Daily Cardiology Research Analysis
A pragmatic cluster randomized trial showed that AI-enabled ECG alerts increased anticoagulant prescribing and atrial fibrillation diagnosis by noncardiologists. A U.S. registry analysis of >2.4 million PCIs using instrumental variable methods confirmed radial access reduces in-hospital mortality and bleeding but slightly increases ischemic stroke. A CARDIA cohort analysis linked cumulative apoB/LDL-P/TRL-P exposure in young adults with later ASCVD and suggested a usual apoB target <75 mg/dL in
Summary
A pragmatic cluster randomized trial showed that AI-enabled ECG alerts increased anticoagulant prescribing and atrial fibrillation diagnosis by noncardiologists. A U.S. registry analysis of >2.4 million PCIs using instrumental variable methods confirmed radial access reduces in-hospital mortality and bleeding but slightly increases ischemic stroke. A CARDIA cohort analysis linked cumulative apoB/LDL-P/TRL-P exposure in young adults with later ASCVD and suggested a usual apoB target <75 mg/dL in early adulthood.
Research Themes
- AI-enabled cardiovascular diagnostics and implementation
- Access-site strategy in PCI and patient safety
- Life-course lipid exposure and ASCVD prevention
Selected Articles
1. Artificial Intelligence-Enabled ECGs for Atrial Fibrillation Identification and Enhanced Oral Anticoagulant Adoption: A Pragmatic Randomized Clinical Trial.
In a cluster randomized trial across two hospitals, AI-ECG alerts to noncardiologists increased new AF diagnoses and non–vitamin K antagonist oral anticoagulant (NOAC) prescriptions within 90 days. There were no differences in echocardiography orders, cardiology referrals, or short-term ischemic/cardiovascular outcomes.
Impact: This pragmatic RCT demonstrates that AI can close care gaps by changing clinician behavior, improving anticoagulation for at-risk AF patients identified outside cardiology services.
Clinical Implications: Health systems can integrate AI-ECG alerts to support AF detection and timely NOAC initiation by noncardiologists, with governance to monitor downstream testing and safety.
Key Findings
- AI-ECG alerts increased NOAC prescriptions (23.3% vs 12.0%; HR 1.85, 95% CI 1.11–3.07).
- AF diagnosis rates increased with alerts (HR 1.40, 95% CI 1.03–1.90).
- No significant differences in echocardiography orders, cardiologist visits, ischemic stroke, cardiovascular death, or all-cause death.
Methodological Strengths
- Pragmatic, cluster randomized design with physician-level randomization across two hospitals.
- Pre-registered trial (NCT05127460) with objective, EHR-captured process outcomes.
Limitations
- Open-label design and limited to two hospitals in Taiwan; generalizability may be constrained.
- Primary benefits were process measures; trial was underpowered for hard clinical outcomes.
Future Directions: Evaluate long-term clinical outcomes, cost-effectiveness, patient-centered outcomes, and scalable governance for AI alert fatigue and appropriate anticoagulation in broader health systems.
BACKGROUND: Atrial fibrillation (AF) is often underdiagnosed and undertreated by noncardiologists. This study evaluated whether artificial intelligence-enabled ECG (AI-ECG) alerts could improve AF diagnosis and non-vitamin K antagonist oral anticoagulant prescriptions by noncardiologists. METHODS: In this open-label, cluster randomized controlled trial (NCT05127460) at 2 hospitals in Taiwan, noncardiologists were randomized to an intervention group (AI-ECG alerts) or control group (usual care). Alerts were sent to physicians when AI-ECG identified AF in emergency or hospitalized patients at risk of stroke (CHA₂DS₂-VASc ≥1 for men, ≥2 for women), excluding those with prior AF or oral anticoagulant use. Primary end points included a non-vitamin K antagonist oral anticoagulant prescription within 90 days after discharge, new AF diagnosis, echocardiogram arrangements, and cardiologist visits. Secondary end points were ischemic stroke, cardiovascular death, and all-cause death. RESULTS: A total of 8857 and 8960 patients were treated by 120 and 113 noncardiologists in the intervention and control groups, respectively; 275 and 245 patients had AI-detected AF. The non-vitamin K antagonist oral anticoagulant prescription rate was significantly higher in the intervention group (23.3% versus 12.0%; hazard ratio [HR], 1.85 [95% CI, 1.11-3.07]). The intervention group also had a higher rate of AF diagnosis (HR, 1.40 [95% CI, 1.03-1.90]). No significant differences were observed in echocardiogram arrangements, cardiologist visits, or the rates of ischemic stroke, cardiovascular death, and all-cause death. CONCLUSIONS: An AI-ECG alert for AF identification promoted non-vitamin K antagonist oral anticoagulant prescriptions among noncardiologists, thus reducing the disparity in AF care quality between cardiologists and noncardiologists. REGISTRATION: URL: https://clinicaltrials.gov/; Unique identifier: NCT05127460.
