Abstract WP382: Hospital Length of Stay in Stroke: Relationship to Feeding Status and Discharge

Abstract

Introduction: Hospital length of stay (LOS) has been associated with clinical outcome and cost of care. Factors influencing LOS provide an opportunity for cost containment and acute inpatient rehab facilitation. In patients with dysphagia, early alternative feeding options may limit malnutrition and improve rehab potential. Data linking percutaneous endoscopic gastrostomy (PEG tube) placement and LOS is limited. In this study we evaluated stroke LOS with respect to clinical severity, PEG tube, and discharge disposition. Methods: Prospective data on comprehensive stroke center patients was analyzed. Data analyzed includes demographics, pre-stroke Rankin Score (mRS), daily National Institutes of Health Stroke Scale (NIHSS), stroke subtype, disposition (home, acute rehab, skilled nursing facility (SNF), Hospice/Death), LOS, and PEG tube placement. Groups of mild (0-4), moderate (5-15), mod-severe (16-20) and severe (>20) were defined based on NIHSS. Results: 390 patients over 6 months had 73% ischemic stroke, 11% ICH, and 15% TIA. Pre-stroke mRS was ≤ 2 in 87%. About 89% had mild or moderate deficits, and 11% had moderate-severe or severe stroke. Mean LOS was 4.8 days in patients without PEG versus 16.8 days in patients with PEG. PEG tube was placed in 19%, 38.5% and 40% in moderate, mod-severe, and severe stroke patients, respectively. 10% (36/361) were palliated without a PEG. Among full support patients, 21.6% moderate, 50% moderate to severe, and 83.3% of severe strokes received a PEG tube prior to discharge. Average time from admission to PEG tube consult was 3.8 ± 2.8 days. Time to PEG tube placement was 10 ± 4.7 days. Recorded reasons for PEG placement delay included family decision/consent process and fever, however the impact of scheduling was not recorded. Conclusion: Both stroke severity and LOS are strongly associated with PEG tube placement. PEG tube placement occurs after patient and family discussion, expert consultation, and a scheduled procedure. There may be modifiable factors such as narrative and decision making / consent, which can reduce LOS in patients with severe strokes. The authors suggest future research on this topic may include an early PEG versus palliation pathway to facilitate early discharge to rehab or SNF.