We characterized a tobramycin aerosol generated by five nebulizers: Mi
cron One(R), Pulmosonic(R), Pulmo-Aide(R), DeVilbiss Model 65(R), and
UltraNeb 100(R) by particle size and drug concentration. The Micron On
e nebulizer did not produce a recoverable aerosol, while the Pulmosoni
c had a minimal output; therefore three machines were examined for the
ir ability to deliver tobramycin to the lower respiratory tract of pat
ients with cystic fibrosis (CF). The DeVilbiss 65 had the greatest out
put: with air as the carrier gas it produced an aerosol with >60% of t
he particles having a mean mass aerodynamic diameter (MMAD) of >5.5 mu
m. Using helox shifted the MMAD so that >65% of the particles were <5
.5 mu m. Increasing the power in the DeVilbiss 65 increased the output
of particles >9.2 mu m, without a change in the particles <3.3 mu m,
With air as the carrier gas the Pulmo-Aide and the UltraNeb 100 produc
ed an aerosol with >60% particles, <3.3 mu m MMAD. Using helox the Ult
raNeb 100 increased the amount of aerosol with a <3.3 mu m MMAD to 98%
. Tobramycin delivery to the lower respiratory tract with the Pulmo-Ai
de and UltraNeb 100 was compared using air or helox by measuring sputu
m drug concentration. Pulmo-Aide failed to produce detectable tobramyc
in in sputum in 2 out of 9 patients with CF. With the UltraNeb 100, al
l patients had measurable sputum tobramycin immediately after administ
ration (range, 16.2-3385 mu g/g), but no statistically significant dif
ference was found when using either compressed air, helox, or ambient
air. There was no correlation between pulmonary function and sputum to
bramycin content. We conclude that the UltraNeb 100 consistently deliv
ered tobramycin to the lower respiratory tract in patients with CF. (C
) 1994 Wiley-Liss, Inc.