Search

To provide effective medical nutrition therapy (MNT) , it is necessary to consider dietary intake, which can change over time. This longitudinal study analyzed changes in the dietary intakes of Japanese patients with type 2 diabetes over a 5-year period. The JDCS is a nationwide study of Japanese patients with type 2 diabetes conducted in 1996. Participants were aged 40-70 years from outpatient clinics in 59 universities and general hospitals. Analyzed were 922 responders to a dietary survey assessed by the Food Frequency Questionnaire based on food groups at baseline and 5 years after registration. Paired samples t-test was used for analysis. Over a 5-year period, patients with type 2 diabetes maintained almost the same body mass index (BMI) (baseline: 22.9±3.0; 5 years later: 22.8±3.0 kg/m2) . Differences in energy intake, fat-to-energy ratio, and carbohydrate-to-energy ratio were small although significant. Neither the mean intake of fiber nor salt met the required intake for MNT for persons with diabetes both at baseline and after 5 years. As for food groups, there was a significant decrease in intakes of vegetables, seaweed, soybeans/soy products, and milk/dairy products. Details are shown in the Table. Results showed that Japanese patients with type 2 diabetes need to pay special attention to dietary fiber intake and salt intake on an ongoing basis. Disclosure C.Horikawa: None. R.Aida: None. C.Kamada: None. K.Fujihara: None. S.Tanaka: None. A.Araki: None. Y.Yoshimura: None. Y.Akanuma: None. H.Sone: Research Support; Astellas Pharma Inc., Eisai Co., Ltd., Kyowa Kirin Co., Ltd., Novo Nordisk, Ono Pharmaceutical Co., Ltd., Taisho Pharmaceutical Holdings Co., Ltd., Takeda Pharmaceutical Company Limited. Funding Japan Society for the Promotion of Science (19H04028)

Elevated serum uric acid reportedly increases the risk of development and progression of diabetic retinopathy with conflicting results. However, whether it also increases the incidence of severe vision-threatening diabetic eye diseases has been scarcely investigated. Also unknown is whether there are sex differences in the incidence of advanced retinopathies. To clarify this, we utilized ICD-and medical treatment codes from a nationwide claims database that finally included 13,721 type 2 diabetes patients (mean age 50 y, HbA1c 7.0%, mean follow-up period 5.8 y, 10,962 men) . During the study period, 388 patients developed vision-threatening treatment-required diabetic eye diseases (4.9/1000 person-years) . Multivariate Cox analysis demonstrated that a serum uric acid ≥7.1 mg/dl was significantly positively associated in men with future occurrence of treatment-required diabetic eye diseases, but not in women. Hazard ratios for those with serum uric acid ≥7.1 mg/dl for incident treatment-required diabetic eye diseases were 1.36 (95% confidence interval, 1.01-1.85) in men and 2. (0.71-6.08) in women, compared with the reference group ( Disclosure M. Yamamoto: None. K. Fujihara: None. H. Ishiguro: None. M. Ishizawa: None. M.H. Yamada: None. T. Osawa: None. M. Kitazawa: None. M. Kaneko: None. M. Iwanaga: None. T. Yamada: None. S. Kodama: None. H. Sone: Research Support; Astellas Pharma Inc., Eisai Co., Ltd., Kyowa Kirin Co., Ltd., Novo Nordisk, Ono Pharmaceutical Co., Ltd., Taisho Pharmaceutical Holdings Co., Ltd., Takeda Pharmaceutical Company Limited. Funding The Japan Society for the Promotion of Science and the Ministry of Health, Labor and Welfare, Japan. (18K17897)

In 2016, to help achieve the goal of avoiding heart failure and preserving cardiac structure and function, the American Heart Association (AHA) designated Life’s Simple 7 as risk factors that can be addressed through lifestyle changes: smoking, weight, physical activity, diet, cholesterol, blood pressure (BP) , and blood glucose. Although most of these factors are also risk factors for the progression of kidney disease, the association between Life’s Simple 7 and risk of dialysis is unknown. We investigated the impact of 6 factors on the initiation of dialysis based on the AHA’s 7 factors using a nationwide claims database involving 294,647 participants during 2008-16. Multivariate Cox regression model identified risks of starting dialysis. Hazard ratios (HRs) were compared among 7 groups who had 0 to 6 of the following risk factors adjusted by age and sex: BMI ≥25, systolic BP (SBP) ≥140 mmHg, LDL-cholesterol ≥140 mg/dl, diabetes mellitus (DM) , no regular exercise, and smoking. Compared with SBP In summary, the risk of dialysis increased with worsening of risk factors derived from AHA’s Life’s Simple 7. These results indicated that factors similar to those used to predict cardiovascular disease might also be useful to predict initiation of dialysis. Use of these approaches might be helpful in clinical practice, especially in patient education. Disclosure T.Osawa: None. K.Fujihara: None. M.H.Yamada: None. Y.Yaguchi: None. T.Sato: None. M.Kitazawa: None. Y.Matsubayashi: None. T.Yamada: None. H.Sone: Research Support; Astellas Pharma Inc., Eisai Co., Ltd., Kyowa Kirin Co., Ltd., Novo Nordisk, Ono Pharmaceutical Co., Ltd., Taisho Pharmaceutical Holdings Co., Ltd., Takeda Pharmaceutical Company Limited. Funding Japan Society for the Promotion of Science (18K17897) .

Although intake of both vegetables and fruits is known to be negatively associated with obesity in people with and without diabetes, the combined effects of these two food groups including cut-off values for intake have yet to be determined. We investigated the association according to combinations of the amount of fruit and vegetable intake determined by a food frequency questionnaire among 1579 patients with type 2 diabetes (990 men, mean age 50y) registered in the Japan Diabetes Clinical Data Management Study. Logistic regression analysis was used to calculate multi-adjusted odds ratios (ORs) for obesity (BMI ≥25) . Intake of fruits and vegetables was analyzed according to quintiles. Patients in the top quintile of fruit intake (≥150g) and the top three quintiles of vegetable intake (>200g) were significantly less obese than those in the bottom quintiles. In patients above/ below the thresholds for intake of fruits and vegetables, ORs for obesity for daily fruit intake of ≥150g and for daily vegetable intake >200g vs. lower amounts significantly decreased to 0.64 (95%CI: 0.50-0.82) and 0.55 (0.44-0.68) , respectively. In addition, ORs for 4 groups of combinations of intake of vegetables and fruits above and below the threshold were calculated. The ORs were significantly reduced regardless of the amount of fruit intake if vegetable intake was >200g. ORs further decreased significantly when fruit intake was ≥150g and that of vegetables was >200g. However, when vegetable intake was 200g of vegetables per day is essential to avoid obesity regardless of the amount of fruit consumed. However, an additional effect on body weight could be expected with simultaneous fruit intake ≥150g. Disclosure M.Takeuchi: None. K.Fujihara: None. H.Sone: Research Support; Astellas Pharma Inc., Eisai Co., Ltd., Kyowa Kirin Co., Ltd., Novo Nordisk, Ono Pharmaceutical Co., Ltd., Taisho Pharmaceutical Holdings Co., Ltd., Takeda Pharmaceutical Company Limited. M.Hatta: None. C.Horikawa: None. Y.Takeda: None. N.Kato: n/a. M.Kato: None. H.Yokoyama: None. T.Saito: None. H.Maegawa: None.

Although we showed a significant association of poor adherence to OHAs with start of dialysis, this relationship according to the estimated glomerular filtration rate (eGFR) was not clarified. Patients with eGFR Disclosure Y.Yaguchi: None. H.Sone: Research Support; Astellas Pharma Inc., Eisai Co., Ltd., Kyowa Kirin Co., Ltd., Novo Nordisk, Ono Pharmaceutical Co., Ltd., Taisho Pharmaceutical Holdings Co., Ltd., Takeda Pharmaceutical Company Limited. K.Fujihara: None. M.H.Yamada: None. Y.Matsubayashi: None. T.Yamada: None. M.Iwanaga: None. M.Kitazawa: None. M.Yamamoto: None. S.Kodama: None.

Muscle strength, balance, and flexibility has been associated with the development of type 2 diabetes and atherosclerosis, but little is known whether the physical score (PS) , which integrates these physical fitness indices, can predict future metabolic diseases. We investigated the longitudinal relationship between baseline and changes in PS and future prediabetes and MetS. Analyzed were 5620 men aged 30 to 69 y without MetS who underwent physical fitness tests. Principal component analysis was performed on the correlation matrix of the physical fitness test results according to age. PS was defined as the first principal component score. Associations between PS values at the start of observation (year 0) , change in the PS one year later (year 1) and the onset of prediabetes and MetS at the end of observation (years 4 or 5) were examined by Cox hazard regression analysis. The same analysis was performed for the 3812 men without prediabetes at baseline. No significant difference was found for the development of prediabetes, but a significant difference was found for development of MetS. The hazard ratio for the development of MetS was 1.30 (1.18,1.44) for every 1 decrease in the PS in year 0 and 1.35 (1.13,1.60) for every 1 decrease in the change in PS from year 0 to year 1. PS was a simple and non-invasive indicator to predict MetS. Disclosure T. Sato: None. K. Fujihara: None. M. Yamamoto: None. M. Kitazawa: None. H. Ishiguro: None. T. Osawa: None. H. Sone: Research Support; Astellas Pharma Inc., Eisai Co., Ltd., Kyowa Kirin Co., Ltd., Novo Nordisk, Ono Pharmaceutical Co., Ltd., Taisho Pharmaceutical Holdings Co., Ltd., Takeda Pharmaceutical Company Limited.

This study aims to examine the impact of the association between metabolic syndrome (MetS) and metabolic dysfunction-associated fatty liver disease (MAFLD) , which both have insulin resistance and are pathophysiologically similar, on the development of new-onset cardiovascular disease, with and without type 2 diabetes mellitus (T2DM) A total of 570,426 participants without a history of cardiovascular disease who were enrolled in the nationwide claims database from 2008-2016 were included in this study. Participants were classified according to the presence or absence of MetS and/or MAFLD, and the risk of developing coronary artery disease (CAD) and/or cerebrovascular disease (CVD) in each category was analyzed using a multivariate Cox proportional hazard model. The same analysis was performed with stratification by the presence or absence of T2DM. During a median follow-up of 5.2 years, 2,252 CAD and 3,128 CVD events occurred. The hazard ratio (HR) (95% CI) for CAD, compared with the group with neither MAFLD nor MetS, was 1.46 (1.32-1.62) for MAFLD only (without MetS) , 2.02 (1.51-2.70) for MetS only (without MAFLD) , and 2.33 (2.10-2. 58) for MAFLD + MetS. For CVD, The HR was 1.41 (1.28-1.55) for MAFLD only, 1.67 (1.38-2.02) for MetS only, and 1.89 (1.72-2.08) for MAFLD + MetS. These results were similar for non-T2DM participants but different for T2DM participants, with no significant increased risk for CAD in MetS only and a significant increased risk for CVD only when both MAFLD and MetS coexisted. The predictive ability for cardiovascular disease development in patients with MetS and MAFLD differed depending on the presence or absence of T2DM. These results suggest that distinguishing between the diagnosis of MetS and MAFLD, taking into account the presence or absence of T2DM, may lead to more accurate identification of patients at high risk of developing cardiovascular disease. Disclosure K. Fujihara: None. M.H. Yamada: None. M. Yamamoto: None. H. Ishiguro: None. M. Kitazawa: None. M. Iwanaga: None. S. Kodama: None. H. Sone: Research Support; Astellas Pharma Inc., Eisai Co., Ltd., Kyowa Kirin Co., Ltd., Novo Nordisk, Ono Pharmaceutical Co., Ltd., Taisho Pharmaceutical Holdings Co., Ltd., Takeda Pharmaceutical Company Limited. Funding The Japan Society for Promotion of Science (JSPS) KAKENHI Grant Number JP 20K19706)

