Applying masked autoencoder-based self-supervised learning for high-capability vision transformers of electrocardiographies.

The generalization of deep neural network algorithms to a broader population is an important challenge in the medical field. We aimed to apply self-supervised learning using masked autoencoders (MAEs) to improve the performance of the 12-lead electrocardiography (ECG) analysis model using limited ECG data. We pretrained Vision Transformer (ViT) models by reconstructing the masked ECG data with MAE. We fine-tuned this MAE-based ECG pretrained model on ECG-echocardiography data from The University of Tokyo Hospital (UTokyo) for the detection of left ventricular systolic dysfunction (LVSD), and then evaluated it using multi-center external validation data from seven institutions, employing the area under the receiver operating characteristic curve (AUROC) for assessment. We included 38,245 ECG-echocardiography pairs from UTokyo and 229,439 pairs from all institutions. The performances of MAE-based ECG models pretrained using ECG data from UTokyo were significantly higher than that of other Deep Neural Network models across all external validation cohorts (AUROC, 0.913-0.962 for LVSD, p < 0.001). Moreover, we also found improvements for the MAE-based ECG analysis model depending on the model capacity and the amount of training data. Additionally, the MAE-based ECG analysis model maintained high performance even on the ECG benchmark dataset (PTB-XL). Our proposed method developed high performance MAE-based ECG analysis models using limited ECG data.

Kidney outcomes with SGLT2 inhibitor vs. DPP4 inhibitor use in older adults with diabetes.

BACKGROUND AND HYPOTHESIS: While the kidney protective effects of sodium glucose co-transporter-2 (SGLT2) inhibitors have attracted much attention, there are limited real-world clinical data examining the effects of SGLT2 inhibitors on kidney function in older individuals. We aimed to compare the kidney outcomes between SGLT2 inhibitor and dipeptidyl peptidase 4 (DPP4) inhibitor use in older adults with diabetes. METHODS: Using a nationwide claims database, we studied 6 354 older adults (≥ 60 years of age) who had diabetes and newly initiated on SGLT2 inhibitors or DPP4 inhibitors. A 1:4 propensity score matching algorithm was used to compare changes in eGFR between SGLT2 inhibitor and DPP4 inhibitor users. The primary outcome was a decline in the rate of estimated glomerular filtration rate (eGFR), which was obtained using a linear mixed-effects model with an unstructured covariance. RESULTS: Following propensity score matching, 6 354 individuals including 1 271 SGLT2 inhibitor users and 5 083 DPP4 inhibitor users (median age: 68 [65-70] years); men, 60.4%; median eGFR:69.0 [59.1-79.0] ml/min/1.73 m2, median hemoglobin A1c [HbA1c]:6.9 [6.5-7.4]%) were analyzed. SGLT2 inhibitor users had a slower eGFR decline than did DPP4 inhibitor users (-0.97 [95% CI, -1.24 to -0.70] ml/min/1.73m2 vs. -1.83 [95% CI, -1.97 to -1.69] ml/min/1.73m2 per year; p for interaction < 0.001). This finding remained consistent across subgroups based on age, sex, body mass index, HbA1c level, renin-angiotensin system inhibitor use, and baseline eGFR. Additionally, the risk of a ≥ 20%, ≥ 30%, and ≥ 40% decrease in eGFR from baseline was significantly lower in SGLT2 inhibitor users than that in DPP4 inhibitor users. CONCLUSIONS: Our analysis, utilizing a nationwide epidemiological dataset, demonstrated that the decline in eGFR was slower in individuals aged ≥ 60 years with diabetes who were prescribed SGLT2 inhibitors compared to those prescribed DPP4 inhibitors, suggesting a potential advantage of SGLT2 inhibitors for kidney outcomes even in older individuals with diabetes.

Differences in the Association of Lifestyle-Related Modifiable Risk Factors with Incident Cardiovascular Disease Between Individuals with and without Diabetes.

