Abstract
BACKGROUND: Post-traumatic epilepsy (PTE) is a common complication following traumatic brain injury (TBI). Early PTE refers to the appearance of seizure symptoms within 7 days of the injury. The glucose-to-potassium ratio (GPR) has emerged as a potential biomarker for predicting Early PTE risk. This study aimed to evaluate the association between GPR and the risk of Early PTE, and to assess the predictive value of GPR through various analyses. METHODS: A total of 2,049 TBI patients were included in the analysis, with the GPR evaluated both as a continuous and categorical variable. Logistic regression, trend tests, and Kaplan-Meier (KM) curve analyses were performed to assess the relationship between GPR and Early PTE. Subgroup analyses were conducted to explore potential effect modifiers, and restricted cubic spline (RCS) analyses were used to examine non-linear associations. Adjustments were made for demographic, clinical, and biochemical factors. RESULTS: The GPR demonstrated a significant non-linear association with Early PTE risk, with a turning point at GPR = 2.835. Patients with a GPR > 2.835 exhibited a higher risk of epilepsy, as indicated by KM curve analysis (P < 0.0001). Logistic regression analysis revealed that GPR was an independent predictor of Early PTE in both unadjusted and adjusted models. In the fully adjusted model, GPR remained significantly associated with early epilepsy (OR: 1.499, 95% CI: 1.188-1.891, P < 0.001). Subgroup analyses identified gender, hypertension, and diabetes as significant effect modifiers. Trend tests revealed a dose-response relationship between GPR quartiles and epilepsy risk, with the highest quartile showing a significantly higher risk in both unadjusted and partially adjusted models (P = 0.017). CONCLUSIONS: The GPR is a robust and independent predictor of Early PTE, with higher GPR levels strongly associated with an increased risk of epilepsy. The non-linear relationship and variations across subgroups underscore the clinical utility of GPR in risk stratification and personalized management of TBI patients.