Evaluation of the cervical spine (c-spine) in obtunded severely injured trauma patients is controversial, and spine immobilization is frequently prolonged. We examined the effect of two different c-spine evaluation protocols on c-spine immobilization and clinical outcomes. We prospectively evaluated consecutive intubated and mechanically ventilated patients admitted to the surgical intensive care unit (ICU) of a Level I academic trauma center with a diagnosis of multiple blunt injuries who had normal findings on high-resolution helical computed tomogram of C1 to T1 with reconstructions (HCTrecon). From July 1, 2003 to June 30, 2005 (n = 140), the findings of HCTrecon and either clinical examination or magnetic resonance imaging (MRI) were required to be normal to discontinue c-spine immobilization (clinical/MRI protocol). From July 1, 2005 to June 30, 2006 (n = 75), the policy was changed to require normal finding only on HCTrecon to discontinue c-spine immobilization (HCTrecon protocol). Patients evaluated by the clinical/MRI and HCTrecon protocols had similar baseline characteristics. Compared with clinical/MRI patients, HCTrecon patients had their c-spines immobilized for fewer days (median, 6 days vs. 2 days; p < 0.001), were less likely to experience a complication of c-spine immobilization (64% vs. 37%, p = 0.010), required shorter periods of mechanical ventilation (median, 4 days vs. 3 days; p = 0.011), and had shorter stays in the ICU (median, 6 days vs. 4 days; p = 0.028) and hospital (median, 16 days vs. 14 days; p = 0.043). There was no difference in hospital mortality (13% vs. 16%, p = 0.920) and no missed c-spine injuries in either group. Discontinuation of c-spine precautions based on the normal findings of HCTrecon decreases the duration of c-spine immobilization in obtunded severely injured patients and is associated with fewer complications, fewer days of mechanical ventilation, and shorter stays in the ICU and hospital.