Differential impact of ultrasonically nebulized versus tracheal-instilled surfactant on ventilation-perfusion (VA/Q) mismatch in a model of acute lung injury.

In a model of acute lung injury, established by saline lavage of isolated perfused rabbit lungs, the effect of ultrasonic surfactant nebulization on gas exchange was compared with that of tracheal instillation, assessed by the multiple inert gas elimination technique (MIGET). Ultrasonic aerosolization provided particles with a mass median aerodynamic diameter of 4.5 microm (geometric SD, 2.3), the pulmonary deposition of which was monitored on-line by a laserphotometric technique. Under baseline conditions, a narrow unimodal distribution of ventilation and perfusion with shunt-flow ranging below 2% and absence of perfusion of low V A/Q (0.01 < V A/Q < 0.1) areas was noted throughout. This physiological V A/Q matching was not affected by lung deposition of 8.6 mg surfactant/kg body weight (bw), forwarded by 1 h ultrasonic nebulization. In contrast, tracheal bolus injection of 80 mg/ kg bw surfactant in control lungs provoked the appearance of low V A/Q areas (maximum approximately 13% of perfusion) and shunt flow (4 to 6%), in addition to marked ventilation-perfusion mismatch (broadening of perfusion and ventilation distribution) in the midrange V A/Q regions. The saline lavage procedure caused progressive development of shunt flow ( approximately 22%) and perfusion of low V A/Q areas ( approximately 7%), associated with severe V A/Q mismatch. "Rescue" tracheal instillation of 80 mg/kg bw surfactant in lavaged lungs reduced the shunt-flow to approximately 4%, but increased the perfusion of low V A/Q areas to 10 to 14%; V A/Q mismatch in the midrange V A/Q regions was not improved. Ultrasonic deposition of 8.8 mg surfactant/kg bw in the injured lungs reduced the shunt flow to approximately 7% and the perfusion of low V A/Q areas to < 2%, coincident with improvement of V A/Q matching in the midrange V A/Q areas. We conclude that low doses of ultrasonically delivered natural surfactant are similarly effective as "conventional" doses of tracheal-instilled surfactant in reducing shunt flow in an acute lung injury model, but exert more advantageous effects on ventilation perfusion matching.