The Science Journal of the American Association for Respiratory Care

2008 OPEN FORUM Abstracts

TRIGGERING AND INITIAL GAS DELIVERY OF CURRENT ICU VENTILATORS

Andrew D. Marchese1, Demet Suleymanci1, Daniel W. Chipman1, Jesus Villar2, Robert M. Kacmarek1



Background: We compared the triggering and initial gas delivery capabilities of a number of sophisticated ICU ventilators during adult ventilation using the IngMar Medical ASL 5000 lung model.

Methods: The Puritan Bennett 840, Maquet Servo i, Newport e500, and GE Engstr�m were evaluated in the pressure support mode. A total of 24 tests were conducted for each ventilator. Lung mechanics combinations evaluated were compliance (C, mL/cmH2O) and resistance (R, cmH2O/L/sec) of 60 and 10, 60 and 5, and 30 and 10. Lung model inspiration lasted 0.8 sec (3.5% of breath cycle time inspiratory, 0.5% hold, and 22.7% release) at an inspiratory muscle effort (Pmus) of 5 and 10 cmH2O. The ventilators were set at pressure support 15 cmH2O, and PEEP 5 and 15 cmH2O. Triggering was set as sensitive as possible without auto-triggering. Rise time and termination criteria were set at the manufacturer's default and an optimal level (rise time as fast as possible without an initial airway pressure spike; termination criteria set to insure lung model and ventilator terminated the breath simultaneously). Standard adult circuits were used without humidifiers. All data was analyzed using SPSS version 15.0, p < 0.05 and mean difference ≥ 10%.

Results: The following variables were evaluated: Tidal Volume (VT), Pressure to Trigger (PT), Time to Trigger (TT), Time to Baseline (TB, from trigger to the reestablishment of PEEP), Inspiratory T90 (T90, time to 90% of target airway pressure), and trigger Pressure Time Product (PTP). Means and standard deviations of 5 representative breaths from all test scenarios were determined. From High to Low Pmus PT, TB, and PTP decreased. From C60 R10 to C30 R10, VT, PT, TT, TB, T90, and PTP decreased. There were large differences between default and optimal settings. The Newport was unable to ventilate the Lung Model under the lung mechanics settings of C60 R5 due to incompatibilities in repositioning frequencies resulting in oscillations, these tests were omitted.

Conclusion: Although there were differences among ventilator performance, the clinical impact of these differences is unknown. All performed exceptionally well.