Abstract
Funding Acknowledgements
Type of funding sources: None.
Aim/Introduction: Assessment of mechanical dyssynchrony by myocardial perfusion gated-SPECT in patients with non-ischemic cardiomyopathy for predict response to cardiac resynchronization therapy (CRT).
Materials and Methods
We examined 32 patients with non-ischemic cardiomyopathy before and six months after CRT. Left ventricular mechanical dyssynchrony and contractility were assessed for all patients by myocardial perfusion gated-SPECT. The phase standard deviation (PSD), histogram bandwidth (HBW), phase histogram skewness (S) and phase histogram kurtosis (K) were used as an indicator of mechanical dyssynchrony for the both ventricles.
Results
Mechanical dyssynchrony of both ventricles before CRT was increased in all patients. Median value PSD 53°(41-61°), HBW 176°(136-202°), S 1,62(1,21-1,89), K 2,81(1,21-3,49). Six months after CRT 22(68%) respondents were identified. We divided the patients into two groups (responders and non-responders) and compared phase parameters. It was found that the PSD (44°(35-54°)) and HBW (158°(118-179°) in the responders were significantly lower than in the non-responders (PSD (68°(58-72°); HBW (205°(199-249°)). The value of phase histogram skewness and kurtosis in responders were significantly higher (Responders: S 1,77(1,62-2,02); K 3,03(2,60-3,58). Non-responders: S 1,21(0,93-1,31); K 1,21(0,19-1,46)). We found that all four indicators of mechanical dyssynchrony can predict CRT response according to the results of univariate logistic regression analysis. Moreover, It was found that only phase histogram kurtosis (OR = 1.196, 95% CI 1.04-1.37) is an independent predictor of CRT response according to multivariate logistic regression.
Conclusion
Radionuclide assessment of mechanical dyssynchrony may be the optimal diagnostic method for selecting patients with non-ischemic cardiomyopathy on CRT.
Gated SPECT left ventricular shape and prediction of super responders to cardiac resynchronization therapy: Not so easy as it (LV) looks
