Sue A. Ravenscraft

Influence of prone position on the extent and distribution of lung injury in a high tidal volume oleic acid model of acute respiratory distress syndrome

OBJECTIVE To evaluate the influence of body position on the extent and distribution of experimental lung damage in an oleic acid canine model of acute respiratory distress syndrome, using mechanical ventilation with high tidal volumes and positive end-expiratory pressure (PEEP). DESIGN Prospective, randomized study. SETTING Experimental animal laboratory. SUBJECTS Twelve anesthetized and paralyzed dogs. INTERVENTIONS Ninety minutes after lung injury was induced by injection of oleic acid, 12 animals were randomized to be ventilated for 4 hrs, in either the supine (supine group, n = 6) or prone (prone group, n = 6) positions, using the same ventilatory pattern (F10(2) 0.6, PEEP > or = 10 cm H2O, and a tidal volume that generated a peak transpulmonary pressure of 35 cm H2O when implemented in the supine position). Regardless of randomization to position, the tidal volumes, F10(2), and PEEP were kept constant and the pulmonary artery occlusion pressure was maintained between 4 and 6 mm Hg for the duration of the study. MEASUREMENTS AND MAIN RESULTS At the end of the protocol, the lungs were excised for gravimetric determination (wet/dry weight ratio) and histologic examination (histologic score). Changes over time in the static pressure-volume curve of the lungs (obtained in the supine position) were also used as end-point variables. At baseline, hemodynamic and respiratory variables did not differ between groups. Just before randomization to position (90 mins after oleic acid injection), both groups presented similar lung static pressure-volume curves. Pulmonary artery occlusion pressure (4.3 +/- 1.9 vs. 4.8 +/- 1.3 mm Hg [supine vs. prone group]), cardiac output (4.1 +/- 0.4 vs. 5.2 +/- 1.3 L/min [supine vs. prone group]), and venous admixture (36.7 +/- 20.7% vs. 28.3 +/- 19.4% [supine vs. prone group]) were also not significantly (p > .05) different when measured in the supine position. At the end of the experiment, lung gravimetric data in the two experimental groups were not statistically different, suggesting a similar extent of edema. Histologic abnormalities, however, were less in the prone group than in the supine group (p < .01), due primarily to marked differences in extent and severity in the dependent regions of the lungs. Static lung compliance improved over time in the prone group (34 +/- 9 to 46 +/- 19 mL/cm H2O)(p = .02), but not in the supine group (34 +/- 6 to 36 +/- 6 mL/cm H2O). CONCLUSIONS After oleic acid-induced lung injury, animals ventilated with high tidal volume and PEEP undergo less extensive histologic change in the prone position than in the supine position. The prone position alters the distribution of histologic abnormalities.

Mean airway pressure: Physiologic determinants and clinical importance - Part 2: Clinical implications

PurposesTo discuss the theoretical relationship of mean alveolar pressure to its most easily measured analog, mean airway pressure, and to describe the key determinants, measurement considerations, and clinical implications of this index. Data SourcesRelevant articles from the medical and physiological literature, as well as mathematical arguments developed in this article from first principles. Study SelectionTheoretical, experimental, and clinical information that elucidates the physiologic importance, measurement, or adverse consequences of mean airway pressure. Data ExtractionMathematical models were used in conjunction with data from the published literature to develop a unified description of the physiological and clinical relevance of mean airway pressure. SynthesisGeometrical and mathematical analyses demonstrate tha t shared elements comprise mean airway pressure and mean alveolar pressure, two variables that are related by the formula: mean alveolar pressure = mean airway pressure + (E/60) ± (RE – RI), where ±E, RE, and RI are minute ventilation and expiratory and in-spiratory resistances, respectively. Clear guidelines can be developed for selecting the site of mean airway pressure determination, for specifying technical requirements for mean airway pressure measurement, and for delineating clinical options to adjust the level of mean airway pressure. Problems in viewing mean airway pressure as a reflection of mean alveolar pressure can be interpreted against the theoretical basis of their interrelationship. In certain settings, mean airway pressure closely relates to levels of ventilation, arterial oxygenation, cardiovascular function, and barotrauma. Because mean airway pressure is associated with both beneficial and adverse actions, a thorough understanding of its theoretical and practical basis is integral to formulating an effective pressure-targeted strategy of ventilatory support. ConclusionsMean airway pressure closely reflects mean alveolar pressure, except when flow-resistive pressure losses differ greatly for the inspiratory and expiratory phases of the ventilatory cycle. Under conditions of passive inflation, mean airway pressure correlates with alveolar ventilation, arterial oxygenation, hemodynamic performance, and barotrauma. We encourage wider use of this index, appropriately measured and interpreted, as well as its incorporation into rational strategies for the ventilatory management of critical illness.

