Training the eye: an online educational course improves the visual assessment of mechanical dyssynchrony in candidates for cardiac resynchronization therapy

Abstract Background/Introduction Cardiac resynchronization therapy (CRT) is a guideline-recommended treatment for heart failure patients with reduced ejection fraction and a broad QRS, yet non-response is reported in 20-45% of cases (1). There is significant evidence suggesting that Mechanical Dyssynchrony (MechDys) is the substrate amenable by CRT. Contemporary research shows that the presence of MechDys – Septal Flash (SF) and/or Apical Rocking (ApRock) – is strongly associated with lower mortality after CRT (2). AMEND-CRT is an ongoing trial testing whether patient selection using the visual assessment of MechDys is non-inferior to current guidelines (3). However, translating the visual assessment from research to clinical practice requires knowledge on how to train non-experienced healthcare professionals, guided by modern educational theory. Purpose To develop and test the educational effect of an online course designed to train healthcare professionals on the visual assessment of MechDys. Methods Professionals with experience in echocardiography but no experience in the assessment of MechDys were invited to participate. An online course was designed consisting of a Pre-Test, a Training module and a Post-Test. The Training module was built according to cognitive load theory. Both Tests contained 35 cases of CRT eligible patients at varying levels of complexity, each including apical 4- and 3-chamber views. All cases were adjudicated by Expert readers. First, all participants were required to assess the presence of SF and ApRock on the Pre-Test cases to assess baseline skill. Then, participants were randomized, stratified by baseline skill, to either the Study group - which followed the Training module - or the Control group, which did not. Finally, both groups completed the Post-Test to assess potential changes in their skill. Results 39 participants from 6 countries completed the online course. Post-Test Sensitivity for SF (49% to 74%; +25%; p0.001) and ApRock (57% to 70%; +13%; p0.001) increased significantly in the Study group, compared to Pre-Test Sensitivity. Post-Test Sensitivity did not change compared to Pre-Test in the Control group. Gains in Sensitivity were not offset by a reduction in the Positive Predictive Value, suggesting a good balance between identifying true MechDys cases (true positives) and limiting false positives (Figure 1). However, Post-Test Sensitivity remained suboptimal, possibly due to the high case complexity. Conclusions Training healthcare professionals on the visual assessment of MechDys using an online course resulted in a significant improvement in performance to detect SF and ApRock. However, further research is warranted to improve Sensitivity in the detection of these motion patterns, particularly for more complex cases.