Since advanced microangiopathy of diabetes severely impairs activities of daily living (ADL) and quality of life (QoL) and macroangiopathy is life-threatening regardless of severity, treatment of diabetes requires appropriate risk management to avoid all such complications. However, current studies have analyzed and reported diabetes complications separately, despite varying incident rates and impacts on ADL/QoL and severity of illness. It is difficult to determine which predictors should be prioritized when extrapolating the results of these studies to the comprehensive care of individual patients. We defined the combined severe diabetic complications as a new life-altering outcome and determined the association of established risk factors with combined severe diabetic complications. We analyzed data from a nationwide database involving 36,994 participants with diabetes mellitus without any prior life-altering complication of diabetes during 2008-19. Combined severe diabetic complications included severe diabetic eye disease, initiating dialysis, coronary artery disease (CAD) , cerebrovascular disease (CVD) , heart failure (HF) , or amputation which were determined according to ICD-codes, medication and medical procedures. During a median follow-up period of 5.0 years, the following events occurred: severe diabetic eye disease, 774; dialysis initiated, 134: CAD, 703; CVD, 535; HF, 161; and amputations, 31. Multivariate Cox regression model showed the HRs per 1SD for combined severe diabetic complications of HbA1c, SBP, current smoking, HDL-C, BMI and LDL- C were 1.54 (1.49-1.59) , 1.37 (1.31-1.42) , 1.35 (1.23-1.48) , 0.83 (0.79-0.88) , 0.85 (0.81-0.90) and 1. (1.06-1.16) , respectively. We found that HbA1c was the most strongly associated with combined severe diabetic complications among established risk factors. Also, our results implied the necessity of revisiting the value of BMI in risk management for combined severe diabetic complications. Disclosure Y.Yaguchi: None. H.Sone: Research Support; Astellas Pharma Inc., Eisai Co., Ltd., Kyowa Kirin Co., Ltd., Novo Nordisk, Ono Pharmaceutical Co., Ltd., Taisho Pharmaceutical Holdings Co., Ltd., Takeda Pharmaceutical Company Limited. K.Fujihara: None. M.H.Yamada: None. Y.Matsubayashi: None. T.Yamada: None. M.Iwanaga: None. M.Kitazawa: None. M.Yamamoto: None. S.Kodama: None.

The association between screen time (ST) , including that for smartphones, and overweight/obesity (OW/OB) in children was examined separately for boys and girls, considering the influence of lifestyle factors including diet, physical activity, and sleep time. A cross-sectional study was conducted in 2242 Japanese children (1278 girls) aged 10-14 years. OW/OB was defined by the International Obesity Task Force. Logistic regression analysis showed that only for girls, total ST (≥4h) , smartphone ST (≥3h) and non-smartphone ST (≥2h) were all independently and significantly associated with overweight/obesity compared to In Japanese girls, smartphone ST, non-smartphone ST, and total ST were all significantly associated with OW/OB. To avoid OW/OB, it is suggested to keep smartphone ST, non-smartphone ST, and total ST to Disclosure K. Fujihara: None. S.Y. Morikawa: None. H. Ishiguro: None. M.H. Yamada: None. C. Horikawa: None. Y. Ogawa: None. H. Sone: Research Support; Astellas Pharma Inc., Eisai Co., Ltd., Kyowa Kirin Co., Ltd., Novo Nordisk, Ono Pharmaceutical Co., Ltd., Taisho Pharmaceutical Holdings Co., Ltd., Takeda Pharmaceutical Company Limited.

Although a linear relationship between SBP and risk of coronary artery disease or cerebrovascular disease was observed regardless of GTS, this association in HF is little known. We analyzed data from a nationwide claims database from 20to 20on 589,621 individuals. A Cox proportional hazards model identified risks of HF among 5 levels of SBP according to GTS [i.e. normal glucose tolerance (NGT) , prediabetes, diabetes (DM) ]. Incidence of HF per 1,000 person-years was 0.for NGT, 0.18 for prediabetes, and 0.80 for DM. Compared with SBP Disclosure K.Fujihara: None. M.H.Yamada: None. M.Yamamoto: None. M.Oe: None. T.Osawa: None. M.Iwanaga: None. T.Yamada: None. S.Kodama: None. H.Sone: Research Support; Astellas Pharma Inc., Eisai Co., Ltd., Kyowa Kirin Co., Ltd., Novo Nordisk, Ono Pharmaceutical Co., Ltd., Taisho Pharmaceutical Holdings Co., Ltd., Takeda Pharmaceutical Company Limited. Funding JSPS (21K11569)

In adults, short sleep duration and late bedtime were associated with hyperglycemia and obesity, but these relationships in children are little known. We evaluated the association of bedtime and sleep duration with overweight/obesity or hyperglycemia in children by sex, considering the effects of diet, physical activity, and screen time. A cross-sectional study was conducted for 1139 Japanese children (553 boys) aged 10-y. Overweight was as defined by the International Obesity Task Force. Cutoff value for hyperglycemia was HbA1c ≥5.6%. Variables associated with hyperglycemia or overweight were based on logistic regression analysis. Multivariate logistic regression analysis showed that compared with an early bedtime (˜21:59) and long sleep duration (weekdays ≥8.67 h, weekends ≥9 h) , short sleep duration (weekdays Disclosure H. Shiozaki: None. K. Fujihara: None. M.H. Yamada: None. T. Osawa: None. M. Kitazawa: None. M. Yamamoto: None. T. Sato: None. T. Yamada: None. H. Sone: Research Support; Astellas Pharma Inc., Eisai Co., Ltd., Kyowa Kirin Co., Ltd., Novo Nordisk, Ono Pharmaceutical Co., Ltd., Taisho Pharmaceutical Holdings Co., Ltd., Takeda Pharmaceutical Company Limited. Funding Patient Centered Outcomes Research Institute

Insulin and its secretagogues are essential for some patients with type 2 diabetes (T2D) to maintain glycemic control (GC) . However, these drugs are inevitably accompanied by hypoglycemia. Avoiding severe hypoglycemia (SH) is essential considering its poor prognosis. This network meta-analysis aimed to find optimal hypoglycemic treatment regimens for T2D in terms of GC and SH. MEDLINE and EMBASE were used to identify trials comparing 2 or more treatments including insulins and/or SU/glinide and included both GC and SH in study outcomes. Treatment hierarchy was summarized as the surface under cumulative ranking probabilities (SUCRA) . There were 137 eligible trials from which 36 treatments were identified. Insulin plus (+) non-insulin drugs except for SU/glinide (others) had a higher SUCRA for changes in hemoglobin A1c (A1C) compared with only insulins (68.2±5.6% vs. 39.9±6.6%; P=0.01) although the difference in the SUCRA for SH was not significant (P=0.59) . Compared with only insulins, SU/glinide + others had a lower SUCRA for SH (42.4±5.8% vs. 58.5±3.4%, P=0.05) and comparable SUCRA for A1C change (P=1.00) . Cluster analysis indicated that premixed insulin + glucagon-like peptide-1 receptor agonist (Mix-ins+GLP1) consistently belonged to the high-efficacy group for A1C change and achievement of A1C Disclosure S.Kodama: None. K.Kato: None. K.Watanabe: None. H.Sone: Research Support; Astellas Pharma Inc., Eisai Co., Ltd., Kyowa Kirin Co., Ltd., Novo Nordisk, Ono Pharmaceutical Co., Ltd., Taisho Pharmaceutical Holdings Co., Ltd., Takeda Pharmaceutical Company Limited. T.Sato: None. M.H.Yamada: None. M.Yamamoto: None. Y.Matsubayashi: None. H.Ishiguro: None. M.Iwanaga: None. K.Fujihara: None. T.Yamada: None.

Breslow’s Health Practice Index (HPI) includes recommendations for a healthy lifestyle and is composed of 7 simple lifestyle factors (sleep, breakfast, snacks, body mass index (BMI) , physical activity, drinking alcohol, and smoking status) . Cardiovascular health metrics (CVHMs) defined according to the American Heart Association are composed of lifestyle factors and clinical laboratory values (smoking status, BMI, physical activity, diet, systolic blood pressure In summary, not only to follow a healthy lifestyle but to achieve laboratory treatment targets may be essential in people with diabetes. Disclosure T.Iijima: None. T.Yamada: None. S.Kodama: None. H.Sone: Research Support; Astellas Pharma Inc., Eisai Co., Ltd., Kyowa Kirin Co., Ltd., Novo Nordisk, Ono Pharmaceutical Co., Ltd., Taisho Pharmaceutical Holdings Co., Ltd., Takeda Pharmaceutical Company Limited. K.Fujihara: None. M.H.Yamada: None. M.Yamamoto: None. T.Osawa: None. Y.Matsubayashi: None. T.Sato: None. Y.Yaguchi: None. M.Iwanaga: None.

Although UP is a very important feature of diabetic nephropathy, decreased eGFR without UP has become phenotypic of renal complications of DM in the new definition of DKD. Although both UP and reduced eGFR are strong predictors of kidney disease, only a few studies have examined the association between these two variables and risk of dialysis in the same cohort that included people with and without DM. We analyzed data from a nationwide claims database involving 335,778 participants (DM 20,223, non-DM 315,555) during 2008-16. Multivariate Cox regression model identified risks of starting dialysis. HRs were compared among 8 groups of combinations of DM+/-, UP (-, ≥+) and eGFR ( In summary, both eGFR Disclosure T. Osawa: None. K. Fujihara: None. M. H. Yamada: None. Y. Yaguchi: None. T. Sato: None. M. Kitazawa: None. Y. Matsubayashi: None. T. Yamada: None. H. Sone: Research Support; Self; Astellas Pharma Inc., Eisai Co., Ltd., Kyowa Kirin Co., Ltd., Novo Nordisk, Ono Pharmaceutical Co., Ltd., Taisho Pharmaceutical Co., Ltd., Takeda Pharmaceutical Co. Funding Japan Society for the Promotion of Science (18K17897)

Though excessive sodium intake is regarded as a crucial factor related to developing and worsening of diabetes complications, data on this topic in elderly patients with diabetes are sparse. We investigated the associations between sodium intake and diabetes complications in a nationwide cohort of elderly Japanese patients with type 2 diabetes aged 65 - 85 years. The present analysis was conducted among Japanese patients with type 2 diabetes from outpatient clinics in 39 university and general hospitals in Japan. Data were analyzed on 912 individuals whose dietary intake at baseline was assessed by the Food Frequency Questionnaire based on food groups. Primary outcomes were times to diabetic retinopathy, overt nephropathy, cardiovascular disease, and all-cause mortality during 6 years. Cox regression analyses estimated hazard ratios (HRs) for dietary intake adjusted for age, gender, body mass index, HbA1c, smoking, energy intake, and other confounders Mean sodium intake in quartiles ranged from 2498 mg to 5910 mg/day and more than three-quarters of participants had sodium intake more than the recommended dietary allowance by American Diabetes Association (< 2300 mg/day). After adjustment for confounders, HRs for complications in patients with sodium intake in the 2nd, 3rd, and 4th quartiles compared with sodium intake in the 1st quartile were analyzed, and no significant differences were shown between sodium intake and complications. Among patients whose vegetable intake was less than the average of 268.7 g, HRs for diabetic retinopathy in patients in the 2nd, 3rd, and 4th quartiles of sodium intake compared with the 1st quartile were 0.87 (95% CI, 0.31-2.41), 2.61 (1.00-6.83), and 3.70 (1.37-10.02), respectively. Findings indicated that high sodium intake under conditions of low vegetable intake is associated with elevated incidence of diabetic retinopathy in elderly patients with type 2 diabetes. Disclosure C. Horikawa: None. T. Moriya: None. H. Sone: Research Support; Self; Astellas Pharma Inc., Eisai Co., Ltd., Kyowa Kirin Co., Ltd., Novo Nordisk, Ono Pharmaceutical Co., Ltd., Taisho Pharmaceutical Co., Ltd., Takeda Pharmaceutical Co. A. Araki: None. R. Aida: None. S. Tanaka: None. C. Kamada: None. S. Tanaka: None. Y. Yoshimura: None. R. Kodera: None. K. Fujihara: None. R. Kawasaki: None. Funding Japan Society for the Promotion of Science (19H04028)

