Published papers

Context: Previous metabolomics studies suggest potential associations between menopausal changes in lipids and an increased risk of metabolic syndrome (MetS). However, longitudinal data on other key metabolites, such as branched-chain amino acids (BCAAs) and homocysteine, remain limited, and most studies lack long-term follow-up across the menopause transition. Objective: This study aimed to investigate longitudinal changes in circulating metabolites during menopause over a mean follow-up of 5 years and assess their associations with subsequent MetS development. Methods: Premenopausal women from the Tsuruoka Metabolomics Cohort Study who participated in at least one follow-up survey were included. Menopausal status, data on MetS, and plasma metabolites profiled using capillary electrophoresis mass spectrometry were assessed at each visit. Thirty-one metabolites were examined for associations with menopausal status using mixed-effects models. The association of these menopause-related metabolites with MetS development was examined via logistic regression analysis adjusted for follow-up duration. Results: Among 953 women (aged 43.8 ± 5.4 years), 316 (33.2%) reached menopause during follow-up (5.0 ± 1.1 years). Eighteen metabolites changed significantly with menopause, particularly those related to BCAA metabolism, urea cycle, and homocysteine metabolism. Of 695 women without MetS at baseline, 65 (9.4%) developed MetS. Glutamate (odds ratio [95% CI]: 1.95 [1.49-2.57]) was associated with higher MetS risk. Higher levels of glutamate, valine, leucine, and cystine were significantly associated with the development of hyperglycemia. Conclusion: Longitudinal changes in charged metabolites occur across the menopausal transition, with specific metabolites such as glutamate possibly contributing to the metabolic alterations underlying increased MetS risk.

BACKGROUND The role of the apoptosis inhibitor of macrophages (AIM) in human lipid metabolism remains unclear. OBJECTIVE This study aimed to investigate the cross-sectional associations between serum AIM and plasma metabolites and to determine if baseline AIM concentrations predict incident dyslipidemia. METHODS This study used a community-based cohort of 3139 participants for cross-sectional metabolomic analysis. Among them, 1292 participants without dyslipidemia at baseline were followed prospectively for up to 8.5 years. Cox proportional hazards models were used to estimate hazard ratios for incident high low-density lipoprotein cholesterol and metabolic dyslipidemia (high triglycerides and/or low high-density lipoprotein cholesterol [HDL-C]) across AIM tertiles. RESULTS Cross-sectionally, higher AIM concentrations were associated with elevated acylcarnitines and tricarboxylic acid cycle intermediates. Longitudinally, higher baseline AIM was associated with an increased risk of developing metabolic dyslipidemia in women (highest vs lowest tertile hazard ratio, 1.84) and showed a similar but smaller nonsignificant trend in men. The risk for high low-density lipoprotein cholesterol was significantly increased in women but not in men. These associations remained robust after multivariable adjustment. CONCLUSION Elevated baseline AIM concentrations were prospectively associated with the development of metabolic dyslipidemia in women, with a similar but nonsignificant trend in men, and with high LDL-C in women. The related metabolomic profile suggests enhanced lipolysis and altered lipid metabolism. Therefore, AIM may serve as a biomarker for future dyslipidemia, particularly for triglyceride and HDL-C dysregulation.

