The role of surgery in the management of Lennox-Gastaut syndrome: A systematic review and meta-analysis of the clinical evidence.

ABSTRACT

Lennox-Gastaut syndrome (LGS) is a severe form of childhood onset epilepsy in which patients require multiple medications and may be candidates for palliative surgical intervention. In this meta-analysis, we sought to evaluate the impact of palliative vagus nerve stimulation (VNS), corpus callosotomy (CC), and resective surgery (RS) by analyzing their impact on seizure control, antiepileptic drug (AED) usage, quality of life (QOL), behavior, cognition, prognostic factors, and complications. A systematic search of PubMed MEDLINE, Scopus, and Cochrane Database of Systematic Reviews was performed to find articles that met the following criteria (1) prospective/retrospective study with original data, (2) at least one LGS surgery patient aged less than 18 years, and (3) information on seizure frequency reduction (measured as percentage, Engel class, or qualitative comment). Seizures were analyzed quantitatively in a meta-analysis of proportions and a random-effects model, whereas other outcomes were analyzed qualitatively. Forty studies with 892 LGS patients met the selection criteria, with 19 reporting on CC, 17 on VNS, four on RS, two on RS + CC, one on CC + VNS, and one on deep brain stimulation. CC seizure reduction rate was 74.1% (95% confidence interval [CI] = 64.5%-83.7%), and VNS was 54.6% (95% CI = 42.9%-66.3%), which was significantly different (p < .001). RS seizure reduction was 88.9% (95% CI = 66.1%-99.7%). Many VNS patients reported alertness improvements, and most had no major complications. VNS was most effective for atonic/tonic seizures; higher stimulation settings correlated with better outcomes. CC patients reported moderate cognitive and QOL improvements; disconnection syndrome, transient weakness, and respiratory complications were noted. Greater callosotomy extent correlated with better outcomes. AED usage most often did not change after surgery. RS showed considerable QOL improvements for patients with localized seizure foci. In the reported literature, CC appeared to be more effective than VNS for seizure reduction. VNS may provide a similar or higher level of QOL improvement with lower aggregate risk of complications. Patient selection, anatomy, and seizure type will inform decision-making.