The Science Journal of the American Association for Respiratory Care

2009 OPEN FORUM Abstracts

OBJECTIVE MEASURES OF VENTILATOR PERFORMANCE USING THE PRESSURE SUPPORT MODE

David M. Wheeler2, Shannon E. Cook1, Robert L. Chatburn1; 1Respiratory Institute, Cleveland Clinic, Cleveland, OH; 2Anesthesia Institute, Cleveland Clinic, Cleveland, OH

Pressure Support (PS) modes on different ventilators are designed with different operational specifications. Asynchrony can happen due to inappropriate thresholds for trigger, limit, and cycle, variables. The purpose of this study was to compare objective variables relevant to ventilator-patient synchrony and comfort in the pressure support mode with ICU ventilators. METHODS: Ventilators tested: Maquet Servo-i, Cardinal Health Avea and Vela, Hamilton G5, Dräger Evita XL, Puritan Bennett 7200 and 840. The ventilators were set in continuous spontaneous ventilation mode with pressure support = 5 cm H2O and PEEP = 5 cm H2O with default values for pressure rise time and cycle threshold. All were flow triggered at maximum sensitivity. They were connected to an ASL 5000 lung simulator (Ingmar Medical Inc.). The lung model was set with the following values: frequency = 15 breaths/minute, resistance = 3 cm H2O/L/s, compliance = 60 mL/cm H2O, sinusoidal effort with Pmax = 3.5 cm H2O, 33% inspiration, 0% hold, 33% release, unassisted tidal volume = 209 mL. All variables were recorded with the ASL: trigger response time (in ms, from start of effort to return to baseline pressure), inspiratory and expiratory T90 (time to reach 90% of steady state inspiratory or expiratory pressure), patient work to trigger inspiration (mJ), peak inspiratory flow (mL/s), flow cycle threshold (% of peak flow), pressure drop during triggering (cm H2O), tidal volume (mL). For each variable, mean values from 10 breaths were compared with one way ANOVA; P < 0.001 considered significant. RESULTS: The Figure shows major outcome variables and the ventilators differed significantly on all. Despite identical pressure settings, the ventilators delivered different tidal volumes (Vela highest, 547 mL; Avea lowest, 417). Trigger work was highest for the Vela (15.3 mJ) and lowest for the Evita XL (0.3). Peak flow was highest for the Vela (940 mL/s) and lowest for the G5 (592 mL/s). Cycling threshold was highest for the Servo i (37 %) and lowest for Avea (24 %). CONCLUSIONS: These results suggest that despite operating in the same mode and pressure settings, there was significant performance difference among these ventilators. We are conducting a further study to determine if there are subjective impressions of comfort that can be correlated with these objective measures. Studies like this may inform future design considerations. Sponsored Research - None

You are here: RCJournal.com » Past OPEN FORUM Abstracts » 2009 Abstracts » OBJECTIVE MEASURES OF VENTILATOR PERFORMANCE USING THE PRESSURE SUPPORT MODE