Single center assessment of the role of Oakland score among patients admitted for acute lower gastrointestinal bleeding.

BACKGROUND/OBJECTIVES: The Oakland score was developed to predict safe discharge in patients who present to the emergency department with lower gastrointestinal bleeding (LGIB). In this study, we retrospectively evaluated if this score can be implemented to assess safe discharge (score ≤ 10) at WellStar Atlanta Medical Center (WAMC).

METHODS

A retrospective cohort study of 108 patients admitted at WAMC from January 1, 2020 to December 30, 2021 was performed. Patients with LGIB based on the ICD-10 codes were included. Oakland score was calculated using 7 variables (age, sex, previous LGIB, digital rectal exam, pulse, systolic blood pressure (SBP) and hemoglobin (Hgb)) for all patients at admission and discharge from the hospital. The total score ranges from 0 to 35 and a score of ≤ 10 is a cut-off that has been shown to predict safe discharge. Hgb and SBP are the main contributors to the score, where lower values correspond to a higher Oakland score. Descriptive and multivariate analysis was performed using SPSS 23 software.

RESULTS

A total of 108 patients met the inclusion criteria, 53 (49.1%) were female with racial distribution was as follows: 89 (82.4%) African Americans, 17 (15.7%) Caucasian, and 2 (1.9%) others. Colonoscopy was performed in 69.4% patients; and 61.1% patients required blood transfusion during hospitalization. Mean SBP records at admission and discharge were 129.0 (95% CI, 124.0-134.1) and 130.7 (95% CI,125.7-135.8), respectively. The majority (59.2%) of patients had baseline anemia and the mean Hgb values were 11.0 (95% CI, 10.5-11.5) g/dL at baseline prior to hospitalization, 8.8 (95% CI, 8.2-9.5) g/dL on arrival and 9.4 (95% CI, 9.0-9.7) g/dL at discharge from hospital. On admission, 100/108 (92.6%) of patients had an Oakland score of > 10 of which almost all patients (104/108 (96.2%)) continued to have persistent elevation of Oakland Score greater than 10 at discharge. Even though, the mean Oakland score improved from 21.7 (95% CI, 20.4-23.1) of the day of arrival to 20.3 (95% CI, 19.4-21.2) at discharge, only 4/108 (3.7%) of patients had an Oakland score of ≤ 10 at discharge. Despite this, only 9/108 (8.33%) required readmission for LGIB during a 1-year follow-up. We found that history of admission for previous LGIB was associated with readmission with adjusted odds ratio 4.42 (95% CI, 1.010-19.348, p = 0.048).

CONCLUSIONS

In this study, nearly all patients who had Oakland score of > 10 at admission continued to have a score above 10 at discharge. If the Oakland Score was used as the sole criteria for discharge most patients would not have met discharge criteria. Interestingly, most of these patients did not require readmission despite an elevated Oakland score at time of discharge, indicating the Oakland score did not really predict safe discharge. A potential confounder was the Oakland score did not consider baseline anemia during calculation. A prospective study to evaluate a modified Oakland score that considers baseline anemia could add value in this patient population.