AIMS: Individuals with diabetes have a high risk of developing cardiovascular disease (CVD). Little was known whether the association between modifiable risk factors and incident CVD would change according to the presence of diabetes. METHODS: In this study, we analyzed 4,132,006 individuals including 173,262 individuals (4.2%) with diabetes registered in the JMDC Claims Database, and compared the association between modifiable risk factors and risk of CVD between individuals with and without diabetes. RESULTS: The median age was 44 years, and 57.5% were men. Multivariable Cox regression analyses showed that the relationship of obesity, hypertension, and dyslipidemia with incident CVD was attenuated in individuals with diabetes, whereas that of non-ideal eating habits, smoking, and physical inactivity with incident CVD was pronounced in those with diabetes. The hazard ratio per 1-point increase in non-ideal lifestyle-related factors was 1.03 [95% confidence interval (CI) 1.03-1.04] in individuals with non-diabetes, whereas 1.09 [95% CI 1.07-1.11] in individuals with diabetes (p-value for interaction < 0.001). Further, hazard ratios for developing CVD were 1.02 [95% 1.01-1.04] in individuals not having diabetes, whereas 1.09 [95% CI 1.04-1.13] in individuals having diabetes for the increase of lifestyle-related factor after 1-year follow-up (p-value for interaction 0.007). CONCLUSION: Our analysis utilizing a nationwide epidemiological dataset presented that the relationship of lifestyle-related factors with incident CVD would be pronounced in people having diabetes, suggesting that the maintenance of a healthy lifestyle would play a more important role in the development of CVD in individuals having diabetes. (244 words).

Our investigation utilizing a nationwide epidemiological cohort showed a pronounced relationship of lifestyle-related factors with incident CVD in individuals with diabetes. The HRs (95% CI) for the occurrence of CVD events showed a progressive increase with each additional lifestyle-related factor. This trend was more prominent among individuals with diabetes than those without diabetes. The association between changes in the number of lifestyle-related factors over a year and the risk of developing CVD was also more pronounced in individuals with diabetes. These results suggest that maintaining healthy lifestyle habits would be more important for the CVD prevention in individuals having diabetes.

BP classification using the 2017 ACC/AHA guideline with risk of cardiovascular events in older individuals.

BACKGROUND: The association between stage 1 hypertension and the risk of cardiovascular disease (CVD) has not been established in older adults. Furthermore, little is known about whether lowering blood pressure (BP) is beneficial in older adults with stage 1 hypertension. METHODS: This cohort study analyzed nationwide data collected from the Japanese DeSC database, including 476,654 individuals aged ≥60 years. Individuals were categorized into four groups according to the 2017 ACC/AHA BP guidelines: normal BP, elevated BP, stage 1 hypertension, and stage 2 hypertension. The primary outcome was a composite CVD event, including myocardial infarction, angina pectoris, stroke, and heart failure. RESULTS: During a mean follow-up of 3.1 years, 53,946 composite CVD events were recorded. Hazard ratios of stage 1 hypertension for composite CVD events, myocardial infarction, angina pectoris, stroke, and heart failure were 1.10 (95 % CI, 1.07-1.13), 1.16 (95 % CI, 1.03-1.31), 1.06 (95 % CI, 1.01-1.10), 1.13 (95 % CI, 1.08-1.18), and 1.13 (95 % CI, 1.09-1.16), respectively. Individuals with a ≥ 5 mmHg decrease in systolic BP over one year had a lower risk of stroke among individuals with stage 1 hypertension. The positive association between stage 1 hypertension and composite CVD events was attenuated in individuals aged ≥75 years. CONCLUSIONS: Stage 1 hypertension is associated with a higher risk of developing CVD events among older adults. The 2017 ACC/AHA BP guidelines could be applied to older populations; however, the applicability of these guidelines to older adults aged ≥75 years requires further investigations.

Body weight change associated kidney outcomes of sodium-glucose cotransporter new users.

