Blood Urea Nitrogen as a Prognostic Indicator of Non-variceal Upper Gastrointestinal Bleed

Abstract

Purpose: Non-variceal upper gastrointestinal tract bleeding is a life-threatening disorder affecting approximately one per 1,000 people per year. Several risk-scoring systems (Rockall, Blatchford) have been developed for the assessment of patients presenting with upper gastrointestinal hemorrhage with some differences, including sources of bleeding and outcome of interest. We hypothesize that simple biochemical parameters, such as the blood urea nitrogen (BUN) level, will be able to predict outcomes in patients with non-variceal upper gastrointestinal bleeds. Methods: We did a retrospective chart review of 100 consecutive patients admitted with a diagnosis of non-variceal upper gastrointestinal bleed. We collected pertinent data related to patient demographics, coexisting co-morbidities, admission laboratory results, endoscopic findings and specific outcomes (listed below). We defined our high BUN cohort as those patients who presented with a level >20. Our primary endpoints were whether elevated BUN could predict the length of hospital stay (LOS) and the need for intensive care unit (ICU) admission. Our secondary endpoints were the association of high BUN with high grade endoscopic findings (defined as Forrest Classification Grade Ia, Ib and IIa), ICU LOS, and a composite endpoint of the need for surgery and occurrence of rebleed. Results: More patients in the high BUN group were admitted to the ICU (6.25% vs. 40.4%, p=0.007). Among patients admitted to the ICU, those with a low BUN had a shorter ICU LOS (1.0 vs. 3.8 days, p=0.02). We were able to predict the high-grade endoscopic findings with the BUN levels with an excellent degree of sensitivity, but poor specificity (100% and 24% respectively). Taking a composite endpoint for need for surgery or rebleed in 30 days, the cohort with a higher BUN had a higher incidence of poor outcomes (46.4% vs. 12.5%, p=0.04).Patients with a high BUN had a longer hospital length of stay (10.2 vs. 8.2 days). However, this difference did not reach statistical significance (p=0.36). Conclusion: Our study concludes that a readily available parameter like BUN can help predict the likelihood of admission to the ICU, ICU LOS and occurrence of complications in patients with non-variceal upper GI Bleed. BUN also predicts the grade of the peptic ulcer findings with excellent sensitivity. Further studies are required to assess the utility of this test in a prospective manner.