Alteration patterns of peripheral concentrations of cytokines and associated inflammatory proteins in acute and chronic stages of schizophrenia: a systematic review and network meta-analysis
Sean Halstead,Dan Siskind,Michaela Amft,Elias Wagner,Vladislav Yakimov,Zoe Shih-Jung Liu,Ken Walder,Nicola Warren
DOI: https://doi.org/10.1016/S2215-0366(23)00025-1
Abstract:Background: Immune system dysfunction is considered to play an aetiological role in schizophrenia spectrum disorders, with substantial alterations in the concentrations of specific peripheral inflammatory proteins, such as cytokines. However, there are inconsistencies in the literature over which inflammatory proteins are altered throughout the course of illness. Through conducting a systematic review and network meta-analysis, this study aimed to investigate the patterns of alteration that peripheral inflammatory proteins undergo in both acute and chronic stages of schizophrenia spectrum disorders, relative to a healthy control population. Methods: In this systematic review and meta-analysis, we searched PubMed, PsycINFO, EMBASE, CINAHL, and the Cochrane Central Register of Controlled Trials from inception to March 31, 2022, for published studies reporting peripheral inflammatory protein concentrations in cases of people with schizophrenia-spectrum disorders and healthy controls. Inclusion criteria were: (1) observational or experimental design; (2) a population consisting of adults diagnosed with schizophrenia-spectrum disorders with a specified indicator of acute or chronic stage of illness; (3) a comparable healthy control population without mental illness; (4) a study outcome measuring the peripheral protein concentration of a cytokine, associated inflammatory marker, or C-reactive protein. We excluded studies that did not measure cytokine proteins or associated biomarkers in blood. Mean and SDs of inflammatory marker concentrations were extracted directly from full-text publshed articles; articles that did not report data as results or supplementary results were excluded (ie, authors were not contacted) and grey literature and unpublished studies were not sought. Pairwise and network meta-analyses were done to measure the standardised mean difference in peripheral protein concentrations between three groups: individuals with acute schizophrenia-spectrum disorder, individuals with chronic schizophrenia-spectrum disorder, and healthy controls. This protocol was registered on PROSPERO, CRD42022320305. Findings: Of 13 617 records identified in the database searches, 4492 duplicates were removed, 9125 were screened for eligibility, 8560 were excluded after title and abstract screening, and three were excluded due to limited access to the full-text article. 324 full-text articles were then excluded due to inappropriate outcomes, mixed or undefined schizophrenia cohorts, or duplicate study populations, five were removed due to concerns over data integrity, and 215 studies were included in the meta-analysis. 24 921 participants were included, with 13 952 adult cases of schizophrenia-spectrum disorder and 10 969 adult healthy controls (descriptive data for the entire cohort were not available for age, numbers of males and females, and ethnicity). Concentration of interleukin (IL)-1β, IL-1 receptor antagonist (IL-1RA), soluble interleukin-2 receptor (sIL-2R), IL-6, IL-8, IL-10, tumour necrosis factor (TNF)-α, and C-reactive protein were consistently elevated in both individuals with acute schizophrenia-spectrum disorder and chronic schizophrenia-spectrum disorder, relative to healthy controls. IL-2 and interferon (IFN)-γ were significantly elevated in acute schizophrenia-spectrum disorder, while IL-4, IL-12, and IFN-γ were significantly decreased in chronic schizophrenia-spectrum disorder. Sensitivity and meta-regression analyses revealed that study quality and a majority of the evaluated methodological, demographic, and diagnostic factors had no significant impact on the observed results for most of the inflammatory markers. Specific exceptions to this included: methodological factors of assay source (for IL-2 and IL-8), assay validity (for IL-1β), and study quality (for transforming growth factor-β1); demographic factors of age (for IFN-γ, IL-4, and IL-12), sex (for IFN-γ and IL-12), smoking (for IL-4), and BMI (for IL-4); and diagnostic factors including diagnostic composition of schizophrenia-spectrum cohort (for IL-1β IL-2, IL-6, and TNF-α), antipsychotic-free cases (for IL-4 and IL-1RA), illness duration (for IL-4), symptom severity (for IL-4), and subgroup composition (for IL-4). Interpretation: Results suggest that people with schizophrenia-spectrum disorders have a baseline level of inflammatory protein alteration throughout the illness, as reflected by consistently elevated pro-inflammatory proteins, hypothesised here as trait markers (eg, IL-6), while those with acute psychotic illness might have superimposed immune activity with increased concentrations of hypothesised state markers (eg, IFN-γ). Further research is required to determine whether these peripheral alterations are reflected within the central nervous system. This research facilitates an entry point in understanding how clinically relevant inflammatory biomarkers might one day be useful to the diagnosis and prognostication of schizophrenia-spectrum disorders. Funding: None.