165 HAM Score: Predicting Outcomes of Initial Conservative Management for Moderate Traumatic Frontal Hemorrhage of Volume Less Than 20 CC

INTRODUCTION: This study compares the outcomes of patients with moderate traumatic frontal hemorrhage of less than 20 cc, treated surgically or conservatively within 72 hours. It identifies early demographic, clinical, radiological, and laboratory factors that can predict outcome differences, providing a basis for a scoring system. METHODS: A four-year multicenter retrospective mixed descriptive and analytical cross-sectional study in Ethiopia reviewed 11,250 CT images and enrolled 271 patients. Using the Extended Glasgow Outcome Score (E-GOS) at 3 months, the outcomes of the two groups were assessed: E-GOS less than six is considered an unfavorable outcome. Univariate analysis of independent factors was conducted over four combined treatment and outcome groups. A multiple logistic regression model was developed based on significant independent variables with a p-value < 0.05. RESULTS: Initial Conservative and Surgical groups account for 68.27% and 31.73% respectively. 74.17% of all patients had a favorable outcome, with 84.88% and 69.19% favorable outcomes in the Surgical and Conservative groups respectively. All 14 mortalities were in the initial conservative management group. The study revealed a significant but weak association between the treatment group and the outcome (p-value: 0.0094, Cramer’s V value: 0.146). The independent variables that significantly affect the clinical outcomes are: Age, admission blood pressure, risk of hypoxia, GCS scores, side and location of the frontal hemorrhage, number and volume of isolated frontal hemorrhages, presence of a mass effect, degree of midline shift, presence of traumatic hemorrhage in other lobes, and presence of other primary traumatic pathology. CONCLUSIONS: A risk outcome prediction score (HAM score) developed showed promising results to identify and prioritize patients likely to have unfavorable outcomes from conservative management, but further external validation is needed.