Serum glial fibrillary acidic protein and protein gene product 9.5 for predicting neurological outcomes in cardiac arrest patients with cortical response to somatosensory evoked potentials

Abstract

Aim

Predicting neurological prognosis after cardiac arrest remains challenging. Somatosensory evoked potential N20 absence is highly specific but lacks sensitivity. Glial fibrillary acidic protein and protein gene product 9.5 are potential biomarkers for brain injury, yet their roles in cardiac arrest patients with preserved somatosensory evoked potential N20 remain underexplored.

Methods

From January 2023 to December 2024, 69 cardiac arrest patients were enrolled, of whom 46 had preserved somatosensory evoked potential N20 responses. Serum glial fibrillary acidic protein, protein gene product 9.5 and neuron-specific enolase levels were measured at 72 h post-resuscitation. Patients were evaluated for neurological outcomes at 3 months using the Glasgow-Pittsburgh Classification of Cerebral Function scale. Receiver operating characteristic analysis determined biomarker thresholds for poor prognosis.

Results

In patients with preserved somatosensory evoked potential N20 responses, glial fibrillary acidic protein and protein gene product 9.5 levels were significantly higher in those with poor outcomes (P < 0.001). Glial fibrillary acidic protein (area under the curve = 0.908) had an optimal cutoff of 64.1 pg/mL (sensitivity 87.5%, specificity 82.4%) and a 100% specificity threshold of 149 pg/mL. Protein gene product 9.5 (area under the curve = 0.864) had an optimal cutoff of 448.4 pg/mL (sensitivity 87.5%, specificity 70.6%) and a 100% specificity threshold of 1253 pg/mL. The prognostic significance of combining serum glial fibrillary acidic protein, protein gene product 9.5, or neuron-specific enolase levels was explored, with glial fibrillary acidic protein + neuron-specific enolase achieving the highest area under the curve of 0.949 (0.882–1.000).

Conclusions

Serum glial fibrillary acidic protein and protein gene product 9.5 could be valuable predictors of poor neurological outcomes in cardiac arrest patients with cortical response to somatosensory evoked potential, though further studies are required to validate these findings.