Prognostic Value of Left Ventricular Dyssynchrony by Myocardial Perfusion Gated SPECT in Patients with LVEF≤ 35%

Background: Patients with LVEF≤ 35% are recommended to be treated by CRT, ICD, LVAD or cardiac transplantation. NYHA classification, QRS morphology and LBBB are used for therapeutic decision. However, mechanical LV dyssynchrony (MLVD) by phase analysis gated SPECT has not been used for this purpose. We aimed to investigate the prognostic value MLVD for predicting mortality in patients with LVEF≤ 35%.

Methods: Patients referred to a tertiary medical center in 2010-2015 underwent a comprehensive MPI assessment including gated SPECT with Tc-99m sestamibi and phase analysis, were followed-up for all cause mortality. We included only patients with LVEF ≤ 35%. The value of clinical variables, NYHA class, myocardial perfusion, LV function, and phase parameters in predicting cardiac events was tested by univariate and multivariate analyses.

Results: The study group included 585 patients (67.2±11yrs, 88% men) with LVEF = 25% ±6, followed for a median of 2.9 years. There were 45 (7.7%) deaths. In this group, patients were significantly older, have more ICD/CRT, diabetes mellitus, abnormal ECG, CLBBB, higher NYHA class, lower LVEF, more and larger scar as well as increased dyssynchrony parameters: phase SD and histogram bandwidth. In multivariate analysis, it was shown that NYHA class, diabetes mellitus and LVEF were the independent predictors for composite endpoint. However, the predictors for mortality were NYHA class and phase SD. Using an abnormal phase SD as > 40°, It was shown by Cox-proportional hazard survival plots that cardiac mortality significantly stratified by normal and abnormal phase SD adjusted for LVEF, but it was not stratified for composite endpoint.

Conclusion: Routine gated SPECT MPI with phase analysis is a useful tool to assess mechanical LV dyssynchrony in patients with LVEF≤ 35%. Together with perfusion and function data, LV dyssynchrony, as measured by phase SD together with NYHA class, can predict mortality in this population.