Noninvasive estimation of pharyngeal airway resistance and compliance in children based on volume-gated dynamic MRI and computational fluid dynamics

Computational fluid dynamics (CFD) analysis was used to model the effect of collapsing airway geometry on internal pressure and velocity in the pharyngeal airway of three sedated children with obstructive sleep apnea syndrome (OSAS) and three control subjects. Model geometry was reconstructed from volume-gated magnetic resonance images during normal tidal breathing at 10 increments of tidal volume through the respiratory cycle. Each geometry was meshed with an unstructured grid and solved using a low-Reynolds number k-ω turbulence model driven by flow data averaged over 12 consecutive breathing cycles. Combining gated imaging with CFD modeling created a dynamic three-dimensional view of airway anatomy and mechanics, including the evolution of airway collapse and flow resistance and estimates of the local effective compliance. The upper airways of subjects with OSAS were generally much more compliant during tidal breathing. Compliance curves (pressure vs. cross-section area), derived for different locations along the airway, quantified local differences along the pharynx and between OSAS subjects. In one subject, the distal oropharynx was more compliant than the nasopharynx (1.028 vs. 0.450 mm(2)/Pa) and had a lower theoretical limiting flow rate, confirming the distal oropharynx as the flow-limiting segment of the airway in this subject. Another subject had a more compliant nasopharynx (0.053 mm(2)/Pa) during inspiration and apparent stiffening of the distal oropharynx (C = 0.0058 mm(2)/Pa), and the theoretical limiting flow rate indicated the nasopharynx as the flow-limiting segment. This new method may help to differentiate anatomical and functional factors in airway collapse.     

Perinatal development of laryngeal function

The resistance of the upper airway is strongly influenced by the action of opposing sets of laryngeal muscles. Expiratory airflow may be retarded by active adduction of the arytenoid cartilages or by a reduction in the activity of abductor muscles. In developing sheep the adductor muscles appear to represent the principal means by which lung recoil is opposed. This mechanism, which is most pronounced during non-rapid-eye-movement sleep, is regulated by afferent traffic from the lungs. In fetal sheep the laryngeal muscles are also influenced by breathing movements and sleep states. The adductor muscles are normally tonically active during non-rapid-eye-movement sleep when rhythmical breathing movements are absent. It is possible that this activity is at least partially responsible for elevated tracheal pressures and depressed flow of tracheal fluid during fetal apnea. This hypothesis has been tested by observing the effects of fetal paralysis and recurrent laryngeal nerve section. These experiments suggest that in the fetus near term the larynx makes a major contribution to upper airway resistance and hence to the maintenance of pulmonary expansion which has been shown to influence lung development.     

Glottic and cervical tracheal narrowing in patients with obstructive sleep apnea

There are several studies showing that patients with idiopathic obstructive sleep apnea (OSA) have a narrow and collapsible pharynx that may predispose them to repeated upper airway occlusions during sleep. We hypothesized that this structural abnormality may also extend to the glottic and tracheal region. Consequently, we measured pharyngeal (Aph), glottic (Agl), cervical tracheal (Atr1), midtracheal (Atr2), and distal (Atr3) tracheal areas during tidal breathing in 66 patients with OSA (16 nonobese and 50 obese) and 8 nonapneic controls. We found that Aph, Agl, and Atr1, but not Atr2 or Atr3, were significantly smaller in the OSA group than in the control group. Obese patients with OSA had the smallest upper airway area, although the nonapneic controls had the largest areas. Multiple linear regression analysis revealed that the pharyngeal area, cervical tracheal area, and body mass index were all independent determinants of the apnea-hypopnea index, accounting for 31% of the variability in apnea-hypopnea index. Aph, Agl, and Atr showed significant correlation with the body mass index. We conclude that sleep-disordered breathing is associated with diffuse upper airway narrowing and that obesity contributes to this narrowing. Furthermore, we speculate that a common pathophysiological mechanism may be responsible for this reduction in upper airway area extending from the pharynx to the proximal trachea.     

