In order to bring you the best possible user experience, this site uses Javascript. If you are seeing this message, it is likely that the Javascript option in your browser is disabled. For optimal viewing of this site, please ensure that Javascript is enabled for your browser.
Did you know that your browser is out of date? To get the best experience using our website we recommend that you upgrade to a newer version. Learn more.

Late-Breaking Science: Insights from the National Echocardiography Database Australia (NEDA)

NEDA, one of the largest echocardiography databases in the world, was set up to capture digital measurements on both a retrospective and prospective basis from each digital echo laboratory across Australia. NEDA is linked with mortality data from the Australian National Deaths Index and this valuable association has provided some noteworthy observations, including those described at Late-Breaking Science presentations by NEDA’s Principal Investigators, Professor David Playford and Professor Geoff Strange (both from University of Notre Dame Australia, Fremantle, Australia).


Yesterday, Prof. Playford described an analysis that examined the relationship between mortality and left ventricular (LV) diastolic function based on American Society of Echocardiography (ASE)/European Association of Cardiovascular Imaging (EACVI) guidelines1 and also based on individual LV diastolic function measurements.

Of 436,360 adults included, ASE/EACVI guidelines could not be applied in 10% of individuals. In patients with LV ejection fraction (LVEF) ≥50% (87.6%), the distribution of normal and abnormal diastolic function was 69.3% and 9.2%, respectively, and indeterminate in 21.5%. In those with LVEF <50% (11.4%), normal or increased filling pressure was recorded in 4.8% and 33.0% of patients. Of note, 62.2% of filling pressure classifications were indeterminate in LVEF <50%.

In total, 100,597 deaths (23.0%) were identified by linkage over a median follow-up of 4.3 years. After adjustment for age and sex, there were minimal long-term mortality differences between the preserved LVEF classifications, whereas the normal filling pressure classification had a lower mortality than other classifications in the impaired LVEF group. In fully adjusted models, 5-year cardiovascular-related mortality odds ratio (OR) was 1.31 for diastolic dysfunction (95% confidence interval [CI] 1.22–1.42; p<0.001) and 1.11 for indeterminate status (95% CI 1.04–1.18; p<0.001) vs. normal diastolic function. For impaired LVEF, increased and indeterminate filling pressure had fully adjusted ORs for cardiovascular-related mortality of 1.51 (95% CI 1.15–1.98; p<0.001) and 1.25 (95% CI 0.96–1.64; p=0.06) for indeterminate status compared with normal filling pressure.

When individual diastolic function parameters were analysed, a strong association with 5-­year cardiovascular-associated and long-term all-cause mortality was noted for increasing mitral E wave velocity (mortality threshold >90 cm/s) and septal E:e’ ratio (mortality threshold >9), decreasing septal e’ velocity (mortality threshold <9 cm/s) and increasing left atrial volume index (LAVi; mortality threshold >32 mL/m2).

These findings suggest that while diastolic dysfunction and increased filling pressure according to current guidelines is associated with increased mortality, many cases remain indeterminate. Individual parameters of diastolic function (mitral E velocity >90 cm/s, septal e’ velocity <9 cm/s, septal E:e’ ratio >9 and LAVi >32 mL/m2) showed clear pivot points of increased mortality and may be clinically useful to classify indeterminate patients.

Prof. Strange’s presentation from NEDA focussed on the impact of changes in LVEF on long-term mortality. Of 493,155 patients with a definitive LVEF, 117,275 patients had baseline and repeat LVEF values recorded ≥6 months apart. During a median follow-up of 7.6 years, 34,101 (29.1%) patients died.

Even the smallest decrease in LVEF (–0.6 to –2.9 units) was associated with an increased adjusted risk of cardiovascular-­related mortality (1.20; 95% CI 1.09–1.33; p<0.001) and all-cause mortality (1.06; 95% CI 1.01–1.12; p<0.01), increasing to 3.18 (95% CI 2.94–3.45; p<0.001) and 2.15 (95% CI 2.05–2.15; p<0.001), respectively, for those with a >16.0-unit decrease in LVEF. Among those with an LVEF increase (range of 5 to 35 units), the adjusted risk of cardiovascular-­related and all-cause mortality decreased by 20–40% (p<0.001). A plateau of mortality risk was seen once a final LVEF level of 54.0–58.0 units had been reached, which persisted when accounting for age, sex and baseline LVEF, direction and extent of change in LVEF.

These data highlight the prognostic importance of even small LVEF changes in either direction and emphasise the benefits of maintaining systolic function above a specific LVEF threshold.


1. Nagueh SF, et al. Eur Heart J Cardiovasc Imaging 2016;17:1321–1360.

The content of this article reflects the personal opinion of the author/s and is not necessarily the official position of the European Society of Cardiology.