Pediatric catatonia across diagnostic contexts: Presentation, treatment course, and factors associated with ECT use

Objective Catatonia in young people is increasingly recognized, but pediatric cohorts remain limited. We characterized clinical presentation and treatment course and examined factors associated with electroconvulsive therapy (ECT) use and exploratory inflammatory correlates of psychomotor dimensions. Methods We retrospectively reviewed 57 consecutive patients aged <18 years treated at a tertiary referral center; one patient with a medical etiology was excluded from inferential analyses, leaving 56. DSM-5-TR signs were operationalized as hypokinetic, hyperkinetic, and parakinetic scores. Associations with ECT use were estimated using Firth penalized logistic regression. Admission neutrophil-to-lymphocyte ratio (NLR), monocyte-to-lymphocyte ratio (MLR), and systemic inflammation response index (SIRI) were examined with false-discovery-rate correction. Results Catatonia occurred across schizophrenia spectrum (n = 28), bipolar (11), depressive (11), and autism spectrum (6) disorders. ECT was used in 21 patients (37.5%). Older age, male sex, higher lorazepam-equivalent dose, and greater catatonia symptom burden were associated with ECT use. Each additional parakinetic sign was associated with higher odds of ECT use (OR 2.94, 95% CI 1.52-7.29), including after diagnostic-group adjustment; however, the parakinetic score overlapped substantially with total symptom burden, and their independent contributions could not be resolved. NLR and SIRI were associated only with the hyperkinetic score, which consisted of documented agitation. Thirty-five patients (62.5%) had a discharge CGI-catatonia score of 1. Conclusion ECT use was associated with greater catatonia symptom burden and higher parakinetic scores, but overlap between these measures and retrospective treatment selection preclude dimension-specific or causal inference. The laboratory findings were limited to agitation and require prospective confirmation.