Pressure-volume relationships in acute lung injury: methodological and clinical implications.

BACKGROUND: Pressure-volume relationships (PV curves) are the only available method for bedside monitoring of respiratory mechanics. Alveolar recruitment modifies the results obtained from the PV curves. We hypothesized that method-related differences may influence PV-curve guided ventilatory management.

METHODS: Twelve acute lung injury (ALI) patients [PaO2/FiO2 13.0 +/- 1.5 kPa (97.6 +/- 11.3 mmHg), bilateral pulmonary infiltrates] were studied. Two PV curves [one at variable, and another at constant level of positive end-expiratory pressure (PEEP)] were obtained from each patient using constant inspiratory flow and end-inspiratory and -expiratory occlusions. Upper and lower inflection points (UIP, LIP) were estimated. Recruitment due to PEEP and during inflation was assessed by respiratory inductive plethysmography (RIP).

RESULTS: (1) Pressure-volume curves at constant PEEP tended to provide higher LIP values compared with curves at variable PEEP (mean difference +/- SEM 5.1 +/- 1.9 cmH2O); and (2) recruitment occurred throughout the PV curve with no relationship with LIP or UIP.

CONCLUSION: Pressure-volume curves obtained using variable PEEP translate a different physiological reality and seem to be clinically more relevant than curves constructed at constant PEEP. If curves constructed at constant PEEP are used to set the ventilator, unnecessarily high PEEP levels may be used. Respiratory inductive plethysmography technology may be used for monitoring of recruitment at the bedside.