The single path model of airway gas transport was incorporated into each of
the seven parallel regions model of Cruz (Cruz, J. C. Respir. Physiol. 86:
1-14, 1991). Thus, the effect of time on the predicted gas fractions in and
out of the lung could be evaluated. Two experimental maneuvers were simula
ted: (1) fast inhalation of an argon-oxygen mixture from a functional resid
ual capacity and fast exhalation to residual volume, including inspiratory
breath holdings of 5-20 s, and (2) the standard single-breath nitrogen wash
out test. Expired argon and nitrogen are predicted within a +/-3% error of
the experimental data with no breath holding. Breath holding predictions we
re at variance with experimental results because the solution of the diffus
ion-convection equation produced even mixing in the alveoli at the end of i
nspiration. The minimum square of the difference between the experimental d
ata (standard single-breath nitrogen washout test) and those provided by th
e model was 0.0016. This model is capable of generating a nitrogen expirogr
am with four phases when a vital capacity of oxygen is inhaled. However, th
e model failed to produce a sharp distinction between phase 3 and phase 4.
Thus, we conclude that uneven emptying of parallel regions generates any ex
pirogram (a fast or slow expiratory maneuver). The alveolar gas stratificat
ion that is created during inspiration disappears at the end of the inspira
tory maneuver. As a result, breath holding maneuvers cannot be predicted in
the anatomical model used. (C) 2000 Biomedical Engineering Society.