Abstracts

Rationale: Lennox-Gastaut Syndrome (LGS), a severe epileptic encephalopathy, is characterized by a classic triad of symptoms: intractable seizures (multiple types, including drop seizures), cognitive impairment, and abnormal electroencephalogram (EEG) with generalized 1.5–2.5 Hz slow spike-wave discharges. Though onset typically occurs by 8 years of age, LGS persists into adulthood; it may be misdiagnosed in older patients as the classic LGS triad is often not present later in life. The accurate diagnosis of LGS—especially in older patients—is complex, and misdiagnosis can result in suboptimal treatment. A group of experts who care for patients with refractory epilepsy and intellectual and developmental disabilities created the Refractory Epilepsy Screening Tool for LGS (REST-LGS) to improve both the identification and treatment of patients with LGS. Methods: Patient data captured on a case report form (CRF, Figure 1) administered in two large, diverse epilepsy centers will be used to identify potential diagnostic areas that require clarification and measure the extent of inter-rater reliability (IRR). De-identified records of patients >12 years of age with refractory epilepsy with ≥2 clinic notes within the previous 2 years will be included; the first note and two most recent notes will be reviewed. Two raters—a specialist (epileptologist/neurology nurse practitioner) and a non-specialist (nurse, social worker, resident, or pre-medical student)—will complete a CRF for each patient. Raters will be blinded to each other’s responses. A sample size of 100 patients per site was selected based on established methodology. Variables will be described in absolute and relative frequencies, mean and standard deviation, or median and interquartile range where appropriate. Comparisons will be made using t-test, Pearson's chi-square test, and/or Fisher’s exact test. The extent of IRR will be analyzed via Cohen’s kappa statistic. Results: Patients with refractory epilepsy are currently being identified at each facility by specialist raters. Both specialist and non-specialist raters are independently completing CRFs for each patient. CRFs will be analyzed to determine the validity of the REST-LGS, and the expert team will meet to discuss the outcomes of the analysis and to refine the tool. Conclusions: Proper treatment of LGS is important to improve patient outcomes; however, LGS is difficult to diagnose, especially in older patients. The REST-LGS was designed to improve the identification and treatment of patients with LGS. The tool is currently being evaluated by expert and novice raters. Funding: Funding: Lundbeck LLC