Translational Research: Multi-Scale Models of the Pulmonary Circulation in Health and Disease - Multiscale Computer Modeling in Biomechanics and Biomedical Engineering - page 273

on blood flow alone, Burrowes et al. [ 34 ] introduced assessment of gas exchange
function in a model of localized pulmonary disease by solving Eqs. 9 and 10 ,
along with the Monod-Wyman-Changeaux model for S(p cO2 ) [ 86 ]. They assumed
a linear distribution of ventilation with vertical height and were able to predict
baseline p AO2 that was consistent with physiological measurements [ 7 ]. This was
expanded upon by Tawhai et al. [ 87 ] who modeled both ventilation [ 55 ] and
perfusion [ 17 ] in subject specific geometries for nine subjects, and estimated gas
transport using the same methodology as Burrowes et al. [ 34 ]. Figure 6 shows a
normal (baseline) V/ Q distribution as predicted in a normal subject using this
model, compared with a traditional MIGET plot in a normal individual [ 88 ]. The
predictions of function made using anatomical geometries and physical laws by
Tawhai et al. compare well to the MIGET measurements, that are essentially
statistical estimates of ventilation and perfusion distributions made following
global measurements of inert gas elimination from the lung. The coupling of
ventilation and perfusion models to predict gas exchange function in anatomical
geometries opened the door for meaningful studies of pathological pulmonary
perfusion.
5.2 Application to Pulmonary Vascular Disease
The advent of anatomically based models of the pulmonary circulation has
enabled model-based investigation of disease states that present heterogeneously
in the lung (i.e. localized tissue abnormalities with relatively normal tissue in
large parts of the lung) and/or in the population (i.e. diseases that affect one
individual more than another for an apparently similar level of regional disease).
These models have so far focused on investigation of pulmonary hypertension
(PH) secondary to pulmonary embolism (PE). In PE, blood clots occlude pul-
monary arteries leading to a redistribution of blood flow and pressures, which can
in severe cases lead to an elevation of pulmonary arterial pressure (PAP) that can
lead to RV dysfunction or failure. The obvious culprit responsible for elevated
PAP is the occlusion of large pulmonary arteries, as this theoretically could
increase PVR substantially, leading to RV dysfunction. However, there are sig-
nificant inter-subject variations in response to PE, even when patients present with
apparently similar levels of occlusion [ 89 , 90 ]. The question as to what is
responsible for this variable response then hinders PE diagnosis and treatment: as
clot load alone is not correlated with survival [ 91 ], is it clot location (e.g. apical,
basal, proximal, distal) or some humoral response that is responsible for variable
outcomes? Or could inter-patient variability in response to clot simply be a result
of underlying pathologies? To date, these questions have been addressed in
anatomically based models of both the acinus [ 32 ] and the whole lung [ 33 , 34 , 87 ]
in order to capture the functional consequence of arterial occlusion across all
relevant spatial scales.