Rationale:
Biochemical markers of brain injury in blood, such as neurofilament light chain (NfL), glial fibrillary acidic protein (GFAP), and amyloid beta 42 and 40 peptide (Aβ42 and Aβ40), have been utilized in diagnosing and managing various neurological conditions. However, the role of these blood biomarkers in epilepsy remains unclear. In this follow up study with a larger sample size, we aimed to evaluate whether serum markers of neurological injury can:
- Differentiate between epileptic and non-epileptic seizures,
- Function as markers of disease severity in patients with epilepsy, and
- Serve as an indicator of recent seizures.
Methods:
This was a prospective observational study conducted in adult patients admitted to the epilepsy monitoring unit at Johns Hopkins Hospital. Blood samples were collected upon admission to EMU as a baseline and before discharge after concluding EMU stay. Serum NfL, GFAP, Aβ42 and Aβ40 were tested using Simoa assay kit with Aβ42/40 ratios calculated. Clinical data included demographics, identification of epilepsy versus psychogenic non-epileptic seizures (PNES), and baseline epilepsy severity in terms of self-reported monthly seizure frequency. The baseline biomarker levels in epilepsy to PNES were compared using logarithmic transformation of values followed by linear regression adjusting for age. The correlation of seizure frequency to biomarker levels were examined with multiple regression analysis. The difference in biomarker levels before and after seizure were evaluated with t-test.
Results:
62 patients were evaluated (epilepsy=48, PNES=14). Of those with epilepsy, 10 had generalized onset while 38 had focal onset seizures (lesional=20, non-lesional=18). There was a significant difference in the NfL (mean epilepsy=11.1 pg/ml, PNES=6.92 pg/ml; p=0.022), GFAP (mean epilepsy=83.4 pg/ml, PNES=59.9 pg/ml; p=0.033), and Aβ42/40 ratio (mean epilepsy=0.064, PNES=0.077; p=0.032) (Table 1). When broken down to Aβ42 and Aβ40, there was a significant difference in Aβ40 specifically (mean epilepsy=106 pg/ml, PNES=87.1 pg/ml; p=0.016). There were no significant correlation between concentration of biomarkers and seizure burden. There was a significant difference in Aβ42/40 ratio in comparing focal epilepsy types (mean non-lesional=0.068, lesional=0.061; p=0.018).
45 patients had repeat samples obtained prior to discharge (epilepsy=34, PNES=11). 23 patients had seizures in the EMU with 11 being focal seizures, 12 bilateral tonic clonic and 5 having a combination. There was a significant difference in the change in GFAP (p=0.015) before and after seizure (Table 2). There was no significant difference in NfL and Aβ42/40 ratios before and after seizure.
Conclusions:
In this larger prospective study of patients admitted to a tertiary EMU, patients with epilepsy had higher levels of NfL and GFAP and lower Aβ42/40 ratio compared to patients with PNES. There was a significant increase in level of GFAP after seizures, but not in the levels of NfL. This study highlights the potential use of serum GFAP as a diagnostic marker following acute seizures. Our study adds further evidence to the higher baseline serum NfL and GFAP concentrations in patients with epilepsy.
Funding: None