The Way the Wind Blows: Implications of Modeling Nasal Airflow
Kai Zhao, PhD, and Pamela Dalton, PhD, MPH
Corresponding authorPamela Dalton, PhD, MPHMonell Chemical Senses Center, 3500 Market Street, Philadelphia, PA 19104-3308, USA. E-mail: [email protected]
Nasal airflow is important for the many physiological functions of the nose, which include the warming and humidifying of inspired air; the filtration of airborne pollutants; and the sense of smell and nasal pungency. Until recently, airflow properties in the nose could only be understood using qualitative in vitro models of humans or in vivo studies in rodents. Recent advances in constructing three-dimensional geometric models of human nasal passages from CT scans, coupled with computational fluid dynamic modeling, has been a valuable tool for quantifying airflow and transport of gases, heat, particles, and aerosols in the human nose. Additionally, these techniques hold significant promise for evaluating and predicting the impact and successful remediation of a variety of clinical conditions on olfac-tion and nasal patency and setting guidelines for safe levels of exposure to inhaled materials.
IntroductionAs the structure that provides access of ambient air to the respiratory tract, the nose serves several important physiological functions [1]: 1) it filters, warms, and humidifies inspired air; 2) it conserves water by retain-ing the moisture in expired air; and 3) it is the initial site for interaction with the chemical senses, where airborne chemicals contact olfactory receptors and/or trigeminal nerve endings. The anatomical design of the nose also reflects its functional needs. Inside the nose of terrestrial mammals lies an intricate internal skeleton of scrolls and plates of bone, collectively known as turbinates. Covered with epithelium and mucus, turbinates provide a large surface for trapping airborne particles and chemicals, for heat and gas exchange, and for the location of olfactory and trigeminal receptors. The turbinates unavoidably also diverge the inspiratory and expiratory nasal airflow
into different parallel channels; the resulting airflow can exhibit dramatic intra- and inter-individual differences due to congenital anatomical features, inflammation aris-ing from acute or chronic conditions (eg, rhinosinusitis or allergic rhinitis), or the presence of polyps. Significantly, even small deviations in the path of airflow may lead to large functional changes in the ability to smell or sense chemical irritation (pungency).
Although clinicians have employed standard mea-surements of airflow (ie, rhinomanometry, acoustic rhinometry) for many years, the results of such measures are often poorly correlated with patients’ subjective symp-toms or post-treatment improvements. The goal of this article is to review recent developments in the realm of nasal airflow modeling for humans and animals and their implications for 1) predicting the degree to which inflam-matory conditions or anatomical features of the airways will affect local airflow patterns and thereby impair olfac-tory function; 2) optimizing treatment plans (surgical and nonsurgical) to improve local airflow to areas that subserve olfaction and perceived nasal patency; 3) evalu-ating the deposition, dosimetry, and toxicity of airborne pollutants in the nose for setting guidelines for safe levels of exposure to inhaled materials; and 4) optimizing the characteristics (aerosol size, flow speed) of nasal drug delivery systems targeting specific nasal regions.
The History of Nasal Airflow InvestigationAlthough the small size and structural complexity of the nasal cavity has prevented detailed in vivo experimental measurements of nasal airflow, a number of in vitro stud-ies have been reported using physical models cast from noses of cadavers or from computed tomography (CT) images for humans [2–5], and for monkeys and rats [6]. However, measurements of airflow properties in these models were generally crude or descriptive, accomplished by visualizing smoke in airflow [2]; or by using miniature Pitot tubes [3], laser Doppler velocimetry [4], or radio-active tracers [5]. In an attempt to increase the spatial resolution and the quantitative accuracy in measurement, enlarged models of the nasal cavity based on coronal magnetic resonance imaging (MRI) were constructed [7,8] where velocities for inspiratory and expiratory flows
118 Rhinitis
under laminar and turbulent conditions were measured using a hot-film anemometer. Particle image velocimetry has also shown the capability to quantify the whole nasal airflow field from a series of parallel snapshots [9].
With the advancement of computational fluid dynam-ics (CFD) techniques, however, it has become possible to simulate the nasal airflow within numerically constructed, anatomically accurate models. Keyhani et al. [10] were among the first to examine nasal airflow in the human nose using a numerical model constructed from CT scans, followed by others [11,12], while a similar technique was also applied for rat [13], monkey [14], and bullfrog [15]. With validation by experimental measurements [8,16], CFD has rapidly become the state-of-the-art methodology for analysis of nasal airflow, largely due to its ability to provide valuable quantitative airflow information at any location within the geometry. However, until recently, the methods employed to construct the CFD models were labor intensive and time consuming, and thus limited the number of individuals or species who could be studied.