2. Cumulative exposure to atherogenic lipoprotein particles in young adults and subsequent incident atherosclerotic cardiovascular disease.
In CARDIA, higher cumulative and usual early-adult exposure to apoB, LDL-P, and TRL-P predicted incident ASCVD after age 40, with adjusted HRs ≈1.30 per SD. Risk rose beyond a usual apoB exposure of ~75 mg/dL/year, supporting earlier life-course lipid targets.
Impact: Quantifies life-course risk from particle-based exposure and proposes a practical apoB threshold for young adults, shifting prevention earlier.
Clinical Implications: Primary prevention should track apoB (and particle metrics) beginning in early adulthood, with consideration of targets <75 mg/dL to minimize cumulative exposure and later ASCVD risk.
Key Findings
- Among 4366 participants, 241 ASCVD events occurred after age 40 over ~19.3 years.
- Each 1 SD higher cumulative apoB/LDL-P/TRL-P exposure associated with adjusted HR ≈1.30 for ASCVD.
- ASCVD hazard increased beyond a usual apoB exposure ≈75 mg/dL/year in ages 18–<40.
Methodological Strengths
- Prospective, population-based cohort with repeated particle measurements and long follow-up.
- Robust modeling of cumulative and usual exposures with adjusted Cox regression.
Limitations
- Observational design limits causal inference; residual confounding possible.
- Generalizability bounded by CARDIA demographics and measurement intervals.
Future Directions: Test apoB-centered early-life interventions and quantify benefits of maintaining apoB <75 mg/dL across diverse populations, integrating lifetime risk modeling and cost-effectiveness.
BACKGROUND AND AIMS: Atherogenic lipoprotein exposure during young adulthood increases the risk of atherosclerotic cardiovascular disease (ASCVD) later in life. The relationships between cumulative and usual yearly apolipoprotein B (apoB), low-density lipoprotein particle (LDL-P), and triglyceride-rich lipoprotein particle (TRL-P) exposure in early adult life and incident ASCVD was quantified. METHODS: Follow-up data of young adults aged 18 to <40 years from the longitudinal population-based Coronary Artery Risk Development in Young Adults (CARDIA) cohort were used. Cumulative early adult exposure of apoB, LDL-P, and TRL-P were defined over a 22-year exposure period (18 to <40 years). 'Usual' exposure to atherogenic lipid particles was calculated by dividing the cumulative exposure to apoB, LDL-P, and TRL-P by 22 years, and the hazard ratio was calculated between a 1 SD higher cumulative lipoprotein exposure with incident ASCVD after age 40 using adjusted Cox regression models. RESULTS: Among 4366 CARDIA participants, there were 241 ASCVD events after age 40 (mean follow-up of 19.3 years). A 1 SD higher cumulative exposure to apoB, LDL-P, and TRL-P was associated with unadjusted HRs of 1.53 [95% confidence interval (CI) 1.36-1.72], 1.54 (95% CI 1.36-1.75), and 1.48 (95% CI 1.30-1.68) for incident ASCVD after age 40, respectively. Adjustment for covariates yielded HRs for each measure of approximately 1.30. The hazard ratio for ASCVD increased after a usual apoB exposure of approximately 75 mg/dL/year from age 18 to <40. CONCLUSIONS: Cumulative exposure to atherogenic lipid particles in young adulthood increases the risk for incident ASCVD later in life. Apolipoprotein B concentration <75 mg/dL may represent a goal to maintain low risk in young adults.
