Estimating left atrial pressure in atrial fibrillation: an ongoing struggle

This editorial refers to ‘Determinants of invasive left atrial pressure in patients with atrial fibrillation’, by B. Andrea et al., https://doi.org/10.1093/ehjci/jeae194. Echocardiography is widely used in the evaluation of left ventricular (LV) diastolic function and algorithms that combine selected echocardiographic parameters can differentiate between normal and elevated LV filling pressure with good accuracy.1,2 These algorithms, however, are not to be applied in patients with atrial fibrillation due to weaker associations between echocardiographic parameters and LV filling pressure than in patients with sinus rhythm.3,4 Most studies on the association between echocardiographic parameters and LV filling pressure have used pulmonary capillary wedge pressure as an indirect measure of left atrial (LA) mean pressure. Due to limited access to the left atrium, few studies have related echocardiographic parameters to directly measured LA pressure. In this issue of the journal, Andrea Bonelli et al.5 present results from a study that investigated the association between LA structural and functional parameters and directly measured LA pressure in patients with persistent or paroxysmal atrial fibrillation who were referred for transcatheter ablation.5 This is a multicentre prospective study in 101 patients with mean age 66 years and mean LV ejection fraction (EF) of 57%. Forty-five patients had atrial fibrillation at the time of measurements, and 56 patients were in sinus rhythm, but had a history of atrial fibrillation. Mean LA pressure ≥ 15 mmHg was considered as increased. Data were analysed both for the total group and separately for patients in sinus rhythm and in atrial fibrillation at the time of measurements. When considering the total group, patients with increased LA pressure showed significantly lower values of LV global longitudinal strain, larger LA volumes, and worse right ventricular function than those with normal LA pressure. For the total group, after multivariable adjustment, higher ratio of early mitral flow velocity to annular velocity and minimal LA volume index (LAVImin), lower peak LA longitudinal strain and right ventricular free wall strain, but not maximal LA volume index (LAVImax) (P = 0.137), were significantly associated with mean LA pressure. Overall, LAVImin showed the best diagnostic accuracy to predict elevated LAP (area under the curve 0.703). The authors are to be commended for undertaking this study on non-invasive estimation of LV filling pressure in atrial fibrillation, as there is a paucity of research on this topic to date. A strength of the study is that LA pressure was measured directly in the left atrium, rather than by indirect estimates as pulmonary capillary wedge pressure (PCWP) or LV diastolic pressures that were used in most previous validation studies of echocardiographic parameters as indicators of LA pressure. When analysing the total group, which included patients in atrial fibrillation as well as those in sinus rhythm at the time of measurement, Bonelli et al.5 using directly measured LA pressure, confirmed previous studies that have tested associations between echocardiographic parameters and PCWP or LV diastolic pressures.1,2,6 Bonelli et al., however, observed that LA minimum volume index (LAVImin) had a stronger association with LA mean pressure than LAVImax. This is consistent with observations in studies of patients in sinus rhythm, which suggest that LAVImin may be a better marker of LV filling pressure and a slightly stronger predictor of adverse cardiovascular events than LAVImax.7,8 As addressed previously, the larger measurement variability for LAVImin than for LAVImax is a challenge,9–11 and therefore it is debateable whether LAVImin should replace LAVImax in the evaluation of LV diastolic function in clinical practice. Since LA size is one of the most widely used cardiac biomarkers, it is important to resolve this issue. Prospective studies in different patient populations are needed to determine the added value of routinely measuring LAVImin as a single LA size parameter, or in addition to LAVImax. When Bonelli et al. analysed the group of patients with atrial fibrillation at the time of measurement, no echocardiographic parameters had a significant association with LA pressure. This may in part be explained by the small sample size and therefore relatively low statistical power. In a recently published multicentre study12 of a much larger population with atrial fibrillation, there were rather weak, but significant correlations between single echocardiographic parameters and LV filling pressure. Therefore, a multiparametric algorithm combining several echocardiographic parameters plus body mass index was proposed. This algorithm differentiated normal from elevated LV filling pressure with an accuracy of 75%. The latter approach was recommended in a guideline that was recently published by the British Society of Echocardiography.13 For a more comprehensive discussion about the assessment of diastolic function in patients with atrial fibrillation, readers are referred to a review article by Beladan et al.14 in this focus issue on heart failure with preserved EF. The study by Bonelli et al. is limited by the rather small size. Furthermore, they only included patients referred for transcatheter ablation, with most patients having normal or only mildly increased LA volumes. Most patients had normal LV EF, and therefore the observed associations between echocardiographic parameters and LA pressure may not be valid for patients with heart failure and reduced LV EF. In conclusion, in the study of Bonelli et al.,5 in patients with ongoing atrial fibrillation, there was no suggestion that any single echocardiographic parameter could be used to estimate LA pressure. In patients with sinus rhythm at the time of measurements, LAVImin was a more sensitive marker of LA pressure than LAVImax. It is not yet clear if the slightly better sensitivity of LAVImin for detecting elevated LV filling pressure outweighs the disadvantage of its weaker reproducibility. The study was limited to patients referred for transcatheter ablation, and the findings should be further explored in larger populations with different causes of atrial fibrillation. None declared.