Although a history of both CVD and glucose abnormality are risk factors for CVD, few large studies have examined the association between prior CVD and glucose abnormality on subsequent CVD in the same cohort. Thus, we compared the impact of prior CVD, glucose abnormality status and their combinations on subsequent CVD using real-world data. Men aged 18-72 years and followed for ≥3 years between 2008 and 2016 were classified as normoglycemia (n = 210,434), borderline glycemia (n = 119,933) or diabetes (DM) (n = 33,260) defined by fasting plasma glucose, HbA1c, and antidiabetic drug prescription. Prior and subsequent CVD were identified according to claims using ICD-10 codes, medical procedures and questionnaires. Cox regression showed that CVD+ conferred a 5- to 8-fold excess risk for CVD regardless of glucose abnormality status (Table). Borderline glycemia did not influence risk of CVD in both prior CVD+ and CVD- status, whereas DM affected subsequent CVD in CVD-. In CVD-/DM, age, current smoking, SBP, HDL-C and HbA1c were associated with risk of CVD, but only SBP was related to CVD risk in CVD+/DM. Impacts of glucose tolerance and glycemic control on risk of CVD were much less than prior CVD. SBP was more strongly associated with CVD risk than glycemia in DM with prior CVD. Individualized treatment strategies should consider glucose tolerance status and prior CVD. Disclosure M. Oe: Employee; Self; Kowa Company, Ltd. T. Yamada: None. H. Sone: Research Support; Self; Astellas Pharma Inc., Eisai Co., Ltd., Kyowa Kirin Co., Ltd., Novo Nordisk, Ono Pharmaceutical Co., Ltd., Taisho Pharmaceutical Co., Ltd., Takeda Pharmaceutical Co. K. Fujihara: None. M. H. Yamada: None. T. Osawa: None. M. Kitazawa: None. Y. Matsubayashi: None. T. Sato: None. Y. Yaguchi: None. M. Iwanaga: None.

We examined the ability of machine learning models to predict insulin initiation by specialists and examined whether the machine learning approach could support decision-making by general physicians for insulin initiation in patients with type 2 diabetes. Prescription data from diabetes specialists’ patient registries on antidiabetic agents from 2009 to 2015 were used for 4,860 patients who received initial monotherapy with either insulin (293 patients) or non-insulin (4,567 patients) and had laboratory data. Neural network output was insulin initiation ranging from 0 to 1 with a cutoff >0.5 for the dichotomous classification. Accuracy, recall and area under the receiver operation characteristic curves (AUCs) were calculated to compare the ability of machine-learning decisions on insulin initiation to decisions by logistic regression and 22 general physicians. Then, in comparing the ability of machine-learning and logistic regression decisions to decisions made by general physicians, the gold standard was defined as 80% agreement among 9 diabetes specialists based on patients’ information. AUCs and accuracy according to logistic regression were higher than those of machine learning (AUC of logistic regression, 0.89-0.90; AUC of machine learning, 0.67-0.74). Recall (accuracy of the cases receiving insulin) in machine learning was similar to that of logistic regression analysis (recall of logistic regression, 0.05-0.68; recall of machine learning, 0.11-0.52). The accuracy of logistic regression and the machine learning model was 1.00 and 0.86, respectively, for down sampling ratios of 1:8, which was higher than that of general physicians (0.43) for 7 gold standard cases. In conclusion, machine learning had higher accuracy in prediction of insulin initiation compared with that of general physicians defined by diabetes specialists’ choice of the gold standard. Assistance by machine learning may be beneficial to any general physician in deciding upon insulin initiation. Disclosure K. Fujihara: None. H. Sone: Research Support; Self; Astellas Pharma Inc., Eisai Co., Ltd., Kyowa Kirin Co., Ltd., Novo Nordisk, Ono Pharmaceutical Co., Ltd., Taisho Pharmaceutical Co., Ltd., Takeda Pharmaceutical Co. Funding Japan Society for the Promotion of Science (19H04028)

Excessive sodium intake is regarded as a crucial factor related to developing and worsening of cardiovascular disease (CVD). However, epidemiological studies on this topic in patients with type 2 diabetes are sparse. We investigated the associations between sodium intake and CVD risk in Japanese patients with type 2 diabetes from the analysis of pooled data from two nation-wide prospective longitudinal studies. Data were analyzed on 2428 individuals whose dietary intake at baseline was assessed by the Food Frequency Questionnaire based on food groups. Primary outcome was the 6-year risk of a CVD event. Cox regression analyses estimated hazard ratios (HRs) for dietary intake adjusted for age, gender, body mass index, HbA1c, smoking, energy intake, and other confounders. Mean sodium intake in quartiles ranged from 2494 mg to 6166 mg/day and more than three-quarters of participants had sodium intake more than the recommended dietary allowance by American Diabetes Association (< 2300 mg/day). After adjustment for confounders, HRs for CVD in patients with sodium intake in the 2nd, 3rd, and 4th quartiles compared with sodium intake in the 1st quartile were analyzed, and no significant differences were shown between sodium intake and CVD [Q2: 0.77 (95%CI, 0.48-1.23); Q3: 1.03 (0.65-1.63); and Q4: 1.31 (0.80-2.14), trend p = 0.131]. Results of subgroup analysis according to age showed that patients aged 39 through 59 years in the higher quartiles of sodium intake compared with the 1st quartile had a significantly higher incidence of CVD [Q2: 1.37 (0.57-3.33); Q3: 1.68 (0.70-4.03); and Q4: 4.16 (1.63-10.58), trend p = 0.004]. However, no significant association between sodium intake and CVD in participants aged over 60 years. Findings indicated that high sodium intake is associated with elevated incidence of CVD in younger patients with type 2 diabetes. Disclosure C. Horikawa: None. H. Sone: Research Support; Self; Astellas Pharma Inc., Eisai Co., Ltd., Kyowa Kirin Co., Ltd., Novo Nordisk, Ono Pharmaceutical Co., Ltd., Taisho Pharmaceutical Co., Ltd., Takeda Pharmaceutical Co. R. Aida: None. S. Tanaka: None. C. Kamada: None. Y. Yoshimura: None. R. Kodera: None. K. Fujihara: None. T. Moriya: None. A. Araki: None. Funding Japan Society for the Promotion of Science (19H04028)

We investigated the impact of medication adherence for a one-year period and subsequent glycemic control and their combinations on the risk of micro/macrovascular diseases among Japanese using nationwide claims data. Analyzed were 19,565 patients with diabetes whose medication records for oral hypoglycemic agents (OHAs) were available for at least 1 year without prior treatment-requiring diabetic eye disease (TRDED), initiation of dialysis (dialysis), coronary artery disease (CAD), or cerebrovascular disease (CVD), respectively (mean age 53y, HbA1c 7.2%). Medication adherence was evaluated by the proportion of days covered (PDC), and PDC Disclosure Y. Yaguchi: None. H. Sone: Research Support; Self; Astellas Pharma Inc., Eisai Co., Ltd., Kyowa Kirin Co., Ltd., Novo Nordisk, Ono Pharmaceutical Co., Ltd., Taisho Pharmaceutical Co., Ltd., Takeda Pharmaceutical Co. K. Fujihara: None. M. H. Yamada: None. Y. Matsubayashi: None. T. Yamada: None. M. Iwanaga: None. M. Kitazawa: None. M. Yamamoto: None. S. Kodama: None.

Heavy total alcohol consumption was reported to increase the risk of development and progression of diabetic retinopathy. However, differing from total alcohol consumption, whether the quantity of alcohol intake per occasion had harmful or beneficial effects related to severe eye diseases caused by diabetes remains unknown. To clarify this, we utilized ICD-10 and medical treatment codes from a nationwide claims database that finally included 21,392 men with type 2 diabetes (mean age 53 y, HbA1c 7.0%, mean follow-up period 4.3 y). During the study period, 425 patients developed vision-threatening treatment-required diabetic eye diseases (TRDED) (4.6/1000 person-years). Multivariate Cox analysis demonstrated that a large quantity of alcohol per occasion defined as ≥3 drinks per occasion was positively associated with future occurrence of TRDED. One drink was defined as a standard Japanese drink with a volume equivalent to 23 g ethanol. Hazard ratio in patients drinking large amounts of alcohol per occasion for incident TRDED was 4.49 (95% confidence interval, 1.98-10.19) compared with the reference group (≤1 drink per occasion) after adjusting for covariates including total alcohol consumption. This finding demonstrated that a large quantity of alcohol per occasion had a significant association with future vision-threatening retinopathy in men with type 2 diabetes. The mechanism of this finding with regard to severe retinopathy awaits further research. Disclosure M. Yamamoto: None. T. Yamada: None. S. Kodama: None. H. Sone: Research Support; Self; Astellas Pharma Inc., Eisai Co., Ltd., Kyowa Kirin Co., Ltd., Novo Nordisk, Ono Pharmaceutical Co., Ltd., Taisho Pharmaceutical Co., Ltd., Takeda Pharmaceutical Co. K. Fujihara: None. Y. Yaguchi: None. T. Osawa: None. M. Kaneko: None. M. Kitazawa: None. T. Sato: None. M. H. Yamada: None. Y. Matsubayashi: None.

Although both hyperglycemia and HT are highly predictive of kidney disease, only a few studies have investigated the associations between the severity of HT and risk of end-stage renal disease especially the initiation of renal replacement therapy in DM+ or DM- in the same cohort at the same time. Thus, we analyzed data from a nationwide claim database involving 258,874 participants during 2008-16. Multivariate Cox regression model was used to identify risks of starting dialysis. HT as a covariate was determined according to systolic blood pressure (SBP)(mmHg) diagnosed by seven different cut-offs (≥110, ≥115, ≥120, ≥125, ≥130, ≥140, ≥150). Hazard ratios (HRs) were compared among 10 groups divided according to combinations of DM+ or DM- and five levels of SBP (mmHg) (≤119, 120-129, 130-139, 140-149, ≥150). Among DM-, baseline age, percent of men, BMI, smoking rate, SBP, and percent of medication for HT were higher in those who started dialysis. In those with DM+, baseline percent of men, SBP, and percent of medication for DM and HT were higher in those who started dialysis. HT was significantly associated with the initiation of dialysis regardless of the presence or absence of DM. The incidence of starting dialysis in those with SBP ≤119 mmHg and DM+ was almost the same as in those with SBP ≥150 mmHg and DM-. In comparison with SBP ≤119 mmHg, SBP ≥150 mmHg significantly increased the risk of the initiation of dialysis about 2.5 times regardless of DM+ or DM-. Compared with DM- and SBP ≤119mmHg, the HR for DM+ and SBP ≥150 mmHg was 6.88 (95% CI 3.66-12.9). Although the risks of HT differed only slightly regardless of the presence or absence of DM, risks for the initiation of dialysis with DM+ and SBP ≤119 mmHg were equivalent to DM- and SBP ≥150 mmHg, indicating stricter blood pressure interventions in DM+ are needed to avoid dialysis. Future studies are needed to clarify the cut-off SBP level to avoid initiation of dialysis considering the risks of strict control of blood pressure. Disclosure T. Osawa: None. K. Fujihara: None. M.H. Yamada: None. T. Sato: None. M. Kitazawa: None. Y. Yaguchi: None. Y. Matsubayashi: None. M. Iwanaga: None. N. Yamanaka: None. H. Seida: None. H. Sone: Research Support; Self; Kyowa Hakko Kirin Co., Ltd., Novartis AG, Ono Pharmaceutical Co., Ltd., Taisho Pharmaceutical Co., Ltd., Takeda Pharmaceutical Co. Funding Japan Society for the Promotion of Science

