Simonetto AI-ECG (cirrhosis)

Detection of Undiagnosed Liver Cirrhosis via AI-Enabled Electrocardiogram: a Pragmatic, Cluster-Randomized Clinical Trial

Patient / Population Intervention / Exposition Comparaison Critère de jugement

In 15,596 adults having routine 12-lead ECGs at 98 primary care teams, an ECG-based machine-learning alert for advanced chronic liver disease increased new diagnoses of chronic liver disease with advanced fibrosis within 180 days compared with usual care.

15 596

Schéma

Pragmatic cluster RCT (98 primary care teams)

Critère

New diagnosis of chronic liver disease with advanced fibrosis within 180 days of the ECG

Pertinence

RésultatAdvanced CLD diagnosed in 1.0% vs 0.5% (OR 2.09, 95% CI 1.22–3.55; P=0.007); among ECG-ML-positive patients 4.4% vs 1.1% (OR 4.37, 95% CI 1.94–9.88; P<0.001); any fibrosis (secondary) 1.7% vs 0.5% (OR 3.17, 95% CI 1.86–5.40; P<0.001).

Simonetto DA, et al. Nat Med. 2026;32(1):160-167. 10.1038/s41591-025-04058-y

Discussion et critique

Extends the EAGLE design beyond the heart: a routine ECG becomes a case-finding test for a non-cardiac disease, with targeted liver testing downstream. The diagnostic yield stayed well below epidemiological estimates of advanced CLD prevalence, which the authors attribute to variable clinician adherence to the AI recommendation — the same adoption bottleneck seen in EAGLE and Rushlow. Diagnosis, not liver outcomes, was the endpoint; open-label cluster design.