ECG Low Ejection Fraction Detection and Guiding in AI Navigated Treatment Era

NCT06968533 | Enrolling By Invitation DMC

• 1 other identifier: TSGHA25002
• Study Type: interventional
• Target: 13,350
• Locations: 1 country
• Sites: 1

Brief Summary

Asymptomatic left ventricular systolic dysfunction (ALVSD), identified as a key component of stage B heart failure (HF) by AHA/ACC guidelines, is a common precursor to clinically overt HF. This progressive chronic disease affects over 23 million people worldwide and leads to significant morbidity, mortality, and healthcare costs. Although ALVSD presents a relatively lower risk compared to symptomatic reduced ejection fraction HF, it remains associated with a 1.6-fold increase in the risk of incident HF, a 2.13-fold increase in cardiovascular mortality, and a 1.46-fold increase in all-cause mortality. The prevalence of ALVSD ranges from 3% to 6%, at least twice that of symptomatic HF. To prevent progression to symptomatic heart failure and associated morbidities and mortalities, guideline-directed medical therapy, including ACEIs/ARBs or beta-blockers, is essential for patients with ALVSD. However, distinguishing individuals with ALVSD from the general population is challenging due to the lack of symptoms. Effective screening methods are crucial to identify individuals with ALVSD. Traditionally, diagnosing ALVSD involves screening asymptomatic populations using transthoracic echocardiography (TTE), which is costly, time-consuming, and inconvenient for patients. Other screening methods, such as laboratory tests for brain natriuretic peptide (BNP) or N- terminal pro-atrial natriuretic peptide (NT-proBNP), have insufficient diagnostic performance. Previous research proposed an AI-based alarm system (AI-S) to screen patients for ALVSD, demonstrating greater accuracy than BNP screening and improved accessibility compared to widespread echocardiography. AI-S demonstrated a sensitivity of 92.6% (standard error \[SE\] 0.042) for detecting medium-risk ALVSD patients and 63% (SE 0.154) for high-risk ALVSD patients, with a specificity of 92.7% (SE 0.003) for medium-risk patients and 98.7% (SE 0.002) for high-risk patients. AI-S is accuracy, noninvasive, highly accessible in local medical clinics, less time-consuming, and cost-effective, making it a valuable screening tool for identifying ALVSD prior to echocardiography or other confirmatory diagnostic methods. To date, no randomized controlled trial has assessed the cost-effectiveness and impact of AI-assisted screening tools for heart failure prevention in Asians. The ECG AI-Guided Screening for Low Ejection Fraction (EAGLE) trial reported a 32% increase in diagnosing of low left ventricular ejection fraction (defined as LVEF ≤50%) within 90 days of the ECG. However, this population was not Asian, and randomization involved primary care teams rather than participants. Therefore, this randomized controlled trial is designed to evaluate the impact of AI-S on diagnosing low ejection fraction in Asians, its cost-effectiveness, and the incidence of worsening HF (defined as admission for HF or HF-related emergency department visits).

Conditions

Heart Failure; Ventricular Dysfunction, Left; Artificial Intelligence; Early Diagnosis; Asymptomatic Diseases; Cost-Benefit Analysis

Keywords

Left ventricular dysfunction; artificial intelligence; Electrocardiography

Outcome Measures

Primary Outcomes (1)

• Detection of mildly reduced or severely reduced LV function

  The endpoint measures the proportion of participants with mildly reduced (LVEF 40-49%) or severely reduced (LVEF \<40%) LV function detected by echocardiography.

  Within 90 days after randomization

Secondary Outcomes (3)

• Severe reduced LVEF

  Within 90 days after randomization

• Heart failure events

  Within 90 days after randomization

• Receiving echocardiography exam

  Within 90 days after randomization

Study Arms (2)

AI-ECG guided diagnosis

EXPERIMENTAL

Participants undergo screening using the AI-ECG system. Participants identified as medium- to high-risk for LV dysfunction (LVEF \<50%) are recommended for echocardiography to confirm the diagnosis and guide subsequent management.

Diagnostic Test: AI-ECG guided diagnosis

Standard clinical care

NO INTERVENTION

Participants undergo screening using the AI-ECG system, but diagnosis and management follow usual clinical practice without immediate echocardiography based on AI-ECG results.

Interventions

AI-ECG guided diagnosis DIAGNOSTIC_TEST

Participants undergo screening using the AI-ECG system. Participants identified as medium- to high-risk for LV dysfunction (LVEF \<50%) are recommended for echocardiography to confirm the diagnosis and guide subsequent management.

AI-ECG guided diagnosis

Eligibility Criteria

• Age: 60 Years - 85 Years
• Sex: all
• Healthy Volunteers: Yes
• Age Groups: Adult (18-64), Older Adult (65+)

You may qualify if:

• Outpatients with at least one 12-lead ECG
• Age between 60-85 years

You may not qualify if:

• Documented echocardiography within the previous 6 months
• Known severe LV dysfunction (LVEF \<40%)
• Known heart failure history
• Scheduled echocardiography exam

Sponsors & Collaborators

• National Defense Medical Center, Taiwan: lead

Study Sites (1)

Tri-Service General Hospital, National Defense Medical Center

Taipei, 11490, Taiwan

Location

MeSH Terms

Conditions

Heart Failure; Ventricular Dysfunction, Left; Disease; Asymptomatic Diseases

Condition Hierarchy (Ancestors)

Heart Diseases; Cardiovascular Diseases; Ventricular Dysfunction; Pathologic Processes; Pathological Conditions, Signs and Symptoms; Disease Attributes

Study Design

• Study Type: interventional
• Phase: not applicable
• Allocation: RANDOMIZED
• Masking: NONE
• Purpose: SCREENING
• Intervention Model: PARALLEL
• Sponsor Type: OTHER
• Responsible Party: PRINCIPAL INVESTIGATOR
• PI Title: Attending physician

Study Record Dates

• First Submitted: May 5, 2025
• First Posted: May 13, 2025
• Study Start: June 1, 2024
• Primary Completion (Estimated): June 1, 2027
• Study Completion (Estimated): June 1, 2027
• Last Updated: May 21, 2025

Record last verified: 2025-05

Locations

TW (1)