3. Radial vs femoral access for percutaneous coronary intervention: temporal trends and outcomes in the USA.
In a national registry (6.66 million PCIs), radial access rose from 20.3% to 57.5% (2013–2022). Instrumental variable analyses in 2.42 million eligible cases linked radial access to lower in-hospital mortality, major bleeding, and vascular complications, but a small absolute increase in ischemic stroke risk.
Impact: Confirms contemporary, nationwide benefits of radial PCI with rigorous causal methods while alerting to a small stroke trade-off, informing guideline implementation and operator training.
Clinical Implications: Radial access should remain the default PCI strategy given mortality and bleeding benefits; teams should mitigate stroke risk via technique optimization (e.g., catheter manipulation, embolic vigilance) and patient selection.
Key Findings
- Radial PCI adoption increased from 20.3% (2013) to 57.5% (2022) across all indications and regions.
- Radial access reduced in-hospital mortality (ARD −0.15%), major access-site bleeding (ARD −0.64%), and other major vascular complications (ARD −0.21%).
- Radial access was associated with a small absolute increase in ischemic stroke (ARD +0.05%) with no signal on a falsification endpoint.
Methodological Strengths
- Massive nationwide registry with contemporary practice spanning a decade.
- Instrumental variable analysis leveraging operator preference and falsification endpoint for causal inference robustness.
Limitations
- Retrospective design; instrumental variable assumptions may be imperfect.
- Outcomes limited to in-hospital events; mechanisms behind stroke risk not elucidated.
Future Directions: Investigate mechanisms and mitigation strategies for periprocedural stroke during radial PCI and evaluate long-term outcomes with patient-level linkage.
BACKGROUND AND AIMS: Radial access site for percutaneous coronary intervention (PCI) is recommended by clinical practice guidelines because of superior outcomes compared with femoral access site. Historically, the adoption of radial access site in the USA has lagged behind much of the rest of the world, but contemporary data on access site selection across the spectrum of clinical presentations and its association with outcomes are lacking. METHODS: A retrospective cohort study from the National Cardiovascular Data Registry's CathPCI Registry was conducted including PCIs performed between 1 January 2013 and 30 June 2022. The comparative safety of radial vs femoral access site for PCI was evaluated with instrumental variable analysis, a technique that can be used to support causal inference, exploiting operator variation in access site preferences as the instrumental variable. RESULTS: Overall, 6 658 479 PCI procedures were performed during the study period, of which 40.4% (n = 2 690 355) were performed via radial access site, increasing from 20.3% in 2013 to 57.5% in 2022. This increase was seen in all geographic regions and across the full spectrum of presentations, with the largest relative increase seen in patients with ST-elevation myocardial infarction. Overall, 2 420 805 PCIs met inclusion criteria for the comparative safety analysis. In instrumental variable analyses, radial access site was associated with lower in-hospital mortality [absolute risk difference (ARD) -.15%, 95% confidence interval (CI) -.20 to -.10], major access site bleeding (ARD -.64%, 95% CI -.68 to -.60), and other major vascular complications (ARD -.21%, 95% CI -.23 to -.18) but a higher risk of ischaemic stroke (ARD .05%, 95% CI .03-.08). There was no association with the falsification endpoint of gastrointestinal or genitourinary bleeding (ARD .00%, 95% CI -.03-.03). CONCLUSIONS: Over the past decade, use of radial access site for PCI has increased 2.8-fold in the USA and now represents the dominant form of access site across all procedural indications. Based on instrumental variable analyses, PCI with radial access site had lower rates of in-hospital mortality, major access site bleeding, and other major vascular complications compared with femoral access site but a slightly higher risk of ischaemic stroke in contemporary practice.