Flow limitation (or choked flow) occurs during expiration when the flow rate reaches a maximum value, and further increases in driving pressure do not result in an increased flow rate. The flow of liquid or gas in airways can be described in terms of the physical properties of the airway and the fluids (gas or liquid), and airway, alveolar, and pleural pressures. Flow limitation has been studied extensively in gas ventilation1–25 and to a lesser extent in liquid ventilation.26–30 In classic flow-limitation papers, Dawson and Elliot, and Shapiro1,2,19 demonstrated that the maximum flow of a fluid through an elastic tube is limited at the speed of propagation of pressure pulse waves along the tube, which depends on local cross-sectional area and local stiffness of the tube, which are properties of the specific airway,11,31–36 the density of the fluid, and the velocity profiles.1,2,19 The airway and fluid properties, along with flow rate, determine in which airway generation flow limitation occurs.
Total liquid ventilation (TLV), which is currently an experimental ventilation strategy for the potential treatment of acute lung injury and acute respiratory distress syndrome, involves totally filling the lungs with a perfluorocarbon (PFC) liquid and ventilating with a liquid tidal volume.37–42
There are 43,000 to107,000 new cases of acute respiratory distress syndrome occurring in children and adults in the United States each year.43–45 Despite improvement in intensive care management, the mortality in the setting of severe respiratory failure in the nonneonatal population remains between 30% and 50%.46–48 TLV has been demonstrated to provide adequate gas exchange when used in animal experiments49 and is a potential means for providing effective pulmonary management in the setting of respiratory failure. Maximal expiratory flow rates during TLV are typically 20 to 100 times lower than those in gas ventilation.29 Because of the higher density of PFC compared with air, the wave speeds are much lower in TLV. To avoid flow limitation, expiratory flow rates are usually set at low, constant levels in animal models of TLV. The occurrence of flow limitation during expiration in TLV limits respiratory rate (3 to 9 breaths/min), minute ventilation, and carbon dioxide clearance.29 Previous work has examined the location at which flow limitation occurs during gas ventilation15,20–24 and has shown the location moves distally as lung volume decreases. To our knowledge, the location of flow limitation during total liquid ventilation has not been investigated. PFC liquids are denser and more viscous than gas. However, their kinematic viscosity is much less than that of air, and therefore the Reynolds number range in TLV can be quite different than in gas ventilation. The Reynolds number indicates the relative importance of inertial and viscous effects and is defined as Re = ρ·D·U/μ, where ρ is fluid density, D is airway diameter, U is the characteristic fluid velocity, and μ is the fluid viscosity. During gas ventilation, Re in the central airways of humans can exceed 10,000.12 Typical values of Re for TLV in rabbits are < 2,500. Although the mechanisms of flow limitation in TLV are expected to be similar to gas ventilation, the different fluid properties may affect where flow limitation occurs. This study investigated the location of flow limitation in PFC-filled rabbit lungs, as a model of TLV, and examined the effects of lung volume and expiratory flow rate. Once the location of flow limitation in TLV is known and its dependence on lung volume and expiratory flow rate are understood, strategies for reducing its occurrence can be better implemented.
Materials and Methods
Pressures were measured along the airways via movable pressure catheters to determine the location of flow limitation. In a second, smaller set of experiments, the location of flow limitation was determined by radiographic imaging to provide validation of the pressure catheter measurements. The animal preparation was similar in both, and only the measured parameters were different. We first describe the procedure for the experiments in which pressure catheter measurements were made and then describe the methods for the imaging studies. These protocols were approved by the University Committee on Use and Care of Animals at the University of Michigan.
