A comparison of biologically variable ventilation to recruitment manoeuvres in a porcine model of acute lung injury

BACKGROUND: Biologically variable ventilation (return of physiological variability in rate and tidal volume using a computer-controller) was compared to control mode ventilation with and without a recruitment manoeuvre – 40 cm H(2)O for 40 sec performed hourly; in a porcine oleic acid acute lung injury model. METHODS: We compared gas exchange, respiratory mechanics, and measured bronchoalveolar fluid for inflammatory cytokines, cell counts and surfactant function. Lung injury was scored by light microscopy. Pigs received mechanical ventilation (F(I)O(2 )= 0.3; PEEP 5 cm H(2)O) in control mode until PaO(2 )decreased to 60 mm Hg with oleic acid infusion (PaO(2)/F(I)O(2 )<200 mm Hg). Additional PEEP to 10 cm H(2)O was added after injury. Animals were randomized to one of the 3 modes of ventilation and followed for 5 hr after injury. RESULTS: PaO(2 )and respiratory system compliance was significantly greater with biologically variable ventilation compared to the other 2 groups. Mean and mean peak airway pressures were also lower. There were no differences in cell counts in bronchoalveolar fluid by flow cytometry, or interleukin-8 and -10 levels between groups. Lung injury scoring revealed no difference between groups in the regions examined. No differences in surfactant function were seen between groups by capillary surfactometry. CONCLUSIONS: In this porcine model of acute lung injury, various indices to measure injury or inflammation did not differ between the 3 approaches to ventilation. However, when using a low tidal volume strategy with moderate levels of PEEP, sustained improvements in arterial oxygen tension and respiratory system compliance were only seen with BVV when compared to CMV or CMV with a recruitment manoeuvre.