THE PRONE POSITION IMPROVES ARTERIAL OXYGENATION AND REDUCES SHUNT IN OLEIC-ACID-INDUCED ACUTE LUNG INJURY

Publisher Website
Google Scholar
Abstract
The arterial oxygen tension (PaO2) may increase when patients with the adult respiratory distress syndrome are turned from supine to prone. We sought to reproduce this observation in dogs with acute lung injury to study the physiologic mechanism by which the improvements in oxygenation might occur. Twenty anesthetized dogs were ventilated with a constant tidal volume (20 ml/kg) of 100% oxygen. Oleic acid (0.09 ml/kg) was injected into the right atrium while rotating the animals through 360 degrees in 4 stages. Animals in Group 1 (n = 5) remained supine for 10 to 120 min until the supine PaO2 fell below 200 mm Hg. Those in Group II (n = 4) were kept prone during this period. Dogs in Group I and II were then turned supine or prone every 30 min 5 times. Cardiac output and pulmonary vascular pressures, functional residual capacity (helium dilution), and regional diaphragmatic motion (determined by dorsal and ventral diaphragmatic markers relative to markers on the chest wall seen on lateral chest radiographies taken at FRC and at end-inspiration) were obtained in each position. Eleven dogs were kept supine (Group III, n = 6) or prone (Group IV, n = 5) for 2 h after oleic acid infusion, after which intrapulmonary shunt (.ovrhdot.Qs/.ovrhdot.Qt) and ventilation-perfusion heterogeneity were measured in the supine and prone positions using the multiple inert gas elimination technique. The PaO2 increased each time the animals were placed prone in all groups (mean .+-. SEM of all measurement times of PaO2, supine to prone were 126 .+-. 8 to 312 .+-. 17 mm Hg in Group 1, 94 .+-. 4 to 196 .+-. 13 mm Hg in Group II, 193 .+-. 29 to 419 .+-. 34 mm Hg in Group III, and 205 .+-. 39 to 340 .+-. 50 mm Hg in Group IV, all p < 0.01). The .ovrhdot.Qs/.ovrhdot.Qt decreased from 22.7 .+-. 3.8 to 8.2 .+-. 2 in Group III and from 32.3 .+-. 6.3 to 18.1 .+-. 4.7 in Group IV when animals were turned from supine to prone. No position-dependent changes in cardiac output, pH, PaCO2, FRC, or pulmonary vascular driving pressure were observed. Sixty percent of the total diaphragmatic movement was localized to the nondependent portion regardless of dog position. The prone position-induced improvement in PaO2 occurs secondary to a reduction in .ovrhdot.Qs/.ovrhdot.QT and is not related to changes in FRC, regional diaphragmatic motion, cardiac output, or pulmonary vascular pressures. Why .ovrhdot.Qs/.ovrhdot.QT decreases in the prone position remains to be determined.