The Science Journal of the American Association for Respiratory Care

2012 OPEN FORUM Abstracts


Gary O. Martin, Christopher Culter, Andrew Weirauch, Allan Andrews, Carl F. Haas; Respiratory Care, University of Michigan, Ann Arbor, MI

Background: It has become a standard practice to conduct an airway or cuff leak test in patients at risk for post-extubation upper airway obstruction. Typically, the ventilator is set to a predetermined VT and the exhaled volume observed. The endotracheal cuff is then deflated and the expired volume assessed. The use of automatic leak compensation on many modern ventilators is intended to increase the inspired volume to maintain a set delivered volume in the face of a system leak. Study Objective: We hypothesize that active automatic leak compensation will provide a false result suggesting a minimal leak when there is actually a significant leak. Methods: A ventilator (Evita Infinity V500, Drager Medical) was connected to a lung simulator (ASL 5000, Ingmar Medical) set to a single lung model with a compliance of 25 mL/cmH2O, a resistance of 10 cmH2O/L/sec, and a passive cycle. A “T” piece with a 12 inch length of corrugated tubing was attached to the simulator and represented the trachea. An endotracheal tube was inserted into the tubing and attached to the ventilator set to a VT of 500 mL with leak compensation inactive. Baseline measurements were obtained with the cuff fully inflated. Cuff pressures were decreased incrementally to 15, 10, and 5 cmH2O and measurements taken at each level. Measurements from the ventilator included: inspired VT (VTi), expired VT (VTe), peak inspiratory pressure (Ppeak) and % Leak. Each condition was studied in triplicate. Leak compensation was activated and the process repeated. The Perceived Leak (Set VT – VTe) and Actual Leak (VTi – VTe) were calculated and compared to each other. Results: With leak compensation off Actual Leak was not different from the Perceived Leak. The results with leak compensation on are shown in the table. A % Leak displayed from the ventilator of >20% seemed to be associated with a leak >110 mL. Conclusions: When leak compensation is active, exhaled volume readings can be altered and affect the clinical decision to extubate patients. Using either an audible assessment or the ventilator displayed % Leak may be more appropriate in this situation. Sponsored Research - None