http://informahealthcare.com/jmt ISSN: 0309-1902 (print), 1464-522X (electronic) J Med Eng Technol, 2015; 39(1): 1–8 ! 2015 Informa UK Ltd. DOI: 10.3109/03091902.2014.968675
INNOVATION
Effect of mechanical ventilation waveforms on airway wall shear Ramana M. Pidaparti*1 and John Swanson2 College of Engineering, University of Georgia, Athens, GA, USA and 2Department of Mechanical and Nuclear Engineering, Virginia Commonwealth University, Richmond, VA, USA
J Med Eng Technol Downloaded from informahealthcare.com by Gazi Univ. on 01/10/15 For personal use only.
1
Abstract
Keywords
Better understanding of airway wall shear stress/strain rate is very important in order to prevent inflammation in patients undergoing mechanical ventilation due to respiratory problems in intensive-care medicine. The objective of this study was to investigate the role of mechanical ventilation waveforms on airway wall shear/strain rate using computational fluid dynamics analysis. Six different waveforms were considered to investigate the airway wall shear stress (WSS) from fluid dynamics analysis for the airway geometry of two-to-three generations. The simulation results showed that Original with Sine Inhale Waveform (OSIW) produced the highest WSS value and the Near True Sine Waveform produced the lowest WSS value. Also, the Original with Sine Inhale Waveform and the Short Sine Inhale with Long Sine Exhale Waveform (SSILSEW) produced a higher shear strain rate in comparison to the Original Waveform (OW). These results, combined with optimization, suggest that it is possible to develop a set of mechanical ventilation waveform strategies to avoid inflammation in the lung.
Airway wall shear, fluid dynamics analysis, mechanical ventilation, strain rate
1. Introduction Patients with acute lung injury (ALI), including acute respiratory distress syndrome (ARDS), airway and other pulmonary diseases, often require mechanical ventilation [1,2]. In the US alone, the incidence of respiratory failure and acute lung injury is 137–253 per 100,000 in the general population [3,4]. Ventilator Assisted Lung Injury (VALI) can be fatal, especially for ARDS patients, and might contribute to multiple organ dysfunction syndrome (MODS) from volutrauma, atelectrauma or biotrauma mechanisms [5,6]. Many techniques have been suggested to prevent VALI and mainly centre around reductions in tidal volumes to reduce airway pressure. However, decreasing tidal volume can have drawbacks such as causing hypercapnia, decreasing aerated lung volume, increase shunting and worsening oxygenation [7]. Increasing positive end-expiratory pressure (PEEP) to compensate for reductions in tidal volume can occasionally lead to transient oxygen desaturation, hypotension, barotrauma, arrhythmia and bacterial translocation [8]. Thammanomai et al. [9] investigated the combined effects of ventilation mode and PEEP and concluded that higher PEEP improved the mechanics and gas exchange, but also increased the epithelial cell injury in mice. Also, inadequate PEEP can *Corresponding author. Email: [email protected]
History Received 30 April 2014 Revised 4 September 2014 Accepted 19 September 2014
result in epithelial lung injury due to mechanical ventilation induced shear and normal stresses [10,11]. Developing new modes of mechanical ventilation is challenging, in part because of the difficulty in predicting the effects. Recently, computational models representing airflow in patient-specific human lungs under mechanical ventilation have been developed [12]. A greater appreciation now exists for the effects of distending forces on the lung from mechanical ventilation and how these forces are distributed via the lung’s fibrous skeleton [13]. The airway wall shear stress due to mechanical ventilation is likely to play a central role in initiating, maintaining and/or exacerbating new or existing inflammatory responses in the lung and the subsequent development of VALI [9]. Computational studies also suggest that high airway wall shear stresses lead to inflammation [14,15]. Many experimental studies have suggested the potential for using computational methods to investigate the effects of mechanical ventilation on lung injury in hopes of developing means to reduce injury in the future [14–18]. However, the role of mechanical ventilation waveforms, especially with non-sinusoidal inhale/exhale shapes on airway wall shear stress and shear strain rate has not been studied quantitatively. The objective of this study was to investigate the role of mechanical ventilation waveform on airway wall shear/strain rate using computational fluid dynamics analysis. Using
2
R. M. Pidaparti and J. Swanson Input: Airflow Velocity from Mechanical Venlaon
J Med Eng Technol, 2015; 39(1): 1–8
the compressibility effects are very small. Also, the diameters of the lung bifurcations are sufficiently small to safely assume that the density variation of air throughout the lung is minimal. In light of this, incompressible flow field is assumed in the simulation. Additionally, the flow was modelled as laminar based on the Reynolds number calculation, as shown below. d at 37 C 1:138 mkg3 4:54 ms ð0:005mÞ Re ¼ 1:9 105 Pa s Re ¼ 1360
J Med Eng Technol Downloaded from informahealthcare.com by Gazi Univ. on 01/10/15 For personal use only.
Re ¼
Figure 1. An overview of the airway bifurcation geometry (generations 2–3) with input of airflow due to mechanical ventilation and pressure outlets.
airway geometry of two-to-three generations, the results of airway wall shear stress obtained from computational analysis for six different waveforms are compared and discussed.
