Intracerebral Proinflammatory Cytokine Increase in Surgically Evacuated Intracerebral Hemorrhage: A Microdialysis Study

Abstract

Background: Treatment options for spontaneous intracerebral hemorrhage (ICH) are limited. A possible inflammatory response in the brain tissue surrounding an ICH may exacerbate the initial injury and could be a target for treatment of subsequent secondary brain injury. The study objective was to compare levels of inflammatory mediators in the interstitial fluid of the perihemorrhagic zone (PHZ) and in seemingly normal cortex (SNX) in the acute phase after surgical evacuation of ICH, with the hypothesis being that a difference could be demonstrated between the PHZ and the SNX.

Methods: In this observational study, ten patients needing surgical evacuation of supratentorial ICH received two cerebral microdialysis catheters: one in the PHZ and one in the SNX that is remote from the ICH. The microdialysate was analyzed for energy metabolites (including lactate pyruvate ratio and glucose) and for inflammatory mediators by using a multiplex immunoassay of 27 cytokines and chemokines at 6-10 h, 20-26 h, and 44-50 h after surgery.

Results: A metabolic crisis, indicated by altered energy metabolic markers, that persisted throughout the observation period was observed in the PHZ when compared with the SNX. Proinflammatory cytokines interleukin (IL) 8, tumor necrosis factor α, IL-2, IL-1β, IL-6 and interferon γ, anti-inflammatory cytokine IL-13, IL-4, and vascular endothelial growth factor A were significantly higher in PHZ compared with SNX and were most prominent at 20-26 h following ICH evacuation.

Conclusions: Higher levels of both proinflammatory and anti-inflammatory cytokines in the perihemorrhagic brain tissue implies a complex role for inflammatory mediators in the secondary injury cascades following ICH surgery, suggesting a need for targeted pharmacological interventions.

Keywords: Brain injury; Cytokine; Inflammation mediators; Intracranial hemorrhage; Microdialysis; Stroke.

Fig. 1

Preoperative and postoperative CT scan. a, Preoperative CT scan of a 68-year-old man presenting with an intracerebral hemorrhage (ICH) (black star) in the right basal ganglia. b, Following surgical evacuation of the ICH, a postoperative CT scan shows the hematoma cavity (black star) and the tip of the microdialysis (MD) catheters placed in the perihemorragic zone (PHZ) open arrow and seemingly normal cortex (SNX) (closed arrow), respectively. CT= computed tomography

Fig. 2

Metabolic crisis in the PHZ. a, Microdialysis (MD) glucose levels were significantly lower in the perihemorrhagic zone (PHZ) when compared with those in seemingly normal cortex (SNX) (p < 0.05), although were consistently above critical (0.2 mmol/L) and warning levels (0.8 mmol/L) in both locations [26]. b, The LPR, however, was pathologically elevated in the PHZ, indicating a metabolic crisis in the brain tissue. In contrast, the SNX LPR normalized within the first hours after surgery, and thereafter remained within the normal range. c MD-glutamate levels decreased with time in the PHZ but were significantly higher than in the SNX (p < 0.05) during the initial 48 h. d, In addition, MD-glycerol levels were persistently higher in the PHZ when compared with the SNX (p < 0.05). Data are presented as mean and standard error of the mean (SEM) for clarity. LPR= lactate pyruvate ratio

Fig. 3

Univariate analysis of cytokines. Significantly higher levels of IL-2 and IL-8 were seen in the perihemorrhagic zone (PHZ) compared with seemingly normal cortex (SNX) at 20–26 h after surgery. At 44–50 h after surgery, there was a significantly higher expression of IL-4, which is considered an anti-inflammatory cytokine, and IL-6, which is considered a proinflammatory cytokine, in the PHZ compared with the SNX. Several of the cytokines contributing to group separation in the supervised OPLS-DA model (highlighted with gray background in this figure) did not reach statistically significant differences when explored using univariate statistical methods. In contrast, IL-1α was significantly higher in PHZ at 20–26 h after surgery in univariate analysis but did not contribute to the multivariate model. Median (line) and individual values presented: *p < 0.05 (Wilcoxon signed-rank test). IFN= interferon, IL= interleukin, IP-10= interferon-gamma induced protein 10, LPR= lactate pyruvate ratio, MCP= monocyte chemoattractant protein, MDC= macrophage derived chemokine, MIP= macrophage inflammatory protein, OPLS-DA= orthogonal projection to latent structures discriminant analysis, TARC= thymus and activation regulated chemokine, TNF= tumor necrosis factor, VEGF-A= vascular endothelial growth factor A

