In patients with acute nonvariceal upper GI bleeding (UGIB), increased blood urea nitrogen (BUN) levels at 24 hours were associated with worse outcomes. The marker, already proven useful in acute pancreatitis, could help physicians determine a patient’s prognosis.
Existing measures of UGIB risk are effective, but only about 30% of physicians ever calculate risk scores when evaluating UGIB patients, perhaps because they require measurements at multiple time points. “We personally think the reason for this is the busyness of clinical practices, especially the acute nature of upper GI bleeding. It’s often hard to step back to calculate a score that has multiple variables,” said study author, a fellow in gastroenterology at Brigham and Women’s Hospital, Boston.
The study was published in Gastrointestinal Endoscopy (2017.).
Like acute pancreatitis, upper GI bleeding requires resuscitation management, which suggested that BUN levels might be a useful marker in this condition as well. To find out, the researchers analyzed data from 357 patients who were treated at the Brigham and Women’s Hospital emergency department and ultimately hospitalized for UGIB during 2004-2014.
The researchers analyzed BUN levels measured at admission and at the time closest to 24 hours after hospitalization, which ranged from 6 hours to 48 hours.
Thirty-seven patients (10%) experienced an increase in BUN level, while all the rest had levels that stayed steady or decreased. Those patients with BUN increases had a lower mean Glasgow-Blatchford(7.8 vs. 9.6; P =.010), but there was no difference in AIMS65 .
Patients with BUN increases had greater odds of the composite outcome, which included inpatient death from any cause, inpatient rebleeding, a need for surgical or radiologic intervention, and/or a need for endoscopic reintervention during hospitalization (22% vs. 9%; P =.014). Inpatient mortality was higher in the increased BUN group (8% vs. 1%; P =.004).
Overall, BUN increase at 24 hours was associated with an odds ratio of 2.75 for the composite outcome (95% confidence interval, 1.13-6.70; P = .026).
A potential limitation to using the BUN is that it could just be catching patients with underlying renal disease. But when researchers adjusted for this, the odds ratio for increased BUN remained significant (OR, 3.00; P =.021).
“The nice part of the study is that it’s so easy to interpret and apply in a clinical setting. You just need two data points: BUN at presentation and at 24 hours. If the BUN level has risen, you need to have a higher degree of suspicion for the prognosis of those patients,” said Dr. Kumar.
The downside to BUN is that it doesn’t provide information for the first 24 hours. For that reason, BUN shouldn’t replace measures like the Glasgow-Blatchford score and the AIMS65 score. “But it’s very helpful to use this change in BUN score to get a sense of where the patient is trending. If it’s rising, there’s a higher risk of worse outcomes, and this could influence decisions about whether the patient should be in the ICU or the medical ward,” said Dr. Kumar.