Comparison between the British and the American/European guidelines for the assessment of left ventricular diastolic function with invasive validation of left ventricular filling pressure

Abstract

Abstract Background Left ventricle diastolic function and filling pressures assessment is still a major challenge to echocardiographer. There are two echo guidelines regarding this issue: the British Society of Echocardiography (BSE 2013) and the American Society of Echocardiography/European Association of Cardiovascular Imaging (ASE/EACVI 2016). The 2016 guidelines, which is an expert consensus and simplified update of 2009 guidelines, needs an invasive validation according to its authors. Recent studies raised questions about the diagnostic accuracy as sensitivity results varied very widely (34% to 87%) and also that of 2009 (43% to 79%). This study validated the diastolic pressure invasively in the cath lab and compared the results with the echo guidline algorithms were done immediately before the catheterization. When possible, it included additional assessment of S/D and Ar-A duration. Purpose Validation of the diagnostic accuracy of the 2009, the updated 2016 ASE/EACVI and 2013 BSE echocardiographic LV filling pressure predicting algorithms, as well as pulmonary veins flow (S/D) and (Ar-A) durations with invasively measured LV-pre-A wave. Methods 124 patients (58.06% males) underwent transthoracic echocardiography immediately before left heart catheterization. A trained echocardiographer obtained E/A mitral flow, E/e', left atrial volume index, TR, EDT, lateral and septal e' to estimate LV filling pressure as normal, elevated or indeterminate using the 2009, 2016 ASE/EACVI algorithms and 2013 BSE algorithm. He also obtained Secondary parameters as (S/D) and (Ar-A) duration. Invasive LV pre-A pressure was the reference of this study, with >12 mm Hg defined as elevated. Results Invasive LV pre-A pressure was elevated in 60 (48.38%) patients. When they could determine LV filling pressure, 2016 sensitivity was 0.36 and specificity 0.94, 2009 had 0.56 sensitivity and 0.90 specificity and 2013 resulted in 0.63 sensitivity and 0.80 specificity. Results of diagnostic accuracy of each algorithm as well as (S/D) and (Ar-A) summarized in tables associated in (picture 1: Tables of results). EDT≥150 msec raised NPV in normal, grade one diastolic dysfunction and indeterminate pressure. Conclusion 2016 was the most specific but the least sensitive with modest NPV and PPV between the 2013 and 2009. 2013 was the most sensitive with the highest indeterminate pressure rate to execlude. Adding S/D or Ar-A duration markedly improved the sensitivity and reduced class indeterminate among all algorithms with more benefit when both combined. EDT had a rule out role in normal, grade one diastolic dysfunction and indeterminate pressure patients. We kindly propose a modification of 2016 algorithm by adding S/D, Ar-A and EDT as optional parameters to increase sensitivity and reduce indeterminate class without affecting simplicity or specificity (picture 2: Proposed algorithms A+B). We recommend future studies to validate the diagnostic accuracy of the proposed algorithms. Funding Acknowledgement Type of funding sources: Public hospital(s). Main funding source(s): Al mouwasat University Hospital and University Heart Surgery Center at Damascus, Syrian Arab Republic. Tables of resultsProposed Algorithms A+B