Predictors of Mortality by an Artificial Intelligence Enhanced Electrocardiogram Model for Cardiac Amyloidosis

Amadio, JM; Grogan, M; Muchtar, E; Lopez-Jimenez, F; Attia, ZI; AbouEzzeddine, O; Lin, G; Dasari, S; Kapa, S; Borgeson, DD; Friedman, PA; Gertz, MA; Murphree, DH; Dispenzieri, A

doi:10.1002/ehf2.15061

Heart failure & cardiomyopathies

Predictors of Mortality by an Artificial Intelligence Enhanced Electrocardiogram Model for Cardiac Amyloidosis

ESC Heart FailPublished 31 Aug 2024Retrospective cohort study

Read the original article ↗PubMed

Cite

Authors' conclusion

Among patients with AL and ATTR amyloidosis, the A2E model helps to stratify risk of CA and adds another dimension of prognostication.

Study context

Population: Patients with cardiac amyloidosis (CA)
Participants: 2,533 patients

Exposure: AI ECG (A2E) score as predictor of mortality
Amyloidosis (AL)n = 1,834
ATTRwtn = 530
ATTRvn = 169
Outcome: All-cause death (prognostic value of A2E score)

Main finding

As reported in the abstract

All-cause death by A2E score (multivariable)

Increasing A2E score, multivariable

Risk ratio (95% CI)

AL

2.0(1.58–2.55)

ATTR

2.7(1.81–4.24)

Observational data: associations, not effects.

Further findings

Median overall survival

ALmedian 22.9 months95% CI 19.2–28.2

ATTRwtmedian 47.2 months95% CI 43.4–52.3

ATTRvmedian 61.4 months95% CI 48.7–75.9

Median overall survival — Whole cohort median overall survival: 35.6 months [95% CI 32.3, 39.5]

Original publication

Predictors of Mortality by an Artificial Intelligence Enhanced Electrocardiogram Model for Cardiac Amyloidosis. ESC Heart Fail. Published 31 August 2024. doi:10.1002/ehf2.15061

Read the original article ↗

Summarised from the published abstract; every number is checked against it