MedStar Georgetown University Hospital Washington, DC
Stephanie Woo, MD1, Krystina A. Johnson, MD1, Rebecca Xi, MD2, Olivia G. d'Aliberti, MS3, Nikki K. Duong, MD1, Joseph J. Jennings, MD1; 1MedStar Georgetown University Hospital, Washington, DC; 2Johns Hopkins University Hospital, North Potomac, MD; 3Georgetown University, Arlington, VA
Introduction: Approximately 300,000 patients are admitted annually for gastrointestinal bleeding (GIB) in the US and many require emergent endoscopy. Prophylactic intubation is used to prevent intra-operative cardiopulmonary complications. While scoring systems assist clinical decision-making, there are no guidelines to determine which patients should be prophylactically intubated. We aim to identify differences in post-procedural outcomes based on pre-procedural intubation status in emergent GI bleeding cases in our intensive care unit. Methods: Retrospective chart review from June 2016 to Dec 2018 of all emergent endoscopies for acute GIB at Medstar Georgetown University Hospital. 567 procedures were identified. Endoscopies indicated for follow-up or reasons other than GIB were excluded. Data are presented as means with 95% ± t-test confidence intervals, or frequency (%). Analysis of variance (ANOVA), or univariate logistic regression was used for continuous variables, the Pearsons c2 and Fischer Exact test was used for categorical variables. Results: 242 patients were included in the final analysis. There was a statistically significant difference between intubated and non-intubated patients for the following pre-endoscopy variables: procedure type, use of blood thinners, cirrhosis, kidney disease, pressor requirement and number of blood transfusions. Intubation was associated with higher APACHE scores (21 vs. 13, p < 0.001). There was a significantly greater number of post-procedural pneumonia (28% vs 14%), cardiovascular outcomes (11% vs 0%), and all-cause mortality (25% vs 18%) in the intubated group than the non-intubated group [Table 1]. Discussion: Risk calculation has the potential to change the standard of GIB management in critical care. Higher APACHE scores were predictive of pre-procedural intubation, rather than GBS. The decision whether or not to intubate a patient must take into account the increased risk of post-endoscopy pneumonia, cardiac arrest and death. Our study is a single-center retrospective study and limited by sample size and inherent selection bias. There is a compelling imperative for future prospective studies to determine which patients benefit from intubation and their outcomes. Ultimately, this shared decision making should involve the gastroenterology, anesthesiology, and critical care teams.
Table 1. Baseline Characteristics and Clinical Outcomes
Disclosures: Stephanie Woo indicated no relevant financial relationships. Krystina Johnson indicated no relevant financial relationships. Rebecca Xi indicated no relevant financial relationships. Olivia d''Aliberti indicated no relevant financial relationships. Nikki Duong indicated no relevant financial relationships. Joseph Jennings indicated no relevant financial relationships.