Association between triglyceride-glucose index and gastric carcinogenesis: a health checkup cohort study.

BACKGROUND: We aimed to investigate the association between the triglyceride-glucose (TyG) index and gastric carcinogenesis, including precancerous conditions such as dysplasia, atrophic gastritis, and intestinal metaplasia. METHODS: Patients who received an upper endoscopic assessment at a medical center were included. The enrolled patients were divided into four categories according to their TyG index quartile (Q). To evaluate the relationship between increase of TyG index and gastric cancer, we analyzed the patients who received a health checkup twice. Moreover, receiver-operating characteristic curve analysis was used to establish cut-off value of the TyG index for gastric cancer. RESULTS: Of 127,564 enrolled patients, 43,525 (34.1%) and 186 (0.1%) were diagnosed with precancerous conditions and gastric cancer, respectively. The odds ratios (ORs) of precancerous conditions given TyG index progressively increased across quartiles: using Q1 as the reference: Q2 (OR = 1.403, P < 0.001), Q3 (OR = 1.646, P < 0.001), and Q4 (OR = 1.656, P < 0.001). The ORs of gastric cancer also increased according to the quartiles: Q2 (OR = 1.619, P = 0.045), Q3 (OR = 2.180, P = 0.004), and Q4 (OR = 2.363, P = 0.001). Moreover, the increase in TyG index between baseline and follow-up tests was more significant in gastric cancer group than in control group (P = 0.001). The optimal cut-off value for predicting gastric cancer was 9.73. CONCLUSIONS: The TyG index may be a novel predictive biomarker for gastric carcinogenesis. Notably, increase in the TyG index is significantly associated with gastric cancer.

MAFLD Predicts the Risk of Cardiovascular Disease Better than NAFLD in Asymptomatic Subjects with Health Check-Ups.

BACKGROUND AND AIM: Metabolic dysfunction-associated fatty liver disease (MAFLD) was proposed to compensate for the conventional concept of nonalcoholic fatty liver disease (NAFLD). We investigated the superiority of MAFLD versus NAFLD in predicting the risk of atherosclerotic cardiovascular disease (ASCVD). METHODS: A total of 2,144 subjects without a history of ASCVD, who underwent a comprehensive medical health check-up, were selected for the study. The associations between fatty liver status and coronary risk surrogates, such as coronary artery calcium score (CACS), coronary artery disease, quantitative stenosis grade, and 10-year ASCVD risk, were analyzed. RESULTS: MAFLD and NAFLD were identified in 995 (46.4%) and 891 (41.6%) subjects, respectively. Subjects with MAFLD or NAFLD were more likely to be male and had a significantly higher prevalence of central obesity, obesity, hypertension, diabetes, and dyslipidemia (all, p < 0.05) than their counterparts. In terms of coronary risk surrogates, the MAFLD or NAFLD population had a significantly higher proportion of subjects with CACS > 100, coronary artery disease, higher grade of coronary artery stenosis, and higher 10-year ASCVD risk (all, p < 0.05) than their counterparts. Multivariable logistic regression models showed an independent association between MAFLD/NAFLD and coronary risk surrogates (all, p < 0.05). However, NAFLD only, defined as 'NAFLD, but not MAFLD,' was not associated with an increased coronary risk, compared to MAFLD. CONCLUSIONS: Although both MAFLD and NAFLD discriminated different ASCVD risks, MAFLD predicted the risk of ASCVD better than NAFLD in asymptomatic subjects who underwent medical health check-ups.

Body mass index and absolute lymphocyte count predict disease-free survival in Korean breast cancer patients.