Reduced plasma volume (PV) to decrease ventricular filling pressure and increased beta-hydroxybutyrate (BHB) as a substitute for the energy source by SGLT2 inhibitors (SGLT2is) might contribute to preserving the heart in type 2 diabetes mellitus (T2DM). However, the impact of SGLT2is on PV and BHB is little known according to cardiac workload. We investigated the effect of long-term treatment with the SGLT2i, tofogliflozin, on PV and BHB according to baseline brain natriuretic peptide (BNP) levels in T2DM. Analyzed were 774 T2DM patients who received tofogliflozin for 52 weeks as mono- and add-on-therapy in two phase-3 studies. Percent change in estimated PV (%ΔePV) was calculated by the Strauss formula: {100*(pre Hb/post Hb)*[(100 - post Ht)/(100 - pre Ht)] - 100}. We divided patients into two groups (Quartiles 1-3[Q1-3] and Quartile 4[Q4]) according to baseline BNP. Differences in characteristics were analyzed using unpaired t-test, wilcoxon rank sum test or Fisher’s exact test. Differences in changes between groups were analyzed using an analysis of covariance model. Multivariate analysis evaluated the association of baseline BNP with changed variables. Baseline characteristics were: male (71 vs. 52% *, * p Disclosure A. Yoshida: Employee; Self; Kowa Company, Ltd. Y. Matsubayashi: None. T. Nojima: Employee; Self; Kowa Company, Ltd. H. Suganami: Employee; Self; Kowa Company, Ltd. M. Oe: Employee; Self; Kowa Company, Ltd. K. Fujihara: None. S. Tanaka: None. K. Kaku: Advisory Panel; Self; Nippon Boehringer Ingelheim Co. Ltd., Novo Nordisk Inc. Consultant; Self; Sanwa Kagaku Kenkyusho. Research Support; Self; Taisho Pharmaceutical Co., Ltd. Speaker’s Bureau; Self; Astellas Pharma Inc., AstraZeneca, Daiichi Sankyo, Eli Lilly Japan K.K., Kowa Company, Ltd., Merck Sharp & Dohme Corp., Mitsubishi Tanabe Pharma Corporation, Ono Pharmaceutical Co., Ltd., Takeda Pharmaceutical Company Limited. H. Sone: Research Support; Self; Kyowa Hakko Kirin Co., Ltd., Novartis AG, Ono Pharmaceutical Co., Ltd., Taisho Pharmaceutical Co., Ltd., Takeda Pharmaceutical Co.

SGLT2 inhibitors (SGLT2is) promote diuresis and reduce plasma volume (PV). However, the impact of long-term administration and subsequent withdrawal of SGLT2is on PV is little known. This study investigated the effect of long-term administration and then withdrawal of the SGLT2i, tofogliflozin (TOFO), on PV and explored correlations with variables in type 2 diabetes mellitus (T2DM). Analyzed were 166 T2DM patients who received TOFO for 52 weeks as initial monotherapy and underwent 2 weeks of withdrawal (week 54) in a phase-3 study. Percent change in estimated PV (%ΔePV) was calculated by the Strauss formula: {100*(pre Hb/post Hb)*[(100 - pre Ht)/(100 - post Ht)] - 100}. Baseline characteristics were: male (66%), age (mean: 58 y), HbA1c (8%), BMI (26kg/m2), BNP (12pg/mL) and eGFR (84mL/min/1.73m2). Paired t-tests analyzed differences from baseline to week 52 and from week 52 to 54. Correlation analysis was performed by Pearson’s product-moment correlation coefficients. Multivariate analysis explored predictors for %ΔePV. Body weight (BW) [mean, -3kg (-5%)*; * p After long-term TOFO therapy, reduced PV might be correlated not with weight loss but with reduced BNP indicating fluid loss and baseline BNP might predict changed PV. Weight gain after TOFO withdrawal might be caused by fluid gain. Disclosure Y. Matsubayashi: None. A. Yoshida: Employee; Self; Kowa Company, Ltd. T. Nojima: Employee; Self; Kowa Company, Ltd. H. Suganami: Employee; Self; Kowa Company, Ltd. K. Fujihara: None. K. Kaku: Advisory Panel; Self; Nippon Boehringer Ingelheim Co. Ltd., Novo Nordisk Inc. Consultant; Self; Sanwa Kagaku Kenkyusho. Research Support; Self; Taisho Pharmaceutical Co., Ltd. Speaker’s Bureau; Self; Astellas Pharma Inc., AstraZeneca, Daiichi Sankyo, Eli Lilly Japan K.K., Kowa Company, Ltd., Merck Sharp & Dohme Corp., Mitsubishi Tanabe Pharma Corporation, Ono Pharmaceutical Co., Ltd., Takeda Pharmaceutical Company Limited. H. Sone: Research Support; Self; Kyowa Hakko Kirin Co., Ltd., Novartis AG, Ono Pharmaceutical Co., Ltd., Taisho Pharmaceutical Co., Ltd., Takeda Pharmaceutical Co.

DPP-4i do not increase weight and SGLT2i are expected to reduce weight; thus, are widely used in daily practice. SGLT2i are recommended as second-line drugs after metformin (Met) in those at high risk of atherosclerotic cardiovascular disease (ASCVD), but suitable second-line drugs for the majority of patients not at high risk are not yet determined. Furthermore, reports of appropriate second-line drugs in East Asians who have a lower risk of ASCVD than Europeans are scarce. We compared the effects of ipraglifrozin (Ipra) and sitagliptin (Sita) after Japanese patients with type 2 diabetes had been taking Met and were without prior ASCVD in a 52-week randomized open-label trial (jRCTs031180205). The primary endpoint (PE) was a HbA1c reduction of ≥0.5% with no weight gain at 52 weeks. Among 111 patients (mean age 59.2 y, mean BMI 26.6 kg/m2, and male 61.3%), 54 and 57 patients received Ipra and Sita, respectively. After 52 weeks, achievement of the PE was statistically the same (37.0% and 40.3%, p=0.72), as was the HbA1c reduction ≥0.5% (38.9% and 54.4%, p=0.13). Lack of weight gain was higher for Ipra than Sita (94.4% and 54.4%, p Disclosure M. Kitazawa: None. T. Katagiri: None. H. Suzuki: None. S. Matsunaga: None. M.H. Yamada: None. T. Ikarashi: None. M. Yamamoto: None. K. Furukawa: None. M. Iwanaga: None. M. Hatta: None. K. Fujihara: None. T. Yamada: None. S. Tanaka: None. H. Sone: Research Support; Self; Kyowa Hakko Kirin Co., Ltd., Novartis AG, Ono Pharmaceutical Co., Ltd., Taisho Pharmaceutical Co., Ltd., Takeda Pharmaceutical Co. Funding Astellas Pharma Inc.

We examined whether the machine learning approach could be used as support for decision-making by general physicians as to whether to use insulin as an initial therapy for patients with type 2 diabetes. Prescription data on different antidiabetic agents from December 2009 to March 2015 from diabetes specialists’ patient registries were used to compare the machine learning decision of initial insulin use to that of general physicians. Gold standard was defined as consensus, that is, 80% agreement among nine diabetes specialists based on patients’ information. Twenty-two general physicians chose the most suitable medication according to the patients’ information. Analyzed were data on 4,860 patients who received initial monotherapy with either insulin (293 patients) or non-insulin (4,567 patients) and had laboratory data. Accuracy was calculated by 5-fold cross validation in machine learning. The average predictive values of initial insulin use by machine learning were 0.40, 0.50 and 0.82, respectively, for under-sampling ratios of 1:2, 1:4 and 1:8. General physicians achieved a mean accuracy for initial insulin use of 51%. The overall accuracy was 43%, 57% and 86% for under-sampling ratios of 1:2, 1:4 and 1:8. The difference in the accuracy between general physicians and machine learning in the under-sampling ratio of 1:8 was significant (P In conclusion, accuracy of machine learning with the sampling ratio of 1:8 was higher compared with that of general physicians in initial insulin use defined by diabetes specialists’ choice as the gold standard. Assistance by machine learning may be beneficial to any general physician in deciding upon initial insulin usage. Disclosure K. Fujihara: None. M.H. Yamada: None. Y. Matsubayashi: None. M. Yamamoto: None. T. Iizuka: None. K. Miyamura: None. Y. Hasegawa: None. T. Yamazaki: None. S. Kodama: None. H. Sone: Research Support; Self; Kyowa Hakko Kirin Co., Ltd., Novartis AG, Ono Pharmaceutical Co., Ltd., Taisho Pharmaceutical Co., Ltd., Takeda Pharmaceutical Co. Funding Japan Society for the Promotion of Science (18K17897)

Although cardiorespiratory fitness (CRF) and muscular fitness (MF) are modifiable factors for metabolic risk, the role of weight status in these factors has not been clarified in the pediatric population. We examined associations between metabolic indices and CRF or MF according to weight categories and compared the prevalence of metabolic abnormalities among four groups stratified by combinations of PF and weight categories. Cross-sectionally analyzed were 1744 Japanese adolescents aged 13-14 years. The PF test included measurements of CRF (20-m shuttle run test), upper-limb strength (handgrip test), lower limb strength (standing long jump test), and muscular endurance (sit-ups). Participants were classified as non-overweight (non-OW) or overweight/obese (OW) according to BMI cutoffs by the International Obesity Task Force. Metabolic risk was defined as ≥1 SD of the clustered metabolic risk (estimated by summing standardized sex-specific Z scores of HbA1c, MAP, and non-HDL-C). After adjusting for BMI and other characteristics, linear regression analysis showed that non-HDL-C was inversely associated with CRF in the non-OW group (P In summary, relationships of metabolic risk with components of PF differed according to weight status. In OW adolescents, MF might be more useful for stratifying metabolic risk than CRF. Disclosure S.Y. Morikawa: None. K. Fujihara: None. R. Nedachi: None. I. Ikeda: None. Y. Takeda: None. M. Takeuchi: None. M. Hatta: None. H. Ishiguro: None. T. Yamada: None. H. Sone: Research Support; Self; Kyowa Hakko Kirin Co., Ltd., Novartis AG, Ono Pharmaceutical Co., Ltd., Taisho Pharmaceutical Co., Ltd., Takeda Pharmaceutical Co.

Machine learning (ML) methods receive increasing attention for diabetes risk assessment. This meta-analysis quantitatively assessed the predictive ability of ML algorithms for incident T2DM. Electronic literature searches were done using MEDLINE and EMBASE for longitudinal studies whereby an index test was a model constructed by ML algorithms for predicting incident T2DM. Reference standard was a blood test or physicians’ diagnosis confirming T2DM. The model was restricted to algorithms that required the aid of ML (e.g., neural network, support vector machine, etc.). The 2x2 contingency data (i.e., true-positive, false-positive, false-negative, and true-negative) were pooled with a hierarchical summary receiver operating characteristic model. Pooled sensitivity and specificity (95% confidence interval (CI)) of 8 eligible studies were 0.70 (0.66-0.74) and 0.78 (0.69-0.85), respectively. (Figure) Positive likelihood ratio (PLR) calculated as sensitivity/(1-specificity) and negative likelihood ratio (NLR) calculated as (1-sensitivity)/specificity were 3.25 (95%CI, 2.25-4.58) and 0.38 (95% CI, 0.34-0.43), respectively. We conclude that existing prediction models are not yet satisfactory considering that PLR>5 moderately increases the probability of a disease given a positive test and NLR Disclosure S. Kodama: None. T. Sato: None. M. Yamamoto: None. H. Ishiguro: None. M. Iwanaga: None. K. Fujihara: None. T. Yamada: None. K. Kato: None. H. Sone: Research Support; Self; Kyowa Hakko Kirin Co., Ltd., Novartis AG, Ono Pharmaceutical Co., Ltd., Taisho Pharmaceutical Co., Ltd., Takeda Pharmaceutical Co.

