Clinical Characteristics and Blood-Based Inflammatory Indices of Psychiatric Hospitalizations Among Children with Autism in China: A Retrospective Electronic Medical Records Study

Background/Objectives: This study aimed to investigate the clinical characteristics associated with psychiatric hospitalizations in children with ASD and to evaluate changes in blood-based Inflammatory Indices following treatment. Methods: A retrospective study of 269 children and adolescents with ASD (≤18 years) admitted to Nanjing Brain Hospital between 2012 and 2023 was conducted. Electronic medical records were reviewed for demographic characteristics, primary reasons for hospitalization, ASD-specific clinical scale scores, and routine laboratory parameters. The systemic immune–inflammation index (SII), neutrophil-to-lymphocyte ratio (NLR), monocyte-to-lymphocyte ratio (MLR), platelet-to-lymphocyte ratio (PLR), and non-enzymatic antioxidant indicators (uric acid, total bilirubin, direct bilirubin, prealbumin) were calculated. Results: The mean age at admission was 12.13 years, and 80.3% of admitted patients were male. The leading reasons for hospitalization were aggression (76.21%), psychosocial stressors (48.32%), and ADHD-related symptoms (46.09%), with distinct patterns observed by intellectual disability status, sex, and developmental stage. Multiple regression analyses revealed that age, sex, and specific admission indications were significantly associated with baseline inflammatory indices (SII, NLR, PLR, WBC, NEU, PLT; all p < 0.05) and antioxidant indices (UA, TBIL, DBIL, PA; all p < 0.05). Among 196 patients with 4-week follow-up data, significant post-treatment reductions were observed in SII, WBC, NEU, UA, TBIL, and DBIL, alongside increased prealbumin (all p < 0.05). Conclusions: Psychiatric hospitalizations of Chinese children with ASD are driven by distinct behavioral and neurodevelopmental profiles. Concurrently, accessible blood-based inflammatory and antioxidant indices show significant baseline clinical associations and post-treatment alterations, suggesting their potential as adjunctive biological markers in acute psychiatric settings. These findings are limited by the retrospective single-center design, absence of a control group, and lack of post-treatment clinical severity assessments.