Impact of scar on SPECT assay of left ventricular contraction dyssynchrony

Purpose: Many patients presently receiving cardiac resynchronization therapy (CRT) do not respond. A disproportionate number of nonresponders have ischemic cardiomyopathy, with significant left ventricular (LV) scar burden. Current selection criteria, such as electrocardiography or echocardiography, may not reliably portray the magnitude of CRT-remediable LV contraction dyssynchrony. Although phase analysis of gated single photon emission computed tomography (SPECT) image data is increasingly appreciated as a tool for quantifying dyssynchrony, its use in the setting of scar has not been adequately evaluated. Methods: Consecutive patients with ischemic (ICM, n = 50) or nonischemic (NICM, n = 39) cardiomyopathy underwent SPECT imaging prior to receiving CRT. In each patient, phase analysis of the raw images was performed to yield a phase standard deviation (PSD), an index which varies directly with the magnitude of dyssynchrony. ICM patient image data were also reanalyzed after scarred segments were stripped away. Results: Raw image analysis demonstrated that PSD was significantly larger among ICM (57 ± 17) than NICM (35 ± 13, p < 0.001) patients. Among ICM patients, PSD after stripping of scarred segments was significantly decreased (40 ± 13, p < 0.001). Signals emanating from scarred segments were of low amplitude and presented a random pattern, suggestive of noise rather than indicating contraction. Conclusion: PSD values may be spuriously increased by scar. These findings may be important when using SPECT in selecting ischemic cardiomyopathy patients for CRT.