Should mechanical ventilation be optimized to blood gases, lung mechanics, or thoracic CT scan?

This study was aimed at providing data for optimization of mechanical ventilation in patients with acute respiratory distress syndrome (ARDS). The effects of ventilation with positive end-expiratory pressure (PEEP) titrated to blood gases were studied by thoracic computed tomographic (CT) scans and lung mechanics measurements in eight patients. CT density histograms at end-expiration were used to investigate the effects of PEEP on three differently aerated zones. Static pressure-volume (P-V) curves were used to determine the deflection point above which baro-volotrauma (a combination of barotrauma and volotrauma) may occur. Peak pressures, plateau pressures, and lung volumes measured by Respitrace were compared with the deflection point. CT scan showed that PEEP increased "normally aerated" areas, decreased "nonaerated" areas, and did not change "poorly aerated" zones. No correlations were found between CT scan and either PaO2 or mechanical data. Pressure at the deflection point was lower than the usually recommended 35 to 40 cm H2O for peak pressure in four patients (range, 28 to 32 cm H2O). With regard to plateau pressures, only one patient was ventilated above the deflection point. However, monitoring of volumes showed that these four patients had an end-inspiratory volume above this point. We conclude that mechanical ventilation may be initially adjusted on the basis of blood gas values and then optimized on the basis of lung mechanics to limit the risk of baro-volotrauma.