Predicting levetiracetam neuropsychiatric adverse effects using graph-theoretical electroencephalographic network metrics in drug-naive focal epilepsy: A multicenter longitudinal study.

Pretreatment EEG network features showed a moderate signal for identifying people who later had important levetiracetam side effects. The model needs prospective external validation before clinical use.

Sferruzzi M, Ricci L, Matarrese MAG, Izzi F, Placidi F, Sancetta BM, Cerulli Irelli E, Pulitano P, Di Lazzaro V, Tombini M, Assenza G.
Epilepsia · Publication date: 2026-09-22
Added to site: 2026-09-24
DOI: 10.1002/epi.70486

Electronic publication date verified in Europe PMC; first indexed 2026-09-23. This source date is separate from the site's addition date.

Modalities

QEEG, EEG

Clinical focus

Epilepsy, Medication tolerability, Predictive biomarkers

Population

Drug-naive people with focal epilepsy treated at three Italian centers

Sample

62 people selected from a 134-person levetiracetam monotherapy cohort: all 31 with clinically significant neuropsychiatric adverse effects and 31 age- and sex-matched controls

Intervention / method

Resting EEG within 30 days before levetiracetam, with theta-band connectivity graph metrics entered into an elastic-net classifier

Comparator

Patients who developed clinically significant adverse effects within one month versus matched patients without those effects

Clinician summary

Theta-band clustering coefficient, path length and node strength were higher in the adverse-effect group. Nested cross-validation produced 0.71 accuracy and 0.71 AUC, with a wide AUC confidence interval of 0.57 to 0.83; spectral power did not differ.

How to interpret the findings

Retrospective matched subset from a modest cohort, internal validation only and imprecise discrimination. Case-control sampling does not provide real-world positive predictive value, and no prospective decision threshold or external clinical utility was established.

Equipment and disclosures

EEG acquisition systems and montage details were not available in the checked abstract.

Funding and competing-interest details were not available in the checked indexed record.

Related evidence

Compare findings from the same clinical area, including mixed results.

• Interictal electric source imaging with open-source software and routine low-density electroencephalography: A systematic comparison of four inverse methods in drug-resistant epilepsy. · Positive with limitations
• Bridging Temporal and Spatial Scales: Contemporary Advances in EEG-fMRI Integration With a Focus on Epilepsy. · Positive with limitations
• Conditional spatial classification of expert-confirmed interictal epileptiform discharge epochs: An EEG-ECG ablation and SHAP analysis. · Positive with limitations