Is Routinely Repeated Brain Computed Tomography Necessary after Traumatic Brain Injury in an Emergency Setting?

Ko, Heuk Sang; Kim, Gab Teog; Kang, Keung Ho

Abstract

PURPOSE: Computed tomography (CT) is the standard diagnostic method employed in cases of blunt head trauma, and repeat CT (RCT) scans are very often obtained in order to monitor for acute progression of intracranial pathology. The purpose of this study was to retrospectively evaluate the utility of the routine RCT scans in patients with traumatic brain injury (TBI), and to suggest the guidelines for RCT scans of the head.
METHODS
The trauma registry and medical records of patients who were admitted to the emergency department of our university hospital from January 2004 to December 2006 were retrospectively reviewed. All patients admitted with TBI who received RCT scans of head were enrolled in this study. Results of initial head CT scans, indications for RCT (routine vs. neurologic change), and neurosurgical interventions (craniotomy and extraventricular drainage) were recorded. For patients who received worsened or unchanged subsequent RCT scans, the scans were compared and independent predictors of a worsened RCT were identified by stepwise logistic regression. Patients were categorized according to the Glasgow Coma Scale (GCS) as having mild (GCS 14-15), moderate (GCS 9-13), or severe (GCS< or =8) head injury.
RESULTS
All inclusion criteria were met in 338 patients. Most (70.7%, n=239) RCT scans were preformed on a routine basis, whereas 29.3% (n=99) were ordered in response to neurological change. One hundred eleven (32.8%) patients showed signs of worsening on RCT, and 62 (13.8%) required neurosurgical intervention. Risk factors associated with worsening on CT evaluations were coagulopathy; skull fracture; multiple lesions; and an initial diagnosis of acute subdural hematoma, acute epidural hematoma, or hemorrhagic contusion. No patient with a mild or moderate TBI without neurological deterioration underwent an neurosurgical intervention after routine RCT. However, of the 48 patients with severe TBI who underwent routine RCT, the repeat scanning led to neurosurgical intervention in 5 patients (10.4%).
CONCLUSION
In patients with mild and moderate TBI without clinical deterioration, routine RCT does not alter management and is unnecessary. Nevertheless, conclusions about patients with moderate TBI should be drawn with caution. Routine RCT is unequivocally indicated for patients with severe TBI, because the results sometimes dictate neurosurgical intervention even in the absence of obvious neurological deterioration.