Brain injury after 50 h of lung-protective mechanical ventilation in a preclinical model

Abstract

Mechanical ventilation is the cornerstone of the Intensive Care Unit. However, it has been associated with many negative consequences. Recently, ventilator-induced brain injury has been reported in rodents under injurious ventilation settings. Our group wanted to explore the extent of brain injury after 50 h of mechanical ventilation, sedation and physical immobility, quantifying hippocampal apoptosis and inflammation, in a normal-lung porcine study. After 50 h of lung-protective mechanical ventilation, sedation and immobility, greater levels of hippocampal apoptosis and neuroinflammation were clearly observed in the mechanically ventilated group, in comparison to a never-ventilated group. Markers in the serum for astrocyte damage and neuronal damage were also higher in the mechanically ventilated group. Therefore, our study demonstrated that considerable hippocampal insult can be observed after 50 h of lung-protective mechanical ventilation, sedation and physical immobility.

Figure 1

Slides of an NV subject’s hippocampus (A) and an MV subject’s hippocampus (B) after staining by doublecortin (DCX) assay, showing normal cells (blue) and doublecortin-positive cells (brown). Greater neurogenesis can be observed in the MV example than in the NV example. (C) Dot plot showing mitotic indices of 19.07 (17.84–21.38) in the NV group and 26.04 (24.41–30.34) in the MV group, p = 0.0022 (n = 6 per group).

Figure 2

Slide of an NV subject’s hippocampus (A) and an MV subject’s hippocampus (B) after staining by TUNEL assay, showing normal cells (blue) and TUNEL-positive cells (brown). Considerably greater hippocampal cellular apoptosis can be observed in the MV example than in the NV example. (C) Dot plot showing apoptotic indices of 0.96 (0.50–1.61) in the NV group and 31.70 (29.79–43.76) in the MV group p = 0.0002 (n = 6–10 per group).

Figure 3

(A) Dot plot showing mitotic indices of 19.07 (17.84–21.38) in the NV group and 26.04 (24.42–30.34) in the MV group, p = 0.0022 (n = 6 per group). (B) Dot plot showing apoptotic indices of 0.96 (0.50–1.61) in the NV group and 31.70 (29.79–43.76) in the MV group p = 0.0002 (n = 6–10 per group). (C) Dot plot showing turnover indices of + 18.59 (16.86–19.92) in the NV group and − 5.32 (− 15.16- − 0.77) in the MV group, p = 0.0022 (n = 6 per group).

Figure 4

Slide of an NV subject’s hippocampus (A) and an MV subject’s hippocampus (B) after staining by IBA-1, showing normal cells (blue) and IBA-1-positive cells (brown). (C) Dot plot showing percentages of IBA-1-positive cells, combining the two behaviors of microglia, 10.12 (8.93–10.65) in the NV group and 36.17 (30.71–48.27) in the MV group, p = 0.0022 (n = 6 per group). (D) Dot plot showing percentages of low-activity microglia behavior, 84.40 (80.49–86.27) in the NV group and 80.63 (71.59–81.93) in the MV group, p = 0.0411 (n = 6 per group). (E) Dot plot showing percentages of high-activity microglia behavior, 15.60 (13.73–19.51) in the NV group and 18.63 (17.37–27.38) in the MV group, p = 0.0649 (n = 6 per group).

Figure 5

Slide of an NV subject’s hippocampus (A) and an MV subject’s hippocampus (B) after staining by GFAP, showing normal cells (blue) and GFAP-positive cells (brown). (C) Dot plot showing percentages of GFAP-positive cells (astrocytes), 10.69 (9.31–12.85) in the NV group and 25.53 (21.21–28.66) in the MV group, p = 0.0022 (n = 6 per group).

Figure 6

(A) Dot plot showing serum concentrations of S100β, 360.90 pg/ml (252.10–803.60) in the NV group and 193.10 pg/ml (129.20–223.30) in the MV group, p < 0.0001 (n = 10–16 per group). (B) Dot plot showing serum concentrations of UCHL1, 76.57 pg/ml (42.48–90.26) in the NV group and 96.96 ng/ml (80.65–109.60) in the MV group, p = 0.0575 (n = 10–13 per group). (C) Dot plot showing serum concentrations of GFAP, 0.15 ng/ml (0.07–0.23) in the NV group and 0.40 ng/ml (0.28–0.57) in the MV group, p = 0.0013 (n = 8–16 per group). (D) Dot plot showing serum concentrations of IL-1β, 13.39 pg/ml (7.72–17.68) in the NV group and 219.60 pg/ml (56.23–505.40) in the MV group, p = 0.0003 (n = 7–8 per group). (E) Dot plot showing serum concentrations of IL-8, 144.50 pg/ml (100.90–397.90) in the NV group and 23.40 pg/ml (10.88–58.51) in the MV group, p = 0.0006 (n = 7–8 per group). (F) Dot plot showing serum concentrations of IL-10, 67.92 pg/ml (55.60–134.80) in the NV group and 217.50 pg/ml (131.30–1,140.00) in the MV group, p = 0.0289 (n = 7–8 per group).

Figure 7

Dot plot showing the PaO₂/FiO₂ ratio of the NV group (568 mmHg; 546–571), the PaO₂/FiO₂ ratio of the MV group at the start of the experiment (521 mmHg; 454–552) and the PaO₂/FiO₂ ratio of MV group at the end of the experiment (403 mmHg; 357–444). No statistically significant difference was found when comparing the PaO₂/FiO₂ ratio of the NV group (568 mmHg; 546–571) to the PaO₂/FiO₂ ratio of the MV group at the start of the experiment (521 mmHg; 454–552, p = 0.3700). The PaO₂/FiO₂ ratio of MV group at the end of the experiment (403 mmHg; 357–444) when compared to the PaO₂/FiO₂ ratio of the MV group at the start of the experiment (521 mmHg; 454–552, p = 0.0164) was statistically significant.

Figure 8

(A) Dot plot showing the lung injury scores of the NV group (0.19; 0.17–0.21) and the lung injury scores of the MV group (0.19; 0.16–0.27) with no statistical difference between the groups, p = 0.8182. Lung histology slides stained by hematoxylin and eosin, NV subject (B) and MV subject (C).