003 Validating markers of mechanical dyssynchrony by experimental manipulation of interventricular timings: what is needed to make them a reasonable prospect for cardiac resynchronisation therapy selection?

Abstract

<h3>Background</h3> Any dyssynchrony marker proposed for selection of patients for cardiac resynchronisation therapy (CRT) should be stable between heartbeats but change markedly when interventricular delay is experimentally manipulated; the marker should also minimise at some “optimal” interventricular delay. <h3>Methods and Results</h3> We performed 3264 echocardiographic measurements in 13 patients with CRT: separate, replicate measurements at interventricular delays from RV-first 40 ms to LV-first 60 ms, in 20-ms intervals of (1) 3D systolic dyssynchrony index (SDI), (2) Tissue Doppler imaging (TDI), (3) aortic pre-ejection time, (4) interventricular mechanical delay (IVMD), (5) LVOT VTI and (6) QRS duration. In each patient, we identified with blinding on several successive repetitions, an apparently-optimal (minimally-dyssynchronous) interventricular delay for each variable. Agreement between successive optimisations was low: κ values of 0.24 for SDI, 0.02 for TDI, 0.36 for aortic pre-ejection time, 0.14 for IVMD, 0.40 for LVOT VTI and 0.47 for QRS duration. Intraclass correlation coefficient, indicating measurement reproducibility, was low when single measurements were taken (ranging across methods from 0.32 to 0.63), but improved when pairs of measurements were averaged (0.51 to 0.74, p=0.0008). Using averages of pairs of measurements reduced the disagreement between replicate optimisations p=0.007. <h3>Conclusions</h3> Under blinded conditions these mechanical dyssynchrony markers cannot reliably discriminate even large changes in interventricular delay, and can be quickly rejected as candidates for predicting clinical benefit from CRT. It would save time and expense if markers considered for clinical trialling under formal scientific conditions first underwent screening for plausibility by such a stage of inexpensive, active experimentation.