Jun Takahashi

Background: Patients with established atherosclerotic cardiovascular disease (ASCVD) remain at substantial risk of recurrent events despite secondary prevention. Risk stratification may identify patients requiring more intensive prevention. However, most secondary prevention risk models were developed in Western populations and may not reflect the risk profile of Japanese patients, whose cardiovascular risk factors may differ in prevalence and impact. Thus, a risk model for Japanese patients with established ASCVD is needed.Objectives: We aimed to develop a 10-year risk prediction model for recurrent cardiovascular events in Japanese patients with established ASCVD.Methods: We analyzed 3,560 patients with coronary artery disease or ischemic stroke without prior heart failure enrolled in CHART-2. The primary endpoint was a composite of cardiovascular death, myocardial infarction, and stroke. Candidate predictors were selected from established secondary prevention models. Independent predictors were identified using multivariable logistic regression to develop a point-based 10-year risk score.Results: Over a median follow-up of 10.1 years, 756 patients experienced the primary endpoint. Independent predictors included age, smoking, systolic blood pressure, non–high-density lipoprotein cholesterol, diabetes mellitus, renal dysfunction, left ventricular dysfunction, atrial fibrillation, polyvascular disease, and statin use. These predictors were incorporated into a point-based score, termed the J-RECUR score. The score demonstrated a >5-fold gradient in 10-year recurrent cardiovascular risk, with a C-index of 0.66 (95% CI, 0.64–0.68).Conclusions: The J-RECUR score stratified patients across a wide range of 10-year recurrent cardiovascular risk and may provide a practical approach to secondary prevention risk stratification in Japanese patients with established ASCVD.

Aims: B-type natriuretic peptide (BNP) levels of 35–100 pg/mL, common in preclinical heart failure (HF), are associated with heterogeneous prognosis and are influenced by age and renal function. We aimed to characterize risk heterogeneity in preclinical HF patients with BNP levels of 35–100 pg/mL and to develop a practical risk stratification model. Methods and results: Data from the Chronic Heart Failure Registry and Analysis in the Tohoku District-2 (CHART-2) Study, a prospective multicentre cohort with long-term follow-up, were analysed. Among patients with Stage B HF (n = 4074; age 67.6 ± 12.0 years), the predictability of BNP for 10-year HF-related events was evaluated across strata of age and estimated glomerular filtration rate (eGFR). In patients with BNP levels of 35–100 pg/mL, a composite risk score—the Pre-Heart Failure Assessment and Screening Evaluation (PHASE) Index—was developed using classification and regression tree analysis and multivariable Cox proportional hazards modelling. BNP discrimination varied significantly by age and eGFR (P < 0.05), indicating that a single BNP threshold is insufficient for risk stratification in preclinical HF. The PHASE Index, incorporating age, BNP, eGFR, left ventricular ejection fraction, body mass index, and hypertension, stratified patients with BNP levels of 35–100 pg/mL into four categories and demonstrated superior discrimination compared with BNP alone (c-index 0.723 vs. 0.587, P < 0.05). Conclusion: Patients with preclinical HF and BNP levels of 35–100 pg/mL exhibit substantial risk heterogeneity. The PHASE Index improves identification of high-risk individuals and provides a pragmatic tool for risk stratification beyond the conventional BNP threshold.

Background: Accurately predicting left ventricular ejection fraction (LVEF) recovery after percutaneous coronary intervention (PCI) in patients with chronic coronary syndrome (CCS) is crucial for clinical decision-making. Objective: This study aimed to develop and compare multiple machine learning (ML) models to predict LVEF recovery and identify key contributing features. Methods: We retrospectively analyzed 520 patients with CCS from the Clinical Deep Data Accumulation System database. Patients were categorized into 4 binary classification tasks based on baseline LVEF (≥50% or 10% compared with ≤0%; and (2) normal recovery, defined as an LVEF increase of 0% to 10% compared with ≤0%. For each task, 3 feature selection strategies (all features, least absolute shrinkage and selection operator [LASSO] regression, and recursive feature elimination [RFE]) were combined with 4 ML algorithms (extreme gradient boosting [XGBoost], categorical boosting, light gradient boosting machine, and random forest), resulting in 48 models. Models were evaluated using 10-fold cross-validation and assessed by the area under the curve (AUC), decision curve analysis, and calibration plots. Results: The highest AUCs were achieved by RFE combined with XGBoost (AUC=0.93) for preserved LVEF with good recovery, LASSO combined with XGBoost (AUC=0.79) for preserved LVEF with normal recovery, LASSO combined with XGBoost (AUC=0.88) for reduced LVEF with good recovery, and RFE combined with XGBoost (AUC=0.84) for reduced LVEF with normal recovery. Shapley Additive Explanation analysis identified uric acid, platelets, hematocrit, brain natriuretic peptide, glycated hemoglobin, glucose, creatinine, baseline LVEF, left ventricular end-diastolic internal diameter, heart rate, R wave amplitude in V5, and R wave amplitude in V6 as important predictive factors of LVEF recovery. Conclusions: ML models incorporating feature selection strategies demonstrated strong predictive performance for LVEF recovery after PCI. These interpretable models may support clinical decision-making and can improve the management of patients with CCS after PCI.