Pressure Catheter Studies
New Zealand White rabbits weighing 2.8 to 3.2 kg were weighed and anesthetized with 5 mg/kg intramuscular xylazine (Lloyd Laboratories, Shenandoah, IA) and 20 mg/kg ketamine (Fort Dodge Animal Health, Fort Dodge, IA). An ear intravenous (IV) catheter was placed after adequate sedation was confirmed, and a tracheostomy was performed. A 3/16-inch diameter, thin-walled steel tracheostomy tube of length 3 cm was advanced into the trachea for a length of 1.5 to 2.0 cm, and the trachea was doubly secured with silk suture to prevent PFC leakage or dislodgement. The tracheostomy tube was connected to 1/4-inch Tygon (Fisher Scientific, Pittsburgh, PA) tubing, which was then connected to a small animal ventilator (Harvard Apparatus, Holliston, MA). Heparin (Elkins-Sinn, Cherry Hill, NJ) (1000 units/kg) was injected into the ear IV and the rabbit was mechanically ventilated using pure oxygen, which is readily absorbed by PFC, for 5 minutes. This was done to minimize gas trapping within the lungs when they were subsequently filled with PFC, resulting in lungs that were completely liquid-filled. The rabbit was euthanized with 0.5 ml/kg Beuthanasia (Schering-Plough Animal Health, Union, NJ), connected to the experimental circuit, and placed on a load cell table (Omega LCEA-5, Omega Engineering Inc., Stamford, CT) to measure changes in animal weight during the PFC instillation and removal process that followed. Previous data from our laboratory have documented that minimal effects upon parameters relative to flow limitation experiments are noted within 40 to 60 minutes of euthanasia in rabbits.26,30
The rabbit’s lungs were filled with the perfluorocarbon FC77 (3M Specialty Materials, St. Paul, MN) to the prescribed end inspiration lung volume (EILV). The filling was by a slow, direct injection of PFC into the Tygon tubing, using gravity to distribute the PFC into the lungs. PFC was instilled into the Tygon tubing at the end of the endotracheal tube until PFC was visible in Tygon tubing. The chest of the rabbit was massaged during filling in an attempt to remove any trapped gas bubbles. The Tygon tubing was connected to the circuit, which was initially primed. The circuit consisted of a stopcock and tubing, connected to a computer-controlled piston pump, and was similar to the circuit we have previously used in other studies.26–28 Additional PFC was instilled into the lungs through a stopcock in the circuit. The circuit was clamped between the stopcock and the rest of the circuit, so that the additional instillation of PFC entered the lungs of the rabbit. Once the lungs were filled to the desired EILV, the instillation was stopped. The volume of instilled PFC was carefully measured, and the change in animal weight during the filling process provided confirmation that the instilled volume was accurately measured.
After verifying all connections, the clamp on the tubing connecting the endotracheal tube with the piston pump was removed, and airway pressure was measured at the tip of the endotracheal tube via a 3.7 Fr polyethylene pressure catheter in the tube lumen. Before beginning measurements, the animal was ventilated (with liquid via the piston pump) for 3 to 4 breaths at 1.25 ml/s to remove gas bubbles from the lungs. Following this debubbling process, the piston position was verified at origin position and trials of exhalation were begun. The pump flow was stopped via an electronic control circuit when the airway pressure, measured at the exit to the endotracheal tube, dropped below –80 mm Hg. Previous investigations26–28 indicated that flow limitation would occur by the time –80 mm Hg was reached, and that was the lower limit of pressure that could be measured by the pressure transducer. The computer controlling the pump piston was programmed by setting the distance for the piston to travel and its velocity. The servo motor (SmartMotor NEMA 23, Animatics Corporation, Santa Clara, CA) responsible for moving the piston was programmed with specific settings that were previously determined to result in constant flow rates of 2.5, 5.0, 7.5 ml/s, as described previously.26–28 A 3.7 Fr polyethylene intratracheal pressure-monitoring catheter was inserted through the endotracheal tube for measurement of airway pressure at locations distal to the tracheostomy tube tip in 0.5 cm increments (from 0 to 6 cm past the tracheostomy tube tip). This was in addition to the pressure catheter placed at the tip of the tracheostomy tube for measurement of pressure there. Pressure monitoring was performed via a Transpac II pressure transducer (Abbott Laboratories, North Chicago, IL) and a Hewlett Packard infant monitor (Hewlett Packard Corp., Andover, MA).
