Prone SPECT myocardial perfusion imaging is associated with less cardiac drift during the acquisition duration than imaging in the supine position

Abstract

Cardiac cranial drift, a slow vertical upward displacement seen during the acquisition of myocardial single photon emission tomography (SPECT), is a source of image artefacts that may lead to erroneous interpretation. Changes in breathing pattern and depth throughout image acquisition are believed to cause cardiac cranial drift. As the physiology of respiration probably differs with postural changes, we hypothesized that cardiac drift may be different for supine vs. prone acquisitions. Our aim was to assess the magnitude of cardiac displacement for prone and supine SPECT acquisitions in patients undergoing stress myocardial perfusion imaging. We enrolled prospectively 15 subjects undergoing exercise myocardial perfusion imaging. Subjects had post-stress images acquired in both the prone and supine positions. Motion was assessed in the horizontal (x) and vertical (y) axes for both camera heads at all 64 projections at which images were obtained. Pixel displacement (number of pixels from the baseline of zero) in either the cranial/caudal or left/right direction was quantified using the automated camera motion correction algorithm. Supine imaging was associated with more cranial drift than prone imaging (1.20+/-0.40 pixels vs. 0.92+/-0.24 pixels, P<0.05). There was no significant difference in cardiac displacement in the horizontal axis (1.03+/-0.5 pixels vs. 1.12+/-0.22 pixels, P=NS). Prone imaging is associated with less cardiac cranial drift than imaging in the supine position, suggesting that the former is associated with a more constant and reliable diaphragmatic breathing pattern. Acquisitions in the prone position may thus be associated with fewer motion artefacts than supine acquisitions for cardiac SPECT imaging.