Heart failure & cardiomyopathiesCohort study
Cardiac magnetic resonance-derived left atrial strain restratifies patients with hypertrophic cardiomyopathy at low-risk of sudden cardiac death
- retrospective longitudinal single-centre study
- external validation cohort
Authors' conclusion
CMR-derived LA-GLS improves risk stratification in HCM patients at low-risk of SCD and may refine management strategies and optimize resource allocation.
Study context
- Population
- Low-risk hypertrophic cardiomyopathy (HCM) patients (HCM Risk-SCD score <4%) who underwent cardiovascular magnetic resonance (CMR) with left atrial global longitudinal strain (LA-GLS) assessment via feature trackingn = 183, low-risk HCM patients, external validation cohort of 202
- Exposure
- CMR-derived LA-GLS for predicting major adverse cardiovascular events (MACE)
- Outcome
- Composite of MACE (ventricular arrhythmias, heart failure, thromboembolic events, or all-cause death)
- Follow-up
- Median 6.1 (2.8-8.5) years
Main finding
As reported in the abstract
Hazard ratio (95% CI)P value
LA-GLS
2.17(1.31–3.58)
P = 0.002maximal left ventricular (LV) wall thickness
1.10(1.02–1.18)
P = 0.015late gadolinium enhancement (LGE) > 5% of total myocardial mass
2.00(1.01–3.97)
P = 0.049LV ejection fraction (LVEF) < 50%
12.86(5.91–33.66)
P < 0.001- patients experienced MACE
- 67 patients (37%)
Observational data: associations, not effects.
Further findings
Baseline characteristics
- mean age
- 64 ± 19 years
- female
- 47%
MACE rate in patients with LV ejection fraction (LVEF) >50%, late gadolinium enhancement (LGE) <5%, and an LA-GLS >37.6%
- MACE rate
- 2%
- identifying an overall low-risk subgroup
External validation
- prognostic value of LA-GLS validated in an external cohort of 202 low-risk HCM patients
Original publication
Cardiac magnetic resonance-derived left atrial strain restratifies patients with hypertrophic cardiomyopathy at low-risk of sudden cardiac death. Eur Heart J Cardiovasc Imaging. Published 18 February 2026. doi:10.1093/ehjci/jeag049
Summarised from the published abstract; every number is checked against it