CONTEXT:Although salivary steroid sampling offers several advantages, the diagnostic potential of salivary steroid metabolites remains largely unexplored. OBJECTIVE:This work aimed to evaluate the diagnostic utility of salivary steroid profiling in patients with adrenal diseases. METHODS:This prospective, multicenter study comprised 313 patients with nonfunctioning adrenal adenoma (NF), primary aldosteronism (PA), Cushing syndrome (CS), and mild autonomous cortisol secretion (MACS). Salivary samples for morning and night steroids were collected using standardized protocols. The liquid chromatography-mass spectrometry-based steroid profiling was applied to quantify cortisol, cortisone, tetrahydrocortisone (THE), 20α-dihydrocortisol (20α-DHF), 18-hydroxycortisol (18-OHF), and dehydroepiandrosterone sulfate (DHEA-S). The primary outcome was the diagnostic performance of morning and night salivary steroids. RESULTS:In the PA group, morning and night salivary levels of 18-OHF were higher compared to the other groups (all P < .05). Morning and night salivary levels of cortisone, THE, and 20α-DHF were significantly elevated in the CS group compared to other groups (all P < .05). Only night-time salivary levels of cortisone (P = .040) and 20α-DHF (P = .029) were elevated in the MACS group compared to the NF group. Receiver operating characteristic analyses indicated that morning salivary 18-OHF was moderately specific for PA, whereas night salivary 20α-DHF and cortisone provided robust diagnostic accuracy for CS. Combined night salivary steroids exhibited superior diagnostic performance compared to morning salivary steroids in the CS group (area under the curve, 0.903 vs 0.754; P = .007). CONCLUSION:Salivary steroid profiling holds promise as a noninvasive tool for the diagnosis of adrenal diseases.

AIMS:To assess the efficacy and safety of three triple-combination therapies in patients with type 2 diabetes (T2D) inadequately controlled on metformin and a dipeptidyl peptidase-4 (DPP-4) inhibitor. MATERIALS AND METHODS:This multicentre, prospective, randomised, open-label, parallel-group, phase 4 study included patients with T2D inadequately controlled on metformin (≥1000 mg) and a DPP-4 inhibitor. Participants were randomised to receive empagliflozin 10 mg/day (n = 61), pioglitazone 15 mg/day (n = 58) or glimepiride 2 mg/day (n = 57). The primary outcome was glycated haemoglobin (HbA1c) level change after 24 weeks of treatment. RESULTS:The mean age, HbA1c level and diabetes duration of the patients were 58.5 ± 10.0 years, 7.8 ± 0.7% and 8.1 ± 5.6 years, respectively. HbA1c level decreased after treatment (-0.78 ± 0.09, empagliflozin; -0.89 ± 0.09, pioglitazone; and -0.93 ± 0.12 glimepiride). No significant differences were observed in HbA1c reduction among the three triple-combination therapies. The proportions of patients with HbA1c < 7.0% were similar across the three regimens (65.6%, empagliflozin; 56.9%, pioglitazone; and 63.2%, glimepiride). Significant weight loss was observed in the empagliflozin group (-1.73 ± 3.14 kg), whereas weight gain was observed in the pioglitazone and glimepiride groups (1.11 ± 3.97 kg and 1.11 ± 4.07 kg, respectively). The glimepiride group reported four hypoglycaemic episodes (6.56%), while none were reported in the other groups. CONCLUSIONS:The addition of empagliflozin, pioglitazone, or glimepiride to metformin and DPP-4 inhibitors significantly improved glycaemic control in patients with T2D. The selection of a third agent should be individualized based on the patient characteristics.

