Objective: During high-frequency oscillatory ventilation, bias flow is the continuous flow of gas responsible for replenishing oxygen and removing CO(2) from the patient circuit. Bias flow is usually set at 20 L/min, but many patients require neuromuscular blockade at this flow rate. The need for neuromuscular blockade may be eliminated by increasing the bias flow rate, but CO(2) retention is a potential concern. We hypothesize that in a swine model of acute lung injury, increased bias flow rates will not affect CO(2) elimination.
Design: Prospective, randomized, experimental study.
Setting: Research laboratory at a university medical center.
Subjects: Sixteen juvenile swine.
Interventions: Sixteen juvenile swine (12-16.5 kg) were studied using a saline lavage model of acute lung injury. During high-frequency oscillatory ventilation, each animal was ventilated with bias flows of 10, 20, 30, and 40 L/min in random sequence. For ten animals, power was set at a constant level to maintain PaCO(2) 50-60 mm Hg, and amplitude was allowed to vary. For the remaining six animals, amplitude was kept constant to maintain PaCO(2) within the same range, while power was adjusted as needed with changes in bias flow. Linear regression was used for data analysis.
Measurements and main results: Median overall PaCO(2) was 53 mm Hg (range: 31-81 mm Hg). Controlling for both power and amplitude, there was no statistically significant change in PaCO(2) as bias flow varied from 10 to 40 L/min.
Conclusions: Changes in bias flow during high-frequency oscillatory ventilation did not affect ventilation. Further clinical investigation is ongoing in infants and children with acute lung injury being managed with high-frequency oscillatory ventilation to assess the impact of alterations of bias flow on gas exchange, cardiopulmonary parameters, sedation requirements, and other clinical outcomes.