Neurophysiological resting-state EEG markers of catatonia in schizophrenia and mood disorders

Key pointsO_ST_ABSQuestionC_ST_ABSIs catatonia, hypothesized to involve disrupted excitation-inhibition balance and cortical dysconnectivity, associated with quantitative abnormalities on resting-state EEG? FindingsIn this retrospective, case-control, transdiagnostic hospital-based cohort of patients with schizophrenia and/or mood disorders, we analyzed resting-state EEG data comparing individuals with (n=102) and without (n=519) catatonia. Patients with catatonia showed significant increase in delta power, decrease in alpha power, increase in gamma power, significant reduction in peak alpha frequency and longer mean duration of microstate C. MeaningPower spectral density, alpha peak frequency, and microstate disruption in catatonia suggest a neurodevelopmentally-related excitation/inhibition dysregulation and support the use of routine clinical EEG for developing diagnostic biomarkers. ImportanceCatatonia is a severe psychomotor syndrome that complicates many psychiatric, neurodevelopmental, and non-psychiatric conditions. Identifying reliable diagnostic biomarkers remains a key challenge to improve early intervention and reduce morbimortality. Since its pathophysiology may involve cortical dysconnectivity, electroencephalography (EEG) could provide accessible disease-associated measures, such as power spectral density (PSD - reflecting excitation/inhibition balance), peak alpha frequency (PAF - related to deviations in neurodevelopmental trajectories), and C and D microstates (previously linked respectively to self-referential and externally-oriented attentional modes). However, EEG is yet to be used for this purpose. ObjectiveTo leverage routine clinical EEG recordings to identify neurophysiological markers associated with catatonia. We postulate to find anomalies in PSD distribution, a decrease in peak alpha frequency, and increases in the representation of C over D microstate classes. DesignThis study is a case-control retrospective transdiagnostic hospital-based cohort of patients with schizophrenia and/or mood disorders. ParticipantsWe analyzed resting-state EEG data from patients diagnosed with schizophrenia or mood disorders, both with (n=102) and without (n=519) catatonia. EEG data were preprocessed using a well-validated multistep automated pipeline. Exposures/measuresLinear regression models assessed associations between catatonia status and PSD, PAF and microstates, adjusting for age, sex, medication (computed as olanzapine, fluoxetine, and diazepam equivalents), and comorbid neurodevelopmental or neurological conditions. ResultsPatients with catatonia showed increased delta power (T = 2.37, pFDR = 0.03), decreased alpha power (T = -3.55, pFDR = 0.002) and increased gamma power (T = 3.14, pFDR = 0.008), reduced PAF (T = -2.60, p = 0.03), and longer mean duration of microstate C (T = 2.17, p = 0.03). Results were consistent in the subgroup not receiving benzodiazepines. Conclusion and relevanceRoutine clinical EEG revealed quantitative neurophysiological differences between patients with and without catatonia in a transdiagnostic population with psychotic and mood disorders. Power spectral density, alpha peak frequency and microstate anomalies in catatonia shed light on its underlying pathophysiology, suggesting a probable neurodevelopmentally-related excitation/inhibition dysregulation. Importantly, this indicates that routine clinical EEG could be used for diagnostic biomarker development, which would ultimately improve early detection and treatment.