AIM:This study evaluated the efficacy and safety of empagliflozin 10 and 25 mg compared to placebo as add-on treatment for people with type 2 diabetes mellitus (T2DM) uncontrolled after ≥8 weeks of treatment with metformin and sitagliptin. MATERIALS AND METHODS:A randomised, double-blind, multicentre, therapeutic confirmatory, phase 3 clinical trial was conducted in 172 patients with T2DM. Participants with glycosylated haemoglobin (HbA1c) levels 7%-10% receiving sitagliptin and metformin were randomised 1:1:1 to empagliflozin 10 mg, empagliflozin 25 mg, or placebo. The primary endpoint was the change in HbA1c from baseline to week 24. RESULTS:After 24 weeks of treatment, HbA1c levels were significantly decreased in the empagliflozin 10 and 25 mg group versus the placebo group; the adjusted mean differences with empagliflozin 10 and 25 mg versus placebo were -0.7% (95% CI -1.0, -0.4; p <.0001) and -0.8% (95% CI -1.1, -0.5; p <.0001), respectively. Fasting plasma glucose levels were also significantly decreased in both empagliflozin groups compared to the placebo group (both p <.0001). More patients reached HbA1c <7% or <6.5% after 24 weeks in the empagliflozin 10 and 25 mg groups versus the placebo group (both p <.05). Efficacy was maintained in the empagliflozin groups during a 28-week extension period. Empagliflozin add-on was associated with improvements in albuminuria and body weight. The incidence of adverse events was similar across groups; add-on empagliflozin was well tolerated. CONCLUSIONS:These results suggest that coadministration of empagliflozin safely improves glycemic control in Korean patients with T2DM uncontrolled by sitagliptin and metformin.

In Harrison’s Principles of Internal Medicine, human understanding is emphasized as one of three necessary characteristics that a physician must have. Inflammation, which is caused by inflammatory inducers (inf-ids), is a fundamental feature of disease at the cellular and molecular levels. Inflammation protects the body, but excessive or prolonged inflammation can be damaging and can cause disease. Humans are repeatedly exposed to external and internal environmental factors that generate inf-ids throughout their lives. External environmental factors include microbial and non-microbial inf-ids, as well as stressors that inevitably arise during social interactions. Internal environmental factors include the adaptive physiological response that is present from birth. Inf-ids may also be produced by the four-step habit loop, which consists of a cue (e.g., stressor), emotions, routine act (adaptive response), and a reward. Immune cells in the circulatory system and in tissues may have positive and negative effects in inflammatory responses. However, low-grade inflammation may be difficult to detect. We propose a model of disease development that integrates external and internal environmental factors from the perspective of human understanding.

AIM:To compare the effectiveness and safety of moderate-intensity pravastatin 40 mg/day and atorvastatin 10 mg/day in patients with dyslipidemia. METHODS:We conducted a retrospective cohort study using electronic health records of 19 million patients across 14 secondary/tertiary hospitals, standardized to a Common Data Model. New users of pravastatin (40 mg/day) and atorvastatin (10 mg/day) were identified. Six distinct cohorts were used to assess the comparative effectiveness in preventing major adverse cardiovascular events (MACE) and the risks of new-onset diabetes mellitus (NODM), myalgia or rhabdomyolysis, and hepatotoxicity (measured by aspartate aminotransferase [AST]/alanine aminotransferase [ALT]). Propensity score matching (PSM) was applied to each cohort for effectiveness and safety analyses, followed by a meta-analysis of hospital-specific results. RESULTS:After PSM, patients were equally assigned to the pravastatin and atorvastatin groups for primary (n = 2,688/group) and secondary MACE prevention (n = 1,258/group) and to assess the risk of NODM (n = 2,391/group), new-onset myalgia or rhabdomyolysis (n = 11,799/group), and hepatotoxicity (AST, n = 4,034/group; ALT, n = 3,655/group). No significant differences were observed in the hazard ratios (HRs) for primary (HR = 0.84; 95% CI, 0.59-1.20) and secondary MACE prevention (HR = 0.89; 95% CI, 0.68-1.16). Similarly, no significant difference was observed in the risk of NODM (HR, 0.99; 95% CI, 0.79-1.23). The risk of new-onset myalgia/rhabdomyolysis (HR = 0.82, 95% CI, 0.69-0.96) and the incidence of abnormal elevations in AST levels (2.35% vs. 3.37%, p＜0.05) were significantly lower in the pravastatin group. CONCLUSION:Moderate-intensity pravastatin (40 mg/day) showed comparable effectiveness to moderate-intensity atorvastatin (10 mg/day) in preventing MACE with a more favorable safety profile.