A recent major advance in CFD modeling occurred when Zhao el al. [17••] (Fig. 1) extended the CFD tech-nique, permitting the creation of an individual numerical model in several days rather than the months previously required. This advance enabled rapid investigations of nasal airflow and mass transport for an individual and holds significant promise for relating disturbances in nasal airflow as a result of anatomical deviations or clinical conditions, to a generalized or selective nasal symptom.
Another advantage of the CFD technique is that, based on the airflow, it can model the transport of gaseous chemicals, aerosols, heat, and water vapor in the nasal cavity. For example, Keyhani et al. [18] first studied the odorant absorption in the mucosal layer by incorporating a quasi-steady one-dimensional diffusion model into the
CFD analysis, and found that the solubility of the odor-ant in water or mucus, its air phase diffusivity, and nasal airflow rate significantly affected the amount and fraction of odorant deposited in the mucosa. This information is important given that, unlike other sensory systems, olfac-tory receptors among most terrestrial mammals do not have direct access to ambient odor sources. The sense of smell only occurs when airborne chemicals are inhaled within the air stream through the nostrils and dissolve into the mucus layer covering the neuroepithelium. This model was later validated by a comparison with experi-ment results, which in turn improved the accuracy of the key parameters used [19].
Human Nasal Airflow: Implications for Olfaction The human nasal cavity consists of three turbinates (inferior, middle, superior), which along with the septum separate the airway into parallel channels, the meatus. In general, the highest nasal airflow occurs along the nasal floor of the lower meatus, while a second peak occurs in the middle meatus close to the septum. Only a small por-tion of the total inspiratory nasal airflow, ranging from 5% to 15% in various reports, flows through the superior meatus where the olfactory epithelium is located. The sizeable variation in fractional airflow rates and airflow patterns reported in various studies could well be due to differences in the anatomical geometry of the nasal cav-ity being studied. To evaluate this experimentally, using a three-dimensional anatomically accurate nasal cavity model based on a normal subject’s CT scan, Zhao et al. [17••] varied the nasal anatomy in two critical regions (the nasal valve and the olfactory cleft) and showed that although the total nasal airflow through the nostril and
Figure 1. (A) Air-mucosa interface of an anatomically accurate numerical nasal model constructed from nasal CT scans (B, lower right) of a normal adult woman. The model is filled with 1.7 million tetrahydral elements suitable for numerical airflow simulation. (B) Coronal section of the same model. In a close-up view (B, top right), layers of small and fine elements can be seen along the wall that can capture the rapid near wall changes of air velocity and chemical concentration. (Adapted from Zhao et al. [17••].)
Implications of Modeling Nasal Airflow Zhao and Dalton 119
nasal resistance did not change significantly, the amount of airflow though the olfactory region could be reduced by more than 700%. This reduction of airflow to the olfactory region was further shown to have even greater impact on the transport of airborne chemicals (odorant) to the olfactory mucosa, especially for chemicals with high mucosal solubility and diffusivity.
More importantly, this study revealed that the overall nasal airflow pattern was highly sensitive to local nasal geometric configurations: Figure 2A shows a smooth streamline pattern for inspiratory steady laminar flow in the right nasal cavity of a normal subject. Figure 2B shows the streamline pattern in the left side of the same subject; eddies can be seen in the anterior part of the nose induced by a small airway constriction in the nasal valve region, along with a secondary eddy in the superior and posterior part of the nasal cavity. If the constriction was artificially removed, the streamline pattern became smooth again, as in the right side. However, if the airway was further constricted in the nasal valve region (Fig. 2C), the anterior eddy was suppressed, while the secondary eddy was enhanced. This finding can resolve previously unexplained reported airflow pattern discrepancies based on different cast models. These different airflow patterns, independent of total nasal resistance or total nasal airflow rate, are likely to have enormous implications for the sen-sory function of the nose: pungency, patency, and in this context, olfaction.
