Aims In addition to giving optimal medical and device therapy, promoting self‐care of chronic heart failure (CHF) patients also plays an important role in comprehensive disease management for better outcomes. The study was aimed to investigate whether short message service (SMS) would help to improve death or readmission‐free survival and self‐care behaviour in CHF patients. Methods and results This was a randomized controlled trial. Between December 2011 and September 2015, patients admitted with decompensated CHF in a tertiary referral hospital who fulfilled the inclusion criteria were enrolled and randomized to receive SMS, structured telephone support (STS), or usual care after discharge. All patients were followed up to 180 days after discharge by phone call or clinic visit. Primary endpoint was the 180 day composite event, defined as all‐cause mortality or readmission. Secondary endpoints included self‐care behaviour and quality of life. Seven hundred sixty‐seven patients (61 ± 15 years, 56.5% male) were finally randomized to receive SMS (n = 252), STS (n = 255), or usual care (n = 260). Baseline characteristics were similar among the three groups. Five hundred twenty‐five (68.4%) patients were in New York Heart Association Class III or IV, and 472 (61.5%) patients had an ejection fraction of <50%. During a 180 day follow‐up, 76 (9.9%) patients died and 274 (35.7%) patients experienced at least one readmission. In a short‐term follow‐up of 30 days, there was no difference in mortality and the composite endpoint among the three groups (SMS vs. STS vs. usual care: 2.8% vs. 3.1% vs. 3.8% for mortality, P = 0.786; 12.3% vs. 14.5% vs. 15.4% for the composite endpoint, P = 0.588). The 180 day composite event rate was significantly lower in the SMS and STS groups (50.4% vs. 41.3% and 36.5%, both P < 0.05) than in the usual care group, but no difference was observed between the two phone‐based intervention groups (P = 0.268). Although there was no difference between the two groups, better self‐care behaviour was reported in the SMS and STS groups than in the control group (medication compliance, 78.9% vs. 81.4% vs. 69.5%, P = 0.011; water restriction, 70.8% vs. 74.5% vs. 61.5%, P = 0.013). Quality‐of‐life score was similar among the three groups at 180 days (P = 0.526). Conclusions In CHF patients, post‐discharge SMS, which appeared as efficient as STS, reduced the 180 day composite event and improved self‐care behaviour. SMS intervention could be integrated into CHF management.

Background: Clinical notes contain contextualized information beyond structured data related to patients’ past and current health status. Objective: This study aimed to design a multimodal deep learning approach to improve the evaluation precision of hospital outcomes for heart failure (HF) using admission clinical notes and easily collected tabular data. Methods: Data for the development and validation of the multimodal model were retrospectively derived from 3 open-access US databases, including the Medical Information Mart for Intensive Care III v1.4 (MIMIC-III) and MIMIC-IV v1.0, collected from a teaching hospital from 2001 to 2019, and the eICU Collaborative Research Database v1.2, collected from 208 hospitals from 2014 to 2015. The study cohorts consisted of all patients with critical HF. The clinical notes, including chief complaint, history of present illness, physical examination, medical history, and admission medication, as well as clinical variables recorded in electronic health records, were analyzed. We developed a deep learning mortality prediction model for in-hospital patients, which underwent complete internal, prospective, and external evaluation. The Integrated Gradients and SHapley Additive exPlanations (SHAP) methods were used to analyze the importance of risk factors. Results: The study included 9989 (16.4%) patients in the development set, 2497 (14.1%) patients in the internal validation set, 1896 (18.3%) in the prospective validation set, and 7432 (15%) patients in the external validation set. The area under the receiver operating characteristic curve of the models was 0.838 (95% CI 0.827-0.851), 0.849 (95% CI 0.841-0.856), and 0.767 (95% CI 0.762-0.772), for the internal, prospective, and external validation sets, respectively. The area under the receiver operating characteristic curve of the multimodal model outperformed that of the unimodal models in all test sets, and tabular data contributed to higher discrimination. The medical history and physical examination were more useful than other factors in early assessments. Conclusions: The multimodal deep learning model for combining admission notes and clinical tabular data showed promising efficacy as a potentially novel method in evaluating the risk of mortality in patients with HF, providing more accurate and timely decision support.