Background: This study investigated the real-world efficacy and safety of insulin degludec/insulin aspart (IDegAsp) in Korean adults with type 2 diabetes mellitus (T2DM), whose insulin treatment was switched to IDegAsp. Methods: This was a multicenter, retrospective, observational study comprising two 26-week treatment periods, before and after switching to IDegAsp, respectively. Korean adults with uncontrolled T2DM treated with basal or premix insulin (+/- oral antidiabetic drugs) were enrolled. The primary objective was to compare the degree of glycosylated hemoglobin (HbA1c) change in each 26-week observation period. The analyses included changes in HbA1c, fasting plasma glucose (FPG), body weight, proportion of participants achieving HbA1c <7.0%, hypoglycemic events, and total daily insulin dose (ClinicalTrials.gov, number NCT04656106). Results: In total, 196 adults (mean age, 65.95 years; mean T2DM duration, 18.99 years) were analyzed. The change in both HbA1c and FPG were significantly different between the pre-switching and the post-switching period (0.28% vs. -0.51%, P< 0.001; 5.21 mg/dL vs. -23.10 mg/dL, P= 0.005), respectively. After switching, the rate of achieving HbA1c <7.0% was significantly improved (5.10% at baseline vs. 11.22% with IDegAsp, P= 0.012). No significant differences (before vs. after switching) were observed in body weight change, and total daily insulin dose. The rates of overall and severe hypoglycemia were similar in the two periods. Conclusion: In real-world clinical practice in Korea, the change of insulin regimen to IDegAsp was associated with an improvement in glycemic control without increase of hypoglycemia, supporting the use of IDegAsp for patients with T2DM uncontrolled with basal or premix insulin.

We developed an attention model to predict future adverse glycemic events 30 min in advance based on the observation of past glycemic values over a 35 min period. The proposed model effectively encodes insulin administration and meal intake time using Time2Vec (T2V) for glucose prediction. The proposed impartial feature selection algorithm is designed to distribute rewards proportionally according to agent contributions. Agent contributions are calculated by a step-by-step negation of updated agents. Thus, the proposed feature selection algorithm optimizes features from electronic medical records to improve performance. For evaluation, we collected continuous glucose monitoring data from 102 patients with type 2 diabetes admitted to Cheonan Hospital, Soonchunhyang University. Using our proposed model, we achieved F1-scores of 89.0%, 60.6%, and 89.8% for normoglycemia, hypoglycemia, and hyperglycemia, respectively.

AIMS:To evaluate the long-term safety and efficacy of enavogliflozin monotherapy (0.3 mg/day) in individuals with type 2 diabetes mellitus (T2DM). MATERIALS AND METHODS:Following a 24-week randomized, double-blind treatment period with enavogliflozin 0.3 mg/day (n = 77) or placebo (n = 69), consenting participants received enavogliflozin 0.3 mg/day for an additional 28 weeks during an open-label extension (OLE) period. The safety and efficacy of enavogliflozin were assessed at Week 52. RESULTS:A total of 37 participants continued enavogliflozin (maintenance group), and 26 participants switched from placebo to enavogliflozin (switch group). No additional adverse drug reactions related to enavogliflozin were observed during the OLE period. At Week 52, glycated haemoglobin (HbA1c) and fasting plasma glucose were significantly lower than at the baseline, by 0.9% and 24.9 mg/dL, respectively, in the maintenance group (p < 0.0001 for both), and by 0.7% and 18.0 mg/dL, respectively, in the switch group (p < 0.0001 and p = 0.002). The proportions of participants reaching HbA1c 7.0% (53 mmol/mol) at Week 52 were 69.4% in the maintenance group and 65.4% in the switch group. A significant increase in urine glucose-to-creatinine ratio was observed at Week 52, by 84.9 g/g and 67.1 g/g in the maintenance and switch groups, respectively (p < 0.0001 for both). Body weight in both groups decreased significantly (p < 0.0001) from baseline to Week 52, by 3.5 kg and 3.8 kg in the maintenance and switch groups, respectively. CONCLUSIONS:Enavogliflozin 0.3 mg monotherapy provides long-term glycaemic control in T2DM and is safe and well tolerated during a 52-week treatment period.