For each animal, the EILV (20, 30, or 40 ml/kg) and expiratory flow rate, Q, (2.5, 5.0, 7.5 ml/s) were both chosen randomly. The position of the movable pressure catheter was selected randomly for each trial. The airway pressure was recorded with 0.01 Hz time resolution at both catheters, and pressure measurement was synchronized to begin when expiration was started. Once an airway pressure of –80 mm Hg was reached and the pump stopped, the lungs were refilled via moving the piston in the pump to its original location. The movable pressure catheter was then repositioned and the expiration process repeated until pressure was measured at all of the 0.5 cm increment positions. These experiments were performed in 54 animals (6 animals at each of the 9 EILV/Q combinations). Each trial for a particular location of the movable pressure catheter lasted from 10 to 40 seconds, with a minimal delay between trials, thus all measurements were completed within 30 minutes of euthanasia of the animal. A necropsy was performed on each animal to assess the presence of PFC in the pleural spaces (perfluorothorax), and to measure the position of the tracheostomy tube relative to the carina so that the position of the movable catheter could then be expressed relative to the carina. If a perfluorothorax was identified, the animal was not included in data analysis.
These studies used the same procedure described above, except that the location of flow limitation was examined using radiographic imaging, rather than the movable pressure catheter. Because the purpose of the imaging study was to validate the location of flow limitation determined in the pressure catheter measurements, fewer animals and parameter combinations were used. Images were acquired from animals with EILV = 30 ml/kg and expiratory flow rate, Q, of 2.5, 5.0, or 7.5 ml/s, and for Q = 5.0 ml/s and EILV = 20, 30, and 40 ml/kg. Five animals, one per EILV/Q combination, were used in the imaging study. Before filling the lungs of the animal with FC77, tantalum dust (300 mesh) was insufflated into the lungs from a dusting device that provided 10 mg tantalum per puff during the inspiratory phase of ventilation. Five to ten puffs were delivered to the animal until sufficient contrast was provided to allow the central airways to be visualized. The animal’s lungs were then filled with PFC to EILV. Following the debubbling process described above, the PFC was actively drained at the prescribed rate. X-ray images were acquired at 30 frames/second before and during PFC removal by a high-resolution fluoroscopy unit (Philips portable c-arm, Philips Medical Systems, Amsterdam, The Netherlands), which is housed at The University of Michigan Medical Center and recorded to an S-VHS video format. Airway diameters were measured from these digital images, and the temporal and spatial evolution of airway size was compared with the pressure data.
The pressure as a function of time and the minimum pressure for each location were outcome measures in the pressure catheter studies. These are indicators of the location of the flow limited airway segment. From the imaging studies, airway diameters were measured at corresponding positions from the images, and the time evolution of airway diameters and final airway diameters were determined. The final time corresponded to end of expiration, which occurred when the tracheostomy tube pressure reached –80 mm Hg and the piston pump was shut off by the regulator. Position, x, was measured axially along the airway, relative to the carina so that positive values are distal to the carina and negative positions are proximal to the carina. Figure 1 shows a lung image with an imposed sketch of the x-coordinate system.
Perfluorothorax was not observed in any of the animals in either the pressure catheter or imaging studies. An example of pressure, P, versus time, t, data is shown in Figure 2 for one particular animal with EILV = 30 ml/kg and expiratory flow rate, Q = 7.5 ml/kg. Each frame, A–L, of the figure indicates the pressure measured at a different position along the airways and the pressure at the tip of the endotracheal tube (dashed line). As shown in the graph, the pressure everywhere decreases gradually at early times. At a later time, corresponding to a sudden change in pressure at the endotracheal tube and the onset of choke, the P versus t curve becomes much steeper in frames A-G of Figure 2, with the most negative pressures at the tip of the endotracheal tube. Note that in Figure 2, time is measured relative to the initiation of expiration.
