Real-world use of the updated refractory epilepsy screening tool for Lennox-Gastaut syndrome

Steven M Wolf,Danielle Boyce,Patricia Peña,Jesus Eric Piña-Garza,Jessica J Roland,Bethany Thomas,Donika Zogejani,Patricia E McGoldrick

DOI: https://doi.org/10.1002/epi4.12952

Epilepsia Open

Abstract:Objective: To evaluate the Refractory Epilepsy Screening Tool for Lennox-Gastaut Syndrome (REST-LGS) for real-world identification of LGS in adults and to develop a scoring system for the tool. Methods: A retrospective chart review of adults with drug resistant epilepsy (DRE) and intellectual development disorder (IDD) was conducted by 2 primary care providers blinded to diagnosis. The REST-LGS was designed via the Modified Delphi Consensus and was previously validated. This tool consists of 8 criteria (4 major, 4 minor) considered indicative of LGS. To account for missing data in the earlier validation study and to evaluate applicability in a real-world setting, the REST-LGS was refined to include a scoring system in which major criteria were more heavily weighted than minor criteria, producing categories of "likely" (>11 points), "possible" (8-11 points), and "unlikely" (<8 points) LGS. Statistical analyses were descriptive. Results: Of the 100 patients included in the analysis, data for slow spike-waves in electroencephalography and seizure onset age - both major REST-LGS criteria - were missing for 46% and 42% of patients, respectively. The majority of patients met 4 of the 8 REST-LGS criteria (cognitive impairment since childhood, 71%; persistent seizures despite a trial of ≥2 antiseizure medications, 65%; seizure onset before the age of 12 years, 57%; ≥2 seizure types, 56%). All 4 major criteria were met in 22 patients (22%). The percentages of patients considered "likely," "possible," or "unlikely" to have LGS were 26%, 30%, and 44%, respectively. Of the 74 patients without a previous LGS diagnosis, 42 (57%) were identified as "possible" or "likely" to have LGS using REST-LGS. Significance: In this analysis, the validated REST-LGS was evaluated in a real-world setting. The majority of previously undiagnosed patients were identified via REST-LGS as "possible" or "likely" to have LGS. Extensive missing data highlights challenges of LGS diagnosis in adults. Plain language summary: There is a need to identify adult patients with Lennox-Gastaut syndrome (LGS) so they can receive appropriate treatment. The Refractory Epilepsy Screening Tool for LGS (REST-LGS) questionnaire was designed by experts to identify whether patients with seizures that are not controlled by medications may have LGS. In this study, 2 physicians completed the REST-LGS using charts for 100 patients who experience seizures not controlled by medications. Of the patients who were previously diagnosed as not having LGS, the majority were "likely" or "possible" to have LGS based on the REST-LGS; therefore, the REST-LGS can identify patients for further evaluation.