David J. Bjorkman, MD, MSPH (HSA), SM (Epid.) Reviewing Kumar NL et al., Gastrointest Endosc 2016 Sep 29;
Earlier endoscopy — performed within 12 versus 24 hours of bleeding presentation — was associated with worse outcomes in low-risk patients.
Most guidelines recommend that patients with nonvariceal upper gastrointestinal bleeding (UGIB) undergo endoscopy within 24 hours of presentation. But whether earlier endoscopy is associated with improved outcomes is unclear.
To examine this issue, investigators at a single academic medical center retrospectively reviewed the records of 361 patients with acute nonvariceal UGIB and compared outcomes between those who received standard endoscopy (SE) within 24 hours and those who received urgent endoscopy (UE) within 12 hours; 25% of patients received UE.
Overall, 10% of patients met the primary endpoint of inpatient mortality, rebleeding, or the need for additional endoscopic, surgical, or radiographic intervention. In a multivariate analysis, time to endoscopy was a predictor of the endpoint, with SE patients less likely to have a bad outcome (adjusted odds ratio, 0.81 per 6 hours; 95% CI, 0.69–0.94). Lower systolic blood pressure was also associated with worse outcomes (aOR, 0.82 per 10 mm Hg; 95% CI, 0.70–0.95). In low-risk patients (Glasgow-Blatchford score <12), but not high-risk patients, UE was associated with the endpoint. Patients undergoing UE were also more likely to have endoscopic therapy and blood transfusions.
The authors conclude that UE is a predictor of worse outcomes in low-risk patients and suggest further study on the timing of endoscopy.
Whereas prior studies have shown no benefit for UE, these results suggest that patients who receive UE have a worse outcome. As the authors note, an important factor not available in the study is the rationale for UE. It is possible that other factors associated with worse outcomes prompted UE in “low-risk” patients, confounding the analysis. The impact of the timing of endoscopy on clinical outcomes in low- and high-risk patients requires additional study.