A Systematic Review and Meta-Analysis of the Role of Peripheral Inflammation in Delirium

INTRODUCTION: The pathophysiology of delirium is poorly understood, but the importance of inflammation is widely accepted. The objective was to investigate whether changes in the peripheral immune system have been associated with the development of delirium in hospitalized adults.

METHODS: Embase and MEDLINE databases were searched. Risk of bias was assessed using the Newcastle-Ottawa Scale. Eligible studies were split into three groups: Peripheral immune response was measured (1) preceding delirium, (2) during delirium, and (3) in both incident and prevalent delirium. Quantitative data was extracted and included in the meta-analysis; otherwise, studies were included in the qualitative analysis. (Prospero 102931) RESULTS: 149 records were included in the qualitative synthesis and 92 in the meta-analysis. Measured preceding delirium, there was the strongest evidence for higher neutrophil-to-lymphocyte ratio (NLR) (mean difference (MD) 1.00, 95% Confidence Interval (CI) 0.53, 1.48, p < 0.00001) in those that developed delirium compared to those that did not. During delirium there was strongest evidence for higher interleukin-6 (IL-6) (MD 21.29, 95% CI 11.78, 30.80, p < 0.00001), cortisol (MD 159.6, 95% CI 120.52, 198.68, p < 0.00001), and leukocyte count (MD 0.79, 95% CI 0.51-1.07, p < 0.000001) in delirium compared to no delirium.

DISCUSSION: These results support a role for peripheral immune response and inflammation in delirium. However, the heterogeneity of the condition was reflected in the meta-analysis, and we should be cautious extrapolating these results to specific populations. Many studies measured the same soluble markers of inflammation, and a new era of delirium research is needed that transcends this to better understand the condition and develop future treatments.