Figure 3 contains plots of the minimum pressure, Pmin, versus position, x, for (A) EILV = 20 ml/kg and Q = 2.5 ml/s, (B) EILV = 20 ml/kg and Q = 5.0 ml/s, (C) EILV = 20 ml/kg and Q = 7.5 ml/s, (D) EILV = 30 ml/kg and Q = 2.5 ml/s, (E) EILV = 30 ml/kg and Q = 5.0 ml/s, (F) EILV = 30 ml/kg and Q = 7.5 ml/s, (G) EILV = 40 ml/kg and Q = 2.5 ml/s, (H) EILV = 40 ml/kg and Q = 5.0 ml/s, (I) EILV = 40 ml/kg and Q = 7.5 ml/s. Each minimum pressure data point represents an average of the values for all animals at that x, Q, EILV combination. The error bars in the Pmin direction indicate the standard error of the mean of the pressure. The error bars in the x direction indicate the standard error of the mean of position, which resulted from variations in the placement of the pressure catheter between animals due to the location of the carina being unknown a priori and only determined after the conclusion of the experiment on a particular animal.
X-ray images of rabbit lungs during expiration are shown in Figure 4. These images are from a single animal with an EILV of 30 ml/kg and an expiratory flow rate of 7.5 ml/s, and are similar to the images obtained at other EILV and Q settings. Airway collapse is not evident in Figures 4A and 4B, which correspond to times after the initiation of expiration, but before the onset of flow limitation. As expiration continues, the first generation airways collapse near the carina, as shown in Figure 4C, and remain collapsed as expiration continues, as shown in Figure 4D. Dimensional airway diameter is plotted as a function of time in Figure 5 for Q = 7.5 ml/s and EILV = 30 ml/kg, which are the same parameter values used in Figures 2 and 4. Each line in Figure 5 corresponds to a different position, x, along the airway, as indicated on the graph (the position of the carina is x = 0). At time t = 0, the airway diameter decreases with increasing x. As PFC is removed from the lungs, the airway diameters decrease with time in a gradual manner, until flow limitation occurs and the airway collapses. In airways distal to the collapse (e.g., x ≥ 2.5 cm), there was not a sudden decrease in diameter. Note that time in Figure 5 is relative to when the image capturing was initiated, and that Figure 5 presents data from the imaging study, whereas Figure 2 presents data from the pressure catheter study with different animals. Thus, t = 0 does not correspond to the initiation of expiration as in Figure 2, e.g., t = 2.5 s in Figure 5 corresponds approximately to t = 4.5 s in Figure 2.
Figure 6 shows dimensionless airway diameter, Dend/Dinit, versus position, x, at the final time during the expiration, e.g., when airway pressure at the endotracheal tube tip was –80 mm Hg. The dimensionless airway diameter was calculated as the diameter at the time of interest, Dend, divided by the initial (preexpiration) airway diameter, Dinit. Each line corresponds to a different Q and EILV combination, as indicated on the graph. This dimensionless airway diameter indicates the final airway diameter normalized by the initial diameter, and is smaller where the airway is collapsed, e.g. in the flow limited airway segment. Figure 6A shows Dend/Dinit versus x for EILV = 30 ml/kg and Q = 2.5, 5.0, and 7.5 ml/s. Figure 6B shows Dend/Dinit versus x for Q = 5.0 ml/s and EILV = 20, 30, and 40 ml/kg. Note that, because the tracheostomy was performed before imaging the lungs the precise location of the tracheostomy tube tip was not known until the lungs were imaged. For the Q = 5.0 ml/s and EILV = 20 animal, the tracheostomy tube tip was placed too close to the carina to allow diameter measurement at the x = –1.5 cm location, and the corresponding curve in Figure 6B begins at x = –1.0 cm.
Based on the pressure versus time data shown in Figure 2, flow limitation appears to occur in airway generation 1, near the carina. For x ≥ 2.3 cm (shown in Figures 2I–2L), the measured airway pressure slowly decreased with time. For x ≤ 1.8 cm (see Figures 1A–1H), the pressure initially decreased gradually and then suddenly became very negative approximately 4.5 seconds after the initiation of expiration. The minimum pressure, Pmin, was more negative at more proximal locations (more negative value of x). Flow limitation occurs when increases in driving pressure do not increase the flow rate, and, therefore, we consider this change in pressure waveform with position to indicate where flow limitation occurred. A similar criterion for determining the location of flow limitation has been used in gas ventilation studies.22 Because constant expiratory flow was prescribed here, a sharp decrease in airway pressure indicated the onset of flow limitation.
