction was 0.96 (95% CI, 0.41–1.51); synergy index = 2.92 (95% CI, 0.97–8.80). Among 951 subjects with ALD, advanced FIB was associated with lead and cadmium (OR, 3.46, 95% CI, 1.84–6.53; OR, 8.50, 95% CI, 2.54–28.42, respectively), but not with mercury. The effect estimates for lead and cadmium remained significant even after adjustment for daily alcohol intake. Conclusion Blood levels of lead, cadmium, and mercury were significantly associated not only with the risk of ALD but also with ALD FIB. Cadmium and lead have synergistic effects that increase the risk of ALD.

Background/Aims: We aimed to develop a deep learning model for the prediction of the risk of advanced colorectal neoplasia (ACRN) in asymptomatic adults, based on which colorectal cancer screening could be customized. Methods: We collected data on 26 clinical and laboratory parameters, including age, sex, smoking status, body mass index, complete blood count, blood chemistry, and tumor marker, from 70,336 first-time colonoscopy screening recipients. For reference, we used a logistic regression (LR) model with nine variables manually selected from the 26 variables. A deep neural network (DNN) model was developed using all 26 variables. The area under the receiver operating characteristic curve (AUC), sensitivity, and specificity of the models were compared in a randomly split validation group. Results: In comparison with the LR model (AUC, 0.724; 95% confidence interval [CI], 0.684 to 0.765), the DNN model (AUC, 0.760; 95% CI, 0.724 to 0.795) demonstrated significantly improved performance with respect to the prediction of ACRN (p < 0.001). At a sensitivity of 90%, the specificity significantly increased with the application of the DNN model (41.0%) in comparison with the LR model (26.5%) (p < 0.001), indicating that the colonoscopy workload required to detect the same number of ACRNs could be reduced by 20%. Conclusions The application of DNN to big clinical data could significantly improve the prediction of ACRNs in comparison with the LR model, potentially realizing further customization by utilizing large quantities and various types of biomedical information.

Background/Aims: We aimed to develop a deep learning model for the prediction of the risk of advanced colorectal neoplasia (ACRN) in asymptomatic adults, based on which colorectal cancer screening could be customized. Methods: We collected data on 26 clinical and laboratory parameters, including age, sex, smoking status, body mass index, complete blood count, blood chemistry, and tumor marker, from 70,336 first-time colonoscopy screening recipients. For reference, we used a logistic regression (LR) model with nine variables manually selected from the 26 variables. A deep neural network (DNN) model was developed using all 26 variables. The area under the receiver operating characteristic curve (AUC), sensitivity, and specificity of the models were compared in a randomly split validation group. Results: In comparison with the LR model (AUC, 0.724; 95% confidence interval [CI], 0.684 to 0.765), the DNN model (AUC, 0.760; 95% CI, 0.724 to 0.795) demonstrated significantly improved performance with respect to the prediction of ACRN (p < 0.001). At a sensitivity of 90%, the specificity significantly increased with the application of the DNN model (41.0%) in comparison with the LR model (26.5%) (p < 0.001), indicating that the colonoscopy workload required to detect the same number of ACRNs could be reduced by 20%. Conclusions The application of DNN to big clinical data could significantly improve the prediction of ACRNs in comparison with the LR model, potentially realizing further customization by utilizing large quantities and various types of biomedical information.

Background/Aims: Data on the trends in the prevalence of chronic liver disease (CLD) in Korea are scarce. This study aimed to evaluate whether the CLD prevalence changed between 1998–2001 and 2016–2017. Methods: Data were extracted from the Korea National Health and Nutrition Examination Survey (1998–2001 to 2016– 2017; n=25,893). Non-alcoholic fatty liver disease (NAFLD) was defined as a hepatic steatosis index >36 in the absence of any other evidence of CLD. The definition of alcohol-related liver disease (ALD) was excessive alcohol consumption (≥210 g/week for men and ≥140 g/week for women) and an ALD/NAFLD index >0. Results: The prevalence of NAFLD increased from 18.6% (95% confidence interval [CI], 17.8–19.5%) in 1998–2001 to 21.5% (95% CI, 20.6–22.6%) in 2016–2017. During the same time period, increases were observed in the prevalence of obesity (27.0 vs. 35.1%), central obesity (29.4 vs. 36.0%), diabetes (7.5 vs. 10.6%), and excessive drinking (7.3 vs. 10.5%). ALD prevalence also increased from 3.8% (95% CI, 3.4–4.2%) to 7.0% (95% CI, 6.4–7.6%). In contrast, chronic hepatitis B decreased from 5.1% (95% CI, 4.6–5.5%) to 3.4% (95% CI, 3.0–3.8%). The prevalence of chronic hepatitis C was approximately 0.3% in 2016–2017. Conclusions The prevalence of NAFLD and ALD increase among Korean adults. Our results suggest potential targets for interventions to reduce the future burden of CLD.

BACKGROUND/AIMS: The aims of this study were to compare the bowel-cleansing efficacy, patient affinity for the preparation solution, and mucosal injury between a split dose of poly-ethylene glycol (SD-PEG) and low-volume PEG plus ascorbic acid (LV-PEG+Asc) in outpatient scheduled colonoscopies. METHODS: Of the 319 patients, 160 were enrolled for SD-PEG, and 159 for LV-PEG+Asc. The bowel-cleansing efficacy was rated according to the Ottawa bowel preparation scale. Patient affinity for the preparation solution was assessed using a questionnaire. All mucosal injuries observed during colonoscopy were biopsied and histopathologically reviewed. RESULTS: There was no significant difference in bowel cleansing between the groups. The LV-PEG+Asc group reported better patient acceptance and preference. There were no significant differences in the incidence or characteristics of the mucosal injuries between the two groups. CONCLUSIONS: Compared with SD-PEG, LV-PEG+Asc exhibited equivalent bowel-cleansing efficacy and resulted in improved patient acceptance and preference. There was no significant difference in mucosal injury between SD-PEG and LV-PEG+Asc. Thus, the LV-PEG+Asc preparation could be used more effectively and easily for routine colonoscopies without risking significant mucosal injury.
Adult ; Ascorbic Acid/administration & dosage/*adverse effects ; Cathartics/administration & dosage/*adverse effects ; Colonoscopy/methods ; Drug Therapy, Combination ; Female ; Humans ; Intestinal Mucosa/drug effects/*injuries ; Male ; Middle Aged ; Patient Compliance ; Patient Satisfaction ; Polyethylene Glycols/administration & dosage/*adverse effects ; Preoperative Care/*adverse effects/methods ; Surveys and Questionnaires ; Vitamins/administration & dosage/adverse effects