GP.05 Intraoperative acquisition of diffusion tensor imaging in cranial neurosurgery: readout-segmented DTI versus standard single-shot DTI

Abstract

Background: Diffusion-tensor imaging (DTI) tractography is commonly used in neurosurgical practice, but is largely limited to the preoperative setting. This is due primarily to image degradation caused by susceptibility artifact when conventional single-shot (SS) echo-planar imaging DTI is acquired for open cranial, surgical position intraoperative DTI (iDTI). A novel, artifact-resistant, readout-segmented (RS) DTI has not yet been evaluated in the intraoperative MRI (iMRI) environment. Our objective was to evaluate the performance of RS-DTI versus SS-DTI for intraoperative white matter imaging. Methods: Pre- and intraoperative 3T, T1-weighted and DTI (RS-iDTI and SS-iDTI) in 22 adults undergoing intraaxial iMRI resections (low-grade glioma: 14, 64%; high-grade glioma: 7, 32%; cortical dysplasia: 1). Regional susceptibility artifact, anatomical deviation relative to T1WI, and tractographic output were compared between iDTI sequences. Results: RS-iDTI resulted in less regional susceptibility artifact and mean anatomic deviation (RS-iDTI: 2.7±0.2 mm versus SS-iDTI 7.5±0.4 mm; p<0.0001). Tractographic failure occurred in 8/22 (36%) patients for SS-iDTI whereas RS-iDTI permitted successful reconstruction in 4 of these 8. Maximal tractographic differences between DTI sequences were substantial (mean 9.7±5.7 mm). Conclusions: Readout-segmented EPI enables higher quality and more accurate DTI for surgically relevant tractography of major white matter tracts in intraoperative, open cranium, neurosurgical applications at 3T.