OBJECTIVE: Body weight variability is linked to cardiometabolic outcomes, but its renal impact in type 2 diabetes remains uncertain. We tested whether the magnitude of seasonal BMI fluctuation is independently associated with kidney function decline. RESEARCH DESIGN AND METHODS: We analyzed a nationwide, multicenter Japanese cohort (2014-2020). Monthly BMI was modeled using seasonal-trend locally estimated scatterplot smoothing to quantify each participant's within-year peak-to-trough difference. The primary outcome was ≥40% decline in estimated glomerular filtration rate (eGFR). Secondary outcomes were ≥30% eGFR decline, creatinine doubling, incident chronic kidney disease (CKD) stage ≥3, and kidney failure. Associations were estimated using multivariable Cox models with clinic as a random effect. RESULTS: Among 6,700 outpatients (median follow-up: 6.8 years), 779 reached the primary end point. Each 1-SD increase in BMI fluctuation was associated with higher risk of ≥40% eGFR decline (hazard ratio [HR] 1.23, 95% CI 1.16-1.31). The highest versus lowest tertile showed a 1.7-fold increased risk (HR 1.72, 95% CI 1.42-2.09). Patterns were consistent for ≥30% eGFR decline (HR 1.18, 95% CI 1.13-1.23), creatinine doubling (HR 1.30, 95% CI 1.17-1.45), and incident CKD stage ≥3 (HR 1.11, 95% CI 1.07-1.16). Longitudinal analyses showed steeper eGFR decline in the highest-fluctuation group. Results were robust across sensitivity analyses, including models for time-varying medication exposure. CONCLUSIONS: In type 2 diabetes, larger intra-annual BMI fluctuations were independently and dose-dependently associated with kidney function decline. Seasonal BMI amplitude may identify higher-risk individuals; whether reducing seasonal BMI fluctuations improves kidney outcomes warrants prospective evaluation.

Purpose: Whether hyperuricemia (HU) independently predicts cardiovascular disease (CVD) remains debated, whereas chronic kidney disease (CKD) is a well-established risk factor. This study assessed the independent and mediated association of HU in CKD on long-term CVD mortality in a 30-year nationwide Japanese cohort. Methods: Data from NIPPON DATA90 participants aged ≥ 30 years enrolled in 1990 and followed for up to 30 years were used. HU was defined as serum uric acid ≥ 416.4 µmol/L, and CKD as an estimated glomerular filtration rate < 60 mL/min/1.73 m² or positive proteinuria. Four baseline categories were created (neither, HU-only, CKD-only, both conditions), and analyzed via Cox models to estimate hazard ratios (HR) for CVD mortality, adjusted for age, sex, hypertension, diabetes mellitus, dyslipidemia, BMI, and history of smoking/alcohol. Results: Among 7,336 participants (58.5% women; mean age 52.5 years), 807 CVD deaths occurred over a mean 24.2 years. At baseline, 86.6% had neither condition, 4.9% HU-only, 7.1% CKD-only, and 1.5% both. HU-only was not significantly associated with CVD mortality (HR 1.14, 0.81–1.61), CKD-only had significant association (1.28, 1.04–1.59), and both HU/CKD conditions had the highest risk (2.18, 1.50–3.18). Further analyses showed that HU-only was significant in women (2.44, 1.13–5.28) but not men (1.04, 0.71–1.52). Conclusion: Overall, CKD and coexistent HU/CKD showed to be a significant CVD mortality risk, and HU’s association with CVD mortality may be partly mediated through CKD. This study observed that HU and CKD coexistence heighten the risk of CVD mortality. Integrated management of serum urate and renal health may aid long-term CVD prevention.

Background & Aims: Aminotransferases are widely used for metabolic dysfunction-associated steatotic liver disease (MASLD) evaluation, especially in type 2 diabetes mellitus (T2DM). Whether within-person seasonal variation affects classification near commonly used thresholds or relates to long-term metabolic outcomes remains unclear. Methods: This registry-based cohort analysed monthly aspartate aminotransferase (AST) and alanine aminotransferase (ALT) measurements from 6039 adults with T2DM in the Japan Diabetes Clinical Data Management registry (2014–2020). Classification discordance across the 30 IU/L threshold was compared between winter-mean and summer-mean values. Individual seasonal amplitude was derived from seasonal-trend decomposition of multiply imputed monthly time series. Final glycated haemoglobin (HbA1c) and non-achievement of HbA1c < 7% were analysed using multivariable regression. Results: Both AST and ALT showed significant seasonal variation, with the highest values in late autumn to early winter and the lowest in summer (p < 0.001). Approximately one in nine patients with borderline ALT values showed discordant winter-summer classification. Each 1-SD increase in AST amplitude was associated with 0.06 percentage points higher final HbA1c (95% CI, 0.04–0.08) and higher odds of not achieving HbA1c < 7% (odds ratio, 1.14; 95% CI, 1.08–1.21; p < 0.001). ALT amplitude showed a similar but weaker, less consistent association. Conclusions: AST and ALT exhibited reproducible seasonal variation peaking in late autumn to early winter, with approximately one in nine patients near the MASLD screening threshold reclassified depending on season. Greater seasonal amplitude, especially AST, was independently associated with poorer glycemic control, supporting season-aware interpretation in routine clinical practice.