Some plasma metabolites in the urea cycle were reportedly associated with DKD, although results were inconsistent among studies. Also, few studies examined the association between urea cycle metabolites and DKD in community dwellers. To address this, we used the metabolite database from the Tsuruoka Metabolomics Cohort Study, which had global metabolomics in Japan, using capillary electrophoresis-mass spectrometry. Among 10,933 participants, cross-sectionally analyzed were 1,005 who had diabetes mellitus (DM) defined by HbA1c ≥6.5%, and/or fasting blood glucose ≥126 mg/dl and/or antidiabetic drug use. Of these 231 had DKD, which was defined by eGFR Disclosure M. Yamamoto: None. S. Harada: None. T. Okamura: None. K. Fujihara: None. Y. Yaguchi: None. T. Komatsu: None. T. Sato: None. M. Kitazawa: None. M.H. Yamada: None. M. Kaneko: None. T. Osawa: None. Y. Matsubayashi: None. T. Yamada: None. S. Kodama: None. H. Sone: Research Support; Self; Kyowa Hakko Kirin Co., Ltd., Novartis AG, Ono Pharmaceutical Co., Ltd., Taisho Pharmaceutical Co., Ltd., Takeda Pharmaceutical Co. T. Takebayashi: None.

The impact of MA on glycemic control may be affected by lifestyle confounders but little is known on this topic. We investigated clinical variables significantly associated with MA for OHA and the association between MA and glycemic control considering the effects of smoking, drinking alcohol, exercise and dietary habits in Japanese with DM using a large claims database. Analyzed were 6651 DM patients with medication records for OHA for at least 1 y (mean age 54 y, HbA1c 7.2%). MA was evaluated by the proportion of days covered (PDC), and PDC Disclosure Y. Yaguchi: None. K. Fujihara: None. M.H. Yamada: None. T. Sato: None. M. Kitazawa: None. M. Yamamoto: None. H. Seida: None. N. Yamanaka: None. Y. Matsubayashi: None. H. Sone: Research Support; Self; Kyowa Hakko Kirin Co., Ltd., Novartis AG, Ono Pharmaceutical Co., Ltd., Taisho Pharmaceutical Co., Ltd., Takeda Pharmaceutical Co.

An increase in screen time (ST) is reportedly associated with insulin resistance and obesity in children. However, most studies are of TV viewing but not smartphone use. Also, little is known of the relationship between ST and diet. In this study, we clarified the relationships among ST including smartphone use, lifestyle factors including dietary content, and obesity. Cross-sectionally analyzed were 1393 children between 11-15 years of age. ST (min/day) was classified into 4 groups: In conclusion, the increase in ST including smartphone usage was suggested to be associated with various unfavorable lifestyles including dietary habits that could be connected to obesity. Disclosure I. Ikeda: None. K. Fujihara: None. R. Nedachi: None. S.Y. Morikawa: None. H. Ishiguro: None. M.H. Yamada: None. Y. Matsubayashi: None. T. Yamada: None. H. Sone: Research Support; Self; Kyowa Hakko Kirin Co., Ltd., Novartis AG, Ono Pharmaceutical Co., Ltd., Taisho Pharmaceutical Co., Ltd., Takeda Pharmaceutical Co.

BMI ≥25.0 and WC ≥90 are components of the International Diabetes Federation (IDF) definition of metabolic syndrome for Asians. We have investigated more suitable BMI and WC levels for risk of CAD using large-scale nationwide claims data on people with and without DM in the same cohort. The sample included 157,220 Japanese men 18-72 years of age without a known history of CAD at baseline (DM: 15,465 patients, non-DM: 141,755 patients). Multivariate Cox regression model examined the association of BMI/WC categories and their combination with CAD. During the study period (median follow-up 5.3 y), 662 and 332 CAD events occurred in non-DM and DM, respectively. Cox regression model showed that BMI ≥22.5 and WC ≥80 conferred a 1.29 (1.07-1.55), and 1.23 (1.00-1.50) times increased risk of CAD, respectively, in non-DM but that those relationships disappeared in people with DM (HR, 0.96 (0.71-1.29) for BMI and 1.13 (0.78-1.62) for WC. Stratified analysis of 4 groups of combinations of BMI and WC showed that in non-DM with BMI ≥22.5/WC ≥80 the risk of future CAD was significantly increased compared with non-DM with BMI In summary, lower BMI and WC values can predict CAD in individuals without DM whereas their influence was modest in people with DM. Disclosure T. Komatsu: None. K. Fujihara: None. M.H. Yamada: None. T. Sato: None. M. Kitazawa: None. M. Yamamoto: None. H. Seida: None. N. Yamanaka: None. Y. Matsubayashi: None. T. Yamada: None. H. Sone: Research Support; Self; Kyowa Hakko Kirin Co., Ltd., Novartis AG, Ono Pharmaceutical Co., Ltd., Taisho Pharmaceutical Co., Ltd., Takeda Pharmaceutical Co.

Although both average BMI and foods vary greatly between Asian and Western countries, few studies have examined the relationship between dietary patterns and obesity in Asians. We investigated dietary patterns significantly associated with obesity in Japanese with diabetes. Cross-sectionally analyzed were 2070 Japanese with type 2 diabetes (mean age, 62 y; men, 62%) who completed the Food Frequency Questionnaire. Principal component analysis was performed on 20 food groups to elucidate dietary patterns significantly associated with obesity. We also divided participants into quintiles according to factor scores of each dietary pattern. The adjusted odds ratios (ORs) for obesity, defined as BMI ≥25 according to the Asian cutoff, were calculated with other covariates including energy intake. Six dietary patterns were determined from eigenvalues (≥1) and screen plots. For Factor 1, characterized by a well-balanced food group with high intake of light-colored vegetables, green and yellow vegetables, sugar, seaweed, beans, fish and seafood, fruit and potato, the OR for obesity in Quintile5 compared to Quintile1 was 0.34 (95% CI: 0.22-0.53). Conversely, that of Factor 2, characterized by high intake of seasoning and spices, sugar-sweetened beverages, rice and eggs, was 2.56 (1.69-3.89) (Table). In conclusion, a balanced diet with various food groups might help to avoid obesity in Japanese with type 2 diabetes. Disclosure M. Hatta: None. K. Fujihara: None. I. Ikeda: None. M. Takeuchi: None. R. Nedachi: None. S.Y. Morikawa: None. C. Horikawa: None. M. Kato: None. N. Kato: None. H. Maegawa: Speaker’s Bureau; Self; Astellas Pharma Inc., Boehringer Ingelheim Pharmaceuticals, Inc., Daiichi Sankyo, Merck Sharp & Dohme Corp., Mitsubishi Tanabe Pharma Corporation, Sanofi K.K., Takeda Pharmaceutical Company Limited. H. Sone: Research Support; Self; Kyowa Hakko Kirin Co., Ltd., Novartis AG, Ono Pharmaceutical Co., Ltd., Taisho Pharmaceutical Co., Ltd., Takeda Pharmaceutical Co.

Despite recent tightening of cutoffs for hypertension (HT) in major global guidelines, guidelines, especially according to GTS, remain debatable. We analyzed data from a nationwide claims database on 293,396 participants without prior CAD or CVD from 2008-16. Multivariate Cox regression model identified risks of CAD and CVD events. Hazard ratios (HRs) were compared among 5 levels of systolic blood pressure (SBP) (≤119, 120-129, 130-139, 140-149, ≥150) according to GTS [i.e., normal glucose tolerance (NGT), prediabetes (pre-DM), diabetes (DM)]. During the study period, 1099 CAD and 1990 CVD events occurred. Those with SBP levels ≥120 mmHg had a significantly higher risk of incident CAD/CVD than those with lower levels. The incidence of CAD and CVD did not differ significantly between NGT and pre-DM in those with SBP ≤119 whereas HRs for incident CAD for DM was almost 2.7, which was similar to NGT and pre-DM with SBP ≥150 mmHg (Table). Compared with NGT and SBP ≤119 mmHg, the HR for incident CAD among DM and SBP ≥150 mmHg was 5.91 (4.22-8.26), while SBP ≥150 mmHg significantly increased the risk of CVD about 4 fold regardless of GTS. These results based on a large database implied the need for even lower BP targets regardless of GTS. The impact of BP was stronger in DM in preventing CAD. Disclosure M.H. Yamada: None. K. Fujihara: None. Y. Yaguchi: None. T. Osawa: None. M. Kitazawa: None. Y. Matsubayashi: None. M. Iwanaga: None. T. Yamada: None. N. Yamanaka: None. H. Seida: None. W. Ogawa: Research Support; Self; Astellas Pharma Inc., Boehringer Ingelheim Pharma GmbH & Co.KG, Eli Lilly Japan K.K., Kowa Company, Ltd., Nippon Boehringer Ingelheim Co. Ltd., Novo Nordisk Inc., Ono Pharmaceutical Co., Ltd., Sumitomo Dainippon Pharma Co., Ltd. Other Relationship; Self; Sumitomo Dainippon Pharma Co., Ltd., Takeda Pharmaceutical Company Limited. H. Sone: Research Support; Self; Kyowa Hakko Kirin Co., Ltd., Novartis AG, Ono Pharmaceutical Co., Ltd., Taisho Pharmaceutical Co., Ltd., Takeda Pharmaceutical Co. Funding 18K17897

Declining healthy life expectancy due to functional disability is relevant and urgent not only because of its association with decreased quality of life but also for its enormous socioeconomic impact. Thus, we analyzed a population-based retrospective cohort study including 9,673 community dwelling Japanese adults (4,420, men) aged 39-98 years in 2012-2017. Functional disability was defined as a condition meeting Japan’s new long-term care insurance certification requirements for the need of assistance in the activities of daily living whether by caregivers or assistive devices. Cox proportional hazards regression model identified variables related to functional disability. Median follow-up was 3.7 years. During the study period, 165 disabilities occurred in the overall study population. Multivariate analysis showed that diabetes (hazard ratio [HR] 1.74 (95% CI: 1.12-2.68)), no physical activity habit (HR 1.83 (1.27-2.65)), and low BMI category (18.5 kg/m2 >) (HR 1.63 (1.02-2.63)) presented increased risks for disability. The prevalence ratio of 0, 1, 2, 3 and 4 conditions (diabetes, hypertension, dyslipidemia, and physical activity habits) was 12.8%, 40.0%, 32.6%, 12.9 and 1.7%, respectively. HR for disability increased with the number of risk factors (HR of individuals with 1, 2, 3 and 4 conditions, 1.34 (0.71-2.52), 1.95 (1.04-3.63), 2.11 (1.07-4.15), and 3.96 (1.59-9.99) vs. individuals with none of those conditions as a ref.). HR for disability among diabetes patients with and without a physical activity habit was 1.68 (0.70-4.04) and 3.19 (1.79-5.70), respectively, compared to individuals without diabetes with a physical activity habit. This shows that the combination of diabetes and lack of habitual physical activity is predictive of functional disability in Japanese. Habitual physical activity attenuates the risk of functional disability in patients with diabetes. Disclosure K. Fujihara: None. M.H. Yamada: None. Y. Matsubayashi: None. M. Kitazawa: None. M. Kaneko: None. M. Yamamoto: None. Y. Yaguchi: None. T. Yahiro: None. S. Kodama: None. K. Kato: None. H. Sone: Research Support; Self; Kyowa Hakko Kirin Co., Ltd., Novartis AG, Ono Pharmaceutical Co., Ltd., Taisho Pharmaceutical Co., Ltd., Takeda Pharmaceutical Co. Funding Japan Society for the Promotion of Science (18K17897)