BACKGROUND: Our study evaluated the association between body mass index (BMI) and absolute lymphocyte count (ALC) in breast cancer patients and healthy females. Additionally, we determined the prognostic value of these factors in breast cancer. METHODS: We retrospectively identified 1225 primary invasive breast cancer patients and 35,991 healthy females. Factors including BMI and complete blood count associated with disease-free survival (DFS) were assessed using a multi-variable Cox proportional hazard model. RESULTS: BMI and ALC were positively correlated in breast cancer patients and healthy females (both P < 0.001). In multi-variable analysis, overweight or obese participants had worse DFS (hazards ratio [HR], 1.98; 95% confidence interval [CI], 1.34-2.92; P = 0.001) than underweight or normal-weight individuals, but patients with high ALC had better DFS than those with low ALC (HR, 0.43; 95% CI, 0.29-0.65; P < 0.001). After risk stratification according to BMI/ALC, high-risk patients with high BMI/low ALC had worse DFS than others (HR, 2.48; 95% CI, 1.70-3.62; P < 0.001). CONCLUSIONS: BMI and ALC were positive correlated, but their effect on breast cancer prognosis was opposite. Patients with high BMI/low ALC had worse DFS than others. Underlying mechanisms for effect of BMI/ALC on breast cancer prognosis should be studied in the future.

Characteristics of Intracranial Aneurysms According to Levels of Coronary Artery Calcium.

Background and Purpose- Several vascular risk factors are known to be associated with the occurrence of intracranial aneurysms (IAs). Coronary artery calcium (CAC), which reflects the atherosclerotic burden of the coronary arteries, is a known predictor of cardiovascular events and stroke. We investigated the relationship between IA and CAC. Methods- We retrospectively enrolled Korean subjects at a single university hospital who had both brain magnetic resonance angiography and cardiac computed tomography as part of health examinations from January 2010 to July 2017. Subjects were categorized into 4 groups according to CAC score as assessed by cardiac computed tomography: zero (0), low (1-99), intermediate (100-399), or high (≥400). Then, the prevalence of IA in each CAC score group was assessed. We also performed subgroup analysis by age, sex, and location of IA. Results- A total of 4934 subjects (mean age, 54.1±9.8 years; %women, 42.2%) were included for analysis. IAs were detected in 258 subjects (5.23%). The prevalence of IA significantly increased as the CAC score increased (4.8%, 5.4%, 6.4%, and 11.1%, respectively; P for trend, 0.004). In subjects over the age of 50 years, this correlation was more prominent in women than in men (7.1% versus 3.7%, 8.8% versus 4.4%, 8.6% versus 6.3%, and 21.1% versus 10.0%, respectively). Subgroup analysis for the aneurysm location showed that nonbifurcation aneurysm was associated with a high CAC score but bifurcation aneurysm was not. Multivariate logistic regression showed high CAC score was an independent risk factor for the presence of IA compared with zero CAC score (adjusted odds ratio, 2.16; 95% CI, 1.18-3.95). Conclusions- A high CAC score was associated with the presence of IA. This relationship was more prominent in females and nonbifurcation aneurysms.

The association between Helicobacter pylori infection and the risk of advanced colorectal neoplasia may differ according to age and cigarette smoking.

BACKGROUND: The association between Helicobacter pylori infection and advanced colorectal neoplasia (ACN) remains controversial. This study aimed to clarify the association between H. pylori infection and ACN according to age groups. METHODS: We retrospectively analyzed the association between H. pylori infection and ACN in patients aged <50 and ≥50 years receiving a health checkup that included colonoscopy. Helicobacter pylori positivity was determined by the results of serum anti-H. pylori immunoglobulin G or rapid urease test, if the anti-H. pylori immunoglobulin G was in the borderline range. RESULTS: Among the 19 337 patients who were included, 56.2% and 3.4% were positive for H. pylori and ACN, respectively. Helicobacter pylori infection independently increased the risk of ACN in patients aged <50 years (odds ratio [OR], 1.602; 95% confidence intervals [CI], 1.194-2.150) but not in patients aged ≥50 years (OR, 1.046; 95% CI, 0.863-1.268). The positive association between H. pylori infection and ACN was affected by smoking history. When stratified by age and smoking history, H. pylori infection conferred an increased risk of ACN in patients aged <50 years with a history of smoking (OR, 1.926; 95% CI, 1.336-2.775) but not in the other 3 groups (3-way interaction test P = .023). Among patients aged <50 years with ACN, ACN in the left colon was found more frequently in patients with H. pylori infection and a history of smoking than in those without (69.3% vs 54.4%, respectively; P = .031). CONCLUSIONS: Helicobacter pylori infection confers an increased risk of ACN, but the association may differ according to age and smoking history.

