Abstract
INTRODUCTION: Assessing filling pressure (FP) remains a clinical challenge despite advancements in non-invasive imaging techniques. This study investigates the utility of echocardiographic left ventricular (LV) to left atrial (LA) volume ratio in estimating the resting FP in patients with dyspnoea and preserved ejection fraction (EF). METHODS: This study is a prospective, single-centre analysis of 53 consecutive patients with dyspnoea (New York Heart Association grade 2 or 3) and LVEF of ≥50% (mean age 71 ± 10 years) who underwent cardiac catheterisation, including direct measurement of LA pressure at rest using retrograde technique. Echocardiographic data were obtained 1.5 ± 1.0 h after cardiac catheterisation. The patients were divided into two groups: Group 1 consisted of individuals with elevated FP, indicated by a mean LA pressure or mean pulmonary capillary wedge pressure of >12 mmHg, and Group 2 comprised of patients with normal FP. The LV and LA volumes were measured at three specific points: the minimum volume (LV(ES), LA(min)), the volume during diastasis (LV(dias), LA(dias)), and the maximum volume (LV(ED), LA(max)). The corresponding LV/LA volume ratios were analysed: end-systole (LV(ES)/LA(max)), diastasis (LV(dias)/LA(dias)), and end-diastole (LV(ED)/LA(min)). RESULTS: The patients in Group 1 exhibited lower LV/LA volume ratios compared with those in Group 2 (LV(ES)/LA(max) 0.44 ± 0.12 vs. 0.60 ± 0.23, P = 0.0032; LV(dias)/LA(dias) 1.13 ± 0.30 vs. 1.56 ± 0.49, P = 0.0007; LV(ED)/LA(min) 2.71 ± 1.57 vs. 4.44 ± 1.70, P = 0.0004). The LV/LA volume ratios correlated inversely with an increased FP (LV(ES)/LA(max), r = -0.40, P = 0.0033; LV(dias)/LA(dias), r = -0.45, P = 0.0007; LV(ED)/LA(min), r = -0.55, P < 0.0001). Among all the measurements, the LV(dias)/LA(dias) ratio demonstrated the highest discriminatory power to distinguish patients with elevated FP from normal FP, with a cut-off value of ≤1.24 [area under the curve (AUC) = 0.822] for the entire group, encompassing both sinus rhythm and atrial fibrillation. For patients in sinus rhythm specifically, the cut-off value was ≤1.28 (AUC = 0.799), with P < 0.0001 for both. The LV(dias)/LA(dias) index demonstrated non-inferiority to the E/e' ratio [ΔAUC = 0.159, confidence interval (CI) = -0.020-0.338; P = 0.0809], while surpassing the indices of LA reservoir function (ΔAUC = 0.249, CI = 0.044-0.454; P = 0.0176), LA reservoir strain (ΔAUC = 0.333, CI = 0.149-0.517; P = 0.0004), and LA(max) index (ΔAUC = 0.224, CI = 0.043-0.406; P = 0.0152) in diagnosing patients with elevated FP. CONCLUSION: The study presents a straightforward and reproducible method for non-invasive estimation of FP using routine TTE in patients with dyspnoea and preserved EF. The LV(dias)/LA(dias) index emerges as a promising indicator for identifying elevated FP, demonstrating comparable or even superior performance to established parameters.