Studies of CT scans showing regional changes in nasal volume among healthy and hyposmic populations have provided additional support for the notion that volume changes at these critical regions will have a greater impact on olfactory function than in other regions [20,21]. Clini-cally, olfactory loss is found in approximately 25% of chronic rhinosinusitis (CRS) patients without nasal pol-yps, but up to 80% of patients with nasal polyps [22]. Nasal polyps can differentially impair orthonasal versus retronasal olfactory acuity [23], which implies a conduc-tive mechanism that was later confirmed by experimental obstruction of the anterior portion of the olfactory cleft [24]. In light of our results using CFD to model nasal airflow, we hypothesized that 1) inflammation-induced constriction of the airway or nasal polyps will alter or impede airflow through the nasal cavity and thereby reduce access of volatile chemicals to the olfactory epi-thelium; and 2) this mechanism is likely to be one of the primary causes of olfactory loss in rhinosinusitis patients. However, these clinical airway constrictions are function-ally different from the naturally occurring spontaneous congestion and decongestion during the nasal cycle among healthy subjects that can increase nasal resistance up to four-fold, comparable to changes seen in clinical condi-tions, which yet have little effect on olfactory acuity [25].
Most nasal airflow study in the past was based on quasi–steady-state laminar assumptions, which simulated resting breathing conditions. However, during an epoch
of odor-sampling, humans typically employ dynamic sniffing behavior that involves short, high airflow rate (> 300 mL/s through each nostril) bouts of inhalation without exhalation in between. Hahn et al. [8] experimen-tally determined that nasal airflow is usually turbulent at these high flow rates. This has often been characterized as a mechanism to overcome obstruction and enable higher amounts of odorant to deposit onto neuroepithelium and thereby aid in olfaction. However, Zhao et al. [26] found essentially no difference in predicted odorant flux (grams per square centimeter per second, g/cm2·s) onto the neuroepithelium under turbulent versus laminar flow for total nasal flow rates between 300 and 1000 mL/s, for odorants of quite different mucosal solubility and with or without slight nasal valve constriction. This lack of differ-ence was shown to be due to the low turbulent intensity and the higher resistance of odorant lateral diffusive transport from air to mucosal nasal airway wall than of the convective transport in the air phase. Nevertheless, the contribution of motor regulation in human sniffing behavior to olfaction [27] should not be neglected, as the simulation also revealed that the increase in airflow rate during sniffing can increase odorant uptake flux to the nasal/olfactory mucosa, but can lower the cumulative total uptake in the olfactory region when the inspired air/odorant volume was held fixed, which is consistent with the observation that sniff duration may be equally impor-tant as sniff strength for optimizing odor detection [28].
Airflow Modeling: Implications for Clinical OutcomesThe results of airflow modeling previously discussed present a challenge to the clinical community. The current standard rhinometric measurements (eg, rhinomanometry, rhinoresistometry, acoustic rhinom-etry, rhinospirometry) directly index global airflow changes (total nasal airflow rate or resistance, nasal cross-sectional area, etc.) but cannot reveal changes in the patterns of local airflow. It is thus not surprising that these measures of global airflow have been shown to correlate poorly with patients’ subjective symptoms of olfactory function [29], and therefore may be of limited value for evaluating the functional impact of various forms of nasal obstruction and the outcome of any surgical or medical interventions, particularly on olfactory function. In contrast, airflow modeling, which allows computation of local airflow changes and odorant mucosa uptake flux as a function of anatomi-cal modifications, may have more success in predicting clinical outcomes.
Using CFD, Zhao et al. [30•] have shown some prelim-inary success in correlating numerical airflow and odorant transport simulation results with olfactory recovery of one patient with CRS polyposis. Following functional endo-scopic sinus surgery (FESS), the patient’s performance
120 Rhinitis
Current Allergy Reports AL07-2-1-07 fig. 2 324 pts. W/ 636 pts. D (27 x 53) Author: Dalton Editor: Virginia Artist: WL
Right nostril
Right
Nasal valveregion
Left
Nasopharynx
Nasopharynx
Olfactory region
Olfactory region
Airflow direction
A
Left nostril
Airflow direction
Open upblockednasalvalve
Narrowedairway
Originalmodel
B
Left nostril
Airflow direction
C
Figure 2. Nasal airflow patterns can be highly sensitive to local airway geometry. (A) Smooth airflow streamline pattern for inspiratory steady laminar flow in the smooth right nasal cavity. (B) In the left nasal cavity of the same subject, airflow eddies in the anterior part are induced by a small airway constriction in the nasal valve region (CT scan), along with a secondary eddy in the superior and posterior parts. If the constriction was artificially removed, the streamline pattern became smooth again (not shown) as in the right side. (C) However, if the airway was further constricted in the nasal valve region, the anterior eddy was suppressed, while the secondary eddy was enhanced. (Adapted from Zhao et al. [17••].)