A simple scoring model for advanced colorectal neoplasm in asymptomatic subjects aged 40-49 years.

BACKGROUND: Limited data are available for advanced colorectal neoplasm in asymptomatic individuals aged 40-49 years. We aimed to identify risk factors and develop a simple prediction model for advanced colorectal neoplasm in these persons. METHODS: Clinical data were collected on 2781 asymptomatic subjects aged 40-49 years who underwent colonoscopy for routine health examination. Subjects were randomly allocated to a development or validation set. Logistic regression analysis was used to determine predictors of advanced colorectal neoplasm. RESULTS: The prevalence of overall and advanced colorectal neoplasm was 20.2 and 2.5% respectively. Older age (45-49 years), male sex, positive serology of Helicobacter pylori, and high triglyceride and low high-density lipoprotein (HDL) levels were independently associated with an increased risk of advanced colorectal neoplasm. BMI (body mass index) was not significant in multivariable analysis. We developed a simple scoring model for advanced colorectal neoplasm (range 0-9). A cutoff of ≥4 defined 43% of subjects as high risk for advanced colorectal neoplasm (sensitivity, 79%; specificity, 58%; area under the receiver operating curve = 0.72) in the validation datasets. CONCLUSION: Older age (45-49 years), male sex, positive serology of H. pylori, high triglyceride level, and low HDL level were identified as independent risk factors for advanced colorectal neoplasm.

Lipoprotein(a) level predicts the development of nonalcoholic fatty liver disease in Korean adults: A retrospective longitudinal study.

Nonalcoholic fatty liver disease (NAFLD) is a highly prevalent condition in the general population. Although recent studies have demonstrated a link between NAFLD and lipoprotein(a), a low-density lipoprotein-like particle synthesized in the liver, its precise physiological role and mechanism of action remain unclear. This study aimed to investigate the relationship between lipoprotein(a) levels and development of NAFLD and hepatic fibrosis in Korean adults. A total of 1501 subjects who underwent abdominal ultrasonography at least twice as part of a health checkup program were enrolled. Biochemical and ultrasonography results were analyzed longitudinally, and the degree of hepatic fibrosis was calculated in subjects with NAFLD using serum biomarkers, such as fibrosis-4 (FIB-4). During the 3.36-year follow-up period, 352 patients (23.5%) were diagnosed with NAFLD. The subjects were categorized into 4 groups based on their lipoprotein(a) levels. Remarkably, the incidence of NAFLD decreased as the lipoprotein(a) levels increased. Following logistic regression analysis and adjustment for various risk factors, the odds ratio for the development of NAFLD was 0.625 (95% CI 0.440-0.888; P = .032) when comparing the highest to the lowest tertile of lipoprotein(a). However, no significant association was observed between the occurrence of hepatic fibrosis and lipoprotein(a) levels in subjects with NAFLD. Lipoprotein(a) levels have been identified as a significant predictor of NAFLD development. Additional large-scale studies with extended follow-up periods are required to better understand the effect of lipoprotein(a) on NAFLD and hepatic fibrosis.

Combined Effect of Body Mass Index and Waist Circumference in Predicting Nonalcoholic Fatty Liver Disease.