Although metabolic syndrome, including visceral fat accumulation, causes kidney and cardiovascular diseases, the impact of visceral fat accumulation on mild decreased renal function remains unclear. This study examines the association between visceral fat area (VFA) measured by bioimpedance methods and the estimated glomerular filtration rate based on serum cystatin C (eGFRcys) in the Japanese urban population. This community-based cross-sectional study enrolled 952 individuals (287 men, 665 women) who participated in the second follow-up survey of the Kobe Orthopedic and Biomedical Epidemiological (KOBE) study. We compared the multivariate-adjusted means of eGFRcys among VFA quartile groups by gender using the analysis of covariance. Models were adjusted for age, high blood pressure, hypercholesterolemia, glucose intolerance, smoking, and alcohol use, and further adjusted for body mass index (BMI). The highest VFA quartile group had lower eGFRcys than the lowest VFA quartile group after adjusted for cardiometabolic risk factors, except for BMI (93.1 [95% confidence interval (CI), 90.1-96.2] vs. 82.1 [95% CI, 79.1-85.0] in men and 95.8 [95% CI, 94.1-97.5] vs. 89.4 [95% CI, 87.8-90.9] in women). Moreover, further adjustment for BMI revealed a similar result in men (93.5 [95% CI, 89.8-97.2] vs. 81.6 [95% CI, 77.9-85.3]), while no significant association was found in women. This study suggests a significant association between increased VFA levels and lower eGFRcys levels independent of cardiometabolic risk factors, such as glucose intolerance and hypercholesterolemia in men and women, as well as independent of BMI in men.

BACKGROUND: There is limited knowledge regarding effective strategies, including information and communication technology (ICT)-based interventions, to reduce sodium intake and increase potassium intake in healthy individuals. METHODS: We conducted a 3-month randomized controlled trial involving healthy adult employees with spot urine sodium-to-potassium ratios (spot UNa/UK) ≥4.0 or estimated 24-hour salt intake ≥10g. Estimated 24-hour UNa and UK were calculated using the Tanaka formula. Participants were assigned to one of four groups: (i) online education, where participants monitored their spot UNa/UK and received feedback from dieticians (n = 84); (ii) messaging, with similar self-monitoring and dietician messages (n = 84); (iii) self-learning, provided with an educational leaflet (n = 87); and (iv) a control group (n = 87). The primary outcome was the change in spot UNa/UK ratios, and secondary outcomes included changes in estimated 24-hour UNa and UK. The trial protocol specified a hierarchical order for testing the interventions, anticipating the highest efficacy in the online education group. RESULTS: After the intervention, the online education group showed a decrease in spot UNa/UK ratios (mean -0.9 (95% CI: -1.8 to 0.0), P = 0.052) compared to the control group. The increase in estimated 24-hour UK excretion was larger in online education compared to the control group (mean + 2.5 mmol/day (95% CI: -0.3 to 5.3), P = 0.085). The difference in estimated 24-hour UNa excretion between the online education and control groups was -4.3 mmol/day (95% CI: -15.5 to 6.9, P = 0.45). CONCLUSIONS: Combining self-monitoring of sodium and potassium intake with ICT-based interventions, including online nutritional education, was associated with a modest reduction in the estimated ratios of sodium and potassium intake in healthy individuals.