2. Modelling and simulations Geometric representation of the airway generations 2–3, which is the 2nd and 3rd bifurcation within the lungs, in three-dimensions based on anatomical Weibel adult human tracheobronchial airway descriptions [19] was considered in this study. These airway generations have significantly fewer cartilage plates and rings [20] and, hence, the presence of cartilage plates in the respiratory generations 2–3 was ignored in this study. The geometry of the airway bifurcation model developed in the current study is shown in Figure 1. The computational model involves modelling the airways bifurcation for fluid dynamics analysis. The airways were considered as rigid (no deformation) in the computational fluid dynamics (CFD) model from which the airway pressures and wall shear stresses were obtained. Numerical solution of the airflow through airway bifurcation during mechanical ventilation was investigated using ANSYS CFX software [21]. The ANSYS workbench was used to conduct the FSI simulations using CFX and ANSYS Mechanical. To obtain the input flow waveform for the airway generation 2, airflow at the level of the trachea from mechanical ventilation was assumed to divide equally at each bifurcation. The input velocity profile is assumed to be uniform and air was modelled as an incompressible fluid in this study. Using the maximum velocity of the air in lung bifurcations, the Mach number is estimated to be much less than 0.2, in which case
where Re is the Reynolds number, is the air density at 37 C, v is the maximum velocity of the air, d is the inlet diameter of the model, and m is the viscosity of air at 37 C. Since the Reynolds number is less than the transition region of Re ¼ 2300 for a circular pipe, laminar flow can be assumed. A zero constant-distribution pressure was applied at the outlets of the fluid domain. This boundary condition was used in our earlier study [18] for fluid dynamics analysis. However, for fluid-solid analysis, we imposed a zero-displacement boundary condition at the outlets. The pressure boundary condition applied at the outlets of the model in relation to the inlet will greatly affect the resulting results of wall shear stresses which are related to pressure gradients between inlet and outlets. Also, at different bifurcations the pressure boundary condition will be different. It is noteworthy to point out that the sensitivity of model results to pressure boundary conditions needs to be investigated for a realistic application. This is one of the limitations of the present study. A mesh-convergence study was performed to confirm that a fine enough element had been used to represent the fluid domain for analysis. Changes in maximum pressure and velocity were used as convergence criteria. A converged model was obtained when changes in those criteria were less than 5%. The converged finite element model consisted of 107,520 hexahedra elements. Six different waveforms of mechanical ventilation, as shown in Figure 2, were chosen for this study. The first waveform was the original waveform (OW), which is the waveform currently used during mechanical ventilation. All other waveforms were constructed by numerically integrating the area underneath the original waveform and adjusting the waveforms so that the area underneath the inhale and underneath the exhale remained constant for each waveform. The second waveform was modified so that the inhale followed a sinusoidal rise and fall, yet the exhale was the same as the original waveform and was designated as the Original with Sine Inhale Waveform (OSIW). The third waveform was modified so that the exhale followed a sinusoidal fall and rise, yet the inhale followed the original waveform and was designated as the Original with Sine Exhale Waveform (OSEW). The fourth waveform, Near True Sine Waveform (NTSW), attempted to follow a true sine wave as closely as possible by having half the time period for inhalation and half for exhalation but, because of the slight difference in areas of inhalation and exhalation, is not quite a true sine wave. The fifth waveform used the sinusoidal inhale
Mechanical ventilation waveforms
DOI: 10.3109/03091902.2014.968675
Original Waveform
1
2
3
Original Waveform with Sine Inhale (b) 5 4 3 2 1 0 5 -1 0 -2 -3 -4 -5 -6 Velocity (m/s)
Velocity (m/s)
(a) 4 3 2 1 0 -1 0 -2 -3 -4 -5 -6
3
4
1
2
Time (s)
3
4
5
3
4
5
3
4
5
Time (s) Near True Sine Waveform
Original Waveform with Sine Exhale
(d) 3
(c) 3
Velocity (m/s)
0 0
1
2
3
4
5
-1
1 0 0
1
2
-1 -2
-2
-3 Time (s)
Time (s) Trapezoid (f) 4
Short Sine Inhale with Long Sine Exhale Waveform
3
Velocity (m/s)
(e) 5 4 3 2 1 0 −1 0 −2 −3
Velocity (m/s)
J Med Eng Technol Downloaded from informahealthcare.com by Gazi Univ. on 01/10/15 For personal use only.
1
Velocity (m/s)
2 2
1
2
3
4
5
2 1 0 −1
0
1
2
−2 Time (s)
Time (s)
Figure 2. Six different types of mechanical ventilation waveforms considered. (a) Original Waveform; (b) Original Waveform with Sine Inhale; (c) Original Waveform with Sine Exhale; (d) Near True Sine Waveform; (e) Short Sine Inhale with Long Sine Exhale; and (f) Trapezoid.
from the second waveform and the sinusoidal exhale from the third waveform and was designated as the Short Sine Inhale with Long Sine Exhale Waveform (SSILSEW). The sixth waveform was a trapezoid with 0.25 second rises and falls for the inhale and 0.9 s rises and falls for the exhale and was designated as the Trapezoid Waveform (TW). These waveforms were constructed and chosen based on their similarity to the original waveform. They are idealized representations utilized to investigate the influence of the shape of the velocity waveform on the WSS. A time step of 0.1 s was used for the numerical solution for each waveform. This computational speed was chosen after conducting a time step study with the Original Waveform in which the maximum wall shear stress only changed 0.71%, despite decreasing the step size by a factor of 50. The results of this study are shown in Table 1. Both the wall shear stress (WSS) and shear strain rate were calculated using computational analysis. The WSS is a tangential stress at an airway wall
Table 1. Effect of time step size on computational simulation. Time step size (s) Time steps Maximum wall shear (Pa) Percentage change
0.1 42 0.1286
0.05 84 0.1311 1.98%
0.01 420 0.1345 4.63%
0.002 2100 0.1295 0.71%
which is obtained as fluid (air) viscosity multipled by the transverse velocity gradient. In contrast, airway pressure acting in the normal direction to the airways creates stresses across the thickness of the airways. After the simulations were run