Background: Body mass index (BMI) and waist circumference (WC) are the most widely used anthropometric indices for identifying obesity. This study aimed to compare and clarify the usefulness of BMI, WC, and the combination of these two indicators in predicting nonalcoholic fatty liver disease (NAFLD). Methods: This cross-sectional study included 15,267 Korean adults. We defined four obesity categories using BMI and WC as follows: BMI nonobese and WC nonobese (BNWN); BMI obese and WC nonobese (BOWN); BMI nonobese and WC obese (BNWO); and BMI obese and WC obese (BOWO). Analysis of variance was used to compare fatty liver severity across each category. The odds ratios (ORs) and 95% confidence intervals (CIs) for NAFLD were calculated using multiple logistic regression. Results: Compared with BNWN, participants with BNWO were 3.235 (95% CI: 2.774-3.773) times more likely and participants with BOWN were 2.344 (95% CI: 2.045-2.687) times more likely to have NAFLD. Participants with BNWO had higher OR for NAFLD than those with BOWN. Moreover, BOWO participants had the highest OR of 4.788 (95% CI: 4.350-5.270) for NAFLD among all obesity categories. Conclusion: Combined obesity classification by BOWO is the most reliable indicator for NAFLD presence in Korean adults.

Association between Sarcopenic Obesity Status and Nonalcoholic Fatty Liver Disease and Fibrosis.

Background/Aims: There are no data regarding the association between sarcopenic obesity status and nonalcoholic fatty liver disease (NAFLD) and NAFLD-associated liver fibrosis. Therefore, we aimed to investigate the relationship between sarcopenic obesity status (sarcopenia only, obesity only, and sarcopenic obesity) and NAFLD and liver fibrosis in Korean adults. Methods: In total, 2,191 subjects completed a health checkup program, including abdominal ultrasonography and FibroScan. Subjects were classified into the following four categories: optimal body composition (nonobese and nonsarcopenic), sarcopenia only (nonobese), obesity only (nonsarcopenic), and sarcopenic obesity. Sarcopenic obesity was stratified by the skeletal muscle mass index and body fat using bioelectrical impedance analysis. NAFLD was diagnosed by ultrasonography, and liver fibrosis was assessed using transient elastography in subjects with NAFLD. Results: The prevalence of NAFLD and liver fibrosis significantly increased according to the sarcopenic obesity status. In the logistic regression analysis, after adjusting for multiple risk factors, the odds ratio (OR) for the risk of NAFLD was largest in the sarcopenic obesity group (OR, 3.68; 95% confidence interval [CI], 2.94 to 4.60), followed by the obesity only (OR, 2.25; 95% CI, 1.67 to 3.03) and sarcopenia only (OR, 1.92; 95% CI, 1.30 to 2.84) groups, when compared with the optimal group. Additionally, liver fibrosis was independently associated with sarcopenic obesity status (OR 4.69, 95% CI 1.95 to 11.29; OR 4.17, 95% CI 1.56 to 11.17; OR 3.80, 95% CI 0.86 to 16.75, respectively). Conclusions: These results demonstrated that sarcopenic obesity was independently associated with NAFLD and liver fibrosis and increased the risk of NAFLD and liver fibrosis more than obesity or sarcopenia alone.

Risk of Metabolic Syndrome and Fatty Liver Diseases in Gastric Cancer Survivors: A Propensity Score-Matched Analysis.