The data in Figure 3 suggest a similar location of the flow limited airway. Distal to the flow limited airway, we expect the airway pressure to be relatively unchanged and to have a relatively high Pmin. At and proximal to the location of flow limitation, we expect Pmin to be very negative, indicating that the driving pressure has to be increased considerably in order to maintain the constant, prescribed flow rate. In comparing the plots of Pmin versus x for each EILV/Q combination, few differences in the suggested location of flow limitation are observed. Although the value of Pmin near the carina, x = 0, varies with EILV and Q, the value of x (2 cm < x < 3 cm) at which Pmin reaches a plateau does not.
This location of flow limitation is confirmed by the lung images shown in Figure 4, which demonstrate that the first-generation bronchi and trachea near the carina collapse. The plot of dimensional airway diameter versus time shown in Figure 5 quantifies this observation. There is little change in airway diameter for x > 2 cm during expiration. However, there is a significant (> 40% at most locations) and sudden change in diameter in airways proximal to x = 2 cm near a time of 2.5 seconds, corresponding to the onset of flow limitation. Similar behavior is observed for all of the EILV and Q combinations considered. Figure 6A shows that the apparent location of flow limitation does not change significantly over the range of Q values considered for EILV = 30 ml/kg. As shown in Figure 6B, the range of EILV considered here did not change the apparent location of flow limitation for Q = 5.0 ml/s. It is evident from Figure 6 that the change (relative to the initial values) in the diameter of the airways distal to x = 2 cm is much smaller than the change in airway diameter for x < 2.0 cm. Proximal to that location, the diameter at the end of the expiration period (the piston pump driving the expiratory flow was stopped when the endotracheal tube pressure reached –80 mm Hg) was significantly reduced, such that airway diameters were < 75% their initial value at each location. For EILV = 30 ml/kg and Q = 7.5 ml/s the change in airway diameter at x = 2.0 and 2.5 is greater than the other EILV/Q combinations, but is still much less than the diameter change for 0.5 ≤ x ≤ 1.5 cm for EILV = 30 ml/kg and Q = 7.5 ml/s.
Although the mechanics of flow limitation are similar in gas ventilation and liquid ventilation, the results of this study indicate a number of differences in the location of flow limited airways in liquid ventilated lungs compared with gas ventilated lungs. Flow limitation typically occurred in the main bronchi and trachea in this study. Previous investigations of flow limitation in gas ventilated human lungs have indicated that it occurs in lobar, segmental, and subsegmental bronchi.20,22 We are not aware of previous work that has used pressure catheters to assess the location of flow limitation during gas ventilation of rabbit lungs, and thus do not directly compare our TLV measurements to corresponding gas ventilation measurements. However, previous investigations have examined the change in respiratory resistance during expiration as an indicator of flow limitation using theory and rabbit experiments.50,51 These studies found that respiratory resistance during flow limitation in rabbits is volume dependent, suggesting an outward progression of collapsed airways. The location of flow limitation in the current study depended minimally on the expiratory flow rate or the EILV. This is in contrast to the finding that its location in gas ventilation may move distally as lung volume is decreased in gas ventilation.4,8,12,22,52 The differences in location of flow limitation between gas and liquid ventilation are likely due to the differences in material properties of the ventilating fluid. In gas ventilation, density effects are most important in determining the onset of flow limitation at high lung volumes, whereas viscosity effects are important at low lung volumes.10,18,25 In TLV in rabbits, which typically has Re < 2500, viscous effects are important in determining the endotracheal tube pressure at which flow limitation occurs for all lung volumes.26–28 Those studies found that the volume of PFC that remains in rabbit lungs when flow limitation occurs depends strongly on EILV and Q,26–28 but the current study demonstrates that EILV and Q have minimal effects on where it occurs. The location of flow limitation in TLV suggests that a strategy of preventing collapse of the trachea and main bronchi, either through bronchodilation or endotracheal tubes designed to stent the airway open, may potentially facilitate high flow rates and therefore higher minute ventilations than currently are possible.