Simulation of upper airway occlusion without and with mandibular advancement in obstructive sleep apnea using fluid-structure interaction

Obstructive Sleep Apnea (OSA) is a common sleep disorder characterized by repetitive collapse of the upper airway (UA). One treatment option is a mandibular advancement splint (MAS) which protrudes the lower jaw, stabilizing the airway. However not all patients respond to MAS therapy and individual effects are not well understood. Simulations of airway behavior may represent a non-invasive means to understand OSA and individual treatment responses. Our aims were (1) to analyze UA occlusion and flow dynamics in OSA using the fluid structure interaction (FSI) method, and (2) to observe changes with MAS. Magnetic resonance imaging (MRI) scans were obtained at baseline and with MAS in a known treatment responder. Computational models of the patients' UA geometry were reconstructed for both conditions. The FSI model demonstrated full collapse of the UA (maximum 5.83 mm) pre-treatment (without MAS). The UA collapse was located at the oropharynx with low oropharyngeal pressure (−51.18 Pa to −39.08 Pa) induced by velopharyngeal jet flow (maximum 10.0 m/s). By comparison, simulation results from the UA with MAS, showed smaller deformation (maximum 2.03 mm), matching the known clinical response. Our FSI modeling method was validated by physical experiment on a 1:1 flexible UA model fabricated using 3D steriolithography. This is the first study of airflow dynamics in a deformable UA structure and inspiratory flow. These results expand on previous UA models using computational fluid dynamics (CFD), and lay a platform for application of computational models to study biomechanical properties of the UA in the pathogenesis and treatment of OSA.     

Computational fluid dynamics modeling of the upper airway of children with obstructive sleep apnea syndrome in steady flow

Computational fluid dynamic (CFD) analysis was used to model the effect of airway geometry on internal pressure in the upper airway of three children with obstructive sleep apnea syndrome (OSAS), and three controls. Model geometry was reconstructed from magnetic resonance images obtained during quiet tidal breathing, meshed with an unstructured grid, and solved at normative peak resting flow. The unsteady Reynolds-averaged Navier-Stokes equations were solved with steady flow boundary conditions in inspiration and expiration, using a two-equation low-Reynolds number turbulence model. Model results were validated using an in-vitro scale model, unsteady flow simulation, and reported nasal resistance measurements in children. Pharynx pressure drop strongly correlated to airway area restriction. Inspiratory pressure drop was primarily proportional to the square of flow, consistent with pressure losses due to convective acceleration caused by area restriction. On inspiration, in OSAS pressure drop occurred primarily between the choanae and the region where the adenoids overlap the tonsils (overlap region) due to airway narrowing, rather than in the nasal passages; in controls the majority of pressure drop was in the nasal passages. On expiration, in OSAS the majority of pressure drop occurred between the oropharynx (posterior to the tongue) and overlap region, and local minimum pressure in the overlap region was near atmospheric due to pressure recovery in the anterior nasopharynx. The results suggest that pharyngeal airway shape in children with OSAS significantly affects internal pressure distribution compared to nasal resistance. The model may also help explain regional dynamic airway narrowing during expiration.     

Contribution of upper airway geometry to convective mixing

We investigated the axial dispersive effect of the upper airway structure (comprising mouth cavity, oropharynx, and trachea) on a traversing aerosol bolus. This was done by means of aerosol bolus experiments on a hollow cast of a realistic upper airway model (UAM) and three-dimensional computational fluid dynamics (CFD) simulations in the same UAM geometry. The experiments showed that 50-ml boluses injected into the UAM dispersed to boluses with a half-width ranging from 80 to 90 ml at the UAM exit, across both flow rates (250, 500 ml/s) and both flow directions (inspiration, expiration). These experimental results imply that the net half-width induced by the UAM typically was 69 ml. Comparison of experimental bolus traces with a one-dimensional Gaussian-derived analytical solution resulted in an axial dispersion coefficient of 200-250 cm(2)/s, depending on whether the bolus peak and its half-width or the bolus tail needed to be fully accounted for. CFD simulations agreed well with experimental results for inspiratory boluses and were compatible with an axial dispersion of 200 cm(2)/s. However, for expiratory boluses the CFD simulations showed a very tight bolus peak followed by an elongated tail, in sharp contrast to the expiratory bolus experiments. This indicates that CFD methods that are widely used to predict the fate of aerosols in the human upper airway, where flow is transitional, need to be critically assessed, possibly via aerosol bolus simulations. We conclude that, with all its geometric complexity, the upper airway introduces a relatively mild dispersion on a traversing aerosol bolus for normal breathing flow rates in inspiratory and expiratory flow directions.     