AIM: To investigate the clinical significance of body weight changes on kidney outcomes among individuals with diabetes using sodium-glucose cotransporter-2 (SGLT2) inhibitors. MATERIALS AND METHODS: This is a retrospective cohort study using a nationwide epidemiological database, and we conducted an analysis involving 11 569 individuals with diabetes who were newly prescribed SGLT2 inhibitors. The main outcome was the rate of decline in estimated glomerular filtration rate (eGFR), determined through a linear mixed-effects model with an unstructured covariance structure. RESULTS: The median age of the patients was 52 (Q1-Q3: 47-58) years, and the median fasting plasma glucose and glycated haemoglobin (HbA1c) levels were 144 (Q1-Q3: 124-175) mg/dL and 7.4 (Q1-Q3: 6.8-8.3)%, respectively. The median estimated eGFR was 77.7 (Q1-Q3: 67.2-89.1) mL/min/1.73 m2. The median follow-up period was 1.7 (Q1-Q3: 1.0-2.6) years. Participants were stratified into three groups based on the body mass index change rate tertiles between baseline and 1 year after (tertile 1: <-4.55%, tertile 2: -4.55% to -1.43%, tertile 3: >-1.43%). The annual change in eGFR was -0.78 (-0.94 to -0.63) mL/min/1.73 m2 in tertile 1, -0.95 (-1.09 to -0.81) mL/min/1.73 m2 in tertile 2, and -1.65 mL/min/1.73 m2 (-1.84 to -1.47) in tertile 3 (pinteraction < 0.001). A variety of sensitivity analyses confirmed the relationship between the 1-year body mass index decrease and favourable kidney outcomes after SGLT2 inhibitor administration. CONCLUSIONS: Our analysis of a nationwide epidemiological cohort revealed that kidney outcomes following the initiation of SGLT2 inhibitors would be more favourable, with greater body weight loss observed after the initiation of SGLT2 inhibitors.

Current Status and Future Perspective of Onco-Hypertension.

Hypertension and cancer are both increasing with age. Recently, the new concept of "Onco-Hypertension" has been proposed to address the mutual risks posed by hypertension and cancer and to provide comprehensive care for patients with these two conditions in an aging society. Hypertension and cancer share common risk factors and may be interrelated in pathogenesis: hypertension is involved in the development of certain cancers, and cancer survivors have a higher incidence of hypertension. With recent advances in cancer therapy, the number of cancer survivors has increased. Cancer survivors not only have a higher risk of incident hypertension but also an increased risk of future cardiovascular events, highlighting the growing importance of comprehensive care. In this review, we provide an overview of the current status and future perspective of the "Onco-Hypertension," including our research findings.

A leadless pacemaker implantation for a patient with systemic right ventricle under ventricular assist device support.

Patients with congenitally corrected transposition of the great arteries (ccTGA) often develop complete atrioventricular block and heart failure due to the abnormal disposition of atrioventricular node and disadvantage of systemic right ventricle. These issues are managed with a pacing system and a ventricular assist device (VAD), respectively. While technological advances offer new treatment strategies, the simultaneous deployment of a leadless pacemaker and a VAD in cases of ccTGA remains unexplored. Here, we present a case of leadless pacemaker implantation for a VAD-supported ccTGA patient. The safety of a leadless pacemaker for a subpulmonary left ventricle and electromagnetic interference between devices are major concerns when implanting a leadless pacemaker; however, the current case overcomes these obstacles. There were no perioperative complications, and both devices were functioning without problems during a one-year follow up. We expect that, even in patients with cardiac complexity such as systemic right ventricle under VAD support, a leadless pacemaker could become the treatment of choice if the indication is appropriate, although careful and close follow up is needed. Learning objective: Technological advances expand treatment strategies and provide significant benefits to patients with adult congenital heart disease (ACHD). However, discussion of the combination of a leadless pacemaker and a ventricular assist device (VAD) is rare. We demonstrated the efficacy of a leadless pacemaker for a subpulmonary left ventricle in a patient with systemic right ventricle on VAD. This approach could be an option even for ACHD patients.

Body Fat Distribution and Left Atrial Reverse Remodeling After Catheter Ablation for Atrial Fibrillation.

