Evaluation of Trigger Asynchrony in Five Adult Ventilators

Abstract

Background: Trigger asynchrony is a common problem during mechanical ventilation and occurs primarily in patients with dynamic hyperinflation and COPD. We were curious if the brand of ventilator platform also played a role in trigger asynchrony. In other words, do some ventilators inherently allow for easier triggering than others. Our null hypothesis is that no variation of missed triggers will occur across 5 different ventilator platforms.

Methods: An Ingmar Medical ASL 5000 breathing simulator was used to evaluate 5 adult ventilators: Servo-U, Servo-i, Hamilton G5, Hamilton C6, and Dräger V500. Three different patient models were also evaluated: normal, ARDS, and COPD. The ASL was set for 20 breaths/min across all model types, but compliance, resistance, and waveform patterns were left pre-set for the given patient model. Four different inspiratory muscle pressures (IMP) of 20, 10, 5, and 2.5 cm H₂O were used to simulate strong to weak muscle movement. The simulator was then set for one of the three patient models, 5 cm H₂O PEEP, and IMP of 20 cm H₂O. Each ventilator was placed in pressure support mode, flow trigger of 2 L/min, and allowed to run for 1 min while the total number of triggered breaths were counted. The process was repeated for IMPs of 15, 5, and 2.5 cm H₂O. Three runs were performed at each IMP level. The PEEP was then increased to 10 cm H₂O and this process was repeated. After collecting data for one of the model types, the entire process was repeated for the other two model types.

Results: When using normal and ARDS models all ventilator platforms triggered 100% of ASL set breaths at all muscle pressures and both PEEP levels. When using the COPD model, the Servo-i and Servo-U triggered 100% of set breaths at all IMPs except 2.5 cm H₂O in which 90% of breaths were triggered. The Hamilton C6 triggered 90% of set breaths at all IMPs except 2.5 cm H₂O in which 67% were triggered. When PEEP was increased to 10 cm H₂O on the Hamilton C6, triggering improved from 67% to 90%. The Hamilton G5 and Dräger V500 triggered 100% of set breaths at all IMPs except 2.5 cm H₂O in which 40% were triggered. When PEEP was increased to 10 cm H₂O on the Hamilton G5, triggering improved from 40% to 52%. Increasing the PEEP on the Dräger V500 did not improve triggering.

Conclusions: We rejected our null hypothesis and concluded that the Servo-i and Servo-U distal flow sensor was the most responsive in detecting flow in the weak COPD patient model when compared to the Hamilton C6, Hamilton G5, and Drager V500.