Resolution of pulmonary edema with variable mechanical ventilation in a porcine model of acute lung injury.

OBJECTIVE

Resolution of the acute respiratory distress syndrome (ARDS) requires clearance of pulmonary edema. Biologically variable ventilation (BVV) strategies that improve gas exchange, lung mechanics, and inflammatory mediators in ARDS may be beneficial in this regard. We used quantitative computed tomography (CT), a single indicator thermodilution system (PiCCO®) to determine extravascular lung water (EVLW), and the change in edema protein concentration over time to compare edema clearance with BVV vs conventional mechanical ventilation (CMV) in a porcine ARDS model.

METHODS

Sixteen pigs with oleic acid lung injury were randomized to four hours of ventilation with either CMV (n = 8) or BVV (n = 8) at identical low tidal volume and minute ventilation over time. Hemodynamic variables, gas exchange, lung mechanics, and PiCCO derived EVLW were determined hourly. Computed tomography images and edema fluid samples were obtained at baseline lung injury and after four hours of ventilation. Wet and dry lung weights were determined postmortem.

RESULTS

At four hours with BVV, peak airway pressure was decreased significantly and lung compliance improved compared with CMV (P = 0.003; P < 0.001, respectively). Hemodynamic variables and gas exchange were not different between groups. Also at four hours, computed tomography revealed an increase in total gas volume (P = 0.001) and a decrease in total lung weight and global lung density (P = 0.005; P = 0.04 respectively) with BVV. These findings were associated with a significant increase in the gas volume of normally aerated lung regions (P < 0.001) and a decrease in the poorly and non-aerated lung regions (P = 0.001). No change in any CT parameter occurred with CMV. The lung weights derived from computed tomography correlated well with postmortem wet weights (R(2) = 0.79; P < 0.01). The decrease in PiCCO derived EVLW from injury to four hours did not differ significantly between BVV and CMV. Extravascular lung water showed no correlation with postmortem wet weights and significantly underestimated lung water. Average alveolar fluid clearance rates were positive (1.4%·hr(-1) (3%)) with BVV and negative with CMV (-2.0%·hr(-1) (4%)).

CONCLUSIONS

In a comparison between BVV and CMV, computed tomography evidence suggests that BVV facilitates enhanced clearance and/or redistribution of edema fluid with improved recruitment of atelectatic and poorly aerated lung regions; no such evidence was seen with either single thermodilution measurement of EVLW or edema clearance rates. The results of computed tomography provide further evidence of the benefit of BVV over conventional ventilation in ARDS.