Fig. 4

Multivariate data analysis showed separation between PHZ and SNX. a, Principal component analysis (PCA) score scatter plot of cytokine and metabolite data from PHZ (black squares) and SNX (white inverted triangles) show a separation between the two groups. The PHZ data points gravitate toward the upper left quadrant, whereas the SNX data points predominantly cluster toward the lower right quadrant (dashed black circles). b Corresponding PCA loading scatter plot shows clustering of proinflammatory cytokines (red squares) and metabolites associated with a deranged metabolic state (black stars) to the left upper quadrant corresponding to the predominant location of PHZ data (black dashed circle) in part a. In contrast, MD-glucose and MD-glycerol levels cluster with SNX data points, as do chemokines (blue hexagons) and anti-inflammatory cytokines (yellow triangles). Time point 2 (20–26 h after surgery; unfilled circle) correlates with higher proinflammatory cytokine expression compared with time points 1 (4–10 h after surgery) and 3 (44–50 h after surgery), respectively, apart from IL-17A, which is distinctly correlated with time point 1. c Score plot of OPLS-DA model (one component) showing model separation of PHZ (black columns) and SNX (white columns) data points. d Loading plot of OPLS-DA model showing which variables contribute strongest (gray bars) to group separation within the model, with 95% confidence interval (error bars). e Volcano plot for the OPLS-DA model shows variables more specific to PHZ to the left on the x-axis and those associated with SNX to the right. Variables that contribute the strongest to group discrimination (i.e., with a |p(corr)|> 0.4 and VIP > 1) are placed within the red circles. Lactate, pyruvate, LPR, and glutamate (black); proinflammatory cytokines IL-8, TNF-α, IL-2, IL-1β, IL-6, and IFN-γ (red); anti-inflammatory cytokines IL-13 and IL-4 (yellow); and growth factor VEGF-A (blue) were higher in PHZ compared with SNX. The majority of the anti-inflammatory cytokines and chemokines did not contribute significantly to the model, suggesting that their concentrations were similar in the PHZ and the SNX. f S-plot indicating strength of contribution of each variable to the OPLS-DA model. GF= growth factor, IFN= interferon, IL= interleukin, IP-10= interferon-gamma induced protein 10, LPR= lactate pyruvate ratio, MCP= monocyte chemoattractant protein, MDC= macrophage derived chemokine, MIP= macrophage inflammatory protein, OPLS-DA= orthogonal projection to latent structures discriminant analysis, PHZ= perihemorrhagic zone, SNX= seemingly normal cortex, TARC= thymus and activation regulated chemokine, TNF= tumor necrosis factor, VEGF-A= vascular endothelial growth factor A, VIP = variable influence on projection.

Fig. 5

Hierarchical cluster analysis dendrogram depicting two major clusters in the data and protein-protein interaction (PPI) network analysis. a Hierarchical cluster analysis revealed two major clusters in the metabolic markers and inflammatory mediators of the OPLS-DA model. b The STRING PPI network analysis of the inflammatory mediators reveals several connections. CCL= C–C motif chemokine ligand, CCR= C–C motif chemokine receptor, CXCL= C-X-X motif chemokine ligand, IL= interleukin, OPLS-DA= orthogonal projection to latent structures discriminant analysis, PHZ= perihemorrhagic zone, SNX= seemingly normal cortex, TNF= tumor necrosis factor