SESSION TITLE: Non Pulmonary Critical Care Posters
SESSION TYPE: Original Investigation Poster
PRESENTED ON: Wednesday, October 29, 2014 at 01:30 PM - 02:30 PM
PURPOSE: The primary objective is to establish the prognostic implication of plasma troponin I level upon admission to the ICU in patients with acute gastrointestinal bleeding. The secondary objective is to establish factors that are associated with increased troponin I levels in these patients.
METHODS: Design: Retrospective cohort study Setting: Intensive Care Unit at The University of Louisville Hospital Patients: The patient population was derived from a database of patients with acute gastrointestinal bleeding admitted to the University of Louisville Hospital. Those who had plasma troponin I level measured were included in this study.
RESULTS: Our study included 165 patients, of whom 75 (45.5%) had a plasma troponin I level > 0.034 mcg/L. Using the latter cutoff, there was no significant increase in the odds of in-hospital mortality for those with elevated level of troponin I (OR: 1.46; 95% CI: 0.68 to 3.18; P = 0.33). A level > 0.1 mcg/L was associated with a significant increase in the odds of in-hospital mortality (OR: 2.96; 95% CI: 1.24 to 7.07; P = 0.015). However, after adjustment for need for mechanical ventilation, need for pressor, and liver cirrhosis, the association was no longer significant. On univariate analysis, the only variable associated with elevated troponin I level was heart failure (OR: 2.64; 95% CI: 1.10 to 6.33; P= 0.03).
CONCLUSIONS: An increase in plasma troponin I level above 0.1 mcg/L is associated with a 3 fold increase in in-hospital death in patients with acute gastrointestinal bleeding admitted to the ICU. However, this association is no longer significant after adjustment for other variables. Off the clinical and laboratory variables that were measured, only heart failure could predict an elevation of troponin I levels.
CLINICAL IMPLICATIONS: This study shows that although significant elevations of troponin I levels are associated with higher mortality in patients who are admitted to the ICU with GI bleed, it is not an independent predictor of mortality. Other clinical and laboratory variables need to be taken into account when attempting to determine a prognosis for these patients.
DISCLOSURE: The following authors have nothing to disclose: Rahul Sinha, Udit Chaddha, Karim El Kersh, Matthew Rayner-Lawren Woodford, Juan Guardiola, Mohamed Saad, Rodrigo Cavallazzi
No Product/Research Disclosure Information