A longitudinal analysis of bidirectional relationships between executive functioning and peripheral inflammation in schizophrenia.

Executive functioning deficits are a hallmark symptom of schizophrenia, and inflammation has been linked to these deficits. While a study from our group explored the relationship between executive dysfunction over time and baseline inflammation markers, a more detailed exploration of the dynamic interrelationships between these variables is needed, as inflammation markers have been shown to fluctuate throughout the stages of the disease, and may have bidirectional associations with cognition. This study aimed to investigate longitudinal trajectories of inflammatory markers (measured multiple times) and executive functioning, as well as their bidirectional associations across time. Specifically, we investigated the bidirectional relationships between various inflammatory markers (high sensitivity C-Reactive Protein (CRP), Interferon-γ (IFN-γ), Tumor Necrosis Factor (TNF)-α, and Interleukin (IL)-6, -8, and -10) and a composite measure of executive functioning among 171 people with schizophrenia (PwS) and 156 non-psychiatric comparison (NC) participants (333 total participants). Fasting blood samples (65 ml) were drawn at baseline and follow-up visits between 7 a.m. and 12p.m. and centrifuged at 3000 rpm, with the resulting plasma stored at -80 °C. A high-sensitivity particle-enhanced immuno-turbidimetric assay was used to detect plasma CRP levels, using a Roche Cobas c-502 instrument. Meso Scale Discovery (MSD) MULTI-SPOT® Assay System on a SECTOR Imager 2400 instrument was used for measuring plasma levels of IL-6, IL-8, IL-10, TNF-α, and IFN-γ. A composite score for executive functioning was created from 3 subtests (Trail Making, Color-Word Inhibition, and F-A-S) of the Delis-Kaplan Executive Function System (D-KEFS). Linear Mixed Models (LMM) analysis was employed to investigate longitudinal trajectories, while Continuous Time Structural Equation Modeling (CTSEM) was used to examine bidirectional relationships between inflammatory markers and executive functioning. The LMM results showed that PwS had lower executive functioning and higher CRP, IL-6, IL-10, and TNF-α levels than NCs across all time points. No changes over time or group-by-time interactions were noted for any of the measures. CTSEM results showed that, in both groups, higher IL-10 was associated with worse later executive functioning, with a more robust association among PwS. No other significant lagged relationships were observed after adjusting for body mass index. Our findings suggest nuanced immune-cognitive relationships, in which specific inflammatory markers may be associated with cognitive functioning changes over time.