Aims: A decline in left ventricular ejection fraction (LVEF) is often observed in patients with heart failure with preserved ejection fraction (HFpEF) and is associated with adverse outcomes. Although baseline LVEF is crucial for assessing the risk of sudden cardiac death in patients with HF, little is known about the relationship between the decline in LVEF and lethal arrhythmic events among patients with HFpEF. Methods and Results: We retrospectively analysed data from the CHART-2 Study. A total of 1,453 patients with HF and LVEF ≥50% at registration (73 years, 39% female) were included and categorised into three groups based on their LVEF at the prespecified 1-year follow-up; 1,316 with LVEF ≥50%, 120 with LVEF 36-50%, and 17 with LVEF ≤35%. The primary endpoint was a composite of ventricular tachycardia, ventricular fibrillation, and sudden cardiac death. All-cause death was also evaluated. During a median follow-up of 7.9 years, 79 (5.4%) patients experienced a composite event. Patients with LVEF <50% at 1 year had a significantly higher incidence of the composite event (11.7 vs. 4.8%, P < 0.001), and all-cause death (62.8% vs 51.8%, P = 0.006), compared to those with LVEF ≥50%. Importantly, a decline in LVEF to <50% was independently associated with an increased risk of the composite event (adjusted hazard ratio 1.99, 95% CI 1.04-3.79, P = 0.04). Conclusion: In patients with HFpEF, even a mild decline in LVEF was associated with a higher risk of ventricular tachyarrhythmia, sudden cardiac death and all-cause death. Continuous clinical assessment and re-evaluation of LVEF is crucial for the management of patients with HFpEF.

BACKGROUND Accurately predicting left ventricular ejection fraction (LVEF) recovery after percutaneous coronary intervention (PCI) in patients with chronic coronary syndrome (CCS) is crucial for clinical decision-making. OBJECTIVE This study aimed to develop and compare multiple machine learning (ML) models to predict LVEF recovery and identify key contributing features. METHODS We retrospectively analyzed 520 patients with CCS from the Clinical Deep Data Accumulation System database. Patients were categorized into 4 binary classification tasks based on baseline LVEF (≥50% or 10% compared with ≤0%; and (2) normal recovery, defined as an LVEF increase of 0% to 10% compared with ≤0%. For each task, 3 feature selection strategies (all features, least absolute shrinkage and selection operator [LASSO] regression, and recursive feature elimination [RFE]) were combined with 4 ML algorithms (extreme gradient boosting [XGBoost], categorical boosting, light gradient boosting machine, and random forest), resulting in 48 models. Models were evaluated using 10-fold cross-validation and assessed by the area under the curve (AUC), decision curve analysis, and calibration plots. RESULTS The highest AUCs were achieved by RFE combined with XGBoost (AUC=0.93) for preserved LVEF with good recovery, LASSO combined with XGBoost (AUC=0.79) for preserved LVEF with normal recovery, LASSO combined with XGBoost (AUC=0.88) for reduced LVEF with good recovery, and RFE combined with XGBoost (AUC=0.84) for reduced LVEF with normal recovery. Shapley Additive Explanation analysis identified uric acid, platelets, hematocrit, brain natriuretic peptide, glycated hemoglobin, glucose, creatinine, baseline LVEF, left ventricular end-diastolic internal diameter, heart rate, R wave amplitude in V5, and R wave amplitude in V6 as important predictive factors of LVEF recovery. CONCLUSIONS ML models incorporating feature selection strategies demonstrated strong predictive performance for LVEF recovery after PCI. These interpretable models may support clinical decision-making and can improve the management of patients with CCS after PCI. CLINICALTRIAL

