Phenytoin versus levetiracetam as prophylaxis for postcraniotomy seizure in patients with no history of seizures: systematic review and meta-analysis.

OBJECTIVE De novo seizure following craniotomy (DSC) for nontraumatic pathology may adversely affect medical and neurological outcomes in patients with no history of seizures who have undergone craniotomies. Antiepileptic drugs (AEDs) are commonly used prophylactically in patients undergoing craniotomy; however, evidence supporting this practice is limited and mixed. The authors aimed to collate the available evidence on the efficacy and tolerability of levetiracetam monotherapy and compare it with that of the classic AED, phenytoin, for DSC. METHODS PubMed, Embase, Web of Science, and the Cochrane Library were searched for studies that compared levetiracetam with phenytoin for DSC prevention. Inclusion criteria were adult patients with no history of epilepsy who underwent craniotomy with prophylactic usage of phenytoin, a comparator group with levetiracetam treatment as the main treatment difference between the two groups, and availability of data on the numbers of patients and seizures for each group. Patients with brain injury and previous seizure history were excluded. DSC occurrence and adverse drug reaction (ADR) were evaluated. Seizure occurrence was calculated using the Peto odds ratio (POR), which is the relative effect estimation method of choice for binary data with rare events. RESULTS Data from 7 studies involving 803 patients were included. The DSC occurrence rate was 1.26% (4/318) in the levetiracetam cohort and 6.60% (32/485) in the phenytoin cohort. Meta-analysis showed that levetiracetam is significantly superior to phenytoin for DSC prevention (POR 0.233, 95% confidence interval [CI] 0.117-0.462, p < 0.001). Subgroup analysis demonstrated that levetiracetam is superior to phenytoin for DSC due to all brain diseases (POR 0.129, 95% CI 0.039-0.423, p = 0.001) and tumor (POR 0.282, 95% CI 0.117-0.678, p = 0.005). ADRs in the levetiracetam group were cognitive disturbance, thrombophlebitis, irritability, lethargy, tiredness, and asthenia, whereas rash, anaphylaxis, arrhythmia, and hyponatremia were more common in the phenytoin group. The overall occurrence of ADR in the phenytoin (34/466) and levetiracetam (26/432) groups (p = 0.44) demonstrated no statistically significant difference in ADR occurrence. However, the discontinuation rate of AEDs due to ADR was 53/297 in the phenytoin group and 6/196 in the levetiracetam group (POR 0.266, 95% CI 0.137-0.518, p < 0.001). CONCLUSIONS Levetiracetam is superior to phenytoin for DSC prevention for nontraumatic pathology and has fewer serious ADRs that lead to discontinuation. Further high-quality studies that compare levetiracetam with placebo are necessary to provide evidence for establishing AED guidelines.