Predicting Aerosol Deposition during Neonatal Ventilation: Feasibility of Bench Testing

Abstract

Study Question

Can in-vitro bench testing of neonatal pressure-controlled ventilators predict aerosol delivery to the lung?

Background

Lung deposition of an aerosolized drug is determined by the amount of drug inhaled (inhaled mass) and by the tendency of the lung to retain rather than exhale the aerosol. For adult aerosol delivery systems, in-vitro bench testing has resulted in increased lung deposition in clinical studies of spontaneously breathing and mechanically ventilated patients. However, it has been reported that bench testing of neonatal ventilator systems is not useful for predicting clinical deposition.

Purpose of the Study

To assess whether perturbation of a bench model of neonatal ventilation in a manner likely to change aerosol delivery would result in proportional changes in deposition in an animal lung.

Methods

We measured aerosol delivery using a bench model of a pressure-controlled, continuous-flow, neonatal ventilator system delivering a typical range of minute ventilation (260-900 mL) to a test lung. A radioactive aerosol of 99mTc bound to human serum albumin was generated using a small-volume ultrasonic nebulizer and collected on a filter positioned distal to a 3.0-mm endotracheal tube (ETT) that determined inhaled mass. The nebulizer was placed at two locations: its conventional position in the inspiratory line (IL position) and just distal to the Y-piece of the ventilator (DY position) a position not clinically applicable but likely to increase aerosol delivery. After the bench studies, the two nebulizer-ventilator circuits and settings were used to deliver the radioaerosol to the lungs of mechanically ventilated rabbits and lung deposition was calculated using a gamma camera and attenuation-coefficient technique.

Results

Bench model, inhaled mass-using the DY configuration, 8.3 ± 0.94% of the activity was delivered to the inhaled-mass filter compared to 1.6 ± 0.6% in the IL configuration (p < 0.0001, DY:IL 5.2:1). Deposition in the ETT was similar in magnitude to the inhaled mass (DY vs IL configuration, 6.4 ± 1.7% vs 1.4 ± 0.6%, p < 0.0001) and correlated with the measured particle distributions. Rabbit model, lung deposition: DY configuration vs IL position 3.31 ± 2.1 % vs 0.56 ± 0.09%, p = 0.008, DY:IL 5.9:1).

Conclusions

Differences in aerosol delivery measured in an in-vitro bench model are found to be proportional to those detected in the lung. A significant quantity of nebulized particles deposit in the ETT. Therefore, bench testing may have a useful role in preliminary testing of aerosol delivery systems and nebulized drugs in the setting of neonatal ventilation. The clinical acceptance and safety of modifications to standard neonatal circuits (eg, dead-space effects and tubing modifications) require additional clinical studies.