Objective: Mild traumatic brain injury (mTBI) constitutes majority of TBI cases. A considerable portion of mTBI cases has intracranial imaging abnormalities (complicated mTBI), which pose challenges in the diagnosis and management to clinicians. Here, we aimed to evaluate the early clinical course and outcomes of pediatric complicated mTBI cases treated at a large volume tertiary referral center. Materials and Methods: A single center retrospective cohort study was conducted at a large volume tertiary trauma referral center for 12 month period between 2017 and 2018. Pediatric mTBI cases with at least one of the following computed tomography (CT) findings were included a depressed skull fracture, pneumocephalus, intracranial hemorrhage, edema, or contusion. Demographic, clinical, and radiological data were collected and analyzed. Results: One hundred and twenty four patients with complicated mTBI were identified. Falls were the leading mechanisms of trauma (71.8%). Most patients (90.3%) had a Glasgow coma score (GCS) of 15 at initial evaluation. Most frequent radiological findings on initial CT scan were epidural hematoma (EDH) (34.7%) and pneumocephalus (31.5%), followed by subdural hematoma (SDH) (19.4%), subarachnoid hemorrhage (16.9%), contusion (14.5%), and depressed skull fracture (8.1%). Radiological findings in the routine repeat CT scan were stable in 55.6% of the patients, whereas the findings progressed in 15.3% and improved in 29% of patients during this interval period (median 7 h). Neurosurgical operation was performed in 7 (5.6%) patients. Thirty six (29%) patients were identified as having clinically important TBI (ciTBI). Average length of stay at emergency department was 9.7 ± 4.9 h, and the average length of hospital stay was 3.6 ± 2.3 days. Multivariate analysis revealed that age, GCS, pneumocephalus, depressed skull fracture, EDH, and SDH were independent predictors of ciTBI. Conclusion: Pediatric complicated mTBI is associated with higher rates of hospitalization and therefore ciTBI but relatively lower rates of need for neurosurgery. Effective decision making tools and algorithms are needed to guide optimal management strategies of these patients.