Background/Aims: To investigate the risk of metabolic syndrome and fatty liver diseases in gastric cancer survivors compared to non-cancer subjects. Methods: The data from the health screening registry of the Gangnam Severance Hospital from 2014-2019 was used. Ninety-one gastric cancer survivors and a propensity-score-matching 445 non-cancer subjects were analyzed. Gastric cancer survivors were divided into those with surgical treatment (OpGC, n=66) and non-surgical treatment (non-OpGC, n=25). Metabolic syndrome, fatty liver by ultrasonography, and metabolic dysfunction-associated fatty liver disease (MAFLD) were assessed. Results: Metabolic syndrome was in 15.4% of gastric cancer survivors (OpGC; 13.6%, non-OpGC; 20.0%). Fatty liver by ultrasonography was in 35.2% in gastric cancer survivors (OpGC; 30.3%, non-OpGC: 48.0%). MAFLD was in 27.5% of gastric cancer survivor (OpGC; 21.2%, non-OpGC; 44.0%). After adjusting for age, sex, smoking, and alcohol, the risk of metabolic syndrome was lower in OpGC than in non-cancer subjects (OR, 0.372; 95% CI, 0.176-0.786, p=0.010). After adjusting, OpGC showed lower risks of fatty liver by ultrasonography (OR, 0.545; 95% CI, 0.306-0.970, p=0.039) and MAFLD (OR, 0.375; 95% CI, 0.197-0.711, p=0.003) than did non-cancer subjects. There were no significant differences in the risks of metabolic syndrome and fatty liver diseases between non-OpGC and non-cancer subjects. Conclusions: OpGC showed lower risks of metabolic syndrome, fatty liver by ultrasonography, and MAFLD than non-cancer subjects, but there were no significant differences in the risks between non-OpGC and non-cancer subjects. Further studies on metabolic syndrome and fatty liver diseases in gastric cancer survivors are warranted.

Comparison of a Machine Learning Method and Various Equations for Estimating Low-Density Lipoprotein Cholesterol in Korean Populations.

BACKGROUND: LDL-C is the primary target of lipid-lowering therapy and used to classify patients by cardiovascular disease risk. We aimed to develop a deep neural network (DNN) model to estimate LDL-C levels and compare its performance with that of previous LDL-C estimation equations using two large independent datasets of Korean populations. METHODS: The final analysis included participants from two independent population-based cohorts: 129,930 from the Gangnam Severance Health Check-up (GSHC) and 46,470 participants from the Korean Initiatives on Coronary Artery Calcification registry (KOICA). The DNN model was derived from the GSHC dataset and validated in the KOICA dataset. We measured our proposed model's performance according to bias, root mean-square error (RMSE), proportion (P)10-P20, and concordance. P was defined as the percentage of patients whose LDL was within ±10-20% of the measured LDL. We further determined the RMSE scores of each LDL equation according to Pooled cohort equation intervals. RESULTS: Our DNN method has lower bias and root mean-square error than Friedewald's, Martin's, and NIH equations, showing a high agreement with LDL-C measured by homogenous assay. The DNN method offers more precise LDL estimation in all pooled cohort equation strata. CONCLUSION: This method may be particularly helpful for managing a patient's cholesterol levels based on their atherosclerotic cardiovascular disease risk.

Triglyceride Glucose-Waist Circumference (TyG-WC) Is a Reliable Marker to Predict Non-Alcoholic Fatty Liver Disease.

The triglyceride and glucose index (TyG index), a marker of insulin resistance, is positively associated with NAFLD. Modified TyG indices, combining body composition markers including body-mass index (BMI) or waist circumference (WC) with the TyG index, are reported to enhance predictability of insulin resistance. This study aimed to compare the usefulness of modified TyG indices for predicting NAFLD with the TyG index and fatty liver index (FLI). This cross-sectional study included 12,757 Korean adults. The TyG index and FLI were calculated using established formulas, and TyG-BMI and TyG-WC were calculated as TyG × BMI and TyG × WC, respectively. All measures were divided into quartiles. NAFLD severity (grade 0-3) was compared using ANOVA by quartiles of each index. Odds ratios (ORs) and 95% confidence intervals (CIs) for NAFLD were calculated using a multiple logistic regression analysis. ROC and AUROC analyses were performed to compare the predictability of NAFLD using WC, BMI, TyG, TyG-BMI, TyG-WC, and FLI. A higher TyG index, TyG-BMI, TyG-WC, and FLI were associated with a higher grade of NAFLD. ORs (CIs) for NAFLD increased in all indices, especially in TyG-WC (39.251 (31.304-49.215)) and FLI (38.937 (31.145-48.678)). AUROC was 0.848 (0.840-0.855) for TyG-WC and 0.850 (0.842-0.857) for FLI. TyG-WC is a reliable indicator for the presence of NAFLD in Korean adults.