Background: Emerging evidence suggests a pathophysiological link between obesity and atrial fibrillation (AF). However, the contribution of body fat distribution to left atrial (LA) remodeling and its reversibility remain unclear in nonobese AF patients. Objectives: The purpose of this study was to investigate the association of body fat distribution with LA size and reverse remodeling (LARR). Methods: In total, 116 nonobese patients with AF (88 men, age 63 ± 11 years) who underwent first catheter ablation (CA) were included. Body fat distribution was assessed with bioelectrical impedance, and body fat percentage (BF%) and central fat percentage (CF%) were calculated. Patients were categorized by body size metrics (body mass index [BMI] and waist-to-hip [W/H] ratio) and fat parameters (BF% and CF%). Echocardiography was performed before and 6 months after CA. Multivariable logistic regression was used to examine the association between the 4 metrics (ie, BMI, W/H ratio, BF%, and CF%) and a lack of LARR (<15% reduction or increase in the LA volume index). Results: Body size metrics and adiposity measures were not independently associated with baseline LA size. Six months after CA, the higher W/H ratio and CF% groups exhibited persistent LA enlargement compared to their counterparts (both P < 0.01). In the multivariable analysis, W/H ratio and CF% were associated with a lack of LARR (adjusted ORs of 3.86 and 2.81 per 0.10 and 10% increase, respectively, both P < 0.01). The combined assessment of CF% with W/H ratio provided complementary risk stratification for persistent LA enlargement. Conclusions: Central adiposity was associated with a lack of LARR after CA, highlighting the importance of assessing body fat distribution even in nonobese patients.

Significance of eGFR and proteinuria for cardiovascular disease in individuals beyond 85 years.

BACKGROUND: There are limited data on how advancing age influences prediction of CVD risk based on the estimated glomerular filtration rate (eGFR) and proteinuria, especially in older adults, including those aged ≥ 85 years. This study aimed to clarify the association of eGFR and proteinuria with CVD outcomes and the impact of age on this association. METHODS: The distribution of eGFR and urine protein in Japan was assessed retrospectively using real-world administrative claims and health checkup data collected between April 2014 and November 2022. We investigated the associations of these two parameters with the incidence of CVD, with an emphasis on the impact of aging. RESULTS: We assessed 1 829 020 individuals for distribution of eGFR and proteinuria; after excluding those with known CVD, their association with CVD risk was examined in 1 040 101 individuals aged ≥ 40 years. The prevalence of impaired kidney function (eGFR <60 mL/min/1.73 m2) increased with age, being 0.7%, 9.2%, 21.9%, 40.2%, and 60.2% at the ages of 18-39, 40-64, 65-74, 75-84, and ≥ 85 years (P for trend < 0.001); similarly, the proportion with positive proteinuria increased with age, being 2.7%, 4.3%, 5.6%, 9.2%, and 15.8%, respectively (P for trend < 0.001). Both eGFR and urine protein were identified to be independent risk factors for CVD. Hazard ratios for CVD increased significantly when eGFR was <45 mL/min/1.73 m2 at the ages of 40-64, 65-74, and 75-84 and <30 mL/min/1.73 m2 at ≥ 85 years, while proteinuria remained significantly associated with a high CVD risk regardless of age. These findings were consistent even when analyzed separately by sex. CONCLUSIONS: This study identified eGFR and urine dipstick proteinuria to be independent risk factors for CVD, even among individuals aged ≥ 85 years. However, the contribution of eGFR to the CVD risk was attenuated by aging, whereas proteinuria remained less affected by advancing age.

Age-Dependent Relationship of Physical Inactivity With Incident Cardiovascular Disease: Analysis of a Large Japanese Cohort.

BACKGROUND: There have been limited studies examining age-dependent associations between physical inactivity and cardiovascular disease (CVD). We aimed to clarify the age-dependent relationship of physical inactivity with incident CVD. METHODS: We analyzed 1,097,424 participants, aged 18 to 105 years, without histories of CVD, enrolled in the DeSC database (median age, 63 years; 46.4% men). We categorized participants into the following 4 groups based on age: ≤ 44 years (n = 203,835); 45 to 64 years (n = 403,619); 65 to 79 years (n = 437,236); and ≥ 80 years (n = 52,734). We used 3 physical inactivity components gained from the self-reported questionnaire during a health checkup. The outcomes were composite CVD events including myocardial infarction, stroke, heart failure, and each CVD event. RESULTS: During a mean follow-up of 3.2 ± 1.9 years, 81,649 CVD events were observed. The hazard ratios of 3 physical inactivity components for CVD events increased with age category (P for interaction < 0.001). For example, the hazard ratio (95% confidence interval) of physical inactivity defined as not doing light sweaty exercise for 30 minutes at least twice a week for incident CVD in the groups aged ≤ 44 years, 45 to 64 years, 65 to 79 years, and ≥ 80 years were 0.97 (0.88-1.05), 1.08 (1.05-1.12), 1.12 (1.10-1.15), and 1.17 (1.12-1.21), respectively (P for interaction < 0.001). This association was consistent across subtypes of CVD including heart failure, myocardial infarction, and stroke. CONCLUSIONS: The association of physical inactivity with a higher risk of developing CVD increased with age. Preventive efforts for physical activity optimization may be more valuable in older people.