Background: Clinical characteristics and outcomes of patients with myocardial infarction (MI) with nonobstructive coronary arteries (MINOCA) are not fully understood, particularly in Japan. Methods and Results: We enrolled a total of 8881 patients with acute MI from the Miyagi Acute Myocardial Infarction Registry Study (2012–2020), with a median age of 69 years. Among them, 239 patients (2.7%) were diagnosed with MINOCA. Compared with those with MI with obstructive coronary artery disease (MI‐CAD), patients with MINOCA were more often women, had a higher incidence of non–ST‐segment–elevation MI and a lower prevalence of dyslipidemia. Compared with patients with MI‐CAD, patients with MINOCA in all age groups (80 years of age) had a higher incidence of non–ST‐segment–elevation MI. Additionally, those ≤59 years of age were more often women and had a lower prevalence of diabetes and dyslipidemia. In‐hospital mortality increased with age in patients with MI‐CAD (3.9% for 80 years of age; P P =0.36). Compared with patients with MI‐CAD, patients with MINOCA had lower in‐hospital mortality for Killip class IV (40.7% versus 20.0%; adjusted odds ratio [OR], 0.31 [95% CI, 0.10–0.94]; P =0.04) and renal dysfunction (20.0% versus 7.1%; adjusted OR, 0.29 [95% CI, 0.09–0.96]; P =0.04). Conclusions: Patients with MINOCA exhibit distinct clinical characteristics and outcomes compared with those with MI‐CAD, particularly in terms of age, sex, prevalence of comorbidities, and in‐hospital mortality. These findings underscore the importance of tailored clinical approaches for patients with MINOCA.

Background: Goldenhar syndrome sometimes displays progressive scoliosis and other spinal deformities that require treatment. However, few reports exist on scoliosis correction in Goldenhar syndrome. We described the rare radiological outcomes of a patient with Goldenhar syndrome who received brace treatment for scoliosis. Case presentation: A 4-year-old boy was diagnosed as having Goldenhar syndrome and referred to our hospital for scoliosis treatment. The deformity deteriorated gradually, and left convex scoliotic angle was 26 degrees (T3-L2) at 11 years of age. Unexpectedly during treatment with an orthopedic brace, the curve had reversed to 21 degrees (T5-L2) at 7 months of therapy. After another adjustment of the brace, his right convex scoliotic angle improved to 13 degrees (T4-L2) at 15 months of treatment. Conclusions: Curve reversal may occur during brace treatment for scoliosis in Goldenhar syndrome. Clinicians may opt to periodically check curve correction despite the risk of increased radiation exposure.

Study Design. Retrospective multicenter cohort study Objective. We evaluated a new upper instrumented vertebra (UIV) selection method that used the modified Shinshu line (MSL) to establish the selected UIV as the MSL vertebra (MSLV). Summary of Background Data. No reports have addressed optimal UIV selection according to the lower instrumented vertebra (LIV) for good trunk balance in Lenke 1A curves. Methods. Forty-five consecutive patients (44 female, 14.4 ± 2.4 years) receiving posterior spinal fusion (PSF) for a Lenke 1A adolescent idiopathic scoliosis (AIS) curve were analyzed. We defined the novel MSL as the line between the center of the spinous process of C7 and that of the spinous process of the LIV. The vertebral body with which the MSL first contacted proximally was defined as the MSLV. The groups in which the UIV was at, proximal to, or distal to the MSLV were defined as the matched group (M-group; 15 cases [15 female], 14.7 ± 2.1 years), proximal group (P-group; 20 cases, [19 female], 15.0 ± 2.2 years), and distal group (D-group; 10 case [10 female], 14.8 ± 2.5 years), respectively. We measured Cobb angle, main thoracic (MT) curve correction rate and C7 plumb line absolute value (C7PL) at pre- and 2 years postoperatively for comparisons using Dunnett's test, with the M-group as the control. Results. In the M-group, P-group, and D-group, the Cobb angle correction rate between pre- and postoperative time points were 65.3 ± 1.3%, 62.4 ± 1.6%, and 52.8 ± 6.8%, respectively, and comparable apart from a smaller correction tendency in the D-group versus the M-group (P = 0.08). At 2 years postoperatively, C7PL was 0.5 ± 0.4 cm, 1.0 ± 0.6 cm, and 1.3 ± 0.9 cm, respectively, and significantly smaller for the M-group (both P < 0.05). Conclusions. Better trunk balance were obtained without reducing correction rate by setting the novel MSLV as the UIV in PSF for Lenke type 1A curves. Level of Evidence: 3

発作時に失算，失書，手指失認を呈した症候性部分てんかんの1例を報告した．症例は34歳の男性である．反復的に，携帯電話でのメールの文字がさがしにくい，文字が書けない，計算ができなくなるエピソードを自覚し，頻回に検査課題を試みることでGerstmann症候群の3徴が明らかとなった．脳波と画像所見より左頭頂部からの部分発作による機能脱落症状と考えられた．