Daily Cardiology Research Analysis
Analyzed 154 papers and selected 3 impactful papers.
Summary
Three impactful cardiology studies stood out today: a deep learning system enabling continuous QTc surveillance from single‑lead signals; engineered FAP‑targeted CAR Tregs that accelerate post‑MI repair via IL‑10–dependent mechanisms; and an AI‑QCT CCTA cohort defining a plaque burden threshold (PAV ≥22%) where revascularization reduces events. Together they showcase advances in AI monitoring, immune‑regenerative therapy, and imaging‑guided decision‑making.
Research Themes
- AI-enabled cardiac safety monitoring
- Immune-regenerative therapy for myocardial repair
- Imaging-derived thresholds to guide revascularization
Selected Articles
1. Deep Learning-Based Continuous QT Monitoring to Identify High-Risk Prolongation Events After Class III Antiarrhythmic Initiation.
A spatially aware deep learning model reconstructed 12‑lead information from 10‑s single‑lead signals to estimate QT/QTc with guideline‑level accuracy and robust external validation. Continuous monitoring identified outpatient QTc prolongation associated with >4‑fold higher risk of serious ventricular arrhythmias after class III antiarrhythmic initiation.
Impact: This work enables continuous QTc surveillance using ubiquitous single‑lead vectors, overcoming a major hardware limitation and uncovering clinically actionable high‑risk QTc prolongation in outpatients.
Clinical Implications: Clinicians can leverage insertable or wearable single‑lead signals with validated AI to detect high‑risk QTc prolongation after dofetilide or sotalol initiation, enabling earlier dose adjustments, electrolyte correction, or drug discontinuation to prevent malignant arrhythmias.
Key Findings
- Single‑lead deep learning (3DRECON-QT) classified prolonged QTc with AUC 0.942 (internal) and 0.943 (external); QTc MAE ~17.5–21.1 ms.
- Continuous monitoring during dofetilide loading showed strong correlation between AI‑predicted and reference QT/QTc across serial ECGs.
- Outpatient QTc prolongation detected by the system was associated with >4‑fold increased serious ventricular arrhythmia risk.
- Model performance generalized across health systems and ECG hardware; device validation paired insertable monitor recordings with 12‑lead ECGs.
Methodological Strengths
- External validation across different ECG hardware and a public dofetilide dataset
- Device‑level validation pairing insertable cardiac monitor signals with clinical 12‑lead ECGs
- Multitask encoder‑decoder reconstructing 12 leads and predicting QT/QTc from single‑lead inputs
Limitations
- Observational development/validation; no randomized outcome testing of AI‑guided interventions
- Incomplete reporting of some performance details in the abstract (e.g., full correlation metrics)
- Generalizability to all wearables and outpatient contexts needs broader prospective evaluation
Future Directions: Prospective trials testing AI‑triggered clinical actions to mitigate arrhythmic events; integration with wearables; expanding to other QT‑prolonging drugs and populations.
BACKGROUND: Drug-induced QT prolongation after successful inpatient loading of class III antiarrhythmics may occur during routine outpatient care. Insertable cardiac monitors offer continuous signals but are limited by single-lead configuration. We hypothesized that a spatially aware deep learning system (3DRECON-QT) can reconstruct spatial information from a single lead vector to quantify QT/QTc and identify high-risk prolongation. METHODS: We developed 3DRECON-QT using a multitask encoder-decoder that ingests a 10-s single-lead signal, reconstructs 12 leads, and predicts QT/QTc. The model was developed using 12-lead ECGs with clinician-adjudicated QT/RR from a large health system and tested in an external center with different ECG hardware. Continuous monitoring performance was assessed in a public dofetilide-loading data set with serial ECGs. In a real-world cohort of outpatients on dofetilide or sotalol presenting to the hospital or emergency room for any reason, rates of ventricular arrhythmias and QT prolongation were assessed. Device validation was tested in patients with insertable cardiac monitor recordings paired with clinical 12-lead ECGs. RESULTS: 3DRECON-QT classified prolonged QTc from single-lead signals with area under the receiver operating characteristics curve, 0.942 (mean absolute error, 17.5 ms) in the internal test set and 0.943 (mean absolute error, 21.1 ms) externally. During continuous dofetilide monitoring, predictions correlated with ground truth ( CONCLUSIONS: A single-lead, deep-learning approach can achieve guideline-level measurement accuracy, enable continuous QTc surveillance from nonstandard ECG vectors, and identify clinically meaningful outpatient QTc prolongation associated with a >4-fold increase in serious ventricular arrhythmias. This strategy may enhance safety monitoring after class III antiarrhythmic initiation and support targeted intervention.