Chronic Tonsillitis and IgA Nephropathy: Findings From a Nationwide Japanese Cohort Study.

RATIONALE & OBJECTIVE: Little is known regarding the association between chronic tonsillitis and the onset of IgA nephropathy (IgAN). In the present study, we examined the potential relationship between chronic tonsillitis and a subsequent risk of developing IgAN. STUDY DESIGN: Observational cohort study. SETTING & PARTICIPANTS: 4,311,393 individuals without a history of IgAN identified between January 2005 and May 2022 within a Japanese nationwide epidemiological database, the JMDC Claims Database, representing health claims to over 60 insurers. EXPOSURE: Comorbid chronic tonsillitis based on diagnosis codes. OUTCOME: IgAN occurrence. ANALYTICAL APPROACH: Cause-specific Cox proportional hazards analysis adjusting for potential confounding factors was employed to estimate hazard ratios (HRs). RESULTS: Comorbid chronic tonsillitis was identified in 12,842 individuals, constituting 0.3% of the cohort. The cohort had a median age of 44 years (IQR, 36-53), and males accounted for 57.9%, with a follow-up of 1,089 days (IQR, 532-1,797), during which 2,653 cases of IgAN developed. Cumulative incidence curve showed a higher cumulative incidence of IgAN in individuals with chronic tonsillitis compared with their counterparts without this condition. Multivariable cause-specific analysis further demonstrated that individuals with chronic tonsillitis had an elevated risk of developing IgAN, with HR of 2.72 (95% CI, 1.79-4.14). LIMITATIONS: Potential residual confounders, and lack of consideration for ethnic distinctions. CONCLUSIONS: Using a large-scale epidemiological dataset, these findings suggest a relationship between chronic tonsillitis and an elevated risk of IgAN development in the general Japanese population. PLAIN-LANGUAGE SUMMARY: IgA nephropathy (IgAN), the most prevalent form of primary glomerulonephritis worldwide, is associated with unfavorable long-term kidney survival and life expectancy. Despite the substantial implications, the early detection of IgAN still remains challenging due to its commonly asymptomatic clinical presentation. Consequently, the exploration of risk factors assumes a critical research priority. Prior studies have reported the potential role of tonsilitis in the pathogenesis of IgAN. In this study, we assessed whether chronic tonsillitis was associated with the subsequent development of IgAN using a nationwide epidemiological dataset incorporating over 4,000,000 individuals. Within this large-scale cohort, our findings revealed an association between a history of tonsillitis and a greater risk of developing IgAN. These findings should heighten awareness of the potential susceptibility of people with chronic tonsilitis to IgAN.

Change in Body Mass Index and Cardiovascular Outcomes in Patients With Cancer.

