Reply: Can LV Dyssynchrony as Assessed with Phase Analysis on Gated Myocardial Perfusion SPECT Preferably Predict Response to CRT?

REPLY: We appreciate the important comments regarding our recent study on left ventricular (LV) dyssynchrony assessment with gated myocardial perfusion SPECT and its application in the prediction of response to cardiac resynchronization therapy (CRT) (1). We agree with Dr. Song that myocardial contraction is a complex 3-dimensional process. In our study, all patients underwent gated myocardial perfusion SPECT before CRT implantation, and data were reconstructed by filtered backprojection and then reoriented to yield gated short-axis images. Accordingly, Dr. Song suggests that in our study, LV dyssynchrony was assessed from circumferential motion. However, this is not entirely correct.

Phase analysis uses a count-based method to extract amplitude and phase from the regional LV count changes throughout the cardiac cycle. These count changes are a function of the change in thickness in the radial, circumferential, and longitudinal directions. From the stacked LV short-axis datasets, 3-dimensional count distributions are extracted and are subsequently submitted to Fourier analysis. Phase analysis uses a first-harmonic fast Fourier transform to extract 3-dimensional regional phases (phase arrays). These calculated phase arrays describe the regional onset of mechanical contraction of the myocardium in 3 dimensions (2).

Dr. Song questions whether sensitivity and specificity for prediction of response to CRT may be better when LV dyssynchrony is assessed from longitudinal motion or from both longitudinal and circumferential motion. Because phase analysis with gated myocardial perfusion SPECT, as applied in our study, takes into account myocardial contraction in all 3 dimensions, we cannot provide this information. This count-based method has at least 2 advantages over wall-motion–based methods: First, because it does not need a frame of reference to assess wall motion, it is more objective; second, because it directly assesses myocardial contraction, it is much less affected by passive wall motion, which can take place in a noncontracting region that is tethered rather than actively contracting. We believe that gated myocardial perfusion SPECT phase analysis, by including all 3 components of myocardial contraction, is an attractive alternative for LV dyssynchrony assessment.