Accuracy of GFAP and UCH-L1 in predicting brain abnormalities on CT scans after mild traumatic brain injury: a systematic review and meta-analysis.

Abstract

Purpose: Traumatic brain injury (TBI) is a leading cause of death and disability worldwide. In recent years, blood biomarkers including glial fibrillary acidic protein (GFAP) and ubiquitin carboxy-terminal hydrolase L1 (UCH-L1) have shown a promising ability to detect head CT abnormalities following TBI. This review aims to combine the existing research on GFAP and UCH-L1 biomarkers and examine how well they can predict abnormal CT results after mild TBI.

Methods: Our study protocol was registered in PROSPERO (CRD42024556264). PubMed, Google Scholar, and Cochrane electronic databases were searched. We reviewed 37 full-text articles for eligibility and included 14 in our systematic review and meta-analysis.

Results: Thirteen studies reported data for GFAP. The optimal cutoff of GFAP was 65.1 pg/mL with a sensitivity of 76% (95% CI 37 ̶ 95) and a specificity of 74% (95% CI 39 ̶ 93). In patients with GCS 13 ̶ 15 the optimal cutoff was 68.5 pg/mL, showing a sensitivity of 75% (95% CI 17 ̶ 98), and a specificity of 73% (95% CI 20 ̶ 97). Seven studies provided data on UCH-L1. The optimal cutoff was 225 pg/mL, with a sensitivity of 86% (95% CI 50 ̶ 97) and a specificity of 51% (95% CI 19 ̶ 83). In patients with GCS 13 ̶ 15, the optimal cutoff was 237.7 pg/mL, with a sensitivity of 89% (95% CI 74 ̶ 96), and a specificity of 36% (95% CI 29 ̶ 44). Modeling the diagnostic performance of GFAP showed that in adult patients with GCS 13-15 for ruling out CT abnormalities, at the threshold of 4 pg/mL, the optimal diagnostic accuracy was achieved with a sensitivity of 98% (95% CI 94-99) and (negative predictive value) NPV of 97%. For UCH-L1, the optimal diagnostic accuracy for ruling out intracranial abnormalities in adults with GCS 13-15 was achieved at the threshold of 64 pg/mL, with a sensitivity of 99% (95% CI 92-100) and NPV of 99%.

Conclusion: Present results suggest that GFAP and UCH-L1 have the clinical potential for screening mild TBI patients for intracranial abnormalities on head CT scans.