Background: Scheduled repeat head computed tomography after mild traumatic brain injury has been shown to have limited use for predicting the need for an intervention. We hypothesized that repeat computed tomography in persons with intracranial hemorrhage and a Glasgow Coma Scale (GCS) score of 13 to 15, without clinical progression of neurologic symptoms, does not impact the need for neurosurgical intervention or discharge GCS scores.
Methods: This prospective cohort study followed all patients presenting to our urban Level I trauma center with intracranial hemorrhage and a GCS score of 13 to 15 from February 2010 to December 2010. Subjects were divided into two groups: those in whom repeat CT scans were performed routinely (ROUTINE) and those in whom they were performed selectively (SELECTIVE) based on changes in clinical examination. CT scanning decisions were made at the discretion of the neurosurgical service attending physician.
Results: One hundred forty-five patients met the inclusion criteria (ROUTINE, n = 92; SELECTIVE, n = 53). Group demographics, including age, sex, and presenting GCS score were not significantly different. Of SELECTIVE patients, six (11%) required a repeat head computed tomography for a neurologic change, with one having a radiographic progression of hemorrhage (16%) versus 26 (28%) of 92 in the ROUTINE group showing a radiographic progression. No patient in either group required medical or neurosurgical intervention based on repeat scan. The number of CT scans performed differed between the two groups (three scans in ROUTINE vs. one scan in SELECTIVE, p < 0.001), as did the intensive care unit (2 days vs. 1 day, p < 0.001) and hospital (5 days vs. 2 days, p < 0.001) lengths of stay. Discharge GCS score was similar for both groups (15 vs. 15, p = 0.37). One death occurred in the SELECTIVE group, unrelated to intracranial findings. The negative predictive value of a repeat CT scan leading to neurosurgical intervention with no change in clinical examination was 100% for both groups.
Conclusion: A practice of selective repeat head CT scans in patients with traumatic brain injury admitted with a GCS score of 13 to 15 decreases use of the test and is associated with decreased hospital length of stay, without impacting discharge GCS scores.
Level of evidence: Diagnostic study, level II.