Prone Positioning in Acute Lung Injury


Synopsis: Positioning patients with ARDS or ALI in the prone position resulted not only in an increase in PaO2, but also a decrease in chest wall compliance. The decrease in chest wall compliance highly correlated with the increase in PaO2.

Source: Pelosi P, et al. Am J Respir Crit Care Med 1998;157:387-393.

Sixteen patients with ali or ards receiving controlled volume, targeted, square-wave flow pattern ventilation were studied in the prone position. Data were collected at baseline (supine), at 30 and 120 minutes in the prone position, and then at 30 and 120 minutes after being repositioned supine. Gas exchange, lung mechanics, and end expiratory lung volume (EELV) were evaluated.

With prone positioning, PaO2 increased from 103.2 ± 23.8 to 129.3 ± 32.9 mmHg (P < 0.05). In 12 patients, the increase ranged from 9-73 mmHg, and, in four patients the PaO2 decrease ranged from -7 to -16 mmHg. No significant change was observed in EELV, total respiratory system compliance (CRS), lung compliance (CL), or airways resistance. However, chest wall compliance (CW) decreased (204 ± 97.4 to 135.9 ± 52.5 mL/cm H2O; P < 0.01) in the prone position. Upon return to the supine position, the CW returned to baseline and CRS increased (42.3 ± 14.4 vs 38.4 ± 13.7 mL/cm H2O; P < 0.01) as did CL (57.5 ± 25.1 vs 52.4 ± 23.3 mL/cm H2O). In addition, the increase in PaO2 during prone positioning was correlated to the decrease in CW (r = 0.80; P < 0.01).


This study helps to explain the mechanism in which prone positioning improves oxygenation and provides data to help identify which patients may benefit from prone positioning. Early studies of prone positioning speculated that the prone position improved overall EELV (Douglas WW, et al. Am Rev Respir Dis 1977;115:559-566). However, in this study, EELV was unchanged. What did change was CW. More importantly, the change in chest wall compliance highly correlated to improvement in oxygenation: the higher the chest wall compliance at baseline the greater the improvement in prone position PaO2.

These findings add data to verify the hypothesis that prone positioning improves oxygenation by improving overall ventilation/perfusion matching with a more uniform distribution of ventilation. Since the chest wall is fixed at the spinal cord but lightly mobile at the sternum, supine positioning results in a more ventral distribution of ventilation in the ventilated, paralyzed, supine patient with ALI or ARDS. By placing the patient prone, regional chest wall compliance is decreased (sternal areas less free to move and spinal area more free to move), resulting in a more uniform distribution of ventilation.

Although Pelosi and associates do not speculate because of the small number of patients studied, it seems reasonable that the data can be used to at least hypothesize about which patients would be expected to respond to prone positioning and which would not. My hypothesis would be that patients with markedly reduced chest wall compliance would be least likely to respond to prone positioning. Unfortunately, Pelosi et al did not list individual patient data. However, based on the strength of the correlation between oxygenation improvement and CW, I would expect that the four patients who showed a decrease in PaO2 when positioned prone had reduced chest wall compliance in the supine position.