Inflammatory and Neuroendocrine Biomarkers of Insomnia in Acute Ischaemic Stroke: A Systematic Review and Meta-Analysis

Abstract

Introduction: Insomnia complicates 40-75 % of acute ischaemic strokes (AIS) yet its biological determinants remain incompletely synthesised. We aimed to quantify the association between inflammatory and neuroendocrine biomarkers and post-stroke insomnia in AIS.

Methods: PubMed/MEDLINE, Scopus, Web of Science and Cochrane Central were searched systematically. Observational studies reporting cortisol, the neutrophil-to-lymphocyte ratio (NLR), composite indices (PLR, SII, MHR) or cytokines (IL-6, IL-18, TNF-alpha, hs-CRP) alongside a validated sleep outcome (PSQI, ISI or HAMD insomnia items) in adult AIS were eligible. Risk of bias was appraised with the Newcastle-Ottawa Scale, JBI checklist and the ROBINS-I framework. Standardised mean differences (Hedges g) were pooled in a DerSimonian-Laird random-effects model, with Hartung-Knapp confirmation.

Results:Thirteen studies entered qualitative synthesis; eight cohorts (n = 2 455) contributed 12 effects to the primary meta-analysis. The pooled g was 0.79 (95 % CI 0.52-1.05; p < 0.0001), with high heterogeneity (I²= 86.2 %). The composite-inflammation subgroup gave the most reproducible estimate (g = 0.53, 95 % CI 0.33-0.72, I² = 17.5 %). Leave-one-out g was 0.69-0.83 (all p < 1 × 10⁻⁷); Egger (p = 0.887) and Begg (p = 0.79) detected no small-study effect.

Conclusion: Cortisol, NLR and composite inflammatory indices are robustly associated with insomnia in AIS, with composite indices the most reproducible biomarker class, supporting early biomarker-based identification of patients at risk for post-stroke insomnia.