BACKGROUND:We compared the efficacy and safety of low-intensity atorvastatin and ezetimibe combination therapy with moderate-intensity atorvastatin monotherapy in patients requiring cholesterol-lowering therapy.METHODS:At 19 centers in Korea, 290 patients were randomized to 4 groups: atorvastatin 5 mg and ezetimibe 10 mg (A5E), ezetimibe 10 mg (E), atorvastatin 5 mg (A5), and atorvastatin 10 mg (A10). Clinical and laboratory examinations were performed at baseline, and at 4-week and 8-week follow-ups. The primary endpoint was percentage change from baseline in low-density lipoprotein (LDL) cholesterol levels at the 8-week follow-up. Secondary endpoints included percentage changes from baseline in additional lipid parameters.RESULTS:Baseline characteristics were similar among the study groups. At the 8-week follow-up, percentage changes in LDL cholesterol levels were significantly greater in the A5E group (49.2%) than in the E (18.7%), A5 (27.9%), and A10 (36.4%) groups. Similar findings were observed regarding the percentage changes in total cholesterol, non-high-density lipoprotein cholesterol, and apolipoprotein B levels. Triglyceride levels were also significantly decreased in the A5E group than in the E group, whereas high-density lipoprotein levels substantially increased in the A5E group than in the E group. In patients with low- and intermediate-cardiovascular risk, 93.3% achieved the target LDL cholesterol levels in the A5E group, 40.0% in the E group, 66.7% in the A5 group, and 92.9% in the A10 group. In addition, 31.4% of patients in the A5E group, 8.1% in E, 9.7% in A5, and 7.3% in the A10 group reached the target levels of both LDL cholesterol < 70 mg/dL and reduction of LDL ≥ 50% from baseline.CONCLUSIONS:The addition of ezetimibe to low-intensity atorvastatin had a greater effect on lowering LDL cholesterol than moderate-intensity atorvastatin alone, offering an effective treatment option for cholesterol management, especially in patients with low and intermediate risks.

Forecasting blood glucose (BG) values for patients can help prevent hypoglycemia and hyperglycemia events in advance. To this end, this study proposes an intelligent ensemble deep learning system to predict BG values in 15, 30, and 60 min prediction horizons (PHs) based on historical BG values collected via continuous glucose monitoring devices as an endogenous factor and carbohydrate intake and insulin administration information (times) as exogenous factors. Although there are numerous deep learning algorithms available, this study applied five algorithms, namely, recurrent neural network (RNN), which is optimized for sequence data (e.g., time-series), and RNN-based algorithms (e.g., long short-term memory (LSTM), stacked LSTM, bidirectional LSTM, and gated recurrent unit). Then, a genetic algorithm (GA) was applied to the five prediction models to optimize their weights through ensemble techniques and to yield (output) the final predicted BG values. The performance of the proposed model was compared to that of the autoregressive integrated moving average (ARIMA) model as a baseline. The results show that the proposed model significantly outperforms the baseline in terms of the root mean square error (RMSE) and continuous glucose error grid analysis. For the valid 29 diabetic patients for the multivariate models, the RMSE was 11.08 ( ± 3.19), 19.25 ( ± 5.28), and 31.30 ( ± 8.81) mg/DL for 15, 30, and 60 min PH, respectively. When the same data were applied to univariate models, the RMSE was 11.28 ( ± 3.34), 19.99 ( ± 5.59), and 33.13 ( ± 9.27) mg/DL for 15, 30, and 60 min PH, respectively. Both the univariate and multivariate models showed a statistically significant difference compared with the baseline at a 5% statistical significance level. Instead of using a model with a single algorithm, applying a GA based on each output of a model with multiple algorithms was found to play a significant role in improving model performance.

BACKGROUND:Thiazolidinediones (TZDs) have been associated with various safety concerns including weight gain, bladder cancer, and congestive heart failure (CHF). This study evaluated the efficacy and safety of lobeglitazone, a novel TZD in patients with type 2 diabetes mellitus (T2DM) in real practice.METHODS:In this non-interventional, multi-center, retrospective, and observational study conducted at 15 tertiary or secondary referral hospitals in Korea, a total of 2,228 patients with T2DM who received lobeglitazone 0.5 mg for more than 1 year were enrolled.RESULTS:Overall adverse events (AEs) occurred in 381 patients (17.10%) including edema in 1.97% (n=44). Cerebrovascular and cardiovascular diseases were identified in 0.81% (n=18) and 0.81% (n=18), respectively. One case of CHF was reported as an AE. Edema occurred in 1.97% (n=44) of patients. Hypoglycemia occurred in 2.47% (n=55) of patients. Fracture occurred in 1.17% (n=26) of all patients. Lobeglitazone significantly decreased HbA1c level, resulting in a mean treatment difference of -1.05%± 1.35% (P<0.001), and decreased total cholesterol, triglyceride, and low-density lipoprotein cholesterol. However, it increased high-density lipoprotein cholesterol, regardless of statin administration. The patients who received lobeglitazone 0.5 mg showed an apparent reduction in glycosylated hemoglobin (HbA1c) from baseline during the first 6 months of treatment. The HbA1c levels remained stable between months 6 and 42.CONCLUSION:Lobeglitazone has long-term safety profile, good glycemic-lowering effect and long-term durability of glycemic control in real-world clinical settings.

