Oxygen Transport in Bioreactors for Engineered Vascular Tissues - Computational Modeling in Tissue Engineering - page 291

Fig. 10 Model results for a DO and b glucose during PFS bioreactor operation. Example tissue
DO profiles determined for a range of peak pulse pressures induced by varying stroke volumes.
For DO, the effect of flow becomes more apparent at pressures of 25 mmHg, where

h Pe r,O2 i & 4, and higher. Glucose profiles exhibit larger shifts at lower pulse pressures

compared to DO

PARDISO solver with time steps of 0.1 s, a convergence tolerance of 10 -6 and a
minimum damping factor of 10 -4 . Examination of step changes in the inputs, such
as changing the flow rate for example, were examined over a period of at least
1 min to verify that a steady solution had been reached.
A summary of the fluid flow model is shown in Fig. 9 . Model results showed
velocities ranging from 15.9 cm/s at peak flow in the central port of the bottom
manifold to 0.07 cm/s in the bioreactor center. Swirl flow was induced after
leaving the bottom manifold, as evidenced by the streamlines in Fig. 9 d. The
streamlines also indicate that fluid flows around the tissue construct to the chamber
interior and then toward the outlet around the top manifold. This is noteworthy
since the fluid at the bottom of the chamber is well-mixed and flow around the
entire construct serves to improve DO uniformity during culture. To verify the
solution, calculation of mass conservation was conducted by boundary integration
of the velocity field on the inlet and outlet streams. Agreement within 0.02% was
obtained with the mesh geometries used to generate solutions shown in Fig. 10 .
Transmural flow was evident in Fig. 9 e. The vast majority of the flow was
found to leave the lower manifold, with approximately 5% being pulsed through
the tissue for L p = 1 9 10 -6 cm/s/Pa. Though stream lines are apparent in Fig. 9 ,
ablumenal flow is approximately an order of magnitude higher than transmural
flow.
For the frequency and stroke volumes within bioreactor capabilities when
pressure and DO measurements were taken, the predicted DO profiles in the tissue,
shown in Fig. 10 a, did not vary greatly. At P peak values of 5 and 10 mmHg,
corresponding to a 0.5 and 1.0 mL syringe stroke volume and h Pe r,O2 i of 0.95 and
1.9, respectively, there was only a slight disparity from a purely diffusive transport
regime. Doubling the pulse frequency from 0.5 to 1.0 Hz did not impact this