Abstract 16233: Detection of Ischemic and Non-ischemic Cardiomyopathy by Gated Single Photon Emission Computed Tomography (SPECT) can be Improved by Novel Application of Phase Analysis Technique

Abstract

Background: Gated SPECT is often abnormal with patchy perfusion pattern in the non-ischemic cardiomyopathy patients. Phase analysis is a novel technique that allows to predict responses to cardiac resynchronization therapy. We investigated if phase analysis differentiates ischemic vs. non-ischemic cardiomyopathy independently of the perfusion.

Methods: Fifty-eight consecutive patients (2009-2011) with gated SPECT, heart failure and left ventricular ejection fraction (EF) ≤ 35% were age and gender matched into ischemic and non-ischemic cardiomyopathy groups based on coronary angiography and medical records. Gated SPECT studies were processed using Emory Toolbox (Atlanta, GA) with automated derivation of summed stress (SSS), rest (SRS), difference (SDS) scores, perfusion deficit size (PDS), EF, and phase analysis parameters of peak phase, phase standard deviation (SD), histogram bandwidth, skewness, and kurtosis. Sensitivity and specificity for detection of ischemic cardiomyopathy were determined for phase SD ≥ 43 degrees and bandwidth ≥ 135 degrees.

Results: Mean age was 57.8±9.8 years, 34 men, 24 women. There were more diabetic (20/29 vs. 10/29, p=0.0086) and dyslipidemic (27/29 vs. 14/29, p=0.0002) patients in the ischemic cardiomyopathy group, with insignificantly lower EF: 33.9±9.7 vs. 35.4±7.5. Significant differences between the groups are in the Table. Sensitivity and specificity of detecting ischemic cardiomyopathy independently of perfusion were 79% and 59% for phase SD, and 83% and 66% for histogram bandwidth.

Conclusion: Non-ischemic cardiomyopathy patients with heart failure were found to have abnormal perfusion pattern on gated SPECT, although with smaller and less severe defects than ischemic cardiomyopathy patients. A novel technique of phase analysis showed good sensitivity and specificity in determining the cause of cardiomyopathy independently of perfusion. The derivation of optimal cut offs is currently in progress.