Performance of the Oakland Score for Safe Discharge in Asian Patients After Acute Presentation With Lower GI Bleeding.

Objectives: The Oakland score, originally developed and validated in the Western population, aims to identify patients with lower gastrointestinal bleeding (LGIB) who are suitable for safe discharge. We aimed to validate the Oakland score in predicting safe discharge in our Asian cohort. Materials and Methods: A retrospective study was conducted on patients presenting with LGIB who underwent emergency colonoscopies at University Malaya Medical Centre, from January 2018 to December 2018. Demographics, clinical, laboratory parameters, and the Oakland score were evaluated to predict safe discharge, which was defined as the absence of blood transfusion, rebleeding, hemostatic intervention, readmission, and death. The performance of the Oakland score was assessed using the area under the ROC curve. Results: Of 133 patients, 33.8% fulfilled the criteria for safe discharge. Mean age was 66.6 years, predominantly males (64.7%) and from Chinese ethnicity (56.4%). For safe discharge, an Oakland score of ≤ 8 and ≤ 10 had a sensitivity of 97.7% and 95.5%, respectively. The specificities for both thresholds were low (4.4% and 24.4%, respectively). Packed cell transfusion showed a strong association with 'not safe for discharge' status (p < 0.001). Oakland score demonstrated good prediction for safe discharge with an AU-ROC of 0.82 (95% CI, 0.76-0.89). Conclusion: Consistent with previous studies, the Oakland score has shown good performance in this Asian-based population. However, the proportion of patients identified as safe for discharge in our study appears significantly lower compared to other studies. Further large-scale prospective validation studies are warranted before the application of this scoring system in our practice.