Peripartum cardiomyopathy (PPCM) is rare and potentially life-threatening; its etiology remains unclear. Imaging characteristics on cardiovascular magnetic resonance (CMR) and their prognostic significance have rarely been studied. We sought to determine CMR’s prognostic value in PPCM by using T1 and T2 mapping techniques. Data from 21 PPCM patients from our CMR registry database were analyzed. The control group comprised 20 healthy age-matched females. All subjects underwent comprehensive contrast-enhanced CMR. T1 and T2 mapping using modified Look-Locker inversion recovery and T2 prep balanced steady-state free precession sequences, respectively. Ventricular size and function, late gadolinium enhancement (LGE), myocardial T1 value, extracellular volume (ECV), and T2 value were analyzed. Transthoracic echocardiography was performed at baseline and during follow-up. The recovered left ventricular ejection fraction (LVEF) was defined as LVEF ≥50% on echocardiography follow-up after at least 6 months of the diagnosis. CMR imaging showed that the PPCM patients had severely impaired LVEF and right ventricular ejection fraction (LVEF: 26.8 ± 10.6%; RVEF: 33.9 ± 14.6%). LGE was seen in eight (38.1%) cases. PPCM patients had significantly higher native T1 and ECV (1345 ± 79 vs. 1212 ± 32 ms, P < 0.001; 33.9 ± 5.2% vs. 27.1 ± 3.1%, P < 0.001; respectively) and higher myocardial T2 value (42.3 ± 3.7 vs. 36.8 ± 2.3 ms, P < 0.001) than did the normal controls. After a median 2.5-year follow-up (range: 8 months-5 years), six patients required readmission for heart failure, two died, and 10 showed left ventricular function recovery. The LVEF-recovered group showed significantly lower ECV (30.7 ± 2.1% vs. 36.8 ± 5.6%, P = 0.005) and T2 (40.6 ± 3.0 vs. 43.9 ± 3.7 ms, P = 0.040) than the unrecovered group. Multivariable logistic regression analysis showed ECV (OR = 0.58 for per 1% increase, P = 0.032) was independently associated with left ventricular recovery in PPCM. Compared to normal controls, PPCM patients showed significantly higher native T1, ECV, and T2. Native T1, ECV, and T2 were associated with LVEF recovery in PPCM. Furthermore, ECV could independently predict left ventricular function recovery in PPCM.

Caregiving burden and depression in family caregivers have been investigated, but little is known about how they affect paid caregivers. The aim of this study was to investigate caregiving burden and depression in paid caregivers of hospitalized patients. A cross-sectional survey study was conducted in a tertiary referral hospital (Chengdu, China) that enrolled 108 paid caregivers who worked in the inpatient department. The Caregiver Burden Inventory (CBI) and the Center for Epidemiologic Studies Depression (CES-D) scale were incorporated into a self-developed questionnaire to gather demographic information on the following four aspects: general, work, income, and family. The mean total CBI score was 29.7 ± 12.5. The time-dependence burden had the highest score of 15.3 ± 4.0, which was followed by the physical burden score of 6.5 ± 4.6, developmental burden score of 3.7 ± 4.0, social burden score of 3.2 ± 4.0, and emotional burden score of 2.4 ± 3.1. Multiple linear regression analysis showed that a higher CBI was associated with a longer time as a paid caregiver [β=7.041, 95% Confidence Interval (CI):1.935 to 12.974, p = 0.009], lower income satisfaction (β= − 6.573, 95% CI: -11.248 to −3.020, p = 0.001), and higher frequency of meeting with their relatives (β=7.125, 95% CI: 2.019 to 12.456, p = 0.006). The mean CES-D score was 11.9 ± 8.7, and significant depression was found in 28 (25.9%) paid caregivers according to the CES-D score ≥ 16 cut-off. There was a moderate positive correlation between the CBI and CES-D scores (Pearson’s r = 0.452, p < 0.001). A high caregiving burden was commonly observed in paid caregivers of hospitalized patients in China, as was a high prevalence of depression symptoms. Several associated factors were identified that could be areas for future interventions.

Immune checkpoints, such as PD-1 and CTLA-4, are crucial regulators of immune responses, acting as gatekeepers to balance immunity against foreign antigens and self-tolerance. These checkpoints play a key role in maintaining cardiac homeostasis by preventing immune-mediated damage to critical organs like the heart. In this study, we explored the involvement of PD-1 and CTLA-4 in cardiovascular complications, particularly atherosclerosis and myocarditis, which can lead to heart failure. We conducted a comprehensive analysis using animal models and clinical data to assess the effects of immune checkpoint inhibition on cardiac function. Our findings indicate that disruption of PD-1 and CTLA-4 pathways exacerbates myocardial inflammation, accelerates atherosclerotic plaque formation, and promotes the development of heart failure. Additionally, we observed that immune checkpoint inhibition in these models led to increased infiltration of T lymphocytes, higher levels of pro-inflammatory cytokines, and enhanced tissue damage. These results suggest that PD-1 and CTLA-4 are critical in preserving cardiac health, and their inhibition can result in severe cardiovascular toxicity. Our study emphasizes the need for careful monitoring of cardiovascular health in patients undergoing immune checkpoint inhibitor therapies.

Large language models are rapidly transitioning from pilot schemes to routine clinical practice. This creates an urgent need for clinicians to develop the necessary skills to strike the right balance between seizing opportunities and taking accountability. We propose a 3-tier competency framework to support clinicians’ evolution from cautious users to responsible stewards of artificial intelligence (AI). Tier 1 (foundational skills) defines the minimum competencies for safe use, including prompt engineering, human–AI agent interaction, security and privacy awareness, and the clinician-patient interface (transparency and consent). Tier 2 (intermediate skills) emphasizes evaluative expertise, including bias detection and mitigation, interpretation of explainability outputs, and the effective clinical integration of AI-generated workflows. Tier 3 (advanced skills) establishes leadership capabilities, mandating competencies in ethical governance (delineating accountability and liability boundaries), regulatory strategy, and model life cycle management—specifically, the ability to govern algorithmic adaptation and change protocols. Integrating this framework into continuing medical education programs and role-specific job descriptions could enhance clinicians’ ability to use AI safely and responsibly. This could standardize deployment and support safer clinical practice, with the potential to improve patient outcomes.