Baseline glycemic status and mortality in 241,499 Korean metropolitan subjects: A Kangbuk Samsung Health Study

Insulin Resistance, Non-Alcoholic Fatty Liver Disease and Type 2 Diabetes Mellitus: Clinical and Experimental Perspective

It has been generally accepted that insulin resistance (IR) and reduced insulin secretory capacity are the basic pathogenesis of type 2 diabetes mellitus (T2DM). In addition to genetic factors, the persistence of systemic inflammation caused by obesity and the associated threat of lipotoxicity increase the risk of T2DM. In particular, the main cause of IR is obesity and subjects with T2DM have a higher body mass index (BMI) than normal subjects according to recent studies. The prevalence of T2DM with IR has increased with increasing BMI during the past three decades. According to recent studies, homeostatic model assessment of IR was increased compared to that of the 1990s. Rising prevalence of obesity in Korea have contributed to the development of IR, non-alcoholic fatty liver disease and T2DM and cutting this vicious cycle is important. My colleagues and I have investigated this pathogenic mechanism on this theme through clinical and experimental studies over 20 years and herein, I would like to summarize some of our studies with deep gratitude for receiving the prestigious 2023 Sulwon Award.

The Influence of Obesity and Metabolic Health on Vascular Health

The prevalence of obesity is rapidly increasing worldwide. Obesity should not be understood only as the accumulation of fat in the body, but instead as a phenomenon that exerts different effects on our health according to the place of fat deposition and its stability. Obesity is the starting point of most metabolic diseases, such as diabetes, hypertension, metabolic syndrome, sleep apnea, and eventually cardiovascular disease. There are different kinds of obesity, ranging from simple obesity to sarcopenic obesity. The main purpose of intervening to address obesity is to decrease the ultimate consequence of obesity-namely, cardiovascular disease. The main mechanism through which obesity, especially abdominal obesity, increases cardiovascular risk is the obesity-induced derangement of metabolic health, leading to the development of metabolic diseases such as diabetes, non-alcoholic fatty liver disease, and metabolic syndrome, which are the main initiators of vascular damage. In this review, I discuss the influence of various types of obesity on the risk of metabolic diseases, and how these diseases increase cardiovascular disease risk.

Application of deep learning to predict advanced neoplasia using big clinical data in colorectal cancer screening of asymptomatic adults

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%.

CONCLUSION

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.

Deep Neural Network-Based Prediction of the Risk of Advanced Colorectal Neoplasia

Background/Aims

Risk prediction models using a deep neural network (DNN) have not been reported to predict the risk of advanced colorectal neoplasia (ACRN). The aim of this study was to compare DNN models with simple clinical score models to predict the risk of ACRN in colorectal cancer screening.

Methods

Databases of screening colonoscopy from Kangbuk Samsung Hospital (n=121,794) and Kyung Hee University Hospital at Gangdong (n=3,728) were used to develop DNN-based prediction models. Two DNN models, the Asian-Pacific Colorectal Screening (APCS) model and the Korean Colorectal Screening (KCS) model, were developed and compared with two simple score models using logistic regression methods to predict the risk of ACRN. The areas under the receiver operating characteristic curves (AUCs) of the models were compared in internal and external validation databases.

Results

In the internal validation set, the AUCs of DNN model 1 and the APCS score model were 0.713 and 0.662 (p<0.001), respectively, and the AUCs of DNN model 2 and the KCS score model were 0.730 and 0.667 (p<0.001), respectively. However, in the external validation set, the prediction performances were not significantly different between the two DNN models and the corresponding APCS and KCS score models (both p>0.1).

Conclusions

Simple score models for the risk prediction of ACRN are as useful as DNN-based models when input variables are limited. However, further studies on this issue are warranted to predict the risk of ACRN in colorectal cancer screening because DNN-based models are currently under improvement.

Increased Mortality Burden in Young Asian Subjects with Dysglycemia and Comorbidities

BACKGROUND

High blood glucose level has a linear relationship with all-cause mortality. However, the influence of glycemic abnormality on mortality differs by age group. We aimed to analyze all-cause mortality according to glycemic status, age groups, and comorbidities using a national health database.

