@article{dave_kleinstreuer_chari_2022, title={An effective PBPK model predicting dissolved drug transfer from a representative nasal cavity to the blood stream}, volume={160}, ISSN={["1879-1964"]}, DOI={10.1016/j.jaerosci.2021.105898}, abstractNote={Predicting the fate of nasally administered drugs is important for the understanding and possible improvement of in vivo performance. When computational fluid-particle dynamics (CFPD) results are coupled with a physiologically based pharmacokinetic (PBPK) model, drug concentrations in the blood stream can be obtained. Specifically, hybrid CFPD-PBPK simulations can predict inhaled particle transport, deposition, and uptake in the nasal mucus layers and subsequently absorbed drug migration from the nasal cavity to the blood stream.The computer simulation results of Chari et al. (2021) were used as input to a basic PBPK model to track the deposited and dissolved drugs from the nasal cavities to the blood stream. Employing the open-source toolbox OpenFOAM, our PBPK model predictions were compared with experimental in vivo data sets for different corticosteroids. The relative differences between experimental and simulated values of PK metrics, following administration of mometasone furoate nasal spray, were all 7% or less. Drug plasma concentrations based on different drug parameters, such as solubility and partition coefficient, were studied as well. The drug concentration in the plasma was found to increase with an increase in drug solubility (Cs = 0.02 mg/ml, 0.1 mg/ml, 0.2 mg/ml). The same trend was observed for different partition coefficients (Kow = 5e-3, 2, 5000), where the plasma concentration curve peaked for a partition coefficient of 5000. It was also observed that drug dosage controls the amount of residual drug concentrations in the plasma with the passage of time. Two different drug dosages were studied, ie, 50 μg and 800 μg, with the former being completely absorbed in the plasma after 8 h; however, in the latter case the drug was not completely absorbed after that time interval. These modeling and simulation results are useful for planning aspects in drug development, as the predictions provide physical insight to differences in device, formulation, and dosage selection.}, journal={JOURNAL OF AEROSOL SCIENCE}, author={Dave, Sujal and Kleinstreuer, Clement and Chari, Sriram}, year={2022}, month={Feb} }
 @article{chari_sridhar_kleinstreuer_2022, title={Effects of subject-variability on nasally inhaled drug deposition, uptake, and clearance}, volume={165}, ISSN={["1879-1964"]}, DOI={10.1016/j.jaerosci.2022.106021}, abstractNote={Accurate and realistic predictions of the fate of nasally inhaled generic drugs provide new physical insight which can be of great importance to toxicologists, drug developers and federal regulators alike. To understand the dynamics of mucociliary clearance (MCC) and subsequent absorption of the dissolved drug by the nasal epithelium, it becomes necessary to model the air-particle-mucus dynamics accurately. The MCC process, including particle dissolution, transport and absorption for a 3-D representative nasal cavity, were established by Chari et al. (2021). In this study, the effects of inter-subject variability of three representative nasal cavities (subjects A, B, C) on deposition and subsequent uptake of the dissolved drug in the nasal epithelium are analyzed for three generic drugs: Mometasone furoate (MF), Flunisolide (FN), and Ribavirin (RB). The computational fluid-particle dynamics (CF-PD) results indicate that smaller sized particles (3 μm) deposit more in the ciliated portion of the nasal cavity where the columnar cells responsible for uptake are present. In contrast, larger particles (10 μm) tend to deposit in the unciliated anterior third of the nose. The epithelial uptake in case of subject A was considerably higher than that in subjects B and C because of the unique anatomical characteristics of subject A. Also, FN and RB were found to have a higher rate of uptake compared to MF due to their considerably higher partition coefficient. As a visualization tool, concentration contours are used to explain regional trends in cumulative drug uptake for all three cases. • The open-source CFD toolbox, OpenFOAM, has been employed for the development of the computer simulation model. • This study illustrates the effects of inter-subject variability on deposition, dissolution and uptake of 3 generic drugs in representative nasal cavity models. • Smaller particles, with their relatively large surface area, tend to dissolve quicker and are absorbed more rapidly than larger particles. • Particles deposited closer to the ciliated portion of the nasal cavity are more readily absorbed when compared to particles deposited closer to the unciliated nasal vestibule.}, journal={JOURNAL OF AEROSOL SCIENCE}, author={Chari, Sriram and Sridhar, Karthik and Kleinstreuer, Clement}, year={2022}, month={Sep} }