This idealized study focused exclusively on investigating the location of flow limitation in liquid filled lungs and did not attempt to investigate other aspects of TLV that may affect its efficacy. For example, gas exchange,37,38,41,53 ventilator device development,30,49 distribution of PFC within the lungs,54 and other challenges of TLV were not investigated. The animal model presented here was focused on the investigation of flow limitation and did not require us to supply the animal’s metabolic requirements. All measurements were performed within the first 30 minutes after euthanizing the animals. Previous investigations have shown no significant changes in flow limitation behavior or properties of rabbit airways during the first 40 minutes postmortem.26,30 This suggests that the location of choke is similar in these experiments as in live animals. Although the repeated collapse during these experiments may have affected the local postmortem airway compliance, we did not specifically measure airway compliance changes. The repeatability of the experimentally determined location of flow limitation despite the randomization of the location of the pressure catheter for measurement (and thus order and time during the experiment) suggests that these effects may also be small. Future studies could investigate such changes. The presence of the catheter in the airway during the pressure measurements may have resulted in errors in measured pressure and could potentially affect the flow in the choke region. However, the catheters were small, and agreement between the choke behavior in the catheter study and the imaging study (in which the intra-airway catheter was not present) suggests the presence of the catheter had minimal effects. Catheters were previously used in gas ventilation studies of flow limitation and its location, and their presence was not considered to significantly affect the results.22 In fact, the agreement between the pressure catheter measurements and imaging measurements, with regard to location of choke, in the current study suggests that pressure catheter measurements are a valid method of determining the location of flow limitation.
These experiments assessed the location of flow limitation during TLV, but did not consider how to avoid it. Future work is needed before the implementation of a comprehensive strategy to delay the onset of flow limitation. The effect of repeated flow limitation on airways is not known, and the potential for injury may affect how close to the flow limited condition lungs can safely be ventilated. The use of bronchodilators may reduce the likelihood of choke by decreasing the local flow velocity at a given flow rate, while increasing the airway diameter. The findings of the current study could be incorporated into such a strategy by targeting the airways that are most likely to collapse—the trachea and main bronchi. The modulation of expiratory flow based on theoretical models of flow limitation in TLV coupled with feedback mechanisms, based in part on measured tracheal and bronchial pressures, may result in an “intelligent” liquid ventilator that can effectively avoid choke while maximizing minute ventilation.
In conclusion, we have demonstrated that flow limitation in liquid-filled rabbit lungs occurs near the carina, and that its location doesn’t move distally as lung volume is decreased. Despite the involvement of similar flow limitation mechanisms in gas ventilation and liquid ventilation, the location of flow limitation in liquid ventilation does not exhibit the same dependence on lung volume as in gas ventilation and is relatively unaffected by these parameters. These findings suggest that the onset of flow limitation in liquid ventilation may be delayed by specifically targeting the trachea and main bronchi with interventions to delay their collapse.
1. Dawson SV, Elliott EA: Wave-speed limitation on expiratory flow: A unifying concept. J Appl Physiol
43: 498–515, 1977.
2. Dawson SV, Elliott EA: Use of the choke point in the prediction of flow limitation in elastic tubes. Federation Proc
39: 2765–2770, 1980.
3. Elad D, Kamm RD, Shapiro AH: Choking phenomena in a lung-like model. J Biomech Eng
109: 1–9, 1987.
4. Elad D, Kamm RD, Shapiro AH: Mathematical simulation of forced expiration. J Appl Physiol
65: 14–25, 1988.
5. Elad D, Kamm RD, Shapiro AH: Steady compressible flow in collapsible tubes: Application to forced expiration. J Fluid Mech
203: 401–418, 1989.
6. Elliott EA, Dawson SV: Test of wave-speed theory of flow limitation in elastic tubes. J Appl Physiol
43: 516–522, 1977.
7. Kamm RD, Pedley TJ: Flow in collapsible tubes: A brief review. J Biomech. Eng
111: 177–179, 1989.
