Abstract TMP36: Anatomical Predictors of Gastrostomy Tube Placement After Large Vessel Occlusion Ischemic Stroke

Abstract

Dysphagia is a common neurologic deficit after ischemic stroke; as a result, patients often require percutaneous endoscopic gastrostomy (PEG) tube placement for safe maintenance of sufficient caloric intake. Right hemispheric strokes have previously been associated with post-stroke dysphagia. We examined whether topographic locations of post-stroke diffusion-weighted MRI (DWI) lesions associated with the need for placement of PEG tube. A registry of 898 patients evaluated for acute treatment of suspected large vessel occlusion (LVO) stroke was used. 65 patients underwent post-stroke PEG placement, and 65 additional patients were selected as propensity matches based on age, baseline NIH Stroke Scale score, and recanalization status. Binary masks of 24-to-72-hour post-stroke DWI lesions were co-registered to standard template space. Voxel-based lesion symptom mapping (V2.6), rewritten to perform logistic regression at each voxel, was used to generate statistical maps of lesion contribution to PEG placement. Uncorrected t-statistic maps demonstrated voxels in the right frontal, parietal, and temporal regions were associated with post-stroke PEG placement. Upon controlling for age and/or recanalization status, lesions in the right parietal lobe were associated with need for PEG. After controlling for lesion volume, this association weakened; once all variables were controlled for, there were no topographical regions associated with PEG placement. In this limited series, there is not a topographic region on post-stroke diffusion MRI that is significantly associated with need for PEG placement in patients with LVO stroke after controlling for appropriate covariates. PEG tube requirement may be a composite outcome, dependent on various inputs notably including lesion volume.