Application of CPAP resulted in a marked decrease in blood flow to shunt units that was most pronounced in the presence of inhaled NO. The observed decrease in intrapulmonary shunt may be explained by two mechanisms: (1) shunt units were recruited and became ventilated, thereby converting to low or normal Va/Q units, or (2) perfusion of shunt units was reduced by redistribution of the blood flow to normal Va/Q, high Va/Q, or previously nonperfused areas. Unfortunately, the results of the inert gas measurements only quantify the effects of CPAP and NO inhalation on the Va/Q distributions and do not allow us to distinguish between the two potential mechanisms.
Alveolar recruitment with CPAP is supported by previous studies demonstrating decrease in intrapulmonary shunt and increase in end-expiratory lung volume with application of positive end-expiratory pressure (PEEP) during mechanical ventilation in humans and experimental animals with severe lung injury. Although end-expiratory lung volumes were not measured in our dogs, it is well accepted that CPAP improves pulmonary gas exchange during acute lung injury by increasing end-expiratory lung volumes and by recruiting poorly ventilated lung units. Increased perfusion of high Va/Q and nonperfused units during CPAP is unlikely, and would be expected to produce changes of mean Va/Q of the blood flow distribution and its dispersion (logSDQ), which we did not see. Instead, decrease in mean Va/Q of the ventilation distribution, reduction in dead space ventilation, and the absence of changes in the dispersion of the ventilation distribution (logSDv) and in high Va/Q units, indicate that CPAP redistributed ventilation to previously nonventilated or poorly ventilated lung areas.
Previous experimental and clinical observations suggest that NO inhalation improves overall Va/Q matching by redistributing blood flow from essentially nonventilated to ventilated lung units. Decrease in blood flow to shunt units without changes in the dispersion of ventilation distribution (logSDv) and in high Va/Q units during CPAP with NO inhalation support that blood flow was distributed from nonventilated to well-ventilated lung areas.