Implications of Modeling Nasal Airflow Zhao and Dalton 121
on standard olfactory function tests returned to normal levels from the pre-surgical status of hyposmic. Airflow simulation based on this patient’s pre-surgical and post-surgical CT scans revealed a dramatic post-surgical increase (> 1000 times, back to normal level) of odorant flux to olfactory region for this patient, which was mainly due to the redistribution of airflow/odorant transport patterns after remodeling of airway in the osteomeatal complex. In contrast to the increases in local airflow, the alterations in global airflow due to decreased nasal resistance were mini-mal (two to three times). This finding also suggested that improved local airflow transport to the olfactory region can sometimes be the result of surgery away from the site, which again is an indication that nasal airflow cannot be fully indexed through volumetric measurements. Further studies are underway to determine the proportion of variance in olfactory acuity among normal and clinical populations that can be accounted for by the calculated olfactory transport ability, based on models constructed for individual patients (from CT scans), along the time course of their treatment and at time points reflecting the nasal cycle dynamics.
Airflow Modeling: Implications for Nasal PatencyThe subjective sensation of nasal airflow (nasal patency) is another important aspect of the normal function of the nose. Subjective perception of nasal obstruction is one of the hall-mark symptoms that drives patients to seek treatment, while relief of the perceived obstruction is crucial to a patient’s evaluation of a successful therapeutic outcome [31,32]. However, there is controversy over whether the total nasal airflow measured by techniques such as rhinomanometry are well correlated with subjective nasal patency [33,34], and whether such measurements have any clinical value [35]. One reason for the disassociation might be that perceived nasal patency is only partly due to alterations in the volume of airflow through the nose, while a major contribution arises from the activation of temperature-sensitive trigeminal free nerve endings located at the mucosa wall, which respond to heat loss when cool air is inspired through the nose. Stud-ies have shown that modulating the cold sensor response by topical application of menthol or local anesthesia results in an increased or decreased sensation of nasal patency without significant changes in total nasal resistance or airflow [25]. Parallel to the odorant transport problem just discussed, quantifying nasal heat transfer through the index of total nasal airflow or resistance may not be adequate; rather, the local airflow and its interaction with local nasal geometry play a more significant role.
Radical surgical interventions (eg, total turbinectomy) to improve nasal patency can frequently result in a wide open nasal airway in which no obstructions to airflow remain. However, due to the removal of significant amounts of mucosal surfaces containing trigeminal nerve endings, the airflow produces little sensation of air movement, and
the result is often an equally or more pronounced obstruc-tive sensation (“empty nose syndrome”) [36]. Extrapolating from the results of our nasal airflow modeling and the unsuccessful outcome of such surgeries, we suggest that restoring nasal patency requires not only removing any obstructions that are judged to impair successful airflow to critical regions, but also preserving 1) nasal mucosa where the trigeminal sensors are located, and 2) the local anatomy (the turbinate structure) where interaction with incoming airflow creates optimal heat transfer efficiency.
Experimental measurements of the gradient of nasal temperature and moisture level during respiration present a challenge due to its small size and complex anatomy. Limited data have shown that the human nose is remarkably efficient in warming and humidify-ing incoming air to the ideal condition for alveoli gas exchange, even at extreme conditions (0°C and 0% relative humidity) [37,38] and the majority of the pro-cess is completed at the turbinate region [39]. Nasal mucosal temperature fluctuates during the respiratory cycle roughly from 30°C to 36°C depending on mea-suring site, environmental conditions, and total nasal airflow rate. CFD technique has shown greater promise to provide details of nasal heat transfer [40•]; however, available models currently assume an unlimited supply of heat and water from the mucosal wall, which awaits further experimental validation. Nasal cycle and clini-cal condition, such as inflammation or congestion, may alter the local submucosal blood supply. We also need to be cautious that heat and water vapor transport can-not be de-coupled in CFD simulation, as the latent heat of evaporator accounts for a major component of the total heat flux, which presented a problem for other studies [41]. Nevertheless, CFD heat transfer analysis based on individual nasal anatomy might in the future provide a better understanding for nasal patency and for evaluating treatment outcome.