Low serum AMY values were previously reported to be associated with incident T2DM. We further investigated that association in detail by showing the combined effects of chronological changes in AMY and baseline AMY as well as results of stratified analyses of baseline age and HbA1c. Analyzed were 5306 individuals who had health examinations yearly for 7 y. We monitored changes in AMY values between baseline1 (first visit) and baseline2 (2 y later). Multivariate Cox analysis was conducted to evaluate the impacts of baseline2 AMY and change in AMY on incident T2DM in consideration of age and HbA1c. During the 5-y follow-up, T2DM occurred in 716 patients. Both low baseline2 AMY and decrease in AMY were independently associated with the development of T2DM. Compared with those whose baseline2 AMY ≥57 (Quintiles (Q) 2-5) and changes in AMY ≥90.4 (Q 2-5), those whose baseline2 AMY ≤56 (Q1) and changes in AMY ≤90.3 (Q1) had a significantly increased risk for T2DM (HR 1.60 [1.22-2.09]). Stratified analyses by age and baseline HbA1c levels showed that the increased risk for T2DM due to low baseline2 AMY or its change occurred only among those These findings imply the clinical relevance of using AMY values and their chronological changes over time for evaluation of T2DM risk among individuals who are relatively young or have low HbA1c. Disclosure I. Ikeda: None. K. Fujihara: None. I. Risa: None. M.H. Yamada: None. R. Nedachi: None. M. Hatta: None. S. Kodama: None. Y. Mori: None. Y. Matsubayashi: None. Y. Arase: None. H. Sone: Research Support; Self; Kyowa Hakko Kirin Co., Ltd., Novartis AG, Ono Pharmaceutical Co., Ltd., Taisho Pharmaceutical Co., Ltd., Takeda Pharmaceutical Co. R. Yamamoto-Honda: None.

Pediatric obesity is associated with clustered cardiometabolic risk and the future incidence of diabetes. However, few studies have determined the effect of pediatric obesity in Asia, where obesity is less common than in Western countries. Thus, we clarified whether metabolic risk factors are associated with weight status including slight overweight and underweight among Japanese adolescents. We performed a cross-sectional analysis of 2241 Japanese adolescents (1180 boys) aged 13-14 years whose height, weight, HbA1c, blood pressure (BP), and serum lipids were measured. Participants were classified as underweight, normal weight, slightly overweight, overweight, or obese according to both the universal and Asian cutoffs of the International Obesity Task Force, which correspond to adult BMI cutoffs. The clustered cardiometabolic risk (Z-CMR) was estimated by summing standardized sex-specific Z scores of HbA1c, mean arterial pressure (MAP), and non-high-density lipoprotein cholesterol (non-HDL-C). Linear regression analysis showed that MAP (sex-specific Z score), non-HDL-C (sex-specific Z score), and Z-CMR were higher in the slightly overweight, overweight, and obese groups than in the normal weight group after adjusting for confounders. Compared with the normal weight group, the slightly overweight, overweight, and obese groups had significantly higher odds ratios (ORs) for high BP [OR = 1.38 (95% CI, 1.03-1.85); OR = 2.63 (1.77-3.91); and OR = 2.39 (1.57-3.64), respectively]. Compared with the normal weight group, underweight boys, but not girls, had a significantly lower OR for high Z-CMR [OR = 0.20 (0.05-0.84)]. Our results indicated that adolescents classified as slightly overweight had higher levels of BP, serum lipids (but not HbA1c), and clustered cardiometabolic risk than those classified as normal weight. This observation demonstrated that significant associations between weight status and cardiometabolic risk factors could be seen during adolescence even in East Asia. Disclosure S.Y. Morikawa: None. K. Fujihara: None. D. Ishii: None. R. Nedachi: None. M. Ishizawa: None. H. Ishiguro: None. Y. Matsubayashi: None. T. Yamada: None. S. Kodama: None. H. Sone: Research Support; Self; Astellas Pharma Inc., Boehringer Ingelheim Pharmaceuticals, Inc., Daiichi Sankyo Company, Limited, Kowa Pharmaceutical Europe Co. Ltd., Kyowa Hakko Kirin Co., Ltd., Novo Nordisk Inc., Ono Pharmaceutical Co., Ltd., Taisho Pharmaceutical Co., Ltd., Takeda Pharmaceutical Company Limited, Teijin Pharma Limited.

Although the 2017 ACC/AHA guidelines indicated a generalized BP goal of less than 130/80 mm Hg, the impact of BP separately on CAD or CVD in people with different GTSs has yet to be clarified. To clarify this, data from a nationwide claims database involving 293,396 participants without prior CAD or CVD during 2008-16 were analyzed. Multivariate Cox regression model was used to identify risks of a CAD and/or CVD event. Hazard ratios (HRs) were compared among 4 groups of participants divided according to stratified levels of BP [i.e., 90 mmHg, ≥140/90 mmHg] according to GTS [i.e., normal glucose tolerance (NGT), prediabetes (pre-DM), diabetes mellitus (DM)]. During the study period, 1099 CAD and 1990 CVD events occurred. Despite that a significant trend between BP and incident CAD was observed regardless of GTS, risk for CVD was dependent on BP only in the NGT and pre-DM groups. Moreover, there were no significant differences in CVD risks between categories of BP Disclosure M. Harada: None. K. Fujihara: None. M. Yamamoto: None. M. Kaneko: None. M. Kitazawa: None. H. Suzuki: None. Y. Matsubayashi: None. T. Yamada: None. N. Yamanaka: None. H. Seida: None. W. Ogawa: Consultant; Self; Abbott. Research Support; Self; Astellas Pharma Inc., AstraZeneca, Boehringer Ingelheim International GmbH, Daiichi Sankyo Company, Limited, Mitsubishi Tanabe Pharma Corporation, Takeda Pharmaceutical Company Limited. Speaker's Bureau; Self; Lilly Diabetes, Novo Nordisk Inc., Ono Pharmaceutical Co., Ltd., Sanofi, Sumitomo Dainippon Pharma Co., Ltd. H. Sone: Research Support; Self; Astellas Pharma Inc., Boehringer Ingelheim Pharmaceuticals, Inc., Daiichi Sankyo Company, Limited, Kowa Pharmaceutical Europe Co. Ltd., Kyowa Hakko Kirin Co., Ltd., Novo Nordisk Inc., Ono Pharmaceutical Co., Ltd., Taisho Pharmaceutical Co., Ltd., Takeda Pharmaceutical Company Limited, Teijin Pharma Limited.

Poor medication adherence is a well-known barrier to meeting therapeutic goals in diabetes mellitus (DM). We aimed to clarify the status of medication adherence and determine clinical variables significantly associated with medication adherence in Japanese patients with DM using a large claims database. Analyzed were data from 7459 patients with DM whose medical records on oral hypoglycemic agents (OHA) were available for at least 1 y (mean age 53 y, HbA1c 7.2%). Medication adherence was evaluated by the proportion of days covered (PDC); PDC Disclosure Y. Yaguchi: None. K. Fujihara: None. T. Osawa: None. M. Yamamoto: None. M. Kaneko: None. M. Kitazawa: None. M. Harada: None. Y. Matsubayashi: None. T. Yamada: None. N. Yamanaka: None. H. Seida: None. H. Sone: Research Support; Self; Astellas Pharma Inc., Boehringer Ingelheim Pharmaceuticals, Inc., Daiichi Sankyo Company, Limited, Kowa Pharmaceutical Europe Co. Ltd., Kyowa Hakko Kirin Co., Ltd., Novo Nordisk Inc., Ono Pharmaceutical Co., Ltd., Taisho Pharmaceutical Co., Ltd., Takeda Pharmaceutical Company Limited, Teijin Pharma Limited.

Diabetic ketoacidosis characterized by lipolysis due to insulin deficiency is a side effect reported for SGLT2 inhibitors (SGLT2is). However, little is known about predictors of ketosis via SGLT2i. The antilipolytic effect of insulin in adipose tissue can be evaluated as adipose tissue insulin resistance (Adipo-IR). This study investigated predictors of ketosis via the SGLT2i, tofogliflozin (TOFO), in patients with type 2 diabetes (T2D) focusing on Adipo-IR. Analyzed were 304 T2D patients who received TOFO as initial monotherapy in two TOFO phase 2 and 3 studies. Adipo-IR was calculated by the product of fasting insulin (IRI) and fasting free fatty acids (FFA). Baseline characteristics were: males (67%), age (mean: 58 y), HbA1c (8%), BMI (26kg/m2), β-hydroxybutyrate (BHB) (75µmol/L), eGFR (84mL/min/1.73m2) and Adipo-IR (33pmol/L mmol/L). Participants were grouped by tertiles of baseline Adipo-IR levels (T1-T3). Differences across tertiles were analyzed by ANOVA, Kruskal Wallis and ANCOVA. Multivariate analysis explored the predictors of ketosis. In the order of T3-T1, younger age, higher baseline HOMA-IR levels and higher baseline HOMA-β levels were observed, while baseline HbA1c and BHB levels were identical across tertiles. At 24 weeks after TOFO administration, reduced HbA1c and body weight and increased urinary glucose excretion were identical among tertiles. Percent changes in FFA were T1, mean +50.9%* (* p In conclusion, high baseline Adipo-IR may attenuate ketosis via SGLT2i through improved Adipo-IR and the antilipolytic effect of insulin despite reduced IRI in T2D. Disclosure Y. Matsubayashi: None. A. Yoshida: Employee; Self; Kowa Pharmaceutical Co. Ltd. T. Nojima: Employee; Self; Kowa Co., Ltd. H. Suganami: Employee; Self; Kowa Co., Ltd. K. Fujihara: None. K. Kaku: Advisory Panel; Self; Astellas Pharma Inc., AstraZeneca, Boehringer Ingelheim Pharmaceuticals, Inc., Novo Nordisk A/S. Speaker's Bureau; Self; Eli Lilly and Company, Merck Sharp & Dohme Corp., Mitsubishi Tanabe Pharma Corporation, Ono Pharmaceutical Co., Ltd., Taisho Pharmaceutical Co., Ltd., Takeda Pharmaceutical Company Limited. H. Sone: Research Support; Self; Astellas Pharma Inc., Boehringer Ingelheim Pharmaceuticals, Inc., Daiichi Sankyo Company, Limited, Kowa Pharmaceutical Europe Co. Ltd., Kyowa Hakko Kirin Co., Ltd., Novo Nordisk Inc., Ono Pharmaceutical Co., Ltd., Taisho Pharmaceutical Co., Ltd., Takeda Pharmaceutical Company Limited, Teijin Pharma Limited.

Obesity is known to increase the risks of cardiovascular events and mortality in patients with diabetes. Low serum magnesium (Mg) levels are reportedly associated with obesity and incident diabetes mellitus among nondiabetic individuals. Since only a few studies have addressed the relationship between intake of Mg and obesity, we analyzed our nationwide registry of diabetic patients to determine if there was an association between intake of Mg and obesity in Japanese patients with T2DM. We analyzed data on 1891 Japanese patients with T2DM (mean age, 62 y; men, 62%) who completed a food frequency questionnaire. Obesity was defined as BMI ≥25. Associations between obesity and intake of Mg were determined by multivariate regression and quintile analyses. After adjusting for confounders including energy intake, the odds ratio (OR) for obesity was 0.93 (0.91,0.96) per 10 mg increase in Mg intake. Quintile analysis revealed a stepwise significant decrease in ORs for obesity from Q1 (lowest, ref.) to Q5 (highest) (Table). Food groups potently related to Mg intake in our patients were beans (correlation coefficients, 0.64), vegetables (0.62), and fish (0.57). These results imply that intake of Mg might be closely associated with obesity in patients with T2DM, which should have clinical relevance in nutrition therapy. Disclosure R. Nedachi: None. K. Fujihara: None. M. Hatta: None. Y. Matsubayashi: None. Y. Takeda: None. D. Ishii: None. C. Horikawa: None. M. Kato: None. H. Maegawa: Research Support; Self; Antares Pharma, Boehringer Ingelheim Pharmaceuticals, Inc., Daiichi Sankyo Company, Limited, Mitsubishi Tanabe Pharma Corporation, Takeda Pharmaceutical Company Limited. Speaker's Bureau; Self; Astellas Pharma Inc., Boehringer Ingelheim Pharmaceuticals, Inc., Daiichi Sankyo Company, Limited, Merck Sharp & Dohme Corp., Mitsubishi Tanabe Pharma Corporation, Takeda Pharmaceutical Company Limited. H. Sone: Research Support; Self; Astellas Pharma Inc., Boehringer Ingelheim Pharmaceuticals, Inc., Daiichi Sankyo Company, Limited, Kowa Pharmaceutical Europe Co. Ltd., Kyowa Hakko Kirin Co., Ltd., Novo Nordisk Inc., Ono Pharmaceutical Co., Ltd., Taisho Pharmaceutical Co., Ltd., Takeda Pharmaceutical Company Limited, Teijin Pharma Limited.