 @article{chari_sridhar_walenga_kleinstreuer_2021, title={Computational analysis of a 3D mucociliary clearance model predicting nasal drug uptake}, volume={155}, ISSN={["1879-1964"]}, DOI={10.1016/j.jaerosci.2021.105757}, abstractNote={Accurate and realistic predictions of the fate of nasally inhaled drugs help to understand the complex fluid-particle dynamics in the nasal cavity. Key elements of such a comprehensive numerical analysis include: (i) inhaled drug-aerosol transport and deposition with air-particle-mucus interactions; and (ii) mucociliary clearance (MCC) dynamics, including drug transport, dissolution and absorption for different nasal inlet conditions. The open-source computational fluid dynamics (CFD) toolbox, OpenFOAM, has been employed for the development of the computer simulation model. As part of the design, a novel 3D meshing technique allows for the smooth capture of both the relatively large flow domain as well as the micron-size mucus layer. This efficient meshing strategy drastically reduces the overall meshing time from hours to a matter of minutes. The effect of pharmacokinetic characteristics of drugs on dissolution, subsequent uptake and clearance were analyzed. A method to impose a boundary-driven flow velocity was introduced in order to mimic the beating of the cilia. Several drug specific parameters, such as solubility, partition coefficient and particle size, were considered. The effects of particle distribution on MCC and uptake were simulated as well. The CFD predictions show that drugs with a high partition coefficient are absorbed rapidly. Similarly, drugs with higher solubility show an appreciable increase in cumulative uptake in the epithelium. Particle size, however, plays a more nuanced role in drug uptake. Specifically, smaller particles with their relatively large surface areas, tend to dissolve quicker and are absorbed more rapidly when compared to larger particles. However, after the initial steeper increase in cumulative uptake of the smaller particles, the difference in the uptake values for the two cases is negligible. Furthermore, the initial deposition locations in the nasal cavity play an important role in overall drug uptake. Particles deposited closer to the ciliated portion of the nasal cavity (i.e. the posterior region) were more readily absorbed when compared to particles deposited closer to the unciliated nasal vestibule.}, journal={JOURNAL OF AEROSOL SCIENCE}, author={Chari, Sriram and Sridhar, Karthik and Walenga, Ross and Kleinstreuer, Clement}, year={2021}, month={Jun} }
 @article{chari_kleinstreuer_2018, title={Convective mass and heat transfer enhancement of nanofluid streams in bifurcating microchannels}, volume={125}, ISSN={["1879-2189"]}, DOI={10.1016/j.ijheatmasstransfer.2018.04.075}, abstractNote={Significantly improved mixing and heat transfer between two nanofluid streams in a Y-shaped sinusoidal microchannel have been achieved via geometric modifications and changes in pulsatile flow conditions. As a result a new mass-and-heat transfer correlation has been obtained as well. The geometry modification was done in two distinct parts. First, the phase shift (ϕ) between the wavy walls of the Y-channel was varied for three different shift values 0°,90°,180°. Once the shift that yielded the highest degree of mixing was determined, the included angle (α) between the input streams was varied 30°,45°,60°. The numerical results show that α=60° and ϕ=90° yield best results. The inlet streams are pulsatile with a velocity of the form V+δVsinωt+Φ where V,δV,ω,Φ are the average velocity, pulse amplitude, pulse frequency and phase shift respectively. Flow variations have been implemented via different phase shifts 45°,90°,135°,180°, different phase amplitudes and different phase frequencies. For the sake of comparison and ease of plotting, non-dimensional parameters have been used. The frequency has been captured by the non-dimensional Strouhal number (St) and the amplitude by the amplitude ratio (δV/V). The average degree of mixing (ζ), which is observed to undergo spatial variations along the exit channel as well as temporal fluctuations over one pulsation cycle is shown to be most sensitive to the amplitude and frequency of pulsations. For a fixed amplitude, the average degree of mixing increases with elevated Strouhal numbers. It reaches a peak at a particular St value and then decreases with further increase in St. The St where the degree of mixing peaks depends on the amplitude of pulsations. For δV/V⩾5, the degree of mixing peaks at St≈0.5. For δV/V⩽5, the degree of mixing peaks at St≈2. Unlike mixing, the heat transfer rate, characterized by the non-dimensional Nusselt number (Nu) peaks at higher frequency values for all δV/V ratios. To generate higher amplitudes in the pulsating flow, a larger pumping power would be required. Hence, to minimize energy cost, low amplitude and high frequency pulsations are most suitable for optimal mixing and heat transfer. Finally, a microchannel with optimized geometry and inlet flow conditions is proposed, which takes advantage of the flow instabilities created by the modified geometry and pulsating flow to yield the highest degree of mixing and heat transfer within the listed constraints. Functional dependencies have been established, based on computer experiments, between non-dimensional parameters such as St,δV/V,ζ,Nu . As a result, a correlation between mixing and heat transfer was developed which allows studying one quantity, say, the Nusset number Nu, to readily obtain the average degree of mixing ζ.}, journal={INTERNATIONAL JOURNAL OF HEAT AND MASS TRANSFER}, author={Chari, Sriram and Kleinstreuer, Clement}, year={2018}, month={Oct}, pages={1212–1229} }