Objective:

Background:

ICH has a high incidence and is often life-threatening, with limited effective treatment options. HE is closely associated with poor prognosis after ICH. Early and effective prediction of poor outcomes is crucial for timely interventions to improve prognosis. Plasma proteomics offers a reliable and powerful method for screening biomarkers.

Design/Methods:

Clinical data were prospectively collected from patients with acute ICH who presented to the Department of Neurology in our hospital within 6 hours of symptom onset between December 2020 and December 2023. TMT proteomics technology was employed to identify differentially expressed plasma proteins between the HE patients and the non-HE patients. Bioinformatics analysis was performed using the GO and KEGG databases to select candidate proteins for predicting HE. ELISA was used to verify the selected proteins. Further univariate and multivariate analyses were conducted to explore their relationship with poor prognosis after ICH, and the area under the ROC curve (AUC) was calculated to assess their predictive ability for poor outcomes.

Results:

A total of 161 patients with ICH were included, with 51 cases (31.6%) experiencing HE and 65 cases (40.4%) having poor outcomes. Among 297 differentially expressed proteins, based on differential expression levels, enriched pathways, and a comprehensive literature review, 9 proteins were selected and verified using ELISA method. Finally, SYND4, Prx2, and ITGB-1 were identified as being associated with HE. Both univariate and multivariate analyses demonstrated their correlation with poor prognosis after ICH. ROC curve analysis revealed that these four proteins exhibited good predictive ability for poor outcomes in ICH.

Conclusions:

SYND4, Prx2, and ITGB-1 are associated with HE and poor prognosis after ICH. These biomarkers hold promise for guiding the development of new therapeutic targets.