Large eddy simulation of the pharyngeal airflow associated with obstructive sleep apnea syndrome at pre and post-surgical treatment

Obstructive Sleep Apnea Syndrome (OSAS) is the most common sleep-disordered breathing medical condition and a potentially life-threatening affliction. Not all the surgical or non-surgical OSAS therapies are successful for each patient, also in part because the primary factors involved in the etiology of this disorder are not completely understood. Thus, there is a need for improving both diagnostic and treatment modalities associated with OSAS. A verified and validated (in terms of mean velocity and pressure fields) Large Eddy Simulation approach is used to characterize the abnormal pharyngeal airflow associated with severe OSAS and its interaction with the airway wall in a subject who underwent surgical treatment. The analysis of the unsteady flow at pre- and post-treatment is used to illustrate the airflow dynamics in the airway associated with OSAS and to reveal as well, the changes in the flow variables after the treatment. At pre-treatment, large airflow velocity and wall shear stress values were found at the obstruction site in all cases. Downstream of obstruction, flow separation generated flow recirculation regions and enhanced the turbulence production in the jet-like shear layers. The interaction between the generated vortical structures and the pharyngeal airway wall induced large fluctuations in the pressure forces acting on the pharyngeal wall. After the surgery, the flow field instabilities vanished and both airway resistance and wall shear stress values were significantly reduced.     

Haemodynamic analysis of vessel remodelling in STA-MCA bypass for Moyamoya disease and its impact on bypass patency

The purpose of this study is to estimate the remodelling characteristics of STA-MCA bypass and its influence on patency via the use of computational fluid dynamic (CFD) technology. The reconstructed three-dimensional geometries from MRA were segmented to create computational domains for CFD simulations. Eleven patients, who underwent regular MRA both immediately following surgery and at the six months follow-up, were studied. The flow velocities at STA were measured via the use of quantitative MRA (QMRA) to validate simulation results. STA-MCA bypass patency was confirmed for each patient immediately following surgery. The simulation indicated that the remodelling of the arterial pedicle in nine patients was associated with a reduction in the resistance to flow through the bypass. For these cases, the modelling of a driving pressure of 10 mmHg through the bypass at 6 months post-surgery resulted in a 50% greater blood flow than those found immediately following surgery. However, two patients were found to exhibit contradictory patterns of remodelling, in which a highly curved bending at the bypass immediately post-surgery underwent progression, with increased resistance to flow through the bypass at 6 months follow-up, thereby resulting in a modelled flow rate reduction of 50% and 25%, respectively. This study revealed that STA-MCA bypass has a characteristic remodelling that usually reduces flow resistance. The initial morphology of the bypass may have had a significant effect on the outcome of vessel remodelling.     

Airflow behavior changes in upper airway caused by different head and neck positions: Comparison by computational fluid dynamics

The feasibility of computational fluid dynamics (CFD) to evaluate airflow characteristics in different head and neck positions has not been established. This study compared the changes in volume and airflow behavior of the upper airway by CFD simulation to predict the influence of anatomical and physiological airway changes due to different head–neck positions on mechanical ventilation. One awake volunteer with no risk of difficult airway underwent computed tomography in neutral position, extension position (both head and neck extended), and sniffing position (head extended and neck flexed). Three-dimensional airway models of the upper airway were reconstructed. The total volume (V) and narrowest area (A_min) of the airway models were measured. CFD simulation with an Spalart–Allmaras model was performed to characterize airflow behavior in neutral, extension, and sniffing positions of closed-mouth and open-mouth ventilation. The comparison result for V was neutral <extension≈sniffing, and for A_min was neutral<extension<sniffing. A_min in sniffing position was nearly 3.0 times that in neutral position and 1.7 times that in extension position. The pressure drop and velocity increasing were more obvious in neutral than sniffing or extension position at the same airflow rate. In sniffing position, pressure differences decreased and velocity remained almost constant. Recirculation airflow was generated near the subglottic region in neutral and extension positions. Sniffing position improves airway patency by increasing airway volume and decreasing airway resistance, suggesting that sniffing position may be the optimal choice for mask ventilation.     

