High tidal volume mechanical ventilation-induced lung injury in rats is greater after acid instillation than after sepsis-induced acute lung injury, but does not increase systemic inflammation: an experimental study

Abstract

Background: To examine whether acute lung injury from direct and indirect origins differ in susceptibility to ventilator-induced lung injury (VILI) and resultant systemic inflammatory responses.

Methods: Rats were challenged by acid instillation or 24 h of sepsis induced by cecal ligation and puncture, followed by mechanical ventilation (MV) with either a low tidal volume (Vt) of 6 mL/kg and 5 cm H2O positive end-expiratory pressure (PEEP; LVt acid, LVt sepsis) or with a high Vt of 15 mL/kg and no PEEP (HVt acid, HVt sepsis). Rats sacrificed immediately after acid instillation and non-ventilated septic animals served as controls. Hemodynamic and respiratory variables were monitored. After 4 h, lung wet to dry (W/D) weight ratios, histological lung injury and plasma mediator concentrations were measured.

Results: Oxygenation and lung compliance decreased after acid instillation as compared to sepsis. Additionally, W/D weight ratios and histological lung injury scores increased after acid instillation as compared to sepsis. MV increased W/D weight ratio and lung injury score, however this effect was mainly attributable to HVt ventilation after acid instillation. Similarly, effects of HVt on oxygenation were only observed after acid instillation. HVt during sepsis did not further affect oxygenation, compliance, W/D weight ratio or lung injury score. Plasma interleukin-6 and tumour necrosis factor-α concentrations were increased after acid instillation as compared to sepsis, but plasma intercellular adhesion molecule-1 concentration increased during sepsis only. In contrast to lung injury parameters, no additional effects of HVt MV after acid instillation on plasma mediator concentrations were observed.

Conclusions: During MV more severe lung injury develops after acid instillation as compared to sepsis. HVt causes VILI after acid instillation, but not during sepsis. However, this differential effect was not observed in the systemic release of mediators.

Figure 1

P_aO₂/F_IO₂ ratio and lung compliance. The P_aO₂/F_IO₂ ratio and compliance in rats ventilated for 4 h after acid instillation or during sepsis with low tidal volume (LVt) (6 mL/kg), with a positive end-expiratory pressure (PEEP) of 5 cm H₂O or high tidal volume (HVt) (15 mL/kg), without PEEP. (A) P_aO₂/F_IO₂ ratio decreased after acid instillation irrespective of ventilation strategy. Mechanical ventilation (MV) after acid instillation decreased the P_aO₂/F_IO₂ ratio in contrast to HVt MV during sepsis. (B) Acid instillation decreased lung compliance as opposed to sepsis. No effects of MV were observed. * p < 0.001 for model compared to LVt and HVt sepsis, # p < 0.001 for interaction between model and MV. Open squares: LVt acid, open triangles: HVt acid, solid squares: LVt sepsis and solid triangles: HVt sepsis.

Figure 2

Lung injury. Wet to dry lung weight (W/D) ratios and lung injury scores in rats ventilated for 4 h after acid instillation or during sepsis with low tidal volume (LVt) (6 mL/kg), with a positive end-expiratory pressure (PEEP) of 5 cm H₂O or high tidal volume (HVt) (15 mL/kg), without PEEP. Acid instillation increased W/D ratio as compared to sepsis. The effect of mechanical ventilation (MV) with HVt on W/D ratio was greater after acid instillation than during sepsis (A). Acid instillation increased lung injury score as compared to sepsis. The effect of HVt MV on lung injury score was greater after acid instillation than during sepsis. Panels (C) to (H) are representative micrographs demonstrating lung injury for: (C) Control rats, (D) and (E) demonstrate increased lung injury observed after acid instillation and MV with (D) LVt and (E) HVt. Sepsis alone increased lung injury (F); MV did not have an effect with either (G) LVt, or (H) HVt. Original magnification × 100. * p < 0.001 for model compared to LVt and HVt sepsis. # p < 0.001 for interaction between model and MV, % p = 0.04 for interaction between model and MV.