S794 In-Hospital Outcomes and Healthcare Burden of Fistulizing Crohn’s Disease Requiring Parenteral Nutrition

Abstract

Introduction: The transmural inflammation characteristic of Crohn’s disease (CD) predisposes patients to the formation of enterocutaneous fistulas (ECF), which is a major cause of morbidity and worse outcomes. Parenteral nutrition (PN) is frequently prescribed to provide bowel rest in efforts for closure, however PN might be associated with an increased risk of adverse events. We sought to analyze the burden and outcomes related to ECF and CD with PN and without PN use. Methods: A retrospective cohort study of the 2016-2017 National Inpatient Sample database was performed, using ICD10-CM/PCS codes to identify patients discharged with a primary diagnosis ECF with CD and those receiving PN. Primary endpoint was in-hospital mortality. Secondary endpoints were length of stay (LOS), in-hospital complications and advanced therapy, hospital related charges. Multivariate regression analysis model to adjust for patient and hospital characteristics, was performed for both primary and secondary outcomes. Results: Of 10,230 admitted with ECF and CD, 10.4% (1,065) received supplemental PN. These patients were more likely to belong to a higher income bracket (21.3% vs 28.2%; P = 0.02), and less likely to be self-pay (4.3% vs 0.1%; P < 0.01). They were more likely to abuse opiates (2.2% vs 6.1%; P < 0.01), have moderate or severe protein-caloric malnutrition (17.5% vs 68.5%; P < 0.01), vitamin D deficiency ( 4.2% vs 7.0%; P = 0.05), hypovolemic hyponatremia ( 4.9% vs 13.2; P < 0.01), hypokalemia (12.7% vs 22.5%; P < 0.01), hypophosphatemia (2.2% vs 11.3%; P < 0.01), upper GI bleeding (0.1% vs 0.5%; P < 0.01) and required transfusion of blood products ( 4.6% vs 13.6%; P < 0.01). On multivariate analysis LOS was longer [6.87 Days; (P < 0.01)] in patients receiving PN, they also had higher healthcare burden including charges [50,732 US$; (P < 0.01) and costs [14,022 US$; (P < 0.01)]. We found higher rates of endotracheal intubation [OR 10.25; (P < 0.01)], prolonged mechanical ventilation (>96 hours) [OR 7.90; (P = 0.02)] and tracheostomy requirement [OR 4.17; (P < 0.01)], however no difference in in-hospital mortality was found [OR 1.25; (P = 0.82)]. Conclusion: PN use in CD with ECF, is not associated with higher in-hospital mortality. However, we found an increased overall in hospital burden, including LOS, healthcare utilization, and prolonged intensive care resource utilization. Further studies are needed to assess why those differences occur between the two groups.Table 1.: Availability of Six FDA-Approved Agents in Commercial vs Governmental Formularies