OBJECTIVE: To investigate the association between body mass index (BMI) changes and the risk of cardiovascular disease (CVD) in patients with cancer. PATIENTS AND METHODS: This retrospective observational study used data from the JMDC Claims Database obtained between January 2005, and April 2021. We included 52,344 individuals (median [IQR] age, 53 years [46 to 60 years]; 23,584 [45.1%] men) with cancer and no prior CVD. Patients were classified into 3 groups based on the percentage change in BMI from the initial health checkup to the checkup 1 year later: -5.0% or less (BMI loss), -5.0% to 5.0% (stable BMI), and 5.0% or more (BMI gain). The primary end point was composite CVD events including heart failure, atrial fibrillation, ischemic heart disease, and stroke. RESULTS: During a median follow-up period of 763 days (IQR, 369 to 1274 days), 3124 composite CVD events were observed. Compared with stable BMI, the hazard ratios (HRs) of BMI loss and gain for CVD events were 1.16 (95% CI, 1.00 to 1.34) and 1.10 (95% CI, 0.96 to 1.25), respectively. A U-shaped association was observed between the BMI changes and CVD events, particularly for nonatherosclerotic CVD outcomes including heart failure and atrial fibrillation. Compared with stable BMI, both BMI loss and gain increased the risk of heart failure (HR, 1.30; 95% CI, 1.08 to 1.57 and HR, 1.22; 95% CI, 1.02 to 1.47, respectively) and atrial fibrillation (HR, 1.70; 95% CI, 1.18 to 2.45 and HR, 1.55; 95% CI, 1.07 to 2.24, respectively). CONCLUSION: Cancer survivors with BMI loss and gain were at greater risk of CVD. Body mass index loss is associated with a higher risk of CVD.

Heart failure promotes multimorbidity through innate immune memory.

Patients with heart failure (HF) often experience repeated acute decompensation and develop comorbidities such as chronic kidney disease and frailty syndrome. Although this suggests pathological interaction among comorbidities, the mechanisms linking them are poorly understood. Here, we identified alterations in hematopoietic stem cells (HSCs) as a critical driver of recurrent HF and associated comorbidities. Bone marrow transplantation from HF-experienced mice resulted in spontaneous cardiac dysfunction and fibrosis in recipient mice, as well as increased vulnerability to kidney and skeletal muscle insults. HF enhanced the capacity of HSCs to generate proinflammatory macrophages. In HF mice, global chromatin accessibility analysis and single-cell RNA-seq showed that transforming growth factor-ß (TGF-ß) signaling was suppressed in HSCs, which corresponded with repressed sympathetic nervous activity in bone marrow. Transplantation of bone marrow from mice in which TGF-ß signaling was inhibited similarly exacerbated cardiac dysfunction. Collectively, these results suggest that cardiac stress modulates the epigenome of HSCs, which in turn alters their capacity to generate cardiac macrophage subpopulations. This change in HSCs may be a common driver of repeated HF events and comorbidity by serving as a key carrier of "stress memory."

Awareness of Being Prescribed Antihypertensive Medications and Cardiovascular Outcomes.

BACKGROUND: Hypertension is a major cause of cardiovascular disease (CVD). In patients with hypertension, unawareness of the disease often results in poor blood pressure control and increases the risk of CVD. However, data in nationwide surveys regarding the proportion of unaware individuals and the implications of such on their clinical outcomes are lacking. We aimed to clarify the association between unawareness of being prescribed antihypertensive medications among individuals taking antihypertensive medications and the subsequent risk of developing CVD.Methods and Results: This retrospective cohort study analyzed data from the JMDC Claims Database, including 313,715 individuals with hypertension treated with antihypertensive medications (median age 56 years). The primary endpoint was a composite of myocardial infarction, angina pectoris, stroke, heart failure, and atrial fibrillation. Overall, 19,607 (6.2%) individuals were unaware of being prescribed antihypertensive medications. During the follow-up period, 33,976 composite CVD endpoints were documented. Despite their youth, minimal comorbidities, and the achievement of better BP control with a reduced number of antihypertensive prescriptions, unawareness of being prescribed antihypertensive medications was associated with a greater risk of developing composite CVD. Hazard ratios of unawareness of being prescribed antihypertensive medications were 1.16 for myocardial infarction, 1.25 for angina pectoris, 1.15 for stroke, 1.36 for heart failure, and 1.28 for atrial fibrillation. The results were similar in several sensitivity analyses, including the analysis after excluding individuals with dementia. CONCLUSIONS: Among individuals taking antihypertensive medications, assessing the awareness of being prescribed antihypertensive medications may help identify those at high risk for CVD-related events.

Sex Differences in the Association Between Depression and Incident Cardiovascular Disease.