Triglyceride-glucose index is associated with gastroesophageal reflux disease and erosive reflux disease: a health checkup cohort study.

The triglyceride-glucose (TyG) index was proposed as a useful marker of metabolic syndrome. Insulin resistance, the main mechanism underlying metabolic syndrome, is related to gastroesophageal reflux disease (GERD). This study aimed to elucidate the association between the TyG index and GERD/erosive reflux disease (ERD). We retrospectively reviewed the electronic medical records of patients who underwent gastroduodenoscopy at a checkup center. The calculation of TyG index used following formula: ln (fasting triglycerides [mg/dL] × fasting glucose [mg/dL]/2). We divided the patients into four groups according to the TyG index quartile (Q). We evaluated the relationship between the alteration of the TyG index and GERD in patients who received health checkups two times. Among the 52,605 enrolled patients, 3073 (5.8%) and 434 (0.8%) were diagnosed with GERD and ERD, respectively. The odds ratios (ORs) for GERD in the TyG index progressively increased across quartiles (P < 0.001): Q2 (OR = 2.477), Q3 (OR = 3.013), and Q4 (OR = 4.027) compared with Q1, which was used as a reference, respectively. Those for ERD also progressively increased across quartiles (P < 0.001): Q2 (OR = 4.264), Q3 (OR = 4.841), and Q4 (OR = 7.390) compared with Q1, respectively. Moreover, the degree of TyG index increase during the first and second tests in the GERD group was more prominent than in the control group (P = 0.001). In conclusion, the higher TyG index was significantly associated with GERD. The TyG index may be a novel predictive biomarker of GERD and ERD.

Comparison of the effectiveness of Martin's equation, Friedewald's equation, and a Novel equation in low-density lipoprotein cholesterol estimation.

Low-density-lipoprotein cholesterol (LDL-C) is the main target in atherosclerotic cardiovascular disease (ASCVD). We aimed to validate and compare a new LDL-C estimation equation with other well-known equations. 177,111 samples were analysed from two contemporary population-based cohorts comprising asymptomatic Korean adults who underwent medical examinations. Performances of the Friedewald (FLDL), Martin (MLDL), and Sampson (SLDL) equations in estimating direct LDL-C by homogenous assay were assessed by measures of concordance (R2, RMSE, and mean absolute difference). Analyses were performed according to various triglyceride (TG) and/or LDL-C strata. Secondary analyses were conducted within dyslipidaemia populations of each database. MLDL was superior or at least similar to other equations regardless of TG/LDL-C, in both the general and dyslipidaemia populations (RMSE = 11.45/9.20 mg/dL; R2 = 0.88/0.91; vs FLDL: RMSE = 13.66/10.42 mg/dL; R2 = 0.82/0.89; vs SLDL: RMSE = 12.36/9.39 mg/dL; R2 = 0.85/0.91, per Gangnam Severance Hospital Check-up/Korea Initiatives on Coronary Artery Calcification data). MLDL had a slight advantage over SLDL with the lowest MADs across the full spectrum of TG levels, whether divided into severe hyper/non-hyper to moderate hypertriglyceridaemia samples or stratified by 100-mg/dL TG intervals, even up to TG values of 500-600 mg/dL. MLDL may be a readily adoptable and cost-effective alternative to direct LDL-C measurement, irrespective of dyslipidaemia status. In populations with relatively high prevalence of mild-to-moderate hypertriglyceridaemia, Martin's equation may be optimal for LDL-C and ASCVD risk estimation.

Deep-learning-based cardiovascular risk stratification using coronary artery calcium scores predicted from retinal photographs.

