The comparison of left ventricular mechanical dyssynchrony obtained by myocardial perfusion scintigraphy and gated blood pool SPECT to predict the result of the cardiac resynchronization therapy

Abstract

Abstract Funding Acknowledgements Type of funding sources: None. Background/Introduction The assessment of left ventricular mechanical dyssynchrony (LV MD) is essential in identifying patients who may benefit from cardiac resynchronization therapy (CRT). Both gated myocardial perfusion scintigraphy (MPS) and gated blood pool SPECT (GBPS) are widely used to LV MD assessment [1,2]. Lack of data is available regarding the comparison of left ventricular MD parameters obtained by MPS and GBPS on cadmium-zinc-telluride (CZT) gamma-cameras and their prognostic value. Purpose to compare two scintigraphic methods – MPS and GBPS in LV MD assessment and to assess the capability of these methods to predict CRT response. Methods This study included 30 heart failure patients referred for CRT. Before CRT all patients underwent rest gated MPS and GBPS with LV MD evaluation on CZT cardiac gamma. Based on the phase analysis, the following indexes were estimated: phase standard deviation (SD) and phase histogram bandwidth (HBW). One year after CRT patients were divided to responders and non-responders. The response criteria were defined as LV end systolic volume decreased by ≥ 15% or LV ejection fraction increase by ≥ 5%, based on echocardiography. Results The correlation between gated MPI indexes and GBPS indexes was suboptimal: SD r = 0.39, p < 0.05; HBW r = 0.48, p < 0.05. Based on Mann-Whitney statistics significant differences between LV MD indexes, assessed by both MPS and GBPS were found: SD LV 56.1 (IQR 50.6 – 64) deg. vs 50.6 (37 - 61) deg., p < 0.05 and HBW LV 188.64 (176 - 213) deg. vs 201.68 (180 - 240) deg., p = 0.03, respectively. The Bland-Altman analysis showed poor agreement between gated MPS and GBPS for SD assessment (p = 0.02) with mean difference value 7.02 (96% CI 1.11 to 12.9). However both MPS and GBPS were comparable in terms of HBW estimation (p = 0.18), with mean difference value of -12.5 (96% CI -31.5 to 6.3).MD indexes derived by MPS differed significantly between CRT responders and non-responders: SD 53.56 (47 – 63.4) deg. vs 62.4 (56-71) deg., p < 0.05; HBW 182.1 (166 - 211) deg. vs 204.3 (179 - 225) deg., p < 0.05. MD indexes obtained by GBPS did not show significant difference in CRT responders and non-responders: SD 51.2 (37 – 62) deg. vs 49.4 (40 – 58) deg., p = 0.92 and HBW 203.4 (186 - 237) deg. vs 198.5 (174 - 240) deg., p = 073. Univariate logistic regression analysis showed that SD and HBW assessed by gated MPS were independent predictors of CRT response: SD (OR = 0.91; 95% CI 0.85-0.97; p < 0.05) and HBW (OR = 0.98; 95% CI 0.96-0.99; p =0.03). However MD indexes obtained by GBPS, did not show statistically significance in prediction of CRT response. Conclusion Gated MPS and GBPS are not interchangeable in terms of left ventricular MD assessment. Left ventricular SD and HBW obtained by gated MPI on CZT gamma-camera showed prognostic significance to predict CRT response.