Although serum phosphate concentrations have been consistently reported to have a negative correlation with BMI, there have been some conflicting reports of the relationship of dietary intake of phosphorus with BMI. Thus, we investigated the association between phosphorus intake and obesity in Japanese patients with diabetes mellitus. Analyzed were 1891 patients with T2DM (mean age, 62 y; men, 62%) in our nationwide registry who had completed a food frequency questionnaire. Obesity was defined according to BMI >25. The association between phosphorus intake and obesity was determined by multivariate regression as well as quintile analyses. Mean daily intake of phosphorus was 996 mg in men and 977 mg in women, values that were quite close to those of the general population in Japan. Intake of phosphorus was significantly associated with decreased risk of obesity with an odds ratio (OR) of 0.90 (0.87-0.94) per 50 mg increase in phosphorus intake. Quintile analysis revealed potent negative correlations in both men and women (Table). Food groups strongly associated with phosphorus among our patients were fish (0.58), dairy products (0.54) and beans (0.52). These results imply the clinical relevance of controlling phosphorus intake as part of medical nutrition therapy for patients with T2DM. Disclosure M. Hatta: None. K. Fujihara: None. Y. Matsubayashi: None. Y. Takeda: None. R. Nedachi: None. D. Ishii: None. C. Horikawa: None. M. Kato: None. H. Maegawa: Research Support; Self; Antares Pharma, Boehringer Ingelheim Pharmaceuticals, Inc., Daiichi Sankyo Company, Limited, Mitsubishi Tanabe Pharma Corporation, Takeda Pharmaceutical Company Limited. Speaker's Bureau; Self; Astellas Pharma Inc., Boehringer Ingelheim Pharmaceuticals, Inc., Daiichi Sankyo Company, Limited, Merck Sharp & Dohme Corp., Mitsubishi Tanabe Pharma Corporation, Takeda Pharmaceutical Company Limited. H. Sone: Research Support; Self; Astellas Pharma Inc., Boehringer Ingelheim Pharmaceuticals, Inc., Daiichi Sankyo Company, Limited, Kowa Pharmaceutical Europe Co. Ltd., Kyowa Hakko Kirin Co., Ltd., Novo Nordisk Inc., Ono Pharmaceutical Co., Ltd., Taisho Pharmaceutical Co., Ltd., Takeda Pharmaceutical Company Limited, Teijin Pharma Limited.

Dietary energy density (DED) is receiving increased attention reflecting recent reports of inverse associations with health outcomes. Although high DED meals are reportedly associated with low diet quality in the general population, such evidence in patients with diabetes is sparse, which led us clarify this issue in Japanese patients with T2DM. Analyzed were 1,615 outpatients with T2DM (men 62.2%) who attended 24 clinics specializing in diabetes care nationwide. Dietary habits were obtained using a food frequency questionnaire (FFQg) based on 18 food groups. DED (kcal/g) was calculated by dividing EI (kcal) from foods (excluding beverages) by weights of these foods (g). Mean DED (±SD) was 1.53±0.23 (kcal/g). Typical dietary patterns comprised of some significant food groups were extracted by principal component analysis. Pearson's correlation analysis was used to evaluate the associations between identified food patterns and DED. We could extract and validate three dietary patterns, which we labeled as “Healthy,” “Western,” and “Animal foods and confectionery.” Two patterns were significantly correlated with DED (Healthy: r=-0.48, Animal foods and confectionery: r=-0.69). A multivariate linear regression model was used to determine food groups that were independently related to DED. DED was positively associated with confectioneries (standardized β=0.19, P In conclusion, lower DED, which is related to higher intakes of vegetables and fruits was associated with a healthy diet pattern in T2DM patients. DED is a useful indicator of diet quality and could be utilized in nutritional education for people with T2DM. Disclosure Y. Takeda: None. K. Fujihara: None. M. Hatta: None. S.Y. Morikawa: None. C. Horikawa: None. D. Ishii: None. R. Nedachi: None. H. Maegawa: Research Support; Self; Antares Pharma, Boehringer Ingelheim Pharmaceuticals, Inc., Daiichi Sankyo Company, Limited, Mitsubishi Tanabe Pharma Corporation, Takeda Pharmaceutical Company Limited. Speaker's Bureau; Self; Astellas Pharma Inc., Boehringer Ingelheim Pharmaceuticals, Inc., Daiichi Sankyo Company, Limited, Merck Sharp & Dohme Corp., Mitsubishi Tanabe Pharma Corporation, Takeda Pharmaceutical Company Limited. H. Sone: Research Support; Self; Astellas Pharma Inc., Boehringer Ingelheim Pharmaceuticals, Inc., Daiichi Sankyo Company, Limited, Kowa Pharmaceutical Europe Co. Ltd., Kyowa Hakko Kirin Co., Ltd., Novo Nordisk Inc., Ono Pharmaceutical Co., Ltd., Taisho Pharmaceutical Co., Ltd., Takeda Pharmaceutical Company Limited, Teijin Pharma Limited.

Although cardiovascular risk and mortality increases in the presence of comorbid obesity among patients with T2D while fish intake is known to decrease such risks, the association between obesity and fish intake is only poorly understood. Thus, we aimed to clarify these possible associations, including the effects of fish-related nutrients. Analyzed were 1891 Japanese patients with T2D (mean age, 62 y; men, 62%) who completed the Food Frequency Questionnaire. BMI>25 was defined as obesity. Associations between obesity and intake of fish and related nutrients were analyzed by quintiles. Vitamin B12 (VB12) (correlation coefficient, 0.97), vitamin D (VD) (0.94), and long-chain omega-3 polyunsaturated fatty acids (n-3) (0.74) were significantly and potently associated with fish intake. Odds ratio (OR) for obesity was significantly decreased to approximately 0.6 in patients in Q4 (men, 80-108g; women, 77-103g) and Q5 (>108g; >103g) of fish intake compared to Q1 ( Disclosure M. Hatta: None. K. Fujihara: None. Y. Matsubayashi: None. Y. Takeda: None. R. Nedachi: None. D. Ishii: None. C. Horikawa: None. N. Kato: None. H. Maegawa: Research Support; Self; Antares Pharma, Boehringer Ingelheim Pharmaceuticals, Inc., Daiichi Sankyo Company, Limited, Mitsubishi Tanabe Pharma Corporation, Takeda Pharmaceutical Company Limited. Speaker's Bureau; Self; Astellas Pharma Inc., Boehringer Ingelheim Pharmaceuticals, Inc., Daiichi Sankyo Company, Limited, Merck Sharp & Dohme Corp., Mitsubishi Tanabe Pharma Corporation, Takeda Pharmaceutical Company Limited. H. Sone: Research Support; Self; Astellas Pharma Inc., Boehringer Ingelheim Pharmaceuticals, Inc., Daiichi Sankyo Company, Limited, Kowa Pharmaceutical Europe Co. Ltd., Kyowa Hakko Kirin Co., Ltd., Novo Nordisk Inc., Ono Pharmaceutical Co., Ltd., Taisho Pharmaceutical Co., Ltd., Takeda Pharmaceutical Company Limited, Teijin Pharma Limited.

Diabetes self-management is an essential part of diabetes care. Although several personality traits were reportedly associated with self-management behaviors or glycemic control in Western countries, clinical evidence of such an association is still sparse, especially among Asian patients with diabetes. Thus, we aimed to clarify the associations among personality, self-management behaviors, and glycemic control in Japanese patients with T2DM. Analyzed were 510 Japanese patients with T2DM (62.2 ± 11.2 y; 57% men) with available information on personality (assessed by NEO Five-Factor Inventory), diabetes self-management activities (assessed by the Japanese version of Summary of Diabetes Self-Care Activities), body mass index (BMI), and HbA1c. Coefficient of correlation indicated that those with high scores for neuroticism, as well as low scores for conscientiousness, had lower levels of diet (neuroticism r = -0.14, P Disclosure S.Y. Morikawa: None. K. Fujihara: None. M. Hatta: None. Y. Takeda: None. D. Ishii: None. J. Yachida: None. C. Horikawa: None. M. Kato: None. H. Maegawa: Research Support; Self; Antares Pharma, Boehringer Ingelheim Pharmaceuticals, Inc., Daiichi Sankyo Company, Limited, Mitsubishi Tanabe Pharma Corporation, Takeda Pharmaceutical Company Limited. Speaker's Bureau; Self; Astellas Pharma Inc., Boehringer Ingelheim Pharmaceuticals, Inc., Daiichi Sankyo Company, Limited, Merck Sharp & Dohme Corp., Mitsubishi Tanabe Pharma Corporation, Takeda Pharmaceutical Company Limited. H. Sone: Research Support; Self; Astellas Pharma Inc., Boehringer Ingelheim Pharmaceuticals, Inc., Daiichi Sankyo Company, Limited, Kowa Pharmaceutical Europe Co. Ltd., Kyowa Hakko Kirin Co., Ltd., Novo Nordisk Inc., Ono Pharmaceutical Co., Ltd., Taisho Pharmaceutical Co., Ltd., Takeda Pharmaceutical Company Limited, Teijin Pharma Limited.

Although both a glucose metabolism abnormality and prior cerebrovascular disease (CVD) are major predictors of CVD, few studies have compared their impact on a subsequent CVD event. Also, few studies evaluated risk of prediabetes separately from nondiabetic categories. Thus, the impact of diabetes mellitus (DM) on subsequent CVD compared to normal glucose tolerance (NGT) could have been underestimated. We analyzed a nationwide claims-based database of 290,865 participants in 2008-14. Participants were classified as having NGT, prediabetes or DM based on fasting plasma glucose, HbA1c and use of antihypoglycemic agents (185,648 with NGT; 87,867 with prediabetes; 17,350 with DM). Participants were also classified as with or without prior CVD based on ICD-10 and questionnaires (1,944 with prior CVD; 288,921 without prior CVD). Cox regression model showed that prediabetes and DM confers about 1.1- and 1.4-fold excess risk for CVD, respectively, compared to NGT without prior CVD. Prediabetes and DM does not increase risk for CVD compared to NGT with prior CVD (Table). Prior CVD confers a 4-7-fold excess risk for CVD regardless of glucose tolerance status. Hazard ratio of CVD in NGT with prior CVD was 4 times higher than in DM without prior CVD. This implies that prediabetes and DM increase CVD only in those without prior CVD. Prior CVD confers a higher risk for subsequent CVD regardless of glucose tolerance status. Disclosure D. Yanagisawa: None. K. Fujihara: None. T. Osawa: None. M. Kitazawa: None. M. Yamamoto: None. Y. Matsubayashi: None. N. Yamanaka: None. S. Kodama: None. H. Sone: Research Support; Self; Astellas Pharma Inc., Boehringer Ingelheim Pharmaceuticals, Inc., Daiichi Sankyo Company, Limited, Kowa Pharmaceutical Europe Co. Ltd., Kyowa Hakko Kirin Co., Ltd., Novo Nordisk Inc., Ono Pharmaceutical Co., Ltd., Taisho Pharmaceutical Co., Ltd., Takeda Pharmaceutical Company Limited, Teijin Pharma Limited.

