Comparing the prognostic efficacy of 2D and 3D left atrial reservoir strain in the prediction of heart failure and cardiovascular death
M Yafasov, F J Olsen, M C H Lassen, N D Johansen, G B Jensen, P Schnohr, P Soegaard, R Moegelvang, T Biering-SoerensenAbstract
Background
Left atrial (LA) strain (LAS) is associated with major adverse cardiovascular events (CV). Three-dimensional echocardiography (3DE) has been proposed as a more accurate and reproducible method for 3D-LAS and to provide incremental prognostic value to the traditional two-dimensional (2D) approach. But 3DE is limited by lower temporal and spatial resolution compared to 2DE, and the incremental benefits of 3DE remain uncertain.
Purpose
Our aim was to compare the prognostic and incremental value of 2D and 3D-LAS in predicting heart failure (HF) and CV death in the general population.
Methods
A total of 1,836 participants from a longitudinal cohort study underwent both comprehensive transthoracic echocardiographic examinations including 2DE and 3DE where LA reservoir strain (LASr) was quantified. The endpoint was a composite of HF and CV death. Cox proportional hazards regression was applied. Multivariable adjustments were made for the ARIC-HF risk score (age, gender, heart rate, systolic blood pressure, antihypertensive medication, diabetes, ischemic heart disease, smoking, and body mass index). Incremental prognostic information was evaluated using C-statistics and the continuous net reclassification index (NRI)
Results
Mean age was 54±17, 43% were male. During a median follow-up time of 4.8 years (IQR 4.3-5.5 years), 77 participants (4.2%) either developed HF or died of CV causes. At baseline 3D-LASr was 29.4±6.6% and mean 2D-LASr was 38.1±11.0%. In univariable analysis both 2D-LASr (HR 1.10 (95% CI 1.08-1.13); p<0.001, per 1% decrease) and 3D-LASr (HR 1.13 (95% CI 1.09-1.17); p<0.001, per 1% decrease) were significantly associated with the composite outcome (Figure 1). After multivariable adjustment both 2D-LASr (HR 1.06 (95% CI 1.03-1.09); p<0.001, per 1% decrease) and 3D-LASr (HR 1.05 (95% CI 1.01-1.09); p=0.01, per 1% decrease) remained significantly associated with the outcome. Using net reclassification analysis, 2D-LASr (Continuous NRI for 2D-LASr 0.48±0.12; p<0.001) yielded a higher degree of incremental prognostic information when added to the ARIC-HF risk score than 3D-LASr (continuous NRI for 3D-LASr 0.21±0.12;p=0.04). When incremental information was evaluated using C-statistics, only 2D-LASr added incremental information when added to the ARIC-HF risk score (improvement in C-stat 0.022±0.008; p=0.004) while 3D-LASr did not (improvement in C-stat 0.007±0.006; p=0.24).
Conclusion
In the general population, both 2D and 3D-LASr are significantly associated with HF and CV death, but 2D-LASr yields higher and more consistent incremental prognostic information compared to clinical risk scores than 3D-LASr.