Background: We evaluated the achievement of low-density lipoprotein cholesterol (LDL-C) targets in patients with type 2 diabetes mellitus (T2DM) according to up-to-date Korean Diabetes Association (KDA), European Society of Cardiology (ESC)/European Atherosclerosis Society (EAS), and American Diabetes Association (ADA) guidelines.Methods: This retrospective cohort study collected electronic medical record data from patients with T2DM (≥20 years) managed by endocrinologists from 15 hospitals in Korea (January to December 2019). Patients were categorized according to guidelines to assess LDL-C target achievement. KDA (2019): Very High-I (atherosclerotic cardiovascular disease [ASCVD]) <70 mg/dL; Very High-II (target organ damage [TOD], or cardiovascular risk factors [CVRFs]) <70 mg/dL; high (others) <100 mg/dL. ESC/EAS (2019): Very High-I (ASCVD): <55 mg/dL; Very High-II (TOD or ≥3-CVRF) <55 mg/dL; high (diabetes ≥10 years without TOD plus any CVRF) <70 mg/dL; moderate (diabetes <10 years without CVRF) <100 mg/dL. ADA (2019): Very High-I (ASCVD); Very High-II (age ≥40+ TOD, or any CVRF), for high intensity statin or statin combined with ezetimibe.Results: Among 2,000 T2DM patients (mean age 62.6 years; male 55.9%; mean glycosylated hemoglobin 7.2%) ASCVD prevalence was 24.7%. Of 1,455 (72.8%) patients treated with statins, 73.9% received monotherapy. According to KDA guidelines, LDL-C target achievement rates were 55.2% in Very High-I and 34.9% in Very High-II patients. With ESC/EAS guidelines, target attainment rates were 26.6% in Very High-I, 15.7% in Very High-II, and 25.9% in high risk patients. Based on ADA guidelines, most patients (78.9%) were very-high risk; however, only 15.5% received high-intensity statin or combination therapy.Conclusion: According to current dyslipidemia management guidelines, LDL-C goal achievement remains suboptimal in Korean patients with T2DM.

Background The choice of an optimal oral hypoglycemic agent in the initial treatment periods for type 2 diabetes mellitus (T2DM) patients remains difficult and deliberate. We compared the efficacy and safety of glimepiride (GLIM), alogliptin (ALO), and alogliptin-pioglitazone (ALO-PIO) in poorly controlled T2DM patients with drug-naïve or metformin failure. Methods In this three-arm, multicenter, open-label, randomized, controlled trial, poorly controlled T2DM patients were randomized to receive GLIM (n=35), ALO (n=31), or ALO-PIO (n=33) therapy for 24 weeks. The primary endpoint was change in the mean glycosylated hemoglobin (HbA1c) levels at week 24 from baseline. Secondary endpoints were changes in HbA1c level at week 12 from baseline, fasting plasma glucose (FPG) levels, lipid profiles at weeks 12 and 24, and parameters of glycemic variability, assessed by continuous glucose monitoring for 24 weeks. Results At weeks 12 and 24, the ALO-PIO group showed significant reduction in HbA1c levels compared to the ALO group (–0.96%±0.17% vs. –0.37%±0.17% at week 12; –1.13%±0.19% vs. –0.18%±0.2% at week 24). The ALO-PIO therapy caused greater reduction in FPG levels and significant increase in high-density lipoprotein cholesterol levels at weeks 12 and 24 than the ALO therapy. Compared to low-dose GLIM therapy, ALO-PIO therapy showed greater improvement in glycemic variability. The adverse events were similar among the three arms. Conclusion ALO-PIO combination therapy during the early period exerts better glycemic control than ALO monotherapy and excellency in glycemic variability than low-dose sulfonylurea therapy in uncontrolled, drug-naïve or metformin failed T2DM patients.

