Neutrophil-to-lymphocyte ratio in schizophrenia: A systematic review and meta-analysis

Background: The immune system appears to be dysregulated in schizophrenia (SZ). The potential prognostic or diagnostic value of neutrophil-to-lymphocyte ratio (NLR), an inexpensive proxy marker for a wide spectrum of conditions, has not been established in SZ. We seek to investigate a) whether NLR is increased in SZ patients, b) if this difference in more prominent in relapsed SZ or first-episode psychosis. Methods: A structured algorithm was applied in MEDLINE (1946-2018), PsychInfo, Scopus, Cochrane Central Register of Controlled Trials (CENTRAL) (1999-2018), PSYNDEX (1984-2017) and Clinicaltrials.gov (2008-2018) databases. Results: Through the database search, 1023 articles were screened. Ten studies (804 SZ patients, 671 controls) were included in the meta-analysis. In SZ patients, the NLR was increased by 0.65 (95% CI: 0.54, 0.86, p < 10(-5)). This difference was significant in both acute relapse and first-episode psychosis subgroups. Studies of moderate and high quality also showed a significant NLR increase in SZ patients (I-2 = 0%). Meta-regression analysis showed that the polymorphonuclear count and antipsychotic use may confound the result. In leave-one-out meta-analysis, no study altered the significance of the result when omitted. Conclusions: NLR in SZ patients is increased, both in chronic disease and in first-episode psychosis. Baseline characteristics, such as polymorphonuclear count and antipsychotic use, may affect its accuracy. The application of this marker in clinical practice requires the description of its normal values in the general population, its potential change after antipsychotic administration and its correlation with disease activity. A large-scale, prospective study design would resolve these issues. (C) 2018 Elsevier B.V. All rights reserved.