516 The Impact of In-Hospital Palliative Unit on Percutaneous Gastrostomy Practice, Volume, and Outcomes

Abstract

INTRODUCTION: Per-cutaneous gastrostomy (PEG) is often preferred for enteral feeding of patients with terminal and critical illness. Implementation of the palliative care often avoids the futile procedures in patients with short life expectancy. Inpatient palliative units have the potential to reduce placement of PEG in terminally ill patients. METHODS: In retrospective review, we analyzed all patients who underwent PEG from 2006 to 2015. The research protocol was approved by Institutional Review Board at our institution. The study population was divided into two groups based on the establishment of in-hospital palliative care unit at our institution in August of 2010. The demographic data including age, gender and ethnicity were reviewed. Endoscopy reports and patient charts were reviewed for the indication of PEG and post-procedural complications. The categorical variables were analyzed with chi-square and continues variable with student t-test. The annual procedural number were extrapolated from our endoscopy EMR. RESULTS: There were 780 PEGS placed on 721 unique patients from 2006 to 2015. Majority of them (62% n = 483) were placed before establishment of in-hospital palliative care unit (August 2010) and the remaining (38% n = 297) were done after Aug 2010. There was no significant difference in total number of hospitalizations or discharges during the study period. After implementation of palliative unit, number of PEGS decreased from 97/year to 52/year (Figure 2). The mean age of patient undergoing PEG was 69 years in both group. There was slight females preponderance (53% vs 50%). African American patients were predominant in both groups. Establishment of in-house palliative care unit led to statically significant decrease (34% to 7%) in number of Hispanic patients and an increase in white Caucasian (from 9% to 42%) undergoing PEGs (Figure 1). There was substantial change noted in the indication of the PEG (Figure 3) after inpatient palliative care unit services, with the predominant decrease for the PEG requirement in patients with dementia (from 42.9% to 24%, P-value < 0.001). There was a statistically significant decrease in the overall mortality (23.3% to 7.4%) and 30-day mortality (9.7% to 4%) for patients undergoing PEGS after the initiation of the palliative unit. However, overall PEG related complications were unaltered. CONCLUSION: The in-house palliative care units impact practice of PEG placement by decreasing procedural volume, lower overall and 30-day mortality after PEG placement.