Cardiac dyssynchrony quantitated by time-to-peak or temporal uniformity of strain at longitudinal, circumferential, and radial level: implications for resynchronization therapy.

BACKGROUND: The standard deviation of time to peak strain (TPS-SD) has been proposed as an index of left ventricular (LV) dyssynchrony in patients to be resynchronized. However, TPS-SD is sensitive to noise, and the influence of outliers on TPS-SD is also relevant. Alternatively, dyssynchrony can be indexed by temporal uniformity of strain (TUS), whereby a time plot of regional strains, arranged for LV location, is subjected to Fourier analysis. If segments shorten simultaneously (synchronously), the plot appears as a straight line, with power only in the zero-order Fourier term, whereas regionally clustered dyssynchrony generates an undulating plot with higher power in the first-order term. TUS index reflects zero-order relative to first-order plus zero-order power.

METHODS: In this study, TUS and TPS-SD were computed in 68 patients (QRS duration >/= 120 ms; ejection fraction
RESULTS: Following CRT, LV volume decreased (diastolic, -10 +/- 20%) and ejection fraction improved from 0.23 +/- 0.07% to 0.30 +/- 0.10% (P < .001 for both). Circumferential strain was ameliorated as well (P = .054). Two-way analysis of variance revealed TUS improvement after CRT (P = .043), with a trend for CRT to contrast asynchrony at the circumferential (+0.06 +/- 0.25) and longitudinal (+0.05 +/- 0.18) levels compared with the radial level (-0.002 +/- 0.18) (interaction P = .06). This was not true for TPS-SD. Multivariate analysis revealed that only TUS, assessed before CRT circumferentially, predicted ejection fraction improvement after CRT. Other asynchrony variables failed in the model.

CONCLUSION: Dyssynchrony indexed by circumferential TUS yields greater CRT benefits than that indexed by TPS-SD, supporting the idea of targeting TUS-measured dyssynchrony as a more informative quantitative measurement in CRT patients.