Objective: Severe posttraumatic lung injury is characterized by impairment of gas exchange and pulmonary densities. The influence of intermittent prone positioning on pulmonary gas exchange and parenchymal densities was investigated prospectively in patients with pulmonary injury after multiple trauma with blunt chest trauma.
Setting: A six-bed trauma intensive care unit in a university hospital.
Design: Prospective, descriptive study.
Patients: Twenty-two consecutive patients with pulmonary injury after multiple trauma with blunt chest trauma and acute lung injury (n = 11) or severe acute respiratory distress syndrome (ARDS) (n = 11) according to the definitions of the consensus conference on ARDS.
Interventions: Pulmonary densities were calculated planimetrically from computed tomographic scans of the chest before the first and after the last cycle of prone positioning. Indications for prone positioning were a) mechanical ventilation with FIO2 >0.5 at positive end-expiratory pressure >10 cm H2O for >24 hrs; or b) pulmonary densities in two or more quadrants being constant or increasing within 48 hrs. Arterial blood gas analysis was performed every 2 hrs. Intrapulmonary right-to-left shunt (Qs/Qt) and alveolar-arterial PO2 difference were calculated 2 hrs after the beginning and end of every prone and supine cycle, respectively. Patients were ventilated in the prone position for 8 hrs each day.
Measurements and main results: Every single posture change from the supine to the prone position resulted in a significant average increase in the oxygenation index of 28+/-8 torr (3.7+/-1.1 kPa) (p<.0001). There was a significant improvement in oxygenation (4.3+/-0.8 torr [0.57+/-0.11 kPa]) with time between two consecutive measurements in the prone as well as the supine position (p<.0001). Alveolar-arterial PO2 difference and Qs/Qt showed a significant decrease of 25+/-7 torr (3.3+/-0.9 kPa) and 1.1+/-0.46%, respectively, for every cycle of prone positioning. Statistical analysis revealed no significant alteration of gas exchange within every prone and supine cycle. Total static lung compliance improved significantly over time (p<.001). However, ventilation of patients in the prone position demonstrated a mean decrease in compliance of 2.1+/-0.72 mL/cm H2O. The response to prone positioning was similar in patients with ARDS and acute lung injury and revealed no significant difference. In both groups, the course of the oxygenation index and Qs/Qt over time was almost parallel. Posture changes were continued for 9.0+/-1.1 days. The oxygenation index showed an overall increase of 129+/-20 torr (17.2+/-2.7 kPa) from baseline supine at the end of prone positioning (p<.0001). Pulmonary densities were reduced significantly from 31.1+/-2.5% to 3.8+/-0.81%, Qs/Qt was reduced from 24.9+/-1.5% to 11.7+/-0.32%, and FIO2 was reduced from 0.43+/-0.04 to 0.26+/-0.02 (p<.01). Gas exchange improved in all patients, and no patient died immediately as a result of respiratory failure.
Conclusion: Repeated prone positioning recruits collapsed lung tissue and improves gas exchange in trauma patients with blunt chest trauma and severe ARDS as well as in trauma patients with acute lung injury.