Getting the Ejection Fraction Right: Reducing Interobserver Variability

In one echocardiography laboratory, a formal teaching intervention significantly improved an important measure of quality.


In many clinical settings, interventions or procedures are predicated on an accurate estimation of left ventricular ejection fraction (LVEF). However, visual interpretation of echocardiograms is inherently inexact. These investigators assessed whether interobserver variability (IOV) in visual estimation of LVEF could be improved by a formal teaching intervention. Fourteen sonographers and 11 physicians from one echocardiography laboratory participated in the study. Participants' years of experience ranged from 1 to 35 (mean, 12 years). The teaching intervention consisted of three 1-hour, facilitator-led, case-based tutorials. The participants completed pre- and post-intervention evaluations of 14 and 13 cases, respectively.

The IOV in estimated LVEF for all cases decreased significantly from ±0.14 EF in the pre-intervention evaluation to ±0.08 EF in the post-intervention evaluation, a 40% reduction. The degree of improvement did not differ between groups of participants categorized broadly by level of experience.


Citation(s):

Johri AM et al. Can a teaching intervention reduce interobserver variability in LVEF assessment? A quality control exercise in the echocardiography lab. J Am Coll Cardiol Img 2011 Aug 1; 4:821.

Marwick TH. When is a number not a number? Quality control arrives in the imaging laboratory. J Am Coll Cardiol Img 2011 Aug 1; 4:830.