CMR-measured Left atrial stiffness index calculation using left atrial volumetric measures in aortic stenosis

Background: Aortic stenosis (AS) is a progressive disease associated with myocardial remodelling and adverse clinical outcomes if left untreated. The left atrial stiffness index (LASI), conventionally derived from E/e′ and LA reservoir strain (LASr) by echocardiography, reflects LV filling pressure, and has been linked to adverse outcomes in AS. LASI can also be derived using CMR LASr, which has been studied in AS¹. However, CMR LASr calculation is computationally intensive and time consuming, whilst LA volumes can be easily derived from routine long-axis cine imaging. We aimed to (i) derive LASI using LA volumetric measures as a surrogate for LASr, (ii) correlate LASI derived from LASr and LA volumes, and (iii) assess clinical and imaging associations of volumetric LASI in asymptomatic AS.

Methods: This post-hoc analysis included 273 participants from prospectively recruited cohorts: symptomatic severe AS (n=78), asymptomatic moderate–severe AS (n=174), and controls (n=21). All underwent same-day echocardiography, venepuncture and CMR. LASr was quantified by Feature Tracking. Bi-plane maximum and minimum LA volumes were measured to calculate LA emptying fraction (LAEF) and expansion index (LAEI). Pulmonary capillary wedge pressure (PCWP), a surrogate for LV filling pressure, was derived using the formula shown in Figure-1. LASI was calculated conventionally using LASr and as PCWP/LAEF or PCWP/LAEI. ANOVA with post-hoc Tukey test was used for LASI comparison between groups. Spearman’s test was used to assess correlations. Univariable and multivariable regression examined associations of volumetric LASI with clinical and imaging variables in asymptomatic AS.

Results: Mean age was comparable across groups (symptomatic AS 67±12 years; asymptomatic AS 66±13; controls 68±9). LASI derived from LAEF and LAEI were significantly higher in symptomatic AS patients compared to asymptomatic AS and controls (Table-1; Figure-2). LASr correlated non-linearly with LAEF (r²=0.7, p< 0.001) and LAEI (r²=0.6, p< 0.001) (Figure 3). Strong correlations were observed between LASr- and volume-derived LASI (LASI-EF r=0.81, p< 0.001; LASI-EI r=0.78, p< 0.001). In asymptomatic AS, multivariable analysis identified independent associations of LAEF-derived LASI with log NT-proBNP, stroke volume and late gadolinium enhancement, while LAEI-derived LASI was independently associated with log NT-proBNP.

Conclusion: CMR LA volumetric-derived LASI strongly correlates with LASr-derived LASI and is significantly increased in symptomatic vs asymptomatic AS. In asymptomatic AS, it is independently associated with markers of adverse remodelling. These findings support volumetric LASI as a practical alternative to LASr-based assessment for evaluating LA stiffness, and its utility as a potential non-invasive marker of early decompensation in asymptomatic AS.

Table 1. LASI derived from LA volumes in three groups of patients

	Controls	Asymptomatic AS	Symptomatic AS	ANOVA p-value
LASI LAEF (Mean±SD)	0.25±0.05	0.30±0.10	0.43±0.40	< 0.001
LASI LAEI (Mean±SD)	0.12±0.05	0.15±0.08	027±0.39	0.002

Figure 1. Study methodology. CMR-measured LASI from reservoir strain and LA volumetrics.
Figure 2. LASI derived from LAEF (A), and LAEI (B), in three groups of patients: controls, asymptomatic moderate-severe AS, and symptomatic severe AS.
Figure 3. Correlation between LA reservoir strain and LAEF (A); LA reservoir strain and LAEI (B).