BACKGROUND: Coronary artery calcium (CAC) score is a clinically validated marker of cardiovascular disease risk. We developed and validated a novel cardiovascular risk stratification system based on deep-learning-predicted CAC from retinal photographs. METHODS: We used 216 152 retinal photographs from five datasets from South Korea, Singapore, and the UK to train and validate the algorithms. First, using one dataset from a South Korean health-screening centre, we trained a deep-learning algorithm to predict the probability of the presence of CAC (ie, deep-learning retinal CAC score, RetiCAC). We stratified RetiCAC scores into tertiles and used Cox proportional hazards models to evaluate the ability of RetiCAC to predict cardiovascular events based on external test sets from South Korea, Singapore, and the UK Biobank. We evaluated the incremental values of RetiCAC when added to the Pooled Cohort Equation (PCE) for participants in the UK Biobank. FINDINGS: RetiCAC outperformed all single clinical parameter models in predicting the presence of CAC (area under the receiver operating characteristic curve of 0·742, 95% CI 0·732-0·753). Among the 527 participants in the South Korean clinical cohort, 33 (6·3%) had cardiovascular events during the 5-year follow-up. When compared with the current CAC risk stratification (0, >0-100, and >100), the three-strata RetiCAC showed comparable prognostic performance with a concordance index of 0·71. In the Singapore population-based cohort (n=8551), 310 (3·6%) participants had fatal cardiovascular events over 10 years, and the three-strata RetiCAC was significantly associated with increased risk of fatal cardiovascular events (hazard ratio [HR] trend 1·33, 95% CI 1·04-1·71). In the UK Biobank (n=47 679), 337 (0·7%) participants had fatal cardiovascular events over 10 years. When added to the PCE, the three-strata RetiCAC improved cardiovascular risk stratification in the intermediate-risk group (HR trend 1·28, 95% CI 1·07-1·54) and borderline-risk group (1·62, 1·04-2·54), and the continuous net reclassification index was 0·261 (95% CI 0·124-0·364). INTERPRETATION: A deep learning and retinal photograph-derived CAC score is comparable to CT scan-measured CAC in predicting cardiovascular events, and improves on current risk stratification approaches for cardiovascular disease events. These data suggest retinal photograph-based deep learning has the potential to be used as an alternative measure of CAC, especially in low-resource settings. FUNDING: Yonsei University College of Medicine; Ministry of Health and Welfare, Korea Institute for Advancement of Technology, South Korea; Agency for Science, Technology, and Research; and National Medical Research Council, Singapore.

Relative fat mass at baseline and its early change may be a predictor of incident nonalcoholic fatty liver disease.

The relationship between changes in body components and the risk of nonalcoholic fatty liver disease (NAFLD) is not fully understood. We investigated the effects of body components and subsequent changes on incident NAFLD at follow-up ultrasound scanning in a longitudinal cohort. We included 9967 participants without NAFLD at baseline who underwent serial health examinations. Sex-specific, weight-adjusted skeletal muscle index (SMI_Wt) was used. Mean follow-up duration was 48.5 ± 33.5 months. NAFLD developed in 2395 participants (24.0%). Body composition was measured using bioelectrical impedance analysis. The following baseline body components were significantly associated with incident NAFLD: the lowest and middle SMI_Wt tertiles in the normal-weight group (adjusted hazard ratio [aHR] = 2.20 and 1.54, respectively), and fat percentage in the normal-weight (aHR = 1.12), overweight (aHR = 1.05), and obese groups (aHR = 1.03) (all P < 0.05). Among 5,033 participants who underwent ≥ 3 health examinations, SMI_Wt increase between the first and second examinations was an independent protective factor against incident NAFLD in non-obese groups (P < 0.05). Increased fat percentage was an independent risk factor for incident NAFLD in all weight categories (P < 0.05). High fat mass at baseline may be a better predictor of incident NAFLD than muscle mass. Reciprocal changes in fat and muscle mass during the first year of follow-up predicted incident NAFLD in non-obese groups.