The Committee of Clinical Practice Guidelines of the Korean Diabetes Association (KDA) updated the previous clinical practice guidelines for Korean adults with diabetes and prediabetes and published the seventh edition in May 2021. We performed a comprehensive systematic review of recent clinical trials and evidence that could be applicable in real-world practice and suitable for the Korean population. The guideline is provided for all healthcare providers including physicians, diabetes experts, and certified diabetes educators across the country who manage patients with diabetes or the individuals at the risk of developing diabetes mellitus. The recommendations for screening diabetes and glucose-lowering agents have been revised and updated. New sections for continuous glucose monitoring, insulin pump use, and non-alcoholic fatty liver disease in patients with diabetes mellitus have been added. The KDA recommends active vaccination for coronavirus disease 2019 in patients with diabetes during the pandemic. An abridgement that contains practical information for patient education and systematic management in the clinic was published separately.

BACKGROUND:Dipeptidyl peptidase-4 (DPP-4) inhibitor has been reported to have kidney-protective benefits. To elucidate how antidiabetic agents prevent diabetic kidney disease progression, it is important to investigate their effect on the kidney environment in type 2 diabetes mellitus (DM) patients. Herein, we investigated the expression pattern of urinary exosome-derived microRNA (miRNA) in patients taking a combination of DPP-4 inhibitor and metformin (DPP-4 inhibitor group) and compared them with patients taking a combination of sulfonylurea and metformin (sulfonylurea group). METHODS:This was a prospective study involving 57 patients with type 2 DM (DPP-4 inhibitor group, n = 34; sulfonylurea group, n = 23) and healthy volunteers (n = 7). We measured urinary exosomal miRNA using the NanoString nCounter miRNA array (NanoString Technologies) across the three groups (n = 4 per each group) and validated findings using real-time polymerase chain reaction. RESULTS:Twenty-one differentially expressed candidate miRNAs were identified, and six (let-7c-5p, miR-23a-3p, miR-26a-3p, miR-30d, miR-205, and miR-200a) were selected for validation. Validation showed no significant difference in miRNA expression between the DPP-4 inhibitor and sulfonylurea groups. Only miR-23a-3p was significantly overexpressed in the diabetes group compared with the control group (DPP-4 inhibitor vs. control, p = 0.01; sulfonylurea vs. control, p = 0.007). This trend was consistent even after adjusting for age, sex, and body mass index. CONCLUSION:There was no significant difference in urine exosome miRNA expression between diabetic participants taking DPP-4 inhibitor and those taking sulfonylurea. The miR-23a levels were higher in diabetic participants than in nondiabetic controls.

Background: Nonalcoholic fatty liver disease (NAFLD) is the most prevalent cause of chronic liver disease worldwide. Type 2 diabetes mellitus (T2DM) is a risk factor that accelerates NAFLD progression, leading to fibrosis and cirrhosis. Thus, here we aimed to develop a simple model to predict the presence of NAFLD based on clinical parameters of patients with T2DM.Methods: A total of 698 patients with T2DM who visited five medical centers were included. NAFLD was evaluated using transient elastography. Univariate logistic regression analyses were performed to identify potential contributors to NAFLD, followed by multivariable logistic regression analyses to create the final prediction model for NAFLD.Results: Two NAFLD prediction models were developed, with and without serum biomarker use. The non-laboratory model comprised six variables: age, sex, waist circumference, body mass index (BMI), dyslipidemia, and smoking status. For a cutoff value of ≥60, the prediction accuracy was 0.780 (95% confidence interval [CI], 0.743 to 0.817). The second comprehensive model showed an improved discrimination ability of up to 0.815 (95% CI, 0.782 to 0.847) and comprised seven variables: age, sex, waist circumference, BMI, glycated hemoglobin, triglyceride, and alanine aminotransferase to aspartate aminotransferase ratio. Our non-laboratory model showed non-inferiority in the prediction of NAFLD versus previously established models, including serum parameters.Conclusion: The new models are simple and user-friendly screening methods that can identify individuals with T2DM who are at high-risk for NAFLD. Additional studies are warranted to validate these new models as useful predictive tools for NAFLD in clinicalpractice.

