Ayla Yildiz, Binnur Karabiyik, Levent Deniz
Differences in SIRI and PNI were observed between psychiatric patients and healthy controls. However, these differences were not consistent across diagnostic and substance-use subgroups, and some associations were attenuated after adjustment for age and sex. These findings indicate variability in inflammatory and immunonutritional indices across psychiatric and substance-use subgroups and warrant further investigation.
BACKGROUND: Systemic inflammatory and nutritional indices derived from hematological parameters have emerged as accessible markers of immune imbalance. Although these indices have been investigated across various medical disciplines, their roles in psychiatric patients with positive urinary drug metabolites have not yet been reported. This study aimed to compare hematological inflammation and nutritional indices between psychiatric patients with positive urinary drug metabolites and healthy controls, to evaluate the Systemic Inflammation Response Index (SIRI), Systemic Immune-Inflammation Index (SII), and Prognostic Nutritional Index (PNI), explore their variations across substance types and usage patterns, and determine their association with immune-nutritional dysregulation.
METHODS: This retrospective study included 760 psychiatric inpatients with positive urinary drug metabolites, 171 psychiatric patients without positive metabolites, and 151 healthy controls treated at X City Hospital between 2022 and 2024. SIRI, SII, and PNI were calculated from hematological and/or biochemical parameters. Patients with infectious, autoimmune, or malignant diseases or incomplete data were excluded. Group comparisons, age- and sex-adjusted regression analyses, multiple-comparison corrections, and ROC analyses were performed.
RESULTS: Patients had higher SIRI [1.24 (0.79-1.97) vs. 0.98 (0.74-1.27)] and lower PNI (56.0 ± 6.12 vs. 58.8 ± 3.87) than controls (both p < 0.001), whereas SII did not differ significantly between the groups (p = 0.108). SIRI was positively associated with disease presence (OR = 1.847, 95% CI: 1.423-2.398; p < 0.001), whereas PNI was inversely associated with disease presence (OR = 0.917, 95% CI: 0.889-0.947; p < 0.001). ROC analysis in the drug-positive schizophrenia subgroup yielded AUCs of 0.727 for SIRI and 0.804 for PNI. After adjustment for age and sex and correction for multiple comparisons, no significant drug-status-related differences in SIRI or SII remained within diagnostic subgroups, whereas lower PNI persisted in drug-positive patients with nonorganic psychosis. Substance-specific associations varied according to psychiatric diagnosis and substance type.
CONCLUSION: Differences in SIRI and PNI were observed between psychiatric patients and healthy controls. However, these differences were not consistent across diagnostic and substance-use subgroups, and some associations were attenuated after adjustment for age and sex. These findings indicate variability in inflammatory and immunonutritional indices across psychiatric and substance-use subgroups and warrant further investigation.