Determinants of left atrial reservoir strain in a large healthy population

Abstract

Abstract Funding Acknowledgements Type of funding sources: Public Institution(s). Main funding source(s): Centre for Innovative Ultrasound Solutions Nord-Trøndelag Hospital Trust. Background Echocardiographic evaluation of the left atrium (LA) has until recently been done by assessment of LA end-systolic volume (LAESV) or Doppler measurements. While these measurements provide prognostic information, they indirectly measure LA function. LA reservoir strain, however, is a direct measurement of LA function and provides additional prognostic information but is affected by the left ventricle (LV) through movement of the atrioventricular plane. Purpose By evaluation of echocardiograms from a large healthy population we aimed to study the associations of LA reservoir strain with age, LA-, and LV volumes, and LV global longitudinal strain (LV GLS). Methods Echocardiographic acquisitions and analyses were performed according to current recommendations. All study personnel were experienced and affiliated with a European Association of Cardiovascular Imaging accredited echocardiographic laboratory. The associations of LA reservoir strain as outcome variable with age, LAESV, LV end-diastolic volume (LVEDV) and LV GLS were assessed by multiple linear regression analyses. Results In total 1412 subjects (56% females) with mean (SD) age 57 (12) years were included. Mean LVEF was 60.4% in females and 59.6% in males. Totally, 39% of the variance in LA reservoir strain between subjects was explained by age, LAESV, LVEDV and LV GLS (Table 1). All four characteristics were significantly associated with LA reservoir strain (all p<0.001). Age and LV GLS were the two most important characteristics for determining LA reservoir strain, indicating lower LA reservoir strain with higher age and lower absolute LV GLS. Six years of higher age had a similar effect as one percent lower absolute LV GLS. Similarly, 28 mL larger LAESV and 35mL smaller LVEDV corresponded to one percent lower absolute LV GLS. Conclusion In this large healthy population, almost 40% of the variance in LA reservoir strain between subjects was explained by age, LV GLS, and the volumes of LA and LV. The most important determinants were age and GLS, contributing equally to a reduction of LA reservoir strain with higher age and lower absolute LV GLS. Correspondingly, the size of LA and LV contributed less to the variability of LA reservoir strain. In conclusion, both age and LV function should be considered when interpreting LA reservoir strain.