Phase-dependent cytokine signatures in alcohol use disorder: A systematic review of psychoneuroimmune links to withdrawal severity and return to drinking

Abstract

Background

Neuroimmune dysregulation is increasingly implicated in alcohol use disorder (AUD), yet the clinical utility of cytokine biomarkers remains unclear.

Objective

To synthesise evidence linking pro- and anti-inflammatory cytokines to AUD severity, withdrawal symptoms, and return-to-drinking risk, with emphasis on temporal staging and translational implications.

Methods

A systematic review and meta-analysis were conducted in accordance with PRISMA 2020 and PROSPERO (CRD420251060888). Six databases and grey literature sources were searched from January 1, 2000 to October 31, 2025 (updated for revision). Eligible studies assessed IL-6, IL-1β, IL-10, TNF-α, or CRP in adults with AUD or alcohol withdrawal, and reported correlations with validated clinical severity scales (e.g., AUDIT, CIWA-Ar) or relapse outcomes. Risk of bias, evidence certainty (GRADE), and heterogeneity were systematically evaluated.

Results

Thirteen studies (N = 1458) were included across active drinking/non-treatment-seeking samples, early abstinence/withdrawal (≤14 days), and post-detoxification/treatment-follow-up contexts where acute withdrawal was absent or not the primary focus. In early abstinence, IL-6 tended to be higher in AUD/AWS than controls; however, the pooled estimate was large but imprecise and highly heterogeneous (random-effects Hedges’ g =1.21, 95 % CI −0.08–2.50; 2 studies; I²≈91 %). Higher baseline CRP was associated with lower subsequent alcohol use and weaker cue-reactivity in individual cohorts, but findings were heterogeneous and not pooled. For return-to-drinking outcomes, a single study provided extractable IL-1β group data comparing participants who returned to drinking with those who maintained abstinence; the single-study standardized effect was non-significant (Hedges’ g=0.40, 95 % CI −0.12–0.92). Lower IL-17RB expression was associated with reduced relapse-free survival in a single study and should be considered preliminary. Methodological variability and incomplete control for hepatic/metabolic confounding limited pooling for several cytokines.

Conclusions

Evidence supports phase-dependent cytokine signals in AUD, with IL-6 most consistently elevated during early abstinence and CRP showing pragmatic translational potential but substantial methodological heterogeneity. Evidence for IL-1β and IL-17RB as relapse-related markers remains preliminary due to limited extractable and non-uniform outcome reporting. Standardised, sex-stratified longitudinal studies with rigorous control for hepatic/metabolic factors and CNS validation are warranted.