Airflow Modeling: Implications for Sinus SurgeryFESS is a common surgical approach to remove osteitic and obstructive material, and is often claimed to improve nasal airflow and mucociliary blanket dynamics. Yet, there are few well-designed studies actually evaluating the functional impact of the various surgical options on nasal airflow. From the advent of such interventions, heated con-troversy has existed among ear, nose, and throat surgeons advocating different procedures (eg, middle turbinate resection versus preservation, or radical inferior turbinate resection versus conservative submucosal debridement). In practice, surgeons often made choices based on previous training and their personal beliefs about the importance of maintaining nasal airflow in the absence of empirical evidence. We believe that the CFD technique can be a valuable tool for evaluating local effects of various sur-
122 Rhinitis
gical procedures on nasal airflow and resistance: inferior turbinectomy [42], septal perforation [43], radical sinus surgery [44], and middle turbinate resection (Zhao et al., Unpublished data). These preliminary investigations were mostly based on experimental modifications to standard normal nasal models. Future CFD studies based on an individual patient’s pre-surgical and/or post-surgical CT scans would enable comparison with an individual’s treat-ment outcome, including their subjective patency, and ultimately lead to a better outcome evaluation. Perhaps even more importantly, we envision that the effects of sur-gical perturbations in the anatomy on nasal airflow can be simulated—“virtual surgery”—which can thus be used to predict surgical outcomes and aid in pre-surgical plan-ning to optimize nasal airflow. Because the CFD model is constructed based on CT or MRI, any pre-surgical plan-ning on the model can easily be superimposed back onto the CT or MRI images and incorporated into the existing image guidance system.
Airflow Modeling: Implications for Evaluating Nasal ToxicityThe human nasal passages serve as an effective filter against inhaled pollutants in gas, aerosol, or particle form to protect the pulmonary airway. Nasal CFD airflow simulations coupled with gas or aerosol transport models have been widely used to evaluate filtration efficacy and assess the risk of inhaled toxicant to nasal mucosa tissue [45,46]. Compared with experimental techniques, nasal CFD models can provide more direct details on the depo-sition distribution and dosimetry of the inhaled pollutants on the mucosa in human nasal cavity. However, to date, the regional dosimetry results from the model simulations have only been correlated with experimental toxicol-ogy data (lesions, tumors) in animals (rat [47], monkey [14]), rather than in humans. Given the important and significant differences in both the anatomy and biochem-istry between the animal and human nasal passages, it is important to consider the degree to which rodent model-
Current Allergy Reports AL07-2-1-07 fig. 3a240 pts. W/ 336 pts. D (20 x 28)
Author: Dalton Editor: Virginia Artist: WL
C
M
P
D
IV
P
CCP
B-LNasal epithelium B-R
MOB
D
V
SO
III
II
I
IIv
IId
DL
LDM
M
V
VL
WM
L
Coronal
MOB
Sagittal
A
Current Allergy Reports AL07-2-1-07 fig. 3b240 pts. W/ 336 pts. D (20 x 28)
Author: Dalton Editor: Virginia Artist: WL
Reference vector 1,0*10-3 g/cm2s
L-carvone Octane
B
Figure 3. (A) Sagittal (top) and coronal (lower) drawing of the hamster nasal epithelium and main olfactory bulb (MOB). Olfactory receptor neurons (ORN) expressing the same receptor gene (dots) are distributed along the inspiratory path within two strictly delimited regions: central (C) and peripheral (P) zones. ORNs in the central domain express both class I and class II genes, whereas ORNs in the peripheral domain express only class II genes. Class I olfactory receptors (ORs) are believed specialized toward detecting hydrophilic odorants. (B) CFD-simulated inspiratory uptake flux for high mucosa-soluble odorant L-carvone and low-soluble octane, in rat nose (coronal section), the patterns of which roughly match the central and peripheral zones in A, implying a structure-functional optimization between airflow and odorant detection. (Adapted from Schoenfeld and Cleland [51] and Zhao et al. [26].)
Implications of Modeling Nasal Airflow Zhao and Dalton 123
ing studies can inform our understanding of the func-tional impact of the pollutants on the human nose.
Airflow Modeling: Implications for Drug DeliveryNasal administration as a means of delivering therapeu-tic agents has garnered increasing attention for several reasons [48]. First, the large absorptive surface area of the highly vascularized mucosa is in close proximity to the nostrils, thereby ensuring adequate dosing for many locally acting drugs (eg, those for allergy, sinusitis) or systemic delivery of insulin, vaccine, and others. Second, many pharmacologic agents targeting the central nervous system may utilize the olfactory nerve tract and gain direct access to the brain, thus bypassing the blood-brain barrier. Of great importance to designing these drug delivery systems is the ability to predict the patterns of aerosol flow and deposition. In this regard, CFD aerosol models have great potential for assisting nasal drug deliv-ery designs, to optimize higher absorption rate at target region (eg, the turbinate region, sinuses, or olfactory cleft) by varying aerosol properties, delivery system, and air-flow rates [49•].
