Actigraphic Correlates of Neuropsychiatric Disability in Adult Patients with Focal Epilepsy

• Published in: bioRxiv 2022-10
• Main Authors: Abboud, Mark A, Kamenn, Jessica L, Bass, John S, Lu, Lin, Gavvala, Jay R, Rao, Sindhu, Smagula, Stephen F, Vaishnav Krishnan
• Format: Article
• Language: English
• Subjects: Activity patterns; Anticonvulsants; Anxiety; Body mass index; Clinical trials; Cognitive ability; Comorbidity; Convulsions & seizures; Epilepsy; Mental depression; Patients; Quality of life; Quantitative psychology; Regression analysis; Seizures; Sleep and wakefulness; Wrist
• DOI: 10.1101/2022.10.18.512750

Disability in patients with epilepsy (PWE) is multifactorial: beyond seizure frequency/severity, PWE are prone to a range of neuropsychiatric, cognitive, and somatic comorbidities that significantly impact quality of life. In this study, we explored how variations in epilepsy severity and the burden of self-reported somatic/neuropsychiatric symptoms are associated with disruptions to 24h activity patterns (rest-activity rhythms, RARs), determined through wrist accelerometry/actigraphy. Continuous multiday recordings were obtained from 59 adult patients with focal epilepsy (44% male, ages 18-72), who contemporaneously provided responses to a range of validated psychometric instruments to measure the burden of anxiety, depression, sleepiness, and somatic symptoms. As a comparator, we conducted a similar psychometric-actigraphic correlation in 1761 subjects of Hispanic origin (35% male, ages 18-65) from the Study of Latinos (SOL) Sueño Ancillary Study. RARs were analyzed via a sigmoidally-transformed cosine model (quantifying RAR amplitude, steepness, acrophase and robustness) and non-parametric measures to estimate RAR stability, fragmentation, and sleep. Compared with age- and sex-matched SOL subjects, RARs from PWE subjects featured a significantly diminished amplitude, a wider rest phase and significantly more total daily sleep. Within PWE, similar RAR distortions were associated with seizure intractability and/or anticonvulsant polytherapy. In contrast, high anxiety, depression, and somatic symptom scores were associated with diminished RAR robustness and a delayed acrophase. We applied the complete SOL Sueño database to train logistic regression models to dichotomously classify anxiety, depression and sleepiness symptoms using age, sex, body mass index and a range of non-collinear RAR parameters. When tested on PWE, these models predicted prevalent anxiety and depression symptoms with modest success (accuracy ~70%) but failed to predict subjective sleepiness. Together, these results demonstrate that RAR features may vary with depression and anxiety symptoms in ambulatory patients with focal epilepsy, potentially offering a set of objective wearable-derived endpoints to adjunct routine clinical care and drug/device treatment trials. With larger actigraphic-psychometric datasets in PWE, we may identify RAR signatures that can more precisely distinguish between variations in seizure risk, the burden of anticonvulsant therapy and prevalent mood/anxiety symp