CMR biomarkers in transthyretin amyloidosis: the emerging role of left atrial strain

Background

Cardiac transthyretin amyloidosis (TTR-CA) is characterised by progressive extracellular deposition of amyloid fibrils, diffuse interstitial expansion, and increased myocardial stiffness. Cardiac magnetic resonance (CMR) enables comprehensive tissue characterisation through quantitative biomarkers—T1 mapping, extracellular volume (ECV), and late gadolinium enhancement (LGE)—with established diagnostic and prognostic implications. However, the relative contribution of CMR-derived feature tracking parameters, particularly left atrial strain (LAS), remains insufficiently defined.

Purpose

To determine the specific contribution of CMR-derived biomarkers in ATTR-CA, using 99mTc-DPD bone scintigraphy as the reference standard.

Methods

Prospective case–control study including 20 consecutive patients referred with suspected amyloidosis, who underwent both CMR and bone scintigraphy within the same year and prior to initiation of transthyretin-stabilising therapy. Eighty per cent were male; mean age was 71 ± 13 years. Fifty per cent were classified as ATTR-CA positive according to scintigraphic criteria (Perugini grade ≥2 combined with a heart-to-contralateral ratio ≥1.25).

CMR was performed on 1.5T and 3T scanners. Post-processing was undertaken using CVI 4.2 (Circle Cardiovascular Imaging), including tissue tracking analysis. Elevated native T1 was defined as ≥1050 ms at 1.5T and ≥1250 ms at 3T. LGE was considered suggestive in the presence of a diffuse subendocardial or transmural pattern. A pulmonary artery diameter >30 mm was categorised as probable pulmonary hypertension, pericardial and pleural effusions were considered absent or present. All remaining variables were analysed as continuous parameters.

Results

Reservoir LAS demonstrated the strongest association with ATTR-CA diagnosis (r = 0.86; p < 0.001), indicating a very high and statistically significant correlation. Native T1 mapping and ECV% showed moderate but significant correlations with ATTR-CA diagnosis (r = 0.59 and r = 0.58, respectively).

Conventional structural and functional parameters—including biventricular ejection fraction, LGE presence, left ventricular longitudinal and radial strain—did not reach statistical significance.

Notably, reservoir LAS exhibited excellent discriminatory performance for identifying ATTR-CA positive cases, with an area under the curve (AUC) of 0.99 (95% CI: 0.96–1.00).

Conclusions

Reservoir LAS emerges as a particularly sensitive biomarker in ATTR-CA, likely reflecting the integrated effects of ventricular stiffness, chronically elevated filling pressures, and direct atrial infiltration. Its early impairment, even in the presence of preserved ejection fraction, suggests a relevant role in the initial stages of disease.

Multiparametric integration of LAS with native T1 mapping and ECV% may optimise non-invasive staging and risk stratification in ATTR, providing incremental value beyond conventional functional parameters.Heatmap Spearman correlation CMR and outcomeBoxplots main CMR variable and results