Mean airway pressure: physiologic determinants and clinical importance--Part 1: Physiologic determinants and measurements.

PurposesTo discuss the theoretical relationship of mean alveolar pressure to its most easily measured analog, the mean airway pressure, and to describe the key determinants, measurement considerations, and clinical implications of this index. Data SourcesRelevant articles from the medical and physiologic literature, as well as mathematical arguments developed in this article from first principles. Study SelectionTheoretical, experimental, and clinical information that elucidates the physiologic importance, measurement, or adverse consequences of mean airway pressure. Data ExtractionMathematical models were used in conjunction with data from the published literature to develop a unified description of the physiological and clinical relevance of mean airway pressure. SynthesisGeometrical and mathematical analyses demonstrate that shared elements comprise mean airway pressure and mean alveolar pressure, two variables that are related by the formula: mean alveolar pressure = mean airway pressure + (VE/60) x (RE-RI), where ±VE, RE, and RI are minute ventilation and expiratory and inspiratory resistances, respectively. Clear guidelines can be developed for selecting the site of mean airway pressure determination, for specifying technical requirements for mean airway pressure measurement, and for delineating clinical options to adjust the level of mean airway pressure. Problems in viewing mean airway pressure as a reflection of mean alveolar pressure can be interpreted against the theoretical basis of their interrelationship. In certain settings, mean airway pressure closely relates to levels of ventilation, arterial oxygenation, cardiovascular function, and barotrauma. Because mean airway pressure is associated with both beneficial and adverse effects, a thorough understanding of its theoretical and practical basis is integral to formulating an effective pressuretargeted strategy of ventilatory support. ConclusionsMean airway pressure closely reflects mean alveolar pressure, except when flow-resistive pressure losses differ greatly for the inspiratory and expiratory phases of the ventilatory cycle. Under conditions of passive inflation, mean airway pressure correlates with alveolar ventilation, arterial oxygenation, hemodynamic performance, and barotrauma. We encourage wider use of this index, appropriately measured and interpreted, as well as its incorporation into rational strategies for the ventilatory management of critical illness.

Low measured auto-positive end-expiratory pressure during mechanical ventilation of patients with severe asthma: hidden auto-positive end-expiratory pressure.

OBJECTIVE To describe the occurrence of low measured auto-end-expiratory pressure (auto-PEEP) during mechanical ventilation of patients severe asthma. DESIGN Observational clinical study. SETTING Medical intensive care unit of a university-affiliated county hospital. PATIENTS Four mechanically ventilated patients with severe asthma who had low measured auto-PEEP despite marked increase in both peak and plateau airway pressures. INTERVENTIONS None. MEASUREMENTS AND MAIN RESULTS Peak pressure, plateau pressure, and auto-PEEP were measured at an early time point, when airflow obstruction was most severe, and again at a later time after clinical improvement. Auto-PEEP was measured by the method of end-expiratory airway occlusion. From the early to the late point, there was a marked decrease in peak pressure (76 +/- 7 to 53 +/- 6 cm H2O; p<.001) and in plateau pressure (28 +/- 2 to 18 +/- 3 cm H2O; p<.001), but only minimal change in auto-PEEP (5 +/- 3 to 4 +/- 3 cm H2O). The difference between plateau pressure and auto-PEEP decreased between the early and late time points (23 +/- 1 to 14 +/- 1 cm H2O; p<.01), even though tidal volume was larger at the late time point. In three patients, low auto-PEEP and a large difference between plateau pressure and auto-PEEP was only seen after expiratory time was prolonged. In these three patients, prolongation of expiratory time resulted in a large decrease in measured auto-PEEP (14 +/- 4 to 5 +/- 4 cm H2O), but a much smaller change in plateau pressure (31 +/- 3 to 29 +/- 3 cm H2O). CONCLUSIONS We conclude that measured auto-PEEP may underestimate end-expiratory alveolar pressure in severe asthma, and that marked pulmonary hyperinflation may be present despite low measured auto-PEEP, especially at low respiratory rates. This phenomenon may be due to widespread airway closure that prevents accurate assessment of alveolar pressure at end-expiration.