Although support from patients’ family members is essential in managing type 2 diabetes, evidence is sparse on family support for medical nutritional therapy (MNT) and habitual dietary intake. This study investigated the association between patients’ perception of family support for MNT and their dietary intake. Analyzed were 154 Japanese patients with type 2 diabetes (men, 45.5%) who completed the Food Frequency Questionnaire and Diabetes Family Behavior Checklist. Multivariate analysis of covariance was used adjusting for age, gender, BMI, duration for diabetes, HbA1c, and use of oral hypoglycemic agents and/or insulin. Participants’ mean age, duration of diabetes and HbA1c were 63.5 y, 12.7 y and 7.0%, respectively. After adjusting for confounders, positive family support, i.e., ‘Praise for following diet,’ was only significantly associated with higher fish intake (none 72.0 g/d, once a month 73.1 g/d, once a week 85.8 g/d, several times a week 97.6 g/d, and at least daily 92.1 g/d; trend p= 0.008). With greater frequency of eating at the same time as the family, salt intake was significantly higher (none 7.8 g/d, once a month 8.3 g/d, once a week 8.7 g/d, several times a week 9.3 g/d, and at least once daily 8.8 g/d; trend p= 0.037). Conversely, negative family support indicated by ‘Nags about following the diet’ was significantly associated with higher fish intake (none 70.3 g/d, once a month 84.4 g/d, once a week 84.3 g/d, several times a week 86.5 g/d, at least once daily 107.9 g/d; trend p= 0.012) and higher intake of vegetables was more frequent in those eating foods not part of the diabetic diet (none 242.1 g/d, once a month 192.1 g/d, once a week 270.4 g/d, several times a week 255.0 g/d, at least once daily 305.5 g/d; trend p= 0.027). These findings implied that patients’ perception of family support for MNT was not always associated with an ideal dietary intake to control diabetes. To facilitate diabetes treatment, sharing patient-family perceptions of MNT could be important. Disclosure C. Horikawa: None. Y. Takeda: None. M. Hatta: None. D. Ishii: None. S.Y. Morikawa: None. K. Fujihara: None. H. Maegawa: Research Support; Self; Antares Pharma, Boehringer Ingelheim Pharmaceuticals, Inc., Daiichi Sankyo Company, Limited, Mitsubishi Tanabe Pharma Corporation, Takeda Pharmaceutical Company Limited. Speaker's Bureau; Self; Astellas Pharma Inc., Boehringer Ingelheim Pharmaceuticals, Inc., Daiichi Sankyo Company, Limited, Merck Sharp & Dohme Corp., Mitsubishi Tanabe Pharma Corporation, Takeda Pharmaceutical Company Limited. H. Sone: Research Support; Self; Astellas Pharma Inc., Boehringer Ingelheim Pharmaceuticals, Inc., Daiichi Sankyo Company, Limited, Kowa Pharmaceutical Europe Co. Ltd., Kyowa Hakko Kirin Co., Ltd., Novo Nordisk Inc., Ono Pharmaceutical Co., Ltd., Taisho Pharmaceutical Co., Ltd., Takeda Pharmaceutical Company Limited, Teijin Pharma Limited. Funding Japan Association for Diabetes Education and Care

Obesity is associated with deterioration of T2DM and its complications. Although low serum zinc levels are reportedly associated with obesity in nondiabetic people, clinical evidence regarding the association between dietary intake of zinc and obesity in diabetic individuals is sparse. Thus, we aimed to clarify this issue utilizing our nationwide registry of Japanese patients with diabetes. Analyzed were 1891 patients with T2DM (mean age, 62 y; men, 62%) whose data from a food frequency questionnaire were available. Obesity was defined as BMI of ≥25. The associations between obesity and intake of zinc were determined by multivariate regression as well as quintile analyses. Odds ratio (OR) for obesity was 0.74 (0.65, 0.84) per 1 mg increase in zinc intake. Quintile analysis revealed a stepwise decrease in ORs for obesity at higher quintiles. ORs in Q3-Q5 were significant compared to Q1 as the reference (Table). Food groups strongly related to zinc intake among our patients were meats (correlation coefficients, 0.64), fish (0.49), and vegetables (0.47). These results suggest close associations between zinc intake and obesity in patients with T2DM. This information would be clinically applicable for nutrition education. Disclosure R. Nedachi: None. K. Fujihara: None. M. Hatta: None. Y. Matsubayashi: None. Y. Takeda: None. D. Ishii: None. C. Horikawa: None. N. Kato: None. H. Maegawa: Research Support; Self; Antares Pharma, Boehringer Ingelheim Pharmaceuticals, Inc., Daiichi Sankyo Company, Limited, Mitsubishi Tanabe Pharma Corporation, Takeda Pharmaceutical Company Limited. Speaker's Bureau; Self; Astellas Pharma Inc., Boehringer Ingelheim Pharmaceuticals, Inc., Daiichi Sankyo Company, Limited, Merck Sharp & Dohme Corp., Mitsubishi Tanabe Pharma Corporation, Takeda Pharmaceutical Company Limited. H. Sone: Research Support; Self; Astellas Pharma Inc., Boehringer Ingelheim Pharmaceuticals, Inc., Daiichi Sankyo Company, Limited, Kowa Pharmaceutical Europe Co. Ltd., Kyowa Hakko Kirin Co., Ltd., Novo Nordisk Inc., Ono Pharmaceutical Co., Ltd., Taisho Pharmaceutical Co., Ltd., Takeda Pharmaceutical Company Limited, Teijin Pharma Limited.

Poor adherence to medication is a clinically relevant issue closely associated with incidence and deterioration of complications in those with diabetes mellitus (DM). Although an elevated HbA1c level is an established risk factor for lower limb amputation, few studies have evaluated the impact of medication adherence combined with HbA1c levels on lower limb amputation. Thus, we analyzed data from 6,957 patients with DM whose records for oral hypoglycemic agents were available for at least 1 y. Medication adherence was evaluated by the proportion of days covered (PDC), with PDC Disclosure M. Kaneko: None. K. Fujihara: None. T. Osawa: None. M. Harada: None. M. Kitazawa: None. Y. Matsubayashi: None. T. Yamada: None. N. Yamanaka: None. H. Seida: None. H. Sone: Research Support; Self; Astellas Pharma Inc., Boehringer Ingelheim Pharmaceuticals, Inc., Daiichi Sankyo Company, Limited, Kowa Pharmaceutical Europe Co. Ltd., Kyowa Hakko Kirin Co., Ltd., Novo Nordisk Inc., Ono Pharmaceutical Co., Ltd., Taisho Pharmaceutical Co., Ltd., Takeda Pharmaceutical Company Limited, Teijin Pharma Limited.

Although obesity is closely associated with development and progression of type 2 diabetes mellitus (T2DM) and its complications, food group intake associated with obesity in diabetic patients has been very poorly investigated to date largely due to lack of large-scale databases with details on dietary intake. Thus, we aimed to determine intake of food groups by utilizing our nationwide database for people with diabetes with adjustment for confounders including energy intake. Analyzed were 1891 Japanese patients with T2DM (mean age, 62 y; men, 62%) who completed the Food Frequency Questionnaire. Obesity was defined by BMI>25. The associations between obesity and food groups were determined by multivariate quartile analysis. Among 15 food groups, both green and yellow and light-colored vegetables had significant negative associations with the presence of obesity among both men and women (Table). In addition, fruit and beans in men and fish in women were also negatively associated with obesity. Conversely, confectionery had positive associations with obesity in both men and women. Other food groups such as rice, meat or eggs did not show significant trends toward the presence of obesity. These results identified favorable dietary patterns for people with T2DM, which could be used for effective nutritional education to prevent obesity. Disclosure M. Hatta: None. K. Fujihara: None. Y. Takeda: None. R. Nedachi: None. D. Ishii: None. S.Y. Morikawa: None. C. Horikawa: None. M. Kato: None. H. Maegawa: Research Support; Self; Antares Pharma, Boehringer Ingelheim Pharmaceuticals, Inc., Daiichi Sankyo Company, Limited, Mitsubishi Tanabe Pharma Corporation, Takeda Pharmaceutical Company Limited. Speaker's Bureau; Self; Astellas Pharma Inc., Boehringer Ingelheim Pharmaceuticals, Inc., Daiichi Sankyo Company, Limited, Merck Sharp & Dohme Corp., Mitsubishi Tanabe Pharma Corporation, Takeda Pharmaceutical Company Limited. H. Sone: Research Support; Self; Astellas Pharma Inc., Boehringer Ingelheim Pharmaceuticals, Inc., Daiichi Sankyo Company, Limited, Kowa Pharmaceutical Europe Co. Ltd., Kyowa Hakko Kirin Co., Ltd., Novo Nordisk Inc., Ono Pharmaceutical Co., Ltd., Taisho Pharmaceutical Co., Ltd., Takeda Pharmaceutical Company Limited, Teijin Pharma Limited.

Hematological markers such as hematocrit (Hct), hemoglobin (Hb) or red blood cell (RBC) count were reported to be associated with the development of diabetic retinopathy with conflicting results. Moreover, which of these markers is the best to predict incident retinopathy remains unknown. To clarify this, we utilized ICD-10 and medical treatment codes from a nationwide claims database that included 7,182 participants (1,490 women, mean age 50.8 y, HbA1c 6.9%, mean follow-up period 5.5y) classified as having diabetes mellitus or vision-threatening treatment-required diabetic eye diseases (TRDED). Multivariate Cox regression model was used to identify predictors of TRDED. During the study period, 177 patients developed TRDED (4.5/1000 person-years). Cox analysis demonstrated that the lower Hcts were significant predictors TRDED and lower Hcts were more predictive than lower RBC counts. The Hb level was not significantly associated with TRDED. Hazard ratios (HRs) per 1 standard deviation (SD) increase in Hct, RBC count and Hb for incident TRDED were 0.792 (95% confidence interval [CI], 0.666-0.942), 0.874(0.765-0.998), and 0.861 (0.725-1.023) as continuous variables, respectively. The interaction between Hct and RBC count was not statistically significant. Multivariate tertile analysis also revealed that a lower Hct was more predictive of TRDED (HR 2.03 [95% CI, 1.33-3.08] for the bottom [≤43.4%] vs. the top [≥46.7%] tertile [P value = 0.001]) than lower RBC counts (HR 1.54 [95% CI, 0.99-2.38] for the bottom [≤468×104/m3] vs. the top [≥507×104/m3] [P value = 0.054]). These findings demonstrated that lower Hcts were significant predictor of vision-threatening retinopathy for people with diabetes and more useful than the RBC count or Hb. Association of the influence of these three markers on severe retinopathy awaits further investigation. Disclosure M. Yamamoto: None. K. Fujihara: None. T. Osawa: None. M. Harada: None. M. Ishizawa: None. H. Suzuki: None. H. Ishiguro: None. H. Seida: None. N. Yamanaka: None. Y. Matsubayashi: None. H. Sone: Research Support; Self; Astellas Pharma Inc., Boehringer Ingelheim Pharmaceuticals, Inc., Daiichi Sankyo Company, Limited, Kowa Pharmaceutical Europe Co. Ltd., Kyowa Hakko Kirin Co., Ltd., Novo Nordisk Inc., Ono Pharmaceutical Co., Ltd., Taisho Pharmaceutical Co., Ltd., Takeda Pharmaceutical Company Limited, Teijin Pharma Limited.