Airflow Models in Animals: Implications for HumansAlthough this review focuses mostly on human studies, data from laboratory animal experiments, especially from rats, are far more abundant and are often used to extrapolate information about humans, on whom some of the studies are impossible to conduct. But as noted previously, there are substantial differences between rodent and human in their nasal anatomy and subse-quently their nasal airflow.
In rodents, CFD models have shown that airflow through the nose is highly structured and laminar due to their more complicated turbinate structure; consequently, inhaled odorants are spatially distributed onto the olfac-tory mucosa depending on their different physiochemical properties, such as solubility and diffusivity [15,26], in a process analogous to gas chromatography (“imposed pattern” [50]). On the other hand, populations of olfac-tory receptor neurons expressing the same receptor gene are distributed in strictly delimited regions along the inspiratory path [51], the pattern of which roughly matches the odorant deposition pattern (Fig. 3), implying a structure-functional optimization between airflow and odorant detection. Supported by electrophysiological findings [52,53], it is becoming clear that the olfactory system in rodents may utilize both imposed and inherent (neuron distribution) patterning to optimize odor per-ception: the turbinate structure distributes the incoming odorant based on physiochemical properties and aero-dynamics to different mucosa regions, where types of
receptors that are more sensitive to or better discrimina-tive of that odorant might be located. Motor regulation of the sniffing behavior of rodents, which involves bouts of both inhalation and exhalation at high frequency (as high as 8 Hz), may contribute to the optimization of odorant distribution by varying sniffing intensity and frequency and utilizing the difference between inspira-tory and expiratory airflow patterns.
From an evolutionary perspective, the human olfactory system appears to have been compromised when com-pared to many other land mammals: the human turbinate structure is much simpler and the olfactory epithelium is confined to a relatively small patch with fewer functional receptor genes, which may account for our poorer olfac-tory acuity compared to other species. Moreover, these features are likely the reason why our olfactory abilities, relative to other species, are far more prone to disrup-tion by even small nasal anatomical changes (eg, disease, developmental, surgery) which restrict odorant access to the limited patch of olfactory epithelium. Nevertheless, it should not deter our effort to save our sense of smell from clinical interruptions, which can often offer us priceless enjoyment from food, flavor, or natural and man-made fragrances and, most essentially, provide early detection of environmental hazards. The use of airflow modeling can be an important tool for interspecies comparison, where animal models and data can help to understand and quantify transport problems in the nose.
ConclusionsNasal airflow is important for the physiological func-tions of the nose, which consist of perception in the chemical senses; air conditioning; and filtration. The use of airflow modeling can be an important tool for interspecies comparison, where animal models and data can help to understand and quantify transport problems in the nose. Airflow modeling in humans can be directly used to determine the impact and successful remediation of clinical conditions on the olfactory sys-tem; to set guidelines for safe levels of human exposure to inhaled materials; and to design palatable food, fla-vor, or drug delivery systems that can target desirable nasal regions.
References and Recommended ReadingPapers of particular interest, published recently, have been highlighted as:• Of importance•• Of major importance
1. Proctor DF: The upper airway. In The Nose. Edited by Proctor DF, Andersen IB. New York: Elsevier Biomedical Press; 1982:23–43.
2. Proetz AW: Air currents in the upper respiratory tract and their clinical importance. Ann Octol Rhinol Laryngol 1951, 60:439–467.
124 Rhinitis
3. Swift DL, Proctor DF: Access of air to the respiratory tract. In Respiratory Defense Mechanism. Edited by Brain JD, Proctor DF, Reid LM. New York: Marcel Dekker; 1977:63–91.
4. Girardin M, Bilgen E, Arbour P: Experimental study of velocity fields in a human nasal fossa by laser anemometry. Ann Otol Rhinol Laryngol 1983, 92:231–236.
5. Hornung DE, Leopold DA, Youngentob SL, et al.: Airflow patterns in a human nasal model. Arch Otolaryngol Head Neck Surg 1987, 113:169–172.
6. Morgan KT, Kimbell JS, Monticello TM, et al.: Studies of inspiratory airflow patterns in the nasal passages of the F344 rat and rhesus monkey using nasal molds: relevance to formaldehyde toxicity. Toxicol Appl Pharmacol 1991, 110:223–240.