8. Kimmel E, Kamm RD, Shapiro AH: Numerical-solutions for steady and unsteady-flow in a model of the pulmonary airways. J Biomech Eng
110: 292–299, 1988.
9. Lambert RK: Sensitivity and specificity of the computational model for maximal expiratory flow. J Appl Physiol
57: 958–970, 1984.
10. Lambert RK: Analysis of bronchial mechanics and density dependence of maximal expiratory flow. J Appl Physiol
61: 138–149, 1986.
11. Lambert RK: Bronchial mechanical-properties and maximal expiratory flows. J Appl Physiol
62: 2426–2435, 1987.
12. Lambert RK, Wilson TA, Hyatt RE, Rodarte JR: A computational model for expiratory flow. J Appl Physiol
52: 44–56, 1982.
13. Mead J: Expiratory flow limitation: A physiologist’s point of view. Federation Proc
39: 2771–2775, 1980.
14. Pedersen OF, Castile RG, Drazen JM, Ingram RH: Density dependence of maximum expiratory flow in the dog. J Appl Physiol
53: 397–404, 1982.
15. Pedersen OF, Thiessen B, Lyager S: Airway compliance and flow limitation during forced expiration in dogs. J Appl Physiol
52: 357–369, 1982.
16. Polak AG: A forward model for maximum expiration. Comput Biol Med
28: 613–625, 1998.
17. Polak AG, Lutchen KR: Computational model for forced expiration from asymmetric normal lungs. Ann Biomed Eng
31: 891–907, 2003.
18. Schilder DP, Fry DL, Roberts A: Effect of gas density and viscosity on maximal expiratory flow-volume relationship. J Clin Invest
42: 1705–1713, 1963.
19. Shapiro AH: Steady flow in collapsible tubes. J Biomech. Eng
99: 126–147, 1977.
20. Smaldone GC, Bergofsky EH: Delineation of flow-limiting segment and predicted airway-resistance by movable catheter. J Appl Physiol
40: 943–952, 1976.
21. Smaldone GC, Messina M, Curran M: Influence of flow-limitation and cough on particle deposition in COPD. Am Rev Respir Dis
125: 258, 1982.
22. Smaldone GC, Smith PL: Location of flow-limiting segments via airway catheters near residual volume in humans. J Appl Physiol
59: 502–508, 1985.
23. Solway J: Properties of steady maximal expiratory flow within excised canine central airways. J Appl Physiol
64: 1650, 1988.
24. Solway J, Fredberg JJ, Ingram RH, et al: Interdependent regional lung emptying during forced expiration: A transistor model. J Appl Physiol
62: 2013–2025, 1987.
25. Staats BA, Wilson TA, Laifook SJ, et al: Viscosity and density dependence during maximal flow in man. J Appl Physiol
48: 313–319, 1980.
26. Baba Y, Brant DO, Brah SS, et al: Assessment of the development of choked flow during total liquid ventilation. Crit Care Med
32: 201–208, 2004.
27. Bull JL, Reickert CA, Tredici S, et al: Flow limitation in liquid-filled lungs: effects of liquid properties. J Biomech Eng
127: 630–636, 2005.
28. Foley DS, Brah R, Bull JL, et al: Total liquid ventilation: dynamic airway pressure and the development of expiratory flow limitation. ASAIO J
50: 485–490, 2004.
29. Koen PA, Wolfson MR, Shaffer TH: Fluorocarbon ventilation: maximal expiratory flows and CO2 elimination. Pediatr Res
24: 291–296, 1988.
30. Meinhardt JP, Ashton BA, Annich GM, et al: The dependency of expiratory airway collapse on pump system and flow rate in liquid ventilated rabbits. Eur J Med Res
8: 212–220, 2003.
31. Shen X, Ramchandani R, Dunn B, et al: Effect of transpulmonary pressure on airway diameter and responsiveness of immature and mature rabbits. J Appl Physiol
89: 1584–1590, 2000.
32. West JB (ed): Respiratory Physiology: The Essentials.
Williams & Wilkins, 1990.
