Observational Study on the Correlation Between Marshall CT Classification and Long-Term Functional Outcomes in Moderate to Severe Head Injury

Introduction: Traumatic brain injury (TBI) is a leading cause of disability and death, with moderate to severe cases often causing long-term impairments. The Marshall CT classification, developed in 1991, grades TBI severity based on initial CT findings like midline shift and mass lesions. While useful for acute prognosis, its role in predicting long-term functional outcomes is unclear. Objective: This study aimed to assess the correlation between Marshall CT classification scores and long-term functional outcomes in moderate to severe TBI patients, using the Glasgow Outcome Scale Extended (GOSE) and Functional Independence Measure (FIM) at one year postinjury, and to evaluate its predictive value for acute outcomes. Methods: Data from 1435 patients with moderate to severe TBI (2010-2024) in the Traumatic Brain Injury Model Systems National Database were analyzed. Marshall CT scores (Grades I-VI) were correlated with GOSE (global recovery) and FIM (motor/cognitive independence) scores at one year, and acute outcomes (mortality, neurosurgery). Multivariate regression adjusted for age, sex, and Glasgow Coma Scale. Results: Higher Marshall scores strongly predicted acute mortality (OR 2.7, 95% CI: 1.9-3.8, p < 0.001) and neurosurgical intervention (68% for Grades V-VI vs. 12% for I-II, p < 0.001). However, they showed weak correlations with one-year GOSE (r = 0.29, p = 0.02) and FIM (r = 0.25, p = 0.03), with age and rehabilitation access being stronger predictors. Discussion: The Marshall CT classification excels in acute prognostication but has limited utility for long-term functional outcomes, likely due to its focus on structural damage rather than factors like diffuse injury or rehabilitation. Complementary tools, such as advanced imaging or biomarkers, are needed for better long-term prognosis. Conclusion: The Marshall CT classification is effective for predicting acute TBI outcomes but weakly predicts long-term functional recovery. A multimodal prognostic approach is needed to improve long-term outcome prediction and personalize TBI care.