How can the left atrial strain be correctly determined by cardiovascular magnetic resonance feature tracking?

Left atrial (LA) strain marks a highly valuable clinical parameter for discrimination of various cardiovascular diseases as consolidated by a plethora of literature. However LA strain assessment by cardiovascular magnetic resonance feature tracking (CMR-FT) is conventionally based on 2-chamber (2Ch) and 4-chamber (4Ch) cine acquisitions, which are aligned for left ventricular (LV) imaging. These slice positions are hypothesized inadequate for identifying the longest LA dimensions required for accurate LA strain quantification.

LA strain was assessed in 21 patients (48.3 ± 15.2 years, 10 females) with various cardiovascular diseases by CMR-FT. Two different planning procedures were compared: (1) the standard planning procedure using the cine steady-state free-precession (SSFP) 2Ch and 4Ch views aligned to the LV and (2) the optimized planning procedure based on cine SSFP 2Ch and 4Ch views, however, aligned to the LA. Strain analysis was performed using CVI42® software. Paired Student's t-test or Wilcoxon test and Bland Altman statistics were used to evaluate differences between both planning procedures.

Indexed maximum, minimum LA volumes and LA volumes before atrial contraction were significantly elevated for optimized planning compared to standard planning (P < 0.001). In contrast, global longitudinal reservoir, conduit and booster LA strain and consecutively strain rates were found reduced for the optimized planning procedure compared to the standard planning procedure (P < 0.007). The total and passive LA ejection fractions remained unchanged, whereas the booster LA ejection fractions was significantly lower for the optimized planning procedure (P = 0.034).

Optimized LA planning procedure for assessment of the longest LA dimensions results in significant alterations in CMR quantifications with increased chamber volumetrics and decreased strain and strain rates compared to standard procedure.