Effect of tracheal gas insufflation on gas exchange in canine oleic acid-induced lung injury.

OBJECTIVE To determine the effect of tracheal gas insufflation on gas exchange in oleic acid-induced lung injury in dogs. DESIGN Prospective, longitudinal study. SETTING University research laboratory. SUBJECTS Five mongrel dogs. INTERVENTIONS The dogs were anesthetized, paralyzed, and mechanically ventilated. Lung injury was induced by infusing 0.09 mL/kg of oleic acid and pulmonary artery occlusion (wedge) pressure (PAOP) was increased to 15 mm Hg by infusing fluids to enhance pulmonary edema formation. After 60 mins, PAOP was allowed to decrease to 5 mm Hg and was maintained at 5 mm Hg for 60 mins to stabilize the pulmonary edema. We studied the effect of tracheal gas insufflation on gas exchange at low and high end-expiratory lung volumes achieved by a positive end-expiratory pressure of 5 and 12 cm H2O, respectively. The FIO2 values of the ventilator and catheter were equivalent (0.6). Each tracheal gas insufflation stage at low and high end-expiratory lung volume was preceded and followed by conventional mechanical ventilation stages without tracheal gas insufflation. During transitions between conventional mechanical ventilation and tracheal gas insufflation, end-expiratory lung volume was maintained constant by adjusting positive end-expiratory pressure while monitoring esophageal pressure and inductive plethysmography. Tidal volume was maintained constant throughout the protocol (0.40 L). MEASUREMENTS AND MAIN RESULTS. At end stage, we measured PaCO2, PaO2, total physiologic deadspace fraction, and venous admixture, which were 43 +/- 4 torr (5.7 +/- 0.5 kPa), 325 +/- 6 torr (43.3 +/- 0.8 kPa), 53 +/- 3%, and 4.0 +/- 0.3% before oleic acid lung injury, respectively. After oleic acid injury at low end-expiratory lung volume, these variables were 55 +/- 4 torr (7.3 +/- 0.5 kPa), 73 +/- 13 torr (9.7 +/- 1.7 kPa), 61 +/- 4%, and 50 +/- 7%, respectively. During tracheal gas insufflation at low end-expiratory lung volume conditions, PaCO2 and the total physiologic deadspace fraction decreased significantly (p < .05) to 45 +/- 4 torr (6.0 +/- 0.5 kPa) and 50 +/- 5%, respectively. Under high end-expiratory lung volume conditions, PaCO2 and the total physiologic deadspace fraction were 55 +/- 7 torr (7.3 +/- 0.9 kPa) and 61 +/- 6%, respectively; during tracheal gas insufflation, these variables decreased to 43 +/- 4 torr (5.7 +/- 0.5 kPa) and 52 +/- 5%, respectively (p < .05). Increasing end-expiratory lung volume improved both PaO2 and venous admixture (p < .05) but tracheal gas insufflation had no significant effect on oxygenation efficiency when end-expiratory lung volume was held constant. CONCLUSIONS Tracheal gas insufflation augmented alveolar ventilation effectively in the setting of oleic acid-induced lung injury in dogs. When end-expiratory lung volume and tidal volume were kept constant, tracheal gas insufflation did not affect oxygenation.