Left atrial coupling index across imaging modalities: TTE vs. CMR in transthyretin cardiac amyloidosis
L Kieviet, I Torre, M G Van Der Meer, R Tarantini, B K Velthuis, P Van Der Harst, M Guglielmo, M I F J OerlemansAbstract
Background
Left atrial coupling index (LACI) is an emerging marker of diastolic function and atrial remodeling in transthyretin cardiac amyloidosis (ATTR-CM) and can be assessed using both transthoracic echocardiography (TTE) and cardiac magnetic resonance (CMR). Potential differences in LACI values between these imaging modalities warrant further investigation, particularly to clarify whether they are related to patient characteristics or primarily reflect differences in measurement technique.
Purpose
To evaluate the agreement between TTE- and CMR-derived LACI and to determine whether patient characteristics affect the difference (ΔLACI) between the two imaging modalities.
Methods
At time of ATTR-CM diagnosis, all patients underwent both TTE and CMR with LACI measured using both modalities. Agreement between TTE- and CMR-derived LACI was assessed using intraclass correlation coefficient (ICC) and Bland-Altman analysis. ΔLACI, calculated as CMR-derived minus TTE-derived LACI, was used to investigate associations with patient characteristics using Pearson correlation analysis and linear regression. Also, potential contributions of left atrial and left ventricular volume differences to ΔLACI were explored.
Results
A total of 92 ATTR-CM patients (85% male, 91% ATTRwt-CM) underwent consecutive TTE and CMR examinations. Mean age at diagnosis was 75.5 ± 7.3 years. Most patients had moderate symptoms, with NYHA class II (57.6%) and NAC stage 1 (48.9%) being the most common. TTE-derived LACI showed moderate agreement with CMR-derived LACI (ICC 0.56, 95% CI 0.402–0.686). Bland–Altman analysis showed a mean bias of −0.10 ± 0.20, with 95% limits of agreement from −0.50 to +0.29, indicating slight overestimation by TTE. The difference in LA/LV ratio between CMR and TTE was mainly driven by slight overestimation of LA volume on TTE (r = 0.70, p < 0.001), with a smaller contribution from differences in left ventricular volume (r = −0.32, p = 0.002). Patient characteristics were well distributed, ΔLACI was not associated with BMI (r = 0.003, p = 0.977), BSA (r = 0.115, p = 0.277) or heart rate, adjusted for atrial fibrillation.(B = 0.001, p = 0.448). ΔLACI did not differ across NAC stage (F(2, 88) = 3.101, p = 0.50) or NYHA class (F(2, 89) = 0.322, p = 0.726).
Conclusion
In ATTR-CM patients, the difference between CMR- and TTE-derived LACI is small and consistent, independent of body size, HR, NAC stage, or functional status and mainly driven by slight overestimation of LA volumes on TTE. TTE-derived LACI demonstrates comparable robustness to CMR, which is important given the emerging role of LACI in ATTR-CM.