7. Schreck S, Sullivan KJ, Ho CM, Chang HK: Correlations between flow resistance and geometry in a model of the human nose. J Appl Physiol 1993, 75:1767–1775.
8. Hahn I, Scherer PW, Mozell MM: Velocity profiles measured for airflow through a large-scale model of the human nasal cavity. J Appl Physiol 1993, 75:2273–2287.
9. Kelly JT, Prasad AK, Wexler AS: Detailed flow patterns in the nasal cavity. J Appl Physiol 2000, 89:323–337.
10. Keyhani K, Scherer PW, Mozell MM: Numerical simulation of airflow in the human nasal cavity. J Biomech Eng 1995, 117:429–441.
11. Subramaniam RP, Richardson RB, Morgan KT, Kimbell JS: Computational fluid dynamics simulations of inspiratory airflow in the human nose and nasopharynx. Inhal Toxicol 1999, 10:91–120.
12. Martonen TB, Quan L, Zhang Z, Musante CJ: Flow simu-lation in the human upper respiratory tract. Cell Biochem Biophys 2002, 37:27–36.
13. Kimbell JS, Gross EA, Joyner DR, et al.: Application of computational fluid dynamics to regional dosimetry of inhaled chemicals in the upper respiratory tract of the rat. Toxicol Appl Pharmacol 1993, 121:253–263.
14. Kepler GM, Richardson RB, Morgan KT, Kimbell JS: Com-puter simulation of inspiratory nasal airflow and inhaled gas uptake in a rhesus monkey. Toxicol Appl Pharmacol 1998, 150:1–11.
15. Yang CC: Numerical modeling of odorant uptake in the rat and bullfrog nasal cavities [PhD dissertation]. Philadelphia: University of Pennsylvania; 1999.
16. Hahn I, Scherer PW, Mozell MM: A mass transport model of olfaction. J Theor Biol 1994, 167:115–128.
17.•• Zhao K, Scherer PW, Hajiloo SA, Dalton P: Effect of anatomy on human nasal airflow and odorant transport patterns: implications for olfaction. Chem Senses 2004, 29:365–379.
This study exemplifies the significant impact of even small anatomi-cal perturbations on human nasal airflow and odorant transport, and the implications for olfactory function. The rapid and advanced CFD technique employed in this study is an important advance for time-sensitive investigations of nasal airflow in individual patients.18. Keyhani K, Scherer PW, Mozell MM: A numerical model of
nasal odorant transport for the analysis of human olfaction. J Theor Biol 1997, 186:279–301.
19. Kurtz DB, Zhao K, Hornung DE, Scherer PW: Experimental and numerical determination of odorant solubility in nasal and olfactory mucosa. Chem Senses 2004, 29:763–773.
20. Leopold DA: The relationship between nasal anatomy and human olfaction. Laryngoscope 1988, 98:1232–1238.
21. Damm M, Vent J, Schmidt M, et al.: Intranasal volume and olfactory function. Chem Senses 2002, 27:831–839.
22. Wolfensberger M, Hummel T: Anti-inflammatory and surgical therapy of olfactory disorders related to sino-nasal disease. Chem Senses 2002, 27:617–622.
23. Landis BN, Giger R, Ricchetti A, et al.: Retronasal olfactory function in nasal polyposis. Laryngoscope 2003, 113:1993–1997.
24. Pfaar O, Landis BN, Frasnelli J, et al.: Mechanical obstruc-tion of the olfactory cleft reveals differences between orthonasal and retronasal olfactory functions. Chem Senses 2006, 31:27–31.
25. Eccles R, Jawad MSM, Morris S: Olfactory and trigeminal thresholds and nasal resistance to airflow. Acta Oto-Laryn-gol 1989, 108:268–273.
26. Zhao K, Dalton P, Yang GC, Scherer PW: Numerical model-ing of turbulent and laminar airflow and odorant transport during sniffing in the human and rat nose. Chem Senses 2006, 31:107–118.
27. Mainland J, Sobel N: The sniff is part of the olfactory percept. Chem Senses 2006, 31:181–196.
28. Sobel N, Khan RM, Hartley CA, et al.: Sniffing longer rather than stronger to maintain olfactory detection threshold. Chem Senses 2000, 25:1–8.
29. Doty RL, Mishra A: Olfaction and its alteration by nasal obstruction, rhinitis, and rhinosinusitis. Laryngoscope 2001, 111:409–423.
