MicroRNAs (miRNAs), which are small noncoding RNAs involved in gene regulation that can be measured reliably in saliva, have emerged as promising biomarker candidates for identifying concussion. However, research thus far is limited, and the role of miRNAs in prognosis following concussion in adolescents remains largely unexplored. This case-cohort study investigated whether salivary miRNAs collected within 10 days of injury could identify concussed adolescents from age- and sex-matched, healthy, uninjured controls, and predict symptom burden at 3-weeks post injury. Saliva samples were collected from 58 concussed adolescents (mean age [SD] = 15.4 [1.6] years; 25 females) within 10 days of injury, and 35 age- and sex-matched, healthy, uninjured controls. Using a neurodegenerative panel of 798 miRNAs, we applied feature selection to identify miRNAs that best identified concussed from healthy, uninjured adolescents. Additional analysis investigated whether these miRNAs were able to predict concussion symptom burden 3-weeks post injury. Results indicated that two miRNAs were significantly upregulated in concussed participants: (1) hsa-miR-26a-5p (F = 8.2, = 0.005, False Discovery Rate [FDR]-corrected = 0.008) and (2) hsa-miR-4286 (F = 9.6, = 0.002, FDR-corrected = 0.008). Higher hsa-miR-4286 levels predicted lower symptom burden at 3-weeks (β = -0.41, = 0.001, FDR-corrected = 0.002). The findings indicated that two salivary miRNAs demonstrated promise as clinically feasible biomarkers for identifying concussion in adolescents, one of which, was also useful for prognosis of symptom burden. The observed miRNA alterations may reflect injury-induced changes in neuroinflammatory and repair pathways, but larger, longitudinal studies are needed to validate these candidates and establish their sensitivity and specificity in this population.