The Science Journal of the American Association for Respiratory Care

2012 OPEN FORUM Abstracts

A COMPARISON OF LEAK COMPENSATION IN ACUTE CARE VENTILATORS DURING INVASIVE VENTILATION; A LUNG MODEL STUDY.

Jun Oto, Andrew D. Marchese, Robert M. Kacmarek; Massachusetts General Hospital, Boston, MA

Background: Patient-ventilator synchrony is mandatory for optimal ventilator support during invasive ventilation in the presence of system leaks. We compared leak compensation in 5 new acute care ventilators. Method: Using an ASL5000 lung simulator, the Maquet Servo-i, Drager V500, Covidien PB840, Respironics V60, and Hamilton C3 were compared during increasing (n=6) and decreasing leaks (n=6). Leak levels used were: BL (baseline < 0.5 L/min), L1 (4-5L/min), L2 (9-10 L/min) and L3 (26-27 L/min). Lung model inspiratory and expiratory resistance were 10, and 20 cmH2O/L/sec with compliance 60 ml/cmH2O (Obstructive model) and inspiratory and expiratory resistance of 5, and 5 cmH2O/L/sec with compliance 20 ml/cmH2O (Restrictive model). Ventilator settings were invasive ventilation, pressure support, PEEP 5, and 10 cmH2O and pressure support level 12 cmH2O. The number of breaths to synchronization was recorded for each leak scenario. Results: Only PB840 and V60 exhibited synchronization following all increasing and decreasing leak scenarios in both obstructive and restrictive models. Servo-i did not synchronize at any leak level and the V500 and C3 could not synchronize to leak L3. Number of breaths to synchronization for increasing leaks differed from decreasing leak with median breaths (25th-75th) of 2 (1, 3) and 0 (0, 0) (p < 0.0001) respectively. Significant differences were observed for number of breaths to synchronization between the obstructive 2 (0, 3) and restrictive model 0 (0, 1) (p < 0.0001), and PEEP 5 cmH2O 0 (0, 2.25) and 10 cmH2O 1 (0, 2) (p =0.03). PB840 required less breaths to synchronize to increasing and decreasing leaks in both obstructive and restrictive lung models and with PEEP 5 cmH2O and 10 cmH2O compared with V60 (p < 0.0001). Conclusions: PB840 and V60 were the only ventilators that adapted well to increasing or decreasing leaks. There are differences in performance between these two ventilators, although the clinical significance of these differences is unclear. In the presence of leaks over 26-27 L/min, most ICU ventilators cannot synchronize even if leak compensation is available. Grant support; COVIDIEN Inc, Boulder, CO Sponsored Research - COVIDIEN Inc, Boulder, CO