PURPOSE: Serum D-dimer has been proposed as a biomarker to aid in the diagnosis of pediatric traumatic brain injury (TBI). We investigated the accuracy of D-dimer in predicting the absence of TBI and evaluated the degree by which D-dimer could limit unnecessary computed tomography scans of the head (CTH). METHODS: Retrospective review of patients with suspected TBI from 2011 to 2013 who underwent evaluation with CTH and quantitative D-dimer. D-dimer levels were compared among patients with clinically-important TBI (ciTBI), TBI, isolated skull fracture and no injury. RESULTS: Of the 663 patients evaluated for suspected TBI, ciTBI was identified in 116 (17.5%), TBI in 77 (11.6%), skull fracture in 61 (9.2%) and no head injury in 409 (61.7%). Patients with no head injury had significantly lower D-dimer values (1531+/-1791pg/muL) compared to those with skull fracture, TBI and ciTBI (2504+/-1769, 2870+/-1633 and 4059+/-1287pg/muL, respectively, p<0.005). Using a D-dimer value <750pg/muL as a negative screen, no ciTBIs would be missed and 209 CTHs avoided (39.7% of total). CONCLUSION: Low plasma D-dimer predicts the absence of ciTBI for pediatric patient with suspected TBI. Incorporating D-dimer into current diagnostic algorithms may significantly limit the number of unnecessary CTHs performed in this population. TYPE OF STUDY: Study of diagnostic test. LEVEL OF EVIDENCE: I.