Hydrodynamic and Longitudinal Impedance Analysis of Cerebrospinal Fluid Dynamics at the Craniovertebral Junction in Type I Chiari Malformation

Elevated or reduced velocity of cerebrospinal fluid (CSF) at the craniovertebral junction (CVJ) has been associated with type I Chiari malformation (CMI). Thus, quantification of hydrodynamic parameters that describe the CSF dynamics could help assess disease severity and surgical outcome. In this study, we describe the methodology to quantify CSF hydrodynamic parameters near the CVJ and upper cervical spine utilizing subject-specific computational fluid dynamics (CFD) simulations based on in vivo MRI measurements of flow and geometry. Hydrodynamic parameters were computed for a healthy subject and two CMI patients both pre- and post-decompression surgery to determine the differences between cases. For the first time, we present the methods to quantify longitudinal impedance (LI) to CSF motion, a subject-specific hydrodynamic parameter that may have value to help quantify the CSF flow blockage severity in CMI. In addition, the following hydrodynamic parameters were quantified for each case: maximum velocity in systole and diastole, Reynolds and Womersley number, and peak pressure drop during the CSF cardiac flow cycle. The following geometric parameters were quantified: cross-sectional area and hydraulic diameter of the spinal subarachnoid space (SAS). The mean values of the geometric parameters increased post-surgically for the CMI models, but remained smaller than the healthy volunteer. All hydrodynamic parameters, except pressure drop, decreased post-surgically for the CMI patients, but remained greater than in the healthy case. Peak pressure drop alterations were mixed. To our knowledge this study represents the first subject-specific CFD simulation of CMI decompression surgery and quantification of LI in the CSF space. Further study in a larger patient and control group is needed to determine if the presented geometric and/or hydrodynamic parameters are helpful for surgical planning.     

Patient-Specific Modeling of Regional Antibiotic Concentration Levels in Airways of Patients with Cystic Fibrosis: Are We Dosing High Enough?

Background

Pseudomonas aeruginosa (Pa) infection is an important contributor to the progression of cystic fibrosis (CF) lung disease. The cornerstone treatment for Pa infection is the use of inhaled antibiotics. However, there is substantial lung disease heterogeneity within and between patients that likely impacts deposition patterns of inhaled antibiotics. Therefore, this may result in airways below the minimal inhibitory concentration of the inhaled agent. Very little is known about antibiotic concentrations in small airways, in particular the effect of structural lung abnormalities. We therefore aimed to develop a patient-specific airway model to predict concentrations of inhaled antibiotics and to study the impact of structural lung changes and breathing profile on local concentrations in airways of patients with CF.

Methods

In- and expiratory CT-scans of children with CF (5–17 years) were scored (CF-CT score), segmented and reconstructed into 3D airway models. Computational fluid dynamic (CFD) simulations were performed on 40 airway models to predict local Aztreonam lysine for inhalation (AZLI) concentrations. Patient-specific lobar flow distribution and nebulization of 75 mg AZLI through a digital Pari eFlow model with mass median aerodynamic diameter range were used at the inlet of the airway model. AZLI concentrations for central and small airways were computed for different breathing patterns and airway surface liquid thicknesses.

Results

In most simulated conditions, concentrations in both central and small airways were well above the minimal inhibitory concentration. However, small airways in more diseased lobes were likely to receive suboptimal AZLI. Structural lung disease and increased tidal volumes, respiratory rates and larger particle sizes greatly reduced small airway concentrations.

Conclusions

CFD modeling showed that concentrations of inhaled antibiotic delivered to the small airways are highly patient specific and vary throughout the bronchial tree. These results suggest that anti-Pa treatment of especially the small airways can be improved.     

