Routine surveillance imaging following mild traumatic brain injury with intracranial hemorrhage may not be necessary

Abstract

BackgroundMild traumatic brain injury (mTBI) comprises the majority of pediatric traumatic brain injury. Children with mTBI even with traumatic intracranial hemorrhage (tICH) rarely experience a clinically significant neurologic decline (CSND). The utility of routine surveillance imaging in the pediatric population also remains controversial, especially owing to concerns about the risks of radiation exposure at a young age. This study aims to identify demographic or injury-related characteristics that may facilitate recognition of children at risk of progression with mTBI. MethodsWe performed a retrospective review of patients <16 years old with mTBI (GCS 13–15) and tICH admitted to a Level I pediatric trauma center between 2009 and 2014. Management of these patients was directed by the Cincinnati Children's Hospital Medical Center Minor Head Injury Algorithm. We reviewed each chart with emphasis on patient demographics, injury specific data, and radiographic or clinical progression. Results154 patients met inclusion criteria with mean age of 4 [0–16]; 116 sustained an tICH and 38 patients had isolated skull fractures. Repeat neuroimaging was obtained in 68 patients (59%). Only 9 patients (13%) with tICH had radiographic progression, none of which resulted in CSND. In addition, 9 patients experienced CSND, leading to neurosurgical intervention in 6 patients. Notably, none of these patients had repeat imaging prior to their neurologic changes. Both CSND and need for intervention were significantly higher in patients with epidural hematomas than other types of tICH (19.2% vs. 1.1%, p = 0.002). Of 154 patients, 19 did not have documented follow-up, 135 were seen as outpatients and 65 (48%) had follow up neuroimaging. All patients who had surveillance imaging in the outpatient setting had stable or resolved tICH. ConclusionFew children with mTBI and tICH experience clinical decline. Importantly, all patients that required neurosurgical intervention were identified by clinical changes rather than via repeat imaging. Our study suggests that in the vast majority of cases, clinical monitoring alone is safe and sufficient in patients in order to avoid exposure to repeat radiographic imaging. Level of EvidenceLevel III, prognostic and epidemiological.