Oronasal and Tracheostomy Delivery of Soft Mist and Pressurized Metered-Dose Inhalers With Valved Holding Chamber

Discussion

We found that in general, albuterol delivery via an oronasal and tracheostomy route using a metallic VHC resulted in a higher lung dose with an SMI than with a pMDI. We also found that in general, the lung dose was higher for tracheostomy than for oronasal delivery for both pMDI and SMI devices used with a metallic VHC. Finally, deposition in the tracheostomy tube was higher for SMI than pMDI when both were used with a metallic VHC.

The finding that the use of an SMI provided a larger lung dose than the pMDI is consistent with the characteristics of the aerosols produced by both inhalers.⁵ The SMI produces an aerosol that is 10 times slower and is ejected for a 3-fold greater duration than the one produced by the pMDI.⁵ We speculate that the difference in magnitude noted among different tidal volumes could be explained by the relative influence of difference in speed and duration of the aerosols. We speculate that a slower aerosol will result in less impaction and will be mainly responsible for the increase in lung dose found with tidal volumes of 155 and 300 mL. We speculate that the higher magnitude of the increase in lung dose noted with the 50-mL tidal volume is heavily influenced by the duration of aerosol because it is only one quarter of the volume of the VHC compared with 75 and 100% of the volume for the tidal volumes of 155 and 300 mL, respectively. Our findings are also consistent with adult studies comparing lung deposition between a pMDI and an SMI but without the use of a VHC.¹³ Wachtel et al¹⁴ reported a lung dose of 33% for an SMI using a different airway model but with a similar VHC (Vortex). The difference with our lung dose (21.1%) could be due to the fact that we used a different airway model and breathing pattern. In another study, Amirav et al¹⁵ reported a 30 and 60% lung dose in a 20-month-old model using breathing patterns with a tidal volume of 80 and 120 mL, respectively. They also reported a 20 and 40% lung dose in a 14-month-old model using breathing patterns with a tidal volume of 50 and 80 mL, respectively.¹⁵ The same group reported a 4.2% lung deposition in a group of 9-month-old infants using a VHC and Respimat.⁹ Their in vitro model overestimated lung deposition by a 4-fold difference.^9,15 Our model yielded a 9% deposition for the breathing pattern with a tidal volume of 155 mL. Again, differences in airway models, breathing patterns, VHCs, and masks could potentially explain these differences.

The finding that during oronasal delivery of pMDI lung dose increased with increasing tidal volumes is consistent with previous reports using anatomically correct and non-anatomically correct models.^11,12,16 During tracheostomy delivery, lung dose was lower for the 50-mL tidal volume but was similar between the breathing patterns with 155 and 300 mL. This finding is in agreement and disagreement with previous reports.^11,12 Cooper et al,¹² using the same model and 2 VHCs made of a nonelectrostatic polymer, reported a direct correlation between tidal volume and lung dose for one (Aerotrach Plus) and a lower lung dose for the 50-mL tidal volume than for other tidal volumes (155 and 300 mL) with the other VHC (Aerochamber Mini). Berlinski et al,¹¹ using a non-anatomically correct model, reported similar findings for the same tracheostomy size and similar breathing patterns (tidal volumes of 80, 155, and 310 mL) for the Aerochamber Mini. However, lung dose was similar among all breathing patterns for the Aerotrach. These differences could be explained in part by the characteristics of the models used. Whereas this study and the one by Cooper et al¹² used a 2-compartment model with bidirectional flow, the study by Berlinski et al¹¹ used a 2-compartment model with unidirectional flow. Their model placed a one-way valve at the vocal cord level that opened during expiration in contrast to the bidirectional flow model that allowed air flow through the vocal cords during inspiration and expiration.^11,12 These findings underscore the fact that data for a specific device should not be extrapolated to others and that choice of model could influence results.

Consistent with a previous report,¹² the amount of drug deposited in the tube increased with increasing tidal volumes. This was true for both the SMI and the pMDI, with the latter being lower than previously reported.¹² However, this was in contrast to findings by Berlinski et al¹¹ that reported for the pMDI similar deposition for different tidal volumes. The difference could be explained in part by the difference in model (unidirectional vs bidirectional flow) and pMDI formulation (Proair HFA vs Ventolin HFA).^11,17 An unexpected finding was that deposition in the tracheostomy tube was higher with the SMI than with the pMDI. We hypothesized that the lower speed would have resulted in lower impaction in the tube. We speculate that the longer duration of the aerosol of the SMI allowed more time to deposit in the tube.

This study adds to the body of knowledge provided by previously reported data using the same model and with experiments done by the same operator and allows a direct comparison with fewer biases than are usually found.¹² During oronasal delivery of pMDI with tidal volumes of 50 and 155 mL, the Aerochamber Mini was better than the others and the Aerotrach VHC was better than the Vortex VHC. However, when the large tidal volume (300 mL) was used, lung dose was higher for the Aerochamber VHC followed by the Vortex VHC and then the Aerochamber Mini VHC. During tracheostomy delivery of pMDI, the Vortex VHC performed similarly to the Aerotrach VHC for the lower tidal volume (50 mL), and both had a lower lung dose than the Aerochamber Mini VHC. However, when larger tidal volumes were used (155 and 300 mL), the Vortex VHC outperformed the other 2. These data suggest the existence of a complex relationship between lung dose obtained with a pMDI, the breathing pattern, and the type of VHC used.

Our findings have several clinical implications: (1) pMDI and SMI are not interchangeable, and the magnitude of the difference in drug delivery varies significantly with different tidal volumes; and 2) recommendations based on data obtained with a specific device/formulation should not be assumed to be equal to other combinations until proven. In the absence of dosing equivalency, clinicians should consider starting at a lower dose and titrating to effect when switching from a less efficient to a more efficient delivery system. Given the number of variables that play a role in drug delivery (VHCs, masks, aerosol generators, breathing patterns, etc), practitioners should consider utilizing products that are cost-efficient and have been studied in conditions similar to the ones in which they will be used.

The major limitation of this study is its in vitro nature, which does not provide the biological variability that occurs during real life. These methods tend to overestimate lung dose, especially the fraction that is smaller than the 0.5 μm that would have been exhaled but is trapped in the filter instead. Nonetheless, these are well-established methods in aerosol research. The findings of our study should not be extrapolated to other devices, tracheostomy size tubes, and breathing patterns. Another limitation is that we did not measure particle size of the aerosols that left the tip of the tracheostomy tube.