2. Engineered Regulatory T Lymphocytes Promote Infarcted Heart Repair.
FAP‑targeted CAR Tregs homed to infarct zones and improved function by reducing fibrosis and inflammation in murine MI and I/R models. Efficacy was IL‑10–dependent, acting via suppression of Smad2/3‑mediated myofibroblast differentiation and promotion of reparative M2 macrophages, with no observed treatment‑related adverse effects.
Impact: This study introduces a first‑in‑class precision immunotherapy for post‑MI repair, combining targeted homing (FAP) with Treg‑mediated IL‑10 signaling to resolve fibrosis and inflammation.
Clinical Implications: While preclinical, CAR Tregs targeting FAP could evolve into a therapeutic platform to prevent adverse remodeling after MI and potentially treat fibrotic diseases beyond the heart.
Key Findings
- FAP CAR Tregs selectively engrafted in infarcted myocardium and improved cardiac function by day 14 in MI and I/R models.
- Therapeutic effects were abrogated by IL‑10 deficiency in CAR Tregs, implicating IL‑10 as necessary for efficacy.
- Mechanistically, IL‑10 suppressed Smad2/3‑dependent myofibroblast differentiation and promoted M2 macrophage polarization.
- No treatment‑related adverse effects were observed in the preclinical models.
Methodological Strengths
- Use of both MI and ischemia‑reperfusion models with functional and histologic endpoints
- Mechanistic dissection via IL‑10 knockout and assessment of downstream Smad2/3 signaling and macrophage polarization
Limitations
- Preclinical murine data; human safety, dosing, and manufacturing feasibility remain unknown
- Potential off‑target effects and long‑term persistence of CAR Tregs require further study
Future Directions: Translation to large‑animal models; safety/toxicology; manufacturing and dosing strategies; early‑phase clinical trials for post‑MI remodeling and fibrotic disorders.
BACKGROUND: Myocardial infarction (MI) initiates a dysregulated healing process characterized by excessive fibrosis and unresolved inflammation, resulting in suboptimal cardiac repair in clinical settings. Regulatory T lymphocytes (Tregs) naturally orchestrate cardiac repair after MI, but their therapeutic potential is limited by inefficient homing to ischemic myocardium. We hypothesize that FAP (fibroblast activation protein)-specific CAR (chimeric antigen receptor) engineering overcomes this barrier by enabling precise delivery of Tregs to FAP⁺-enriched infarct zones, thereby focally amplifying reparative activity within injured myocardium. METHODS: In murine MI and ischemia-reperfusion models, C57BL/6J mice were injected with lentivirus-engineered FAP CAR Tregs (FCTRs) or mock Tregs derived from wild-type, IL-10 (interleukin-10) knockout ( RESULTS: Intravenous injections of FCTRs on day 3 after injury led to targeted engraftment in the damaged cardiac tissue. Compared with controls treated with vehicle or mock Tregs, mice receiving FCTRs exhibited remarkable cardiac functional recovery in both MI and ischemia-reperfusion models by day 14, accompanied by reduced fibrosis and decreased inflammation, all achieved without compromising the integrity of cardiac tissue. Absence of IL-10 in the engineered CAR Tregs abrogated their therapeutic efficacy, whereas the ablation of Areg showed no functional impairment. We further demonstrated that the beneficial effects of FCTRs depended on IL-10 production, which inhibited pathogenic myofibroblast differentiation by suppressing Smad2/3-dependent signaling. In addition, IL-10 secretion by these engineered Tregs promoted the polarization of inflammatory monocytes into reparative M2 macrophages and resolved excessive inflammatory responses. No treatment-related adverse effects were observed. CONCLUSIONS: We pioneered FAP-targeted CAR Tregs as a dual-action precision therapy resolving post-MI fibrosis and inflammation through IL-10-dependent mechanisms. By spatiotemporally suppressing myofibroblast differentiation and remodeling immune niches, this strategy prevents maladaptive remodeling while accelerating functional recovery, establishing a translational platform for fibrotic diseases across organ systems.