Background: Depression is a known risk factor for cardiovascular disease (CVD), but the potential sex differences in this association remain unclear. Objectives: The aim of this study was to investigate the association between depression and subsequent CVD events, and to explore potential sex differences. Methods: The authors conducted a retrospective analysis using the JMDC Claims Database between 2005 and 2022. The study population included 4,125,720 individuals aged 18 to 75 years without a history of cardiovascular disease or renal failure and missing data at baseline. Participants were followed up for a mean of 1,288 days to assess the association between depression and subsequent CVD events, such as myocardial infarction, angina pectoris, stroke, heart failure, and atrial fibrillation. Results: Our analysis revealed a significant association between depression and subsequent composite CVD events in both men and women, with a stronger association observed in women. The HR for the composite endpoint was 1.64 (95% CI: 1.59-1.70) in women and 1.39 (95% CI: 1.35-1.42) in men after multivariable adjustment (P for interaction <0.001). Furthermore, the individual components of the composite endpoint were also associated with depression in both men and women, each of which was also observed to be more strongly associated in women. Conclusions: Our study provides evidence of a significant association between depression and subsequent CVD events in both men and women, with a more pronounced association observed in women. These findings highlight the importance of addressing depression and tailoring prevention and management strategies according to sex-specific factors.

Prediction of adverse cardiovascular events in children using artificial intelligence-based electrocardiogram.

BACKGROUND: Convolutional neural networks (CNNs) have emerged as a novel method for evaluating heart failure (HF) in adult electrocardiograms (ECGs). However, such CNNs are not applicable to pediatric HF, where abnormal anatomy of congenital heart defects plays an important role. ECG-based CNNs reflecting neurohormonal activation (NHA) may be a useful marker of pediatric HF. This study aimed to develop and validate an ECG-derived marker of pediatric HF that reflects the risk of future cardiovascular events. METHODS: Based on 21,378 ECGs from 8324 children, a CNN was trained using B-type natriuretic peptide (BNP) and the occurrence of major adverse cardiovascular events (MACEs). The output of the model, or the electrical heart failure indicator (EHFI), was compared with the BNP regarding its ability to predict MACEs in 813 ECGs from 295 children. RESULTS: EHFI achieved a better area under the curve than BNP in predicting MACEs within 180 days (0.826 versus 0.691, p = 0.03). On Cox univariable analyses, both EHFI and BNP were significantly associated with MACE (log10 EHFI: hazard ratio [HR] = 16.5, p < 0.005 and log10 BNP: HR = 4.4, p < 0.005). The time-dependent average precisions of EHFI in predicting MACEs were 32.4%-67.9% and 1.6-7.5-fold higher than those of BNP in the early period. Additionally, the MACE rate increased monotonically with EHFI, whereas the rate peaked at approximately 100 pg/mL of BNP and decreased in the higher range. CONCLUSIONS: ECG-derived CNN is a novel marker of HF with different prognostic potential from BNP. CNN-based ECG analysis may provide a new guide for assessing pediatric HF.

Comparison of incident hypertension between SGLT2 inhibitors vs. DPP4 inhibitors.

Although several randomized clinical trials have reported the potential benefit of sodium-glucose cotransporter 2 inhibitors (SGLT2i) in reducing blood pressure (BP), whether SGLT2i can reduce incident hypertension is unknown. We analyzed individuals with diabetes who were newly prescribed SGLT2i or dipeptidyl peptidase 4 inhibitors (DPP4i) in a large-scale epidemiological database. The primary outcome was the incidence of hypertension. A propensity score matching algorithm was employed to compare the subsequent development of hypertension between the SGLT2i and DPP4i groups. After propensity score matching, 5708 well-balanced pairs of SGLT2i and DPP4i users were identified. SGLT2i administration was associated with a reduced risk of hypertension (HR 0.91, 95% CI: 0.84-0.97). The advantage of SGLT2i use over DPP4i use for incident hypertension was generally consistent in several sensitivity analyses, and subgroup analyses showed that SGLT2i use was significantly associated with a lower risk of hypertension in men, patients with baseline HbA1c of <7.5%, and baseline systolic blood pressure ≥127 mmHg. Our investigation using nationwide real-world data demonstrated the potential advantage of SGLT2i over DPP4i in reducing the development of hypertension in individuals with diabetes.