Utility of the Rockall Score in Identifying Very Low-Risk Patients with Non‑variceal Upper Gastrointestinal Bleeding

Objective: The Rockall score (RS) together with the Glasgow-Blatchford score (GBS) are commonly used to predict adverse clinical outcomes (ACO) in non-variceal upper gastrointestinal bleeding (NV-UGIB). However, their ability to identify patients who remain free of ACOs is uncertain. This study aimed to evaluate the usefulness of RS and GBS for identifying patients at low risk of ACOs. Methods: In this cross-sectional analysis, 134 individuals diagnosed with NV-UGIB were enrolled. The GBS was assessed prior to endoscopy, while the RS was determined following the procedure. Participants were monitored for adverse clinical outcomes (mortality, rebleeding, transfusion requirement, endoscopic therapy, surgical intervention, or ICU admission). Based on the presence or absence of ACOs, patients were categorized accordingly, and their clinical, laboratory, and scoring characteristics were compared. Results: Of the study cohort, fifteen patients (11%) did not experience any adverse clinical outcome (ACO). In comparison with those who developed ACOs, patients without ACO demonstrated significantly lower white blood cell counts and blood urea nitrogen levels, along with higher hemoglobin and hematocrit values (all p < 0.05). Mean Rockall score (RS) and Glasgow-Blatchford score (GBS) were also markedly reduced in the non-ACO group (p < 0.05). Multivariate regression identified RS and hematocrit as independent predictors of the absence of ACO. A one-point reduction in RS was associated with a 37.5% greater probability of remaining free from ACO. Receiver operating characteristic (ROC) analysis yielded optimal cutoff points of RS ≤ 4 and GBS ≤ 7, providing acceptable sensitivity and specificity. Conclusion: Our study showed that initial RS and GBS evaluation in patients with NV-UGIB may be useful in identifying low-risk cases as well as high-risk cases.