METHODS

The 6,330,369 participants who underwent Korean National Health Screening in 2009 were followed up until 2016, with a median follow-up of 7.3 years. All-cause mortality rates were analyzed according to glycemic status (normoglycemia, impaired fasting glucose [IFG], newly diagnosed diabetes, diabetes duration <5 years, diabetes duration ≥5 years), age groups (20-39, 40-65, and ≥65 years), and comorbidities using the Korean National Health Insurance System database.

RESULTS

At baseline, 712,901 (11.3%) subjects had diabetes. Compared with subjects without diabetes, those with diabetes at baseline showed increased mortality risk after adjustment for multiple risk factors (hazard ratio [HR] 1.613; 95% confidence interval [CI] 1.598,1.629), and those with IFG showed a significantly increased mortality risk compared with normoglycemic subjects (HR 1.053; 95% CI 1.042,1.064). Mortality risk associated with glycemic status decreased gradually from younger to older age groups and was consistently higher in those with diabetes with coronary heart disease, ischemic stroke or decreased renal function than those without comorbidities.

CONCLUSION

Compared with normoglycemic subjects, subjects with diabetes and IFG had an increased mortality risk and the mortality risk was higher in the younger age group than in the older age group. The presence of diabetes and comorbid diseases synergistically increased mortality risk.

Basal insulin initiation use and experience among people with type 2 diabetes mellitus with different patterns of persistence: results from a multi-national survey

BACKGROUND AND OBJECTIVE

People with type 2 diabetes mellitus (T2DM) often interrupt basal insulin treatment soon after initiation. This study aimed to describe the experiences during and after basal insulin initiation among people with T2DM with different persistence patterns.

METHODS

Adults with T2DM from France, Germany, Spain, UK, US, Brazil, and Japan were identified from consumer panels for an online survey. Respondents who initiated basal insulin 3-24 months prior to survey date were categorized as continuers (no gaps of ≥7 days in insulin treatment); interrupters (first gap ≥7 days within 6 months of initiation and restarted insulin); and discontinuers (stopped insulin for ≥7 days within 6 months of initiation without restarting).

RESULTS

Among 942 participants, continuers were older than interrupters and discontinuers (46, 37, and 38 years, respectively, p < .01). Continuers reported having fewer concerns before and after insulin initiation than interrupters and discontinuers, while interrupters had the most concerns. Continuers also reported fewer challenges during the first week of insulin use. Continuers were more likely to respond that insulin use had a positive impact on specific aspects of life than interrupters and discontinuers, for example on glycemic control (73.0%, 63.0%, and 61.8%, respectively; p < .01 vs. continuers).

CONCLUSION

Among people with T2DM with different persistence patterns after basal insulin initiation there were significant differences in patient characteristics and experience during and after insulin initiation. Interrupters and discontinuers more frequently reported having concerns and challenges during the initiation process, negative impacts after initiation, and less improvement in glycemic control than continuers.

Derivation and validation of a risk scoring model to predict advanced colorectal neoplasm in adults of all ages

BACKGROUND AND AIMS

Little is known about how to include adults < 50 years in colonoscopy screening. This study aimed to derive a risk-scoring model incorporating laboratory indicators for metabolic risks to predict advanced colorectal neoplasia (ACN) in asymptomatic Korean adults both younger and older than 50 years.

METHODS

In this cross-sectional study, 70 812 consecutive adult recipients of a screening colonoscopy in a single health check-up center in Korea between 2003 and 2012 were enrolled. A risk score model was developed using multiple logistic regression model and internally validated.

RESULTS

Overall prevalence of ACN was 1.4% (956/70 812). A 15-point score model was developed to comprise age, sex, family history of colorectal cancer, smoking, body mass index, serum levels of fasting glucose, low-density lipoprotein cholesterol, and carcinoembryonic antigen. Based on the score, the validation cohort could be categorized into five risk groups (low, borderline, moderate, high, and very high) with an ACN prevalence of 0.7%, 1.3%, 2.7%, 6.6%, and 13.2%, respectively. Compared with the borderline risk group, the low-risk group showed a 50.3% reduced risk of ACN. Meanwhile, the moderate, high, and very high risk groups showed 2, 5, and 10-fold increased risk of ACN. The score showed significantly superior discriminative power than the Asian-Pacific colorectal screening score (P = 0.003).

CONCLUSIONS

Our scoring model based on both clinical and laboratory risk factors is useful for the prediction of ACN. This score may be used to include adults < 50 years in colonoscopy screening.