Objective: This study was performed to test the hypothesis that derecruitment itself might accentuate lung injury during mechanical ventilation.
Setting: Randomized, controlled trial.
Setting: Experimental laboratory.
Subjects: New Zealand White rabbits (2.8-3.5 kg).
Intervention: Twenty-four rabbits were ventilated in pressure-controlled mode with constant tidal volume (10 mL/kg). After lung injury was induced by repeated saline lavage (PaO2 <100 torr, 13.3 kPa), a pressure-volume curve was drawn to calculate the lower inflection point (Pflex), and randomization was done. The control group (n = 8) received ventilation with positive end-expiratory pressure (PEEP) fixed at Pflex for 3 hrs. The nonderecruitment group (n = 8) was ventilated at PEEP of 2 mm Hg (2.7 cm H2O) for the initial hour and then PEEP of Pflex for the remaining 2 hrs. The derecruitment group (n = 8) was ventilated for 3 hrs with six 30-min cycles consisting of 10 mins at PEEP of 2 mm Hg (2.7 cm H2O) and 20 mins at PEEP of Pflex to induce repeated derecruitments.
Measurements and main results: Variables of gas exchange, mechanics, and hemodynamics were measured, and histologic evaluation was done. In the control group, Pao2 remained >500 torr (66.7 kPa) for 3 hrs. In the nonderecruitment group, PaO2 was 40 +/- 16 (mean +/- SD) torr (5.3 +/- 2.1 kPa) at 1 hr but increased to >500 torr (66.7 kPa) for the remaining 2 hrs after increase in PEEP to Pflex. In the derecruitment group, there was progressive decline in Pao2 with each derecruitment to 220 +/- 130 torr (29.3 +/- 17.3 kPa) at 3 hrs (p <.05 compared with other groups). Histologically there was more hyaline membrane formation in the derecruitment group compared with control (p <.05) and significantly higher mean bronchiolar injury score in the derecruitment group (1.92 +/- 0.78) than both control (0.50 +/- 0.50) and nonderecruitment (0.78 +/- 0.42) groups (p <.05).
Conclusion: Repeated derecruitments can accentuate lung injury during mechanical ventilation.