30.• Zhao K, Pribitkin EA, Cowart BJ, et al.: Numerical modeling of nasal obstruction and endoscopic surgical intervention: outcome to airflow and olfaction. Am J Rhinol 2006, 20:308–316.
This study is the first to apply CFD technique to a model con-structed from CT scans from an individual patient. The technique demonstrates an important innovation for relating the changes in nasal airflow as sequelae to pathology and subsequent treatment to a specific symptom outcome.31. Damm M, Quante G, Jungehuelsing M, Stennert E: Impact
of functional endoscopic sinus surgery on symptoms and quality of life in chronic rhinosinusitis. Laryngoscope 2002, 112:310–315.
32. Giger R, Landis BN, Zheng C, et al.: Objective and subjec-tive evaluation of endoscopic nasal surgery outcomes. Am J Rhinol 2003, 17:327–333.
33. Jones AS, Willatt DJ, Durham LM: Nasal airflow: resistance and sensation. J Laryngol Otol 1989, 103:909–911.
34. Clarke JD, Hopkins ML, Eccles R: Evidence for correla-tion of objective and subjective measures of nasal airflow in patients with common cold. Clin Otolaryngol 2005, 30:35–38.
35. Yaniv E, Hadar T, Shvero J, Raveh E: Objective and sub-jective nasal airflow. Am J Otolaryngol 1997, 18:29–32.
36. Moore EJ, Kern EB: Atrophic rhinitis: a review of 242 cases. Am J Rhinol 2001, 15:355–361.
37. Wolf M, Naftali S, Schroter RC, Elad D: Air-conditioning characteristics of the human nose. J Laryngol Otol 2004, 118:87–92.
38. Rouadi P, Baroody FM, Abbott D, et al.: A technique to measure the ability of the human nose to warm and humidify air. J Appl Physiol 1999, 87:400–406.
39. Keck T, Leiacker R, Heinrich A, et al.: Humidity and temperature profile in the nasal cavity. Rhinology 2000, 38:167–171.
40.• Naftali S, Rosenfeld M, Wolf M, Elad D: The air-condi-tioning capacity of the human nose. Ann Biomed Eng 2005, 33:545–553.
This is one of the most detailed CFD studies on heat transfer in the human nose.41. Lindemann J, Keck T, Wiesmiller K, et al.: A numerical
simulation of intranasal air temperature during inspiration. Laryngoscope 2004, 114:1037–1041.
43. Grant O, Bailie N, Watterson J, et al.: Numerical model of a nasal septal perforation. Medinfo 2004, 11:1352–1356.
44. Lindemann J, Brambs HJ, Keck T, et al.: Numerical simu-lation of intranasal airflow after radical sinus surgery. Am J Otolaryngol 2005, 26:175–180.
Implications of Modeling Nasal Airflow Zhao and Dalton 125
45. Martonen TB, Schroeter JD: Risk assessment dosimetry model for inhaled particulate matter: I. Human subjects. Toxicol Lett 2003, 138:119–132.
46. Kimbell JS, Subramaniam RP: Use of computational fluid dynamics models for dosimetry of inhaled gases in the nasal passages. Inhal Toxicol 2001, 13:325–334.
47. Kimbell JS, Gross EA, Richardson RB, et al.: Correlation of regional formaldehyde flux predictions with the distribution of formaldehyde-induced squamous metaplasia in F344 rat nasal passages. Mutat Res 1997, 380:143–154.
48. Illum L: Nasal drug delivery: possibilities, problems and solutions. J Control Release 2003, 87:187–198.
49.• Schroeter JD, Kimbell JS, Asgharian B: Analysis of particle deposition in the turbinate and olfactory regions using a human nasal computational fluid dynamics model. J Aerosol Med 2006, 19:301–313.
A well-designed CFD study on the impact of aerosol sizes and airflow speeds on localized inhaled particle deposition in human nasal passage, with special interest in nasal drug delivery.
50. Mozell MM, Jagodowicz M: Chromatographic separa-tion of odorants by the nose: retention times measured across in vivo olfactory mucosa. Science 1973, 181:1247–1249.
51. Schoenfeld TA, Cleland TA: Anatomical contributions to odorant sampling and representation in rodents: zoning in on sniffing behavior. Chem Senses 2006, 31:131–144.
52. Scott JW, Acevedo HP, Sherrill L: Effects of concentration and sniff flow rate on the rat electroolfactogram. Chem Senses 2006, 31:581–593.
53. Scott JW: Sniffing and spatiotemporal coding in olfaction. Chem Senses 2006, 31:119–130.