Background: Hypertension is a major risk factor for cardiovascular disease. Mobile health applications may contribute to the primary prevention of hypertension. This systematic review and meta-analysis evaluated the association between the use of mobile health applications and the reduction of blood pressure (BP) in the non-patient adult population. Methods: Original articles published in English up to and including December 31, 2023, were searched on 4 databases. Studies in which the effect of mobile health applications on BP was estimated among non-patient adults were included. Random-effects meta-analyses were conducted using the standardized mean difference (SMD), which indicates the magnitude of change in BP relative to the variability within each study, rather than the absolute change in mm Hg. Analyses were conducted separately in 2 groups according to whether the applications were used in coordination with personalized counseling. Results: Among 1,244 studies initially identified, 10 studies with 1,464 participants met the inclusion criteria. Mobile health applications were associated with a reduction in diastolic BP (SMD: −0.27, 95% confidence interval: −0.42 to −0.12, 7 studies) for applications used in isolation, whereas applications used in coordination with counseling were associated with a reduction in both systolic BP (−0.24, −0.34 to −0.15, 3 studies) and diastolic BP (−0.26, −0.47 to −0.04, 3 studies), between baseline and follow-up among intervention groups. No significant difference was observed in BP at follow-up between intervention and control groups, probably due to the short follow-up period. Conclusions: Mobile health smartphone applications, especially those coupled with counseling, may reduce BP among non-patient adults.

Little is known about the associations between changes in the urinary sodium-to-potassium (Na/K) ratio and blood pressure (BP) in healthy individuals. Using survey data, urinary data, and BP data from the KOBE Study, this longitudinal study aimed to assess the associations between changes in the urinary Na/K ratio and BP in a healthy Japanese population over an 8-year follow-up period. We analyzed 567 participants aged 40–74 years who did not initiate antihypertensive or cardiovascular disease treatment during follow-up. Changes in the spot urinary Na/K ratio and BP were calculated by subtracting the baseline values from the follow-up measurements, and their associations were examined using a multivariable linear regression analysis that adjusted for sex, age, body mass index, baseline Na/K ratio and BP, low-density lipoprotein, hemoglobin A1c, ethanol intake, smoking status, salt taste sensitivity, years of education, employment status, baseline survey season, and 8-year follow-up survey season. The study cohort had a mean spot urinary Na/K ratio of 2.1 at baseline. While BP increased significantly over the 8-year follow-up period, the urinary Na/K ratio remained unchanged. However, urinary Na/K ratio changes were positively associated with changes in systolic BP (β = 1.48, p = 0.001) and diastolic BP (β = 0.76, p = 0.004). In conclusion, increases in spot urinary Na/K ratios were predictive of long-term BP elevation in a healthy, normotensive population with near-optimal urinary Na/K ratios at baseline. These findings indicate that maintaining a low urinary Na/K ratio is important for BP control, even among healthy individuals. (Figure presented.)

Aims: The association between alcohol consumption and the risk of cardiovascular disease (CVD) varies according to the presence of underlying cardiovascular risk factors. Incorporating such risk factors may be important to effectively reduce harmful alcohol use through health guidance. However, whether the risk of alcohol consumption is affected due to renal impairment remains unclear. Methods: A total of 10,583 community-dwelling Japanese adults (mean age 59.4 (SD 10.1) years; 46% men) were followed for 10 years. Alcohol consumption was categorized into five groups for men: never drinkers, former drinkers, light drinkers (＜20 g/day), moderate drinkers (20-39 g/day), and heavy drinkers (≥ 40 g/day), and four groups for women with moderate and heavy combined. Cox proportional hazards models estimated hazard ratios for incident CVD stratified by the presence or absence of reduced estimated glomerular filtration rate (eGFR) (＜60 mL/min/1.73m2) and proteinuria, with adjustment for confounders. Results: Among men with proteinuria, alcohol consumption was associated with a higher risk of atherosclerotic CVD, whereas an inverse association was observed in men without proteinuria [hazard ratio (95% confidence interval): moderate drinkers without proteinuria, 0.58 (0.34-0.97); moderate drinkers with proteinuria, 3.49 (1.15-10.56)]. Moderate to heavy drinking increased the risk of intracerebral hemorrhage irrespective of the renal status. In women, moderate to heavy drinking was associated with an increased CVD risk only when proteinuria was present. In contrast, a reduced eGFR did not clearly affect the association in either sex. Conclusions: The CVD risk associated with alcohol consumption may differ according to the renal status, particularly depending on the presence or absence of proteinuria.