The positional characteristics of patients with obstructive sleep apnea: a single institute retrospective study in Japan

Sleep and Biological Rhythms - Obstructive sleep apnea (OSA) causes sleep-disordered breathing (SDB) due to upper airway obstruction. The severity of OSA changes with position during sleep....     

Correlation between severity of sleep apnea and upper airway morphology based on advanced anatomical and functional imaging

Determination of the apnea hypopnea index (AHI) as a measure of the severity of obstructive sleep apnea/hypopnea syndrome (OSAHS) is a widely accepted methodology. However, the outcome of such a determination depends on the method used, is time consuming and insufficient for prediction of the effect of all treatment modalities. For these reasons more methods for evaluating the severity of OSAHS, based on different imaging modalities, have been looked into and recent studies have shown that anatomical properties determined from three-dimensional (3D) computed tomography (CT) images are good markers for the severity of the OSAHS. Therefore, we correlated anatomical measurements of a 3D reconstruction of the upper airway together with flow simulation results with the severity of OSAHS in order to find a combination of variables to indicate the severity of OSAHS in patients. The AHI of 20 non-selected, consecutive patients has been determined during a polysomnography. All patients also underwent a CT scan from which a 3D model of the upper airway geometry was reconstructed. This 3D model was used to evaluate the anatomical properties of the upper airway in OSAHS patients as well as to perform computational fluid dynamics (CFD) computations to evaluate the airflow and resistance of this upper airway. It has been shown that a combination of the smallest cross-sectional area and the resistance together with the body mass index (BMI) form a set of markers that predict very well the severity of OSAHS in patients within this study. We believe that these markers can be used to evaluate the outcome of an OSAHS treatment.     

Laryngeal influences on breathing pattern and posterior cricoarytenoid muscle activity

Receptors responding to transmural pressure, airflow, and contraction of laryngeal muscles have been previously identified in the larynx. To assess the relative contribution of these three types of receptors to the reflex changes in breathing pattern and upper airway patency, we studied diaphragmatic (DIA) and posterior cricoarytenoid muscle (PCA) activity in anesthetized dogs during spontaneous breathing and occluded efforts with and without bypassing the larynx. Inspiratory duration (TI) was longer, mean inspiratory slope (peak DIA/TI) was lower, and PCA activity was greater with upper airway occlusion than with tracheal occlusion (larynx bypassed). Bilateral section of the superior laryngeal nerves eliminated these differences. When respiratory airflow was diverted from the tracheostomy to the upper airway the only change attributable to laryngeal afferents was an increase in PCA activity. These results confirm the importance of the superior laryngeal nerves in the regulation of breathing pattern and upper airway patency and suggest a prevalent role for laryngeal negative pressure receptors.     

Mandible position and activation of submental and masseter muscles during sleep.

Movement of the mandible could influence pharyngeal airway caliber because the mandible is attached to the tongue and to muscles that insert on the hyoid bone. In normal subjects and patients with obstructive sleep apnea (OSA) we measured jaw position during sleep with strain gauges, as well as masseter and submental electromyograms, airflow, esophageal pressure, oximetry, electroencephalograms, and electrooculograms. Jaws of patients with OSA were open more than those of normal subjects at end expiration and opened further at end inspiration, particularly at the termination of apneas when the masseter and submental muscles contracted. Masseter activation occurred only in patients with OSA and in a pattern similar to that of submental muscles. Jaw opening at end expiration could narrow the upper airway, whereas opening at end inspiration could reflect efforts to expand the airway with tracheal tug and with submental muscle activation and efforts to open the mouth to allow mouth breathing. Masseter contraction does not close the jaw but may serve to stabilize it.     

Neural control of airway pressures in rabbits

In twenty anaesthetized and spontaneously breathing rabbits airway pressures were measured above and below the larynx during tidal respiration through the larynx. Peak inspiratory and expiratory pressures at both sites were recorded in control conditions and then compared to values obtained in the course of progressive denervation of the airways. The two methods of denervation consisted of (1) bilateral section of superior and recurrent laryngeal nerves and of the midcervical vagotomy (horizontal method); (2) right-sided sections of the three nerves followed by left-sided sections (vertical method). Motor denervation of the larynx due to RLNs neurotomy (horizontal method) produced significant increases in intratracheal pressures in both phases of the respiratory cycle. Less prominent increments in pressures were achieved on RLNs neurotomy in the vertical method. SLNs section and vagotomy had little additional effect on airway pressures. Our results indicate that unilateral laryngeal palsy poses far smaller obstruction to breathing than simultaneous bilateral denervation, and that afferent denervation of the larynx has no effect on airway pressures.     

