Peak left atrial strain as a single measure for the non-invasive assessment of left ventricular filling pressures.

ABSTRACT

Echocardiographic assessment of left ventricular (LV) filling pressures is performed using a multi-parametric algorithm. Left atrial (LA) strain was recently found to accurately classify the degree of diastolic dysfunction. We hypothesized that LA strain could be used as a stand-alone marker and sought to identify and test a cutoff, which would accurately detect elevated LV pressures. We studied 76 patients with a spectrum of LV function who underwent same-day echocardiogram and invasive left-heart catheterization. Speckle tracking was used to measure peak LA strain. The protocol involved a retrospective derivation group (N = 26) and an independent prospective validation cohort (N = 50) to derive and then test a peak LA strain cutoff which would identify pre-A-wave LV diastolic pressure > 15 mmHg. The guidelines-based assessment of filling pressures and peak LA strain were compared side-by-side against invasive hemodynamic data. In the derivation cohort, receiver-operating characteristic analysis showed area under curve of 0.76 and a peak LA strain cutoff < 20% was identified as optimal to detect elevated filling pressure. In the validation cohort, peak LA strain demonstrated better agreement with the invasive reference (81%) than the guidelines algorithm (72%). The improvement in classification using LA strain compared to the guidelines was more pronounced in subjects with normal LV function (91% versus 81%). In summary, the use of a peak LA strain to estimate elevated LV filling pressures is more accurate than the current guidelines. Incorporation of LA strain into the non-invasive assessment of LV diastolic function may improve the detection of elevated filling pressures.