Polishing the diastolic dysfunction measurement stick.

Diastolic heart failure currently accounts for more than 50% of all heart failure cases in Western societies. Because of normal left ventricular (LV) systolic function, diastolic heart failure is mainly attributed to diastolic LV dysfunction evident from slow LV relaxation and high diastolic LV stiffness. High diastolic LV stiffness is the most important cause of the repetitive heart failure episodes occurring in these patients, while high arterial elastance, atrial remodelling, and impaired chronotropic or vasomotor responses are additional contributors. Because of the pathophysiological importance of diastolic LV dysfunction, all previous diagnostic guidelines for diastolic heart failure considered evidence of diastolic LV dysfunction to be essential for the diagnosis. The recently updated diagnostic guidelines for diastolic heart failure provided by the Echocardiography and Heart Failure Associations of the European Society of Cardiology continued to adhere to this principle and required three conditions to be simultaneously satisfied for the diagnosis of diastolic heart failure: (i) signs or symptoms of fluid congestion; (ii) a normal systolic LV function evident from a LV ejection fraction (EF) .50% and a LV end-diastolic volume index ,97 mL/m, and (iii) evidence of diastolic LV dysfunction. The latter can be acquired by cardiac catheterization, by Doppler echocardiography, and by biomarkers such as brain natriuretic peptide (BNP) and NT-proBNP. A pulmonary capillary wedge pressure larger than 12 mmHg, an LV end-diastolic pressure (LVEDP) larger than 16 mmHg, and a ratio of mitral early diastolic flow velocity over tissue Doppler mitral annular lengthening velocity (E/E0) in excess of 15 provide stand-alone evidence of diastolic LV dysfunction. An E/E0 ratio ranging from 8 to 15 (E/E0 8–15), BNP larger than 200 pg/mL, and NT-proBNP larger than 220 pg/mL require additional investigations to establish presence of diastolic LV dysfunction. For E/E0 8–15, the second line of additional investigations consists of mitral flow velocity Doppler, pulmonary venous flow velocity Doppler, echocardiographic measures of left atrial volume index (LAVI) or of LV mass index (LVMI), and electrocardiographic evidence of atrial fibrillation. E/E0 8– 15 provides diagnostic evidence of diastolic LV dysfunction, when the ratio of early (E) to late (A) mitral flow velocity (E/A) is less than 0.5 and the E-wave deceleration time larger than 280 ms or when the duration of the pulmonary venous retrograde A-wave exceeds the duration of the mitral A-wave by more than 30 ms (Ard-Ad .30 ms) or when LAVI exceeds 40 mL/m or when LVMI exceeds 122 g/ m in females and 149 g/m in males or when the ECG shows atrial fibrillation. The study by Emery et al. reported in this issue of the journal explores the validity of the cutoff criteria for LAVI, LVMI, and flow velocity Doppler measurements, which were used as additional, second line evidence for diastolic LV dysfunction in the presence of an equivocal E/E0 8–15. Such a validation of the recently proposed diagnostic flowchart for diastolic heart failure is important as it could lead to adjustment by daily clinical practice of cutoff criteria, which were derived from the statistical analysis of published data. Adjustments are indeed proposed in the study by Emery et al., which demonstrates the used cutoff of LAVI (.40 mL/m) to provide optimal sensitivity and specificity, the used cutoffs of LVMI (122 g/m; 149 g/m) to provide only optimal specificity, and the flow velocity Doppler measurements not to contribute to the diagnosis of diastolic heart failure. Straightforward implementation of these findings into the recently proposed diagnostic flowchart for diastolic heart failure seems however premature. Emery et al. based their conclusions on a retrospective analysis of consecutive echocardiograms excluding severe valvular heart disease, transplant recipients, and presence of atrial fibrillation. Extrapolation of findings derived from such a poorly identified study population to patients suspected of diastolic heart failure is questionable. Although the study population was split up in subgroups with normal (.50%) or reduced (,50%) LVEF, the subgroup with normal LVEF still missed essential evidence for the diagnosis of diastolic heart failure because of lack of signs or symptoms of fluid congestion and lack of a normal LV end-diastolic volume index (, 97 mL/m). Failure to assess LV end-diastolic volumes is especially worrysome because concentric LV remodelling is typical for diastolic heart failure, whereas eccentric LV remodelling is characteristic for heart failure The opinions expressed in this article are not necessarily those of the Editors of EJECHO, the European Heart Rhythm Association or the European Society of Cardiology. * Corresponding author. Tel: þ31 20 4448110; fax: þ31 20 4448255. E-mail address: [email protected]