Induction of airway collapse with subatmospheric pressure in awake patients with sleep apnea

To determine whether the pharyngeal airway is abnormal in awake patients with obstructive sleep apnea (OSA), we measured the ability of the pharyngeal airway to resist collapse from subatmospheric pressure applied to the nose in awake subjects, 12 with OSA and 12 controls. Subatmospheric pressure was applied to subjects placed in the supine position through a tightly fitting face mask. We measured airflow at the mask as well as mask, pharyngeal, and esophageal pressures. Ten patients developed airway obstruction when subatmospheric pressures between 17 and 40 cmH2O were applied. Obstruction did not occur in two patients with the least OSA. Obstruction did not occur in 10 controls; one obese control subject developed partial airway obstruction when -52 cmH2O was applied as did another with -41 cmH2O. We conclude that patients with significant OSA have an abnormal airway while they are awake and that application of subatmospheric pressure may be a useful screening test to detect OSA.     

Novel Approach to Simulate Sleep Apnea Patients for Evaluating Positive Pressure Therapy Devices

Bench testing is a useful method to characterize the response of different automatic positive airway pressure (APAP) devices under well-controlled conditions. However, previous models did not consider the diversity of obstructive sleep apnea (OSA) patients’ characteristics and phenotypes. The objective of this proof-of-concept study was to design a new bench test for realistically simulating an OSA patient’s night, and to implement a one-night example of a typical female phenotype for comparing responses to several currently-available APAP devices. We developed a novel approach aimed at replicating a typical night of sleep which includes different disturbed breathing events, disease severities, sleep/wake phases, body postures and respiratory artefacts. The simulated female OSA patient example that we implemented included periods of wake, light sleep and deep sleep with positional changes and was connected to ten different APAP devices. Flow and pressure readings were recorded; each device was tested twice. The new approach for simulating female OSA patients effectively combined a wide variety of disturbed breathing patterns to mimic the response of a predefined patient type. There were marked differences in response between devices; only three were able to overcome flow limitation to normalize breathing, and only five devices were associated with a residual apnea-hypopnea index of <5/h. In conclusion, bench tests can be designed to simulate specific patient characteristics, and typical stages of sleep, body position, and wake. Each APAP device behaved differently when exposed to this controlled model of a female OSA patient, and should lead to further understanding of OSA treatment.     

Recent advances in obstructive sleep apnea pathophysiology and treatment

Obstructive sleep apnea (OSA) is a sleep disorder characterized by recurring collapse of the pharyngeal airway leading to restricted airflow. OSA is becoming increasingly common with at least moderate disease now evident in 17% of middle aged men and 9% of women. The list of recognized adverse health consequences associated with OSA is growing and includes daytime symptoms of sleepiness, impaired cognition and risk of motor vehicle accidents as well as associations with hypertension, cardiovascular morbidity, malignancy and all-cause mortality. In this context adequate treatment of OSA is imperative; however, there are well-recognized pitfalls in the uptake and usage of the standard treatment modality, Continuous Positive Airway Pressure (CPAP). A broad range of pathophysiological mechanisms are now recognized beyond an anatomically smaller pharyngeal airway and impaired compensatory pharyngeal muscle responsiveness. Perturbations in ventilatory control stability, low arousal threshold, sleep-related decrease in lung volume and fluid redistribution as well as upper airway surface tension have all been shown to variously contribute to sleep-disordered breathing. Many new therapies are emerging from these advances in understanding of the mechanisms of OSA. Although many may not be universally effective, the promise of phenotyping patients according to their individual pathophysiology in order to target one or more therapies may prove highly effective and allow the treatment of OSA towards a personalized medicine approach.