Prediction of systemic biomarkers from retinal photographs: development and validation of deep-learning algorithms.

BACKGROUND: The application of deep learning to retinal photographs has yielded promising results in predicting age, sex, blood pressure, and haematological parameters. However, the broader applicability of retinal photograph-based deep learning for predicting other systemic biomarkers and the generalisability of this approach to various populations remains unexplored. METHODS: With use of 236 257 retinal photographs from seven diverse Asian and European cohorts (two health screening centres in South Korea, the Beijing Eye Study, three cohorts in the Singapore Epidemiology of Eye Diseases study, and the UK Biobank), we evaluated the capacities of 47 deep-learning algorithms to predict 47 systemic biomarkers as outcome variables, including demographic factors (age and sex); body composition measurements; blood pressure; haematological parameters; lipid profiles; biochemical measures; biomarkers related to liver function, thyroid function, kidney function, and inflammation; and diabetes. The standard neural network architecture of VGG16 was adopted for model development. FINDINGS: In addition to previously reported systemic biomarkers, we showed quantification of body composition indices (muscle mass, height, and bodyweight) and creatinine from retinal photographs. Body muscle mass could be predicted with an R2 of 0·52 (95% CI 0·51-0·53) in the internal test set, and of 0·33 (0·30-0·35) in one external test set with muscle mass measurement available. The R2 value for the prediction of height was 0·42 (0·40-0·43), of bodyweight was 0·36 (0·34-0·37), and of creatinine was 0·38 (0·37-0·40) in the internal test set. However, the performances were poorer in external test sets (with the lowest performance in the European cohort), with R2 values ranging between 0·08 and 0·28 for height, 0·04 and 0·19 for bodyweight, and 0·01 and 0·26 for creatinine. Of the 47 systemic biomarkers, 37 could not be predicted well from retinal photographs via deep learning (R2≤0·14 across all external test sets). INTERPRETATION: Our work provides new insights into the potential use of retinal photographs to predict systemic biomarkers, including body composition indices and serum creatinine, using deep learning in populations with a similar ethnic background. Further evaluations are warranted to validate these findings and evaluate the clinical utility of these algorithms. FUNDING: Agency for Science, Technology, and Research and National Medical Research Council, Singapore; Korea Institute for Advancement of Technology.

Prevalence of and Risk Factors for Diseases in Korean Americans and Native Koreans Undergoing Health Checkup.

BACKGROUND: Korean Americans constitute the fifth largest subgroup in the Asian American population. Despite their increasing population, research and guidelines regarding their health status assessment and disease screening are lacking. This study aimed to compare the prevalence of diseases in Korean Americans and native Koreans to determine the risk factors and guidelines for disease screening. METHODS: Patients who visited the Gangnam Severance Hospital from February 2010 to May 2015 for a health checkup were enrolled in this study. Baseline characteristics, laboratory data, and the organs (stomach, colon, thyroid, brain, prostate, lung, liver, kidney, pancreas, adrenal gland, and heart) of patients were examined. Data regarding patients' dietary patterns were also obtained. Overall, 1,514 Korean Americans (group 1) and 1,514 native Koreans (group 2) were enrolled. RESULTS: The following diseases were more prevalent in group 1 than in group 2: reflux esophagitis (12.9% vs. 10%), gastric ulcer (3.0% vs. 5.5%), colorectal polyp (37.7% vs. 28.7%), hemorrhoids (32.2% vs. 29.9%), and benign prostatic hyperplasia (30.2% vs. 14.3%). Although not statistically significant, coronary artery disease has a high prevalence rate of >20% in both groups. Dietary patterns were not significant between the two groups. CONCLUSION: This study showed that the prevalence of several diseases in Korean Americans differed from that observed in native Koreans. Therefore, a foundation for setting up new guidelines for disease screening among Korean Americans is established.