Both individual patient-related and injury-related factors predict functional outcomes following moderate-to-severe traumatic brain injury (M/S TBI). Other than binary outcomes such as death, little is known about the role of day-of-injury neuroimaging in predicting long- term outcomes. Classification systems for assessing the severity of injury using computerized tomography (CT) scans, such as the Marshall Classification System (MCS) or Rotterdam scale, have not been systematically studied to see how they relate to long-term rehabilitation and functional outcomes following M/S TBI. The MCS consists of six categories based on information about midline shift, basal cistern compression, surgery evacuation, and lesion size. The Rotterdam scale, however, is a summed score ranging from 1-6 based on the extent of basal cistern compression, extent of midline shift, presence/absence of an epidural lesion, and presence/absence of traumatic subarachnoid hemorrhage (tSAH) or intraventricular blood. The differences between these two CT scales suggest the possibility that MCS and Rotterdam scales may differ in their ability to predict subsequent rehabilitation outcomes. Thus, we compared the relative predictive value of MCS and Rotterdam scores on long-term rehabilitation functional outcomes using the Functional Independence Measure (FIM) at rehabilitation discharge and nine-month post-discharge follow up. The study included 88 participants (25 females, mean age: 42.0 [SD: 21.3]) with M/S TBI. Day-of-injury CT images were scored using both MCS and Rotterdam criteria. Functional outcomes were measured by the cognitive and motor subscales on the FIM at discharge and after nine-month follow up, and length of stay in rehabilitation. Data were analyzed using multiple linear regression models. Neither MCS nor Rotterdam scores nor rehabilitation length of stay significantly predicted motor or cognitive outcomes at discharge or nine-month follow-up. MCS and Rotterdam scales may have limited utility in predicting long- term functional outcome in a rehabilitation setting, but instead appear to be good predictors of acute outcomes, especially regarding mortality and elevated intracranial pressure (ICP). Future research could focus on CT characteristics such as midline shift to predict long-term rehabilitation outcomes to guide treatment instead of CT rating scales.



College and Department

Family, Home, and Social Sciences; Psychology



Date Submitted


Document Type





traumatic brain injury, rehabilitation, computerized tomography