ST-elevation myocardial infarction (STEMI) remains a leading cause of cardiovascular morbidity and mortality worldwide, and accurate early risk stratification is critical for implementing precision therapies in clinical practice. However, existing clinical risk scores and manually derived imaging biomarkers have limited accuracy in predicting post-STEMI outcomes. To address this gap, we developed DeepSTEMI, an end-to-end deep learning system that integrates multi-sequence cardiac magnetic resonance (CMR) images with clinical parameters for predicting 2-year major adverse cardiovascular events (MACE). The system comprised two key algorithmic modules: a U-Net module that automatically segments heart regions from raw CMR images and a Transformer-based module that predicted future cardiovascular events. DeepSTEMI was developed using a multicenter dataset (n = 610; 20,618 images) from STEMI patients enrolled in the EARL-MYO-CMR registry (NCT03768453), with external validation performed in 334 patients (9944 images) from three independent cardiac centers. In external validation, DeepSTEMI demonstrated superior predictive performance compared to conventional clinical risk scores and manual CMR parameters (AUC 0.894, 95% CI: 0.823-0.965; overall accuracy 94.3%). The model identified high-risk patients who exhibited a 20-fold MACE risk compared to low-risk counterparts (HR 20.43, log-rank P < 0.001). SHapley Additive exPlanations (SHAP) analysis revealed that DeepSTEMI's predictive power stems from clinical-imaging synergy, enabling it to capture complex pathological patterns. DeepSTEMI achieved consistently superior performance over the Eitel score across all subgroups, with the greatest benefit observed in women (NRI 1.597) and in patients imaged 4-7 d post-STEMI (NRI 1.442). Overall, DeepSTEMI serves as an automated, scalable, and interpretable clinical copilot, which advances post-STEMI risk stratification beyond the limitations of current paradigms.
Introduction and objectives: The aging population is expected to reach 2 billion by 2050, but the impact of somatic symptom disorder (SSD) on the elderly has been insufficiently addressed. We aimed to clarify the prevalence of SSD in China and to identify physical and psychological differences between the elderly and non-elderly. Methods: In this prospective multi-center study, 9020 participants aged (2206 non-elderly adults and 6814 elderly adults) from 105 communities of Shanghai were included (Assessment of Somatic Symptom in Chinese Community-Dwelling People, clinical trial number NCT04815863, registered on 06/12/2020). The Somatic Symptom Scale-China (SSS-CN) questionnaire was used to measure SSD. Depressive and anxiety disorders were assessed by the Patient Health Questionnaire-9 (PHQ-9) and Generalized Anxiety Disorder-7 (GAD-7), respectively. Results: The prevalence of SSD in the elderly was higher than that in the non-elderly (63.2% vs. 45.3%). The elderly suffered more severe SSD (20.4% moderate and severe in elderly vs. 12.0% in non-elderly) and are 1.560 times more likely to have the disorder (95%CI: 1.399–1.739; p < .001) than the non-elderly. Comorbidity of depressive or anxiety disorders was 3.7 times higher than would be expected in the general population. Additionally, the results of adjusted multivariate analyses identified older age, female sex, and comorbid physical diseases as predictive risk factors of SSD in the elderly group. Conclusions: With higher prevalence of common physical problems (including hypertension, diabetes mellitus and cardio/cerebrovascular disease), the elderly in Shanghai are more vulnerable to have SSD and are more likely to suffer from comorbid depressive and anxiety disorders. SSD screening should be given more attention in the elderly, especially among older females with several comorbid physical diseases.
Mesenchymal stromal cells (MSCs) are considered to have a modest benefit on left ventricular ejection fraction (LVEF) in patients with acute myocardial infarction (AMI). However, the optimal injection timing and dose needed to induce beneficial cardiac effects are unknown. The purpose of this meta-analysis was to identify an optimal MSC transplantation time and cell dose in the setting of AMI to achieve better clinical endpoints. The authors conducted a systematic review of studies published up to June 2016 by searching PubMed, EMBASE, MEDLINE, and the Cochrane Library for relevant randomized controlled trials (RCTs). Eight prospective RCTs with 449 participants were included. The pooled results revealed that patients in the MSC group had no significant increase in LVEF from baseline compared with that in the control group (1.47% increase, 95% confidence interval (CI) −4.5 to 7.45; I 2 = 97%; P > 0.05). A subgroup analysis was conducted to explore the results according to differences in transplantation time and dose of MSCs injected. For transplantation timing, the LVEF of patients accepting a MSC infusion within 1 week was significantly increased by 3.22% (95% CI 1.31 to 5.14; I 2 = 0; P < 0.05), but this increase was insignificant in the group that accepted an MSC infusion after 1 week (−0.35% in LVEF, 95% CI −10.22 to 9.52; I 2 = 99%; P > 0.05). Furthermore, patients accepting a MSC dose of less than 107 cells exhibited an LVEF improvement of 2.25% compared with the control (95% CI 0.56 to 3.93; I 2 = 9%; P < 0.05). Combining transplantation time and cell dose indicates that a significant improvement of LVEF of 3.32% was achieved in the group of patients injected with <107 MSCs within 1 week (95% CI 1.14 to 5.50; I 2 = 0; P = 0.003). Transplantation time and injected cell dose are key factors that determine the therapeutic effect of stem cell therapy. The injection of no more than 107 MSCs within 1 week for AMI after percutaneous coronary intervention might improve left ventricular systolic function. Further studies on the mechanism and the effectiveness of MSCs for long-term therapy are warranted.
The significance of oral/gut microbiota in acute myocardial infarction (AMI) has been increasingly appreciated. However, correlations between oral/gut microbiota and AMI parameter, as well as the key microbiota that may have a crucial function in this process, remain unclear. To investigate the composition and structure of oral and gut microbiota associated with AMI and explore the roles of specific bacterial species in the progression of AMI. We conducted a case-control study with 37 AMI patients and 36 controls. Oral and gut sample were collected and sequenced. Using correlation analysis, we combined bioinformatics data with AMI clinical parameters and obtained heatmaps of correlation coefficients. Additionally, we used antibiotics to eliminate the gut microbiota of C57BL/6J mice, followed by the transplantation of selected bacteria to verify the gut colonization of oral bacteria and their impact on AMI. The component of oral and gut microbiota of AMI group showed significant alterations when compared to the control group. 17 salivary genera, 21 subgingival genera, and 8 gut genera in AMI group substantially differed from those in control group. Additionally, 19 genera from saliva, 19 genera from subgingival plaque, and 11 genera from feces substantially correlated with AMI clinical parameters. Orally administrated S.o (Streptococcus oralis subsp. dentisani), S.p (Streptococcus parasanguinis), and S.s (Streptococcus salivarius) were able to colonize in the gut and exacerbate myocardial infarction. There is a strong correlation between oral/gut microbiota and AMI. Streptococcus spp. is capable to transmit from oral to gut and exacerbate myocardial infarction in mice. Monitoring and control of specific oral microbiota may be an effective new strategy for improving the therapy of AMI.