The Science Journal of the American Association for Respiratory Care

2002 OPEN FORUM Abstracts


Dana A Simonson BA, John R Hotchkiss MD, Alexander B Adams RRT, John Marini MD, David J Dries MD. Regions Hospital?HealthPartners, Univ. of MN, St. Paul, MN.

Background: At 1,3 and 6 days after enrollment, the ARDSnet study reported similar PaO2 values in patients receiving 6 mL/kg and 12 mL/kg tidal volume yet mortality rates were 25% different between the two groups. While PaO2 or PaO2/FIO2 are used to determine lung injury status, these indices may not faithfully track the development or progression of ventilator induced lung injury (VILI). We examined PaO2 during as VILI developed in swine ventilated with a known injurious ventilatory pattern.

Methods: Normal pigs were ventilated at PCV = 35/3, I:E=1:2, FIO2=0.6 for 6 hours. Pulmonary mechanics and ABGs were continuously monitored. Post-study the lungs were extracted for wet weight/dry weight (WW/DW) determination.

Results: While PaO2 clearly decreased in 4 animals, PaO2 remained relatively stable throughout the experiment in 4 animals (Figure). WW/DW ratio (displayed to the right of the PaO2 data) was elevated in all animals (normal=5.5) indicating edema accumulation. Vt decreased after 3 hours of ventilation in all animals indicating a reduction in compliance.

Conclusions: PaO2 may not be a dependable index for the development of VILI. We speculate that hypoxic vasoconstriction can mask a progression of injury that could be tracked by a reduction in compliance. AHA SDG 9930184N; NIH SCOR 50152


You are here: » Past OPEN FORUM Abstracts » 2002 Abstracts » VILI MAY BE CONCEALED BY ACCEPTABLE PAO2 LEVELS.