Predictors Of Surgical Feeding Tube Placement In Acute Ischemic Stroke

Introduction: Delays in decisions to pursue percutaneous endoscopic gastrostomy (PEG) tube placement may prolong hospitalization in patients with acute ischemic stroke (AIS). Prediction models in AIS have had inconsistent findings and have included race as a predictor despite there being no clear biological bases. We hypothesized that stroke severity would predict PEG placement and race would not be a significant predictor after controlling for other variables. Methods: Consecutive cases of AIS were identified from our inpatient registry (6/14- 4/15). Patients transferred to hospice or dying within 3 days of hospitalization were excluded. NIHSS, NIHSS level of consciousness 1A score, medical history, and demographic data were included in the model. Backward stepwise elimination using logistic regression was used to assess predictors of PEG placement. We also explored relationships between PEG placement and Penetration-Aspiration (PEN-ASP) scores (1-2, normal; 3-5, penetration, 6-8, aspiration) calculated using Fiberoptic Endoscopic Evaluation of Swallowing and Modified Barium Swallow Studies. Results: Among 757 AIS patients, 17.9% (136) of patients received a PEG tube. There was no association between race and PEG tube placement in univariate (Table 1; p=0.205 pearson Chi2) or multivariable models. In the multivariable model, significant predictors of PEG placement included arrival NIHSS (p Conclusions: Our study represents the largest cohort of AIS patients undergoing PEG placement. Stroke severity and medical history, but not race, predict PEG placement in this study. Future AIS dysphagia studies should include objective dysphagia scales.