33. Costantino ML, Bagnoli P, Dini G, et al: A numerical and experimental study of compliance and collapsibility of preterm lamb tracheae. J Biomech
37: 1837–1847, 2004.
34. Ramchandani R, Shen X, Gunst SJ, Tepper RS: Comparison of elastic properties and contractile responses of isolated airway segments from mature and immature rabbits. J Appl Physiol
95: 265–271, 2003.
35. Ramchandani R, Bates JHT, Shen X, et al: Airway branching morphology of mature and immature rabbit lungs. J Appl Physiol
90: 1584–1592, 2001.
36. Ramchandani R, Shen X, Elmsley CL, et al: Differences in airway structure in immature and mature rabbits. J Appl Physiol
89: 1310–1316, 2000.
37. Hirschl RB, Tooley R, Parent A, et al: Evaluation of gas exchange, pulmonary compliance, and lung injury during total and partial liquid ventilation in the acute respiratory ristress syndrome. Crit Care Med
24: 1001–1008, 1996.
38. Greenspan JS, Wolfson MR, Rubenstein SD, Shaffer TH: Liquid ventilation of human preterm neonates. J Pediatr
117: 106–111, 1990.
39. Hirschl RB: Current experience with liquid ventilation. Paediatr Respir Rev
5: S339–345, 2004.
40. Hirschl RB, Pranikoff T, Gauger P, et al: Liquid ventilation in adults, children, and full-term neonates: preliminary report. Lancet
346: 1201–1202, 1995.
41. Jackson JC, Standaert TA, Truog WE, Hodson WA: Full-tidal liquid ventilation with perfluorocarbon for prevention of lung injury in newborn non-human primates. Artif Cells Blood Substit Immobil Biotechnol
22: 1121–1132, 1994.
42. Wolfson MR, Shaffer TH: Liquid ventilation: an adjunct for respiratory management. Paediatr Anaesth
14: 15–23, 2004.
43. Goss CH, Brower RG, Hudson LD, Rubenfeld GD: Incidence of acute lung injury in the United States. Crit Care Med
31: 1607–1611, 2003.
44. Rubenfeld GD: Epidemiology of acute lung injury. Crit Care Med
31: S276–S284, 2003.
45. Arroliga AC, Ghamra ZW, Trepichio AP, et al: Incidence of ARDS in an adult population of northeast Ohio. Chest
121: 1972–1976, 2002.
46. Milberg JA, Davis DR, Steinberg KP, Hudson LD: Improved survival of patients with acute respiratory distress syndrome (ARDS): 1983-1993. JAMA
273: 306–309, 1995.
47. Brun-Buisson C, Minelli C, Bertolini G, et al: Epidemiology and outcome of acute lung injury in European intensive care units: Results from the ALIVE study. Intensive Care Med
30: 51–61, 2004.
48. Neff MJ, Rubenfeld GD: Clinical epidemiology of acute lung injury. Semin Respir Crit Care Med
22: 237–246, 2001.
49. Tredici S, Komori E, Funakubo A, et al: A prototype of a liquid ventilator using a novel hollow-fiber oxygenator in a rabbit model. Crit Care Med
32: 2104–2109, 2004.
50. Vassiliou M, Saunier C, Duvivier C, et al: Volume dependence of respiratory system resistance during artificial ventilation in rabbits. Intensive Care Med
27: 898–904, 2001.
51. Vassiliou M, Peslin R, Saunier C, Duvivier C: Expiratory flow limitation during mechanical ventilation detected by the forced oscillation method. Eur Respir J
9: 779–786, 1996.
52. West JB (ed): Pulmonary Pathophysiology: The Essentials.
Baltimore, MD, Williams and Wilkins, 1990.
53. Shaffer TH, Wolfson MR, Clark LC Jr: Liquid ventilation. Pediatr Pulm
14: 102–109, 1992.
Copyright © 2005 by the American Society for Artificial Internal Organs
54. Bull JL, Tredici S, Komori E, et al: Distribution dynamics of perfluorocarbon delivery to the lungs: an intact rabbit model. J Appl Physiol
96: 1633–1642, 2004.