Reliability of the 2021 BASED Score and Hypsarrhythmia for Assessment of Infantile Spasms
Abstract number :
2.138
Submission category :
4. Clinical Epilepsy / 4B. Clinical Diagnosis
Year :
2023
Submission ID :
704
Source :
www.aesnet.org
Presentation date :
12/3/2023 12:00:00 AM
Published date :
Authors :
Presenting Author: Jeff Swarz, MD – University of California, Los Angeles
Oana Shingo, MD PhD – University of California, Los Angeles; Nassim Naderi, MD – University of California, Los Angeles; Benjamin Edmonds, MD – University of California, Los Angeles; Atsuro Daida, MD PhD – University of California, Los Angeles; Rajsekar Rajaraman, MD – University of California, Los Angeles; Hiroki Nariai, MD PhD – University of California, Los Angeles; Shaun Hussain, MD – University of California, Los Angeles
Rationale: Infantile epileptic spasms syndrome (IESS) is typically characterized by epileptic spasms and hypsarrhythmia at onset, usually between two and 24 months of age. However, the identification hypsarrhythmia and the classification of its component electrographic features (i.e., voltage, spike abundance, etc.) exhibit inadequate inter-rater reliability. The Burden of AmplidtudeS and Epileptiform Discharges (BASED) score (2021 revision) has been proposed as a reliable measure to aid diagnosis and assessment of treatment response for IESS. In this study, we set out to independently evaluate the inter-rater reliability of the BASED score.
Methods: A total of 20 extended ( >16 h duration each) EEG studies from patients with IESS were selected for review, including 10 pre and 10 post-treatment studies. EEG recordings were anonymized and converted to .EDF format. Seven raters, including four established pediatric electroencephalographers (‘experts’) and three pediatric epilepsy/clinical neurophysiology fellows (‘trainees’), independently reviewed all 20 EEGs in a blinded fashion, and in random order. For each study, each rater was asked to assign a BASED score and evaluate slow-wave amplitude as well as the abundance and distribution of epileptiform discharges. Finally, for each study, each rater reported confidence in their scoring and the time required for review. The reliability of scores and EEG attribute classification was evaluated using Kappa statistics and intraclass correlation coefficients.
Results: Overall, BASED scores exhibited favorable reliability across all raters (ICC=0.78; 95%CI 0.65–0.89), with a trend suggesting higher reliability among experts (ICC=0.85, 95%CI 0.73-0.93) than trainees (ICC=0.68, 95%CI 0.45–0.84). Similarly, after dichotomizing BASED scores (0-3 vs 4-5), inter-rater reliability was favorable across all raters (K=0.71), with higher reliability among experts (K=0.77) than trainees (K=0.66). Despite this overall impression of reliability, there were significant BASED score discrepancies among the 4 ‘expert’ raters in 8/20 EEGs. In contrast, reliability was marginal for identification of ‘hypsarrhythmia’ (K=0.57), ‘voltage > 300 µV’ (K=0.56), and ‘any epileptiform discharges’ (K=0.57). Reliability was favorable for identification of abundant spikes (K=0.76). Raters’ description of confidence in determinations and time required for assessment were inconsistent, regardless of expertise (all ICC < 0.2).
Clinical Epilepsy