Background Conventional diagnostic approaches for adrenal tumors require multi-step processes, including imaging studies and dynamic hormone tests. Therefore, this study aimed to discriminate adrenal tumors from a single blood sample based on the combination of liquid chromatography-mass spectrometry (LC-MS) and machine learning algorithms in serum profiling of adrenal steroids. Methods The LC-MS-based steroid profiling was applied to serum samples obtained from patients with nonfunctioning adenoma (NFA, n=73), Cushing’s syndrome (CS, n=30), and primary aldosteronism (PA, n=40) in a prospective multicenter study of adrenal disease. The decision tree (DT), random forest (RF), and extreme gradient boost (XGBoost) were performed to categorize the subtypes of adrenal tumors. Results The CS group showed higher serum levels of 11-deoxycortisol than the NFA group, and increased levels of tetrahydrocortisone (THE), 20α-dihydrocortisol, and 6β-hydroxycortisol were found in the PA group. However, the CS group showed lower levels of dehydroepiandrosterone (DHEA) and its sulfate derivative (DHEA-S) than both the NFA and PA groups. Patients with PA expressed higher serum 18-hydroxycortisol and DHEA but lower THE than NFA patients. The balanced accuracies of DT, RF, and XGBoost for classifying each type were 78%, 96%, and 97%, respectively. In receiver operating characteristics (ROC) analysis for CS, XGBoost, and RF showed a significantly greater diagnostic power than the DT. However, in ROC analysis for PA, only RF exhibited better diagnostic performance than DT. Conclusion The combination of LC-MS-based steroid profiling with machine learning algorithms could be a promising one-step diagnostic approach for the classification of adrenal tumor subtypes.

In this study, we propose a personalized glucose prediction model using deep learning for hospitalized patients who experience Type-2 diabetes. We aim for our model to assist the medical personnel who check the blood glucose and control the amount of insulin doses. Herein, we employed a deep learning algorithm, especially a recurrent neural network (RNN), that consists of a sequence processing layer and a classification layer for the glucose prediction. We tested a simple RNN, gated recurrent unit (GRU), and long-short term memory (LSTM) and varied the architectures to determine the one with the best performance. For that, we collected data for a week using a continuous glucose monitoring device. Type-2 inpatients are usually experiencing bad health conditions and have a high variability of glucose level. However, there are few studies on the Type-2 glucose prediction model while many studies performed on Type-1 glucose prediction. This work has a contribution in that the proposed model exhibits a comparative performance to previous works on Type-1 patients. For 20 in-hospital patients, we achieved an average root mean squared error (RMSE) of 21.5 and an Mean absolute percentage error (MAPE) of 11.1%. The GRU with a single RNN layer and two dense layers was found to be sufficient to predict the glucose level. Moreover, to build a personalized model, at most, 50% of data are required for training.

The Committee of Clinical Practice Guidelines of the Korean Diabetes Association revised and updated the 6th Clinical Practice Guidelines in 2019. Targets of glycemic, blood pressure, and lipid control in type 2 diabetes mellitus (T2DM) were updated. The obese and overweight population is increasing steadily in Korea, and half of the Koreans with diabetes are obese. Evidence-based recommendations for weight-loss therapy for obesity management as treatment for hyperglycemia in T2DM were provided. In addition, evidence from large clinical studies assessing cardiovascular outcomes following the use of sodium-glucose cotransporter-2 inhibitors and glucagon-like peptide 1 receptor agonists in patients with T2DM were incorporated into the recommendations.

Background This study was a multicenter, parallel-group, double-blind, double-dummy, randomized, noninferiority trial to evaluate the efficacy and safety of γ-linolenic acid (GLA) relative to α-lipoic acid (ALA) over a 12-week treatment period in type 2 diabetes mellitus (T2DM) patients with painful diabetic peripheral neuropathy (DPN). Methods This study included 100 T2DM patients between 20 and 75 years of age who had painful DPN and received either GLA (320 mg/day) and placebo or ALA (600 mg/day) and placebo for 12 weeks. The primary outcome measures were mean changes in pain intensities as measured by the visual analogue scale (VAS) and the total symptom scores (TSS). Results Of the 100 subjects who initially participated in the study, 73 completed the 12-week treatment period. Per-protocol analyses revealed significant decreases in the mean VAS and TSS scores compared to baseline in both groups, but there were no significant differences between the groups. The treatment difference for the VAS (95% confidence interval [CI]) between the two groups was −0.65 (−1.526 to 0.213) and the upper bound of the 95% CI did not exceed the predefined noninferiority margin (δ1=0.51). For the TSS, the treatment difference was −0.05 (−1.211 to 1.101) but the upper bound of the 95% CI crossed the noninferiority margin (δ2=0.054). There were no serious adverse events associated with the treatments. Conclusion GLA treatment in patients with painful DPN was noninferior to ALA in terms of reducing pain intensity measured by the VAS over 12 weeks.