2007 OPEN FORUM Abstracts
EFFECT OF LUNG IMPEDANCE ON TIDAL VOLUME OUTPUT OF TWO HIGH FREQUENCY VENTILATORS IN A NEONATAL TEST LUNG MODEL
R. Gillette1, S. Messier1
Background: Gas exchange by high frequency (HF) ventilators is highly dependent on tidal volume (TV). The SensorMedics 3100A (Viasys) is widely used but TV is difficult to estimate clinically and effect of load (impedance) on TV, while studied, is not fully explored. Use of Bronchotron and VDR HF "percussive" ventilators (Percussionaire) especially for neonatal transport is spreading rapidly but not well studied, nor are differences between these ventilators.
Objective: To investigate the dependence of TV output of the Bronchotron compared to the SM 3100A on various combinations of compliance and resistance, endotracheal tube (ETT) size, ventilator frequency, and proximal pressure amplitude.
Methods: A HF ventilator was connected to a 1.5 L calibration syringe (Hans Rudolph) used as a test lung with compliance of 0.05-1.1 ml/cmH2O set by varying its volume of compressible air, via a 2.5 or 3.5 mm ETT and calibrated airway resistor of 0-250 cmH2O/L/s, simulating sick newborn lungs. TV and pressure amplitude (PA) were measured with a Florian hot wire anemometer and pressure transducer (Acutronic).
Results: For fixed values of other variables, TV increased with PA, linearly on SM3100A, often less linearly on Bronchotron. TV fell with higher frequency, smaller ETT, and higher resistance, but varied little with compliance at high frequency and resistance. At low frequency and resistance, TV fell rapidly with compliance below 0.3-0.4 but plateaued above that; at high frequency/low resistance, or low frequency/high resistance, that pattern was present but less pronounced. Similar patterns were seen with each ventilator but TV was nearly always higher, often substantially, for the 3100A (example: 5 Hz, 2.5 ETT, resistance 20, compliance 1.0, PA 40: TV 6.8 ml for Bronchotron vs 8.9 for 3100A).
Conclusion: TV depends markedly on lung impedance as well as set frequency and PA, consistent with principles of fluid mechanics, and suggesting that both ventilators act more nearly as pressure than flow sources, with relatively low internal impedance. We speculate the TV difference between ventilators is due to different operating mechanisms and pressure/flow waveforms. TV and gas exchange may be predictable if load can be estimated. Altered gas exchange should be anticipated when switching HF ventilator types e.g. for transport. Further studies of ventilator differences, clinical ways to determine TV, and relation to gas exchange are warranted.