OBJECTIVE Over the years, blood biomarkers have been extensively applied for diagnostic and prognostic assessment of traumatic brain injury (TBI). Herein, we conducted a meta-analysis to evaluate the diagnostic and prognostic value of glial fibrillary acidic protein (GFAP) for TBI patients. METHODS The online databases, including PubMed, Embase, Cochrane Library, CNKI, and WFSD, were systematically retrieved from inception until May 2021. The RevMan 5.3 software and Stata 15 were used to conduct data analysis. RESULTS A total of 22 eligible studies comprising 3709 patients were included in this meta-analysis. The pooled results indicated that serum GFAP had a diagnostic value in detecting traumatic intracranial lesions (AUC 0.81; 95% CI 0.77-0.84; p < 0.00001). The pooled sensitivity and specificity were 0.93 (95% CI 0.81-0.98), and 0.66 (95% 0.53-0.77; p < 0.00001), respectively. For assessment of unfavorable outcome, the pooled sensitivity, specificity and AUC value were 0.66 (95% CI 0.54-0.76; p < 0.00001), 0.82(95% CI 0.72-0.90; p < 0.00001), and 0.82 (95% CI 0.76-0.88; p < 0.00001), respectively. Besides, GFAP exhibited a significant value in predicting mortality (AUC 0.81; 95% CI 0.77-0.84; p < 0.00001), with high sensitivity and specificity (0.86, 95% CI 0.79-0.92, p < 0.00001, and 0.66, 95% CI 0.52-0.77, p < 0.00001). The subgroup analysis indicated that the type of TBI and cut-off value were potential sources of heterogeneity, which influenced the pooled AUC values for mortality prediction. CONCLUSIONS Our meta-analysis indicated that GFAP had diagnostic and prognostic value for TBI patients, especially during the early TBI.