3. Outcomes with Revascularization vs. Medical Therapy According to Plaque Burden from Coronary Computed Tomography Angiography.
In a 2‑center cohort (n=2233, median follow‑up 6.8 years), AI‑QCT quantified plaque burden and revealed a significant interaction: revascularization reduced death/MI/unstable angina when per‑vessel PAV was ≥22%. This threshold did not translate as clearly at the per‑patient level after adjustment.
Impact: Defines an imaging‑derived per‑vessel plaque burden threshold (PAV ≥22%) for which revascularization confers benefit, moving toward precision selection for invasive therapy.
Clinical Implications: CCTA with AI‑QCT can stratify vessels/patients for revascularization; when per‑vessel PAV ≥22%, adding revascularization to medical therapy may reduce long‑term adverse events.
Key Findings
- Significant interaction between plaque burden and revascularization at both patient and vessel levels (p‑interaction=0.042 and 0.026).
- Revascularization associated with lower events when per‑vessel PAV ≥22% (HR 0.64, 95% CI 0.29–0.99); subgroup adjusted HR 0.50 (0.27–0.91).
- Cohort of 2233 patients with 206 events over median 6.8 years; models adjusted for calcium score, ischemia, risk factors, symptoms, and medications.
Methodological Strengths
- Multicenter cohort with long follow‑up and 206 events
- AI‑guided quantitative CT enabling continuous PAV metrics and interaction modeling with revascularization
- Comprehensive covariate adjustment including calcium score and perfusion ischemia
Limitations
- Observational design with potential residual confounding and selection bias
- Generalizability limited to patients referred for CCTA at tertiary centers
- Threshold requires prospective validation and clinical workflow integration
Future Directions: Prospective randomized trials testing PAV‑guided revascularization strategies; integration of AI‑QCT into multidisciplinary heart team decision‑making.
AIMS: We aimed to investigate, whether plaque burden from coronary computed tomography angiography (CCTA) could be used to identify patients potentially benefitting from revascularization. METHODS AND RESULTS: We assessed consecutive patients undergoing CCTA and selective 15O-water perfusion positron emission tomography for evaluation of coronary artery disease (CAD) at two tertiary care centers in Finland and the Netherlands. Per-patient percent atheroma volume (PAV) and maximum per-vessel PAV in each patient was quantified by artificial intelligence-guided quantitative computed tomography (AI-QCT). We constructed a Cox regression for death, myocardial infarction (MI), or unstable angina pectoris (uAP) including continuous PAV, revascularization, and their interaction, adjusted for calcium score, ischemia, cardiovascular risk factors, symptoms, and medication in a subcohort of 2233 patients (206 events;median follow-up 6.8 years). There was significant interaction between revascularization and continuous PAV on patient-level (p-interaction=0.042) and vessel-level (p-interaction=0.026). Revascularization was associated with a significantly lower event rate at per-patient PAV 22% (HR 0.70, 95% CI 0.43-0.98) and per-vessel PAV 22% (HR 0.64, 95% CI 0.29-0.99) or higher. In subgroup analyses, after adjustment for age, sex, cardiovascular risk factors, ischemia, antiplatelet and lipid-lowering drugs, revascularization in patients with per-vessel PAV ≥22% was associated with a significantly reduced event rate (HR 0.50, 95% CI 0.27-0.91, p=0.024) (p-interaction=0.016), whereas patient-level results remained non-significant (HR 0.62, 95% CI 0.35-1.10, p=0.104) (p-interaction<0.001). CONCLUSION: In this cohort study of patients referred for CCTA, revascularization on top of medical therapy was associated with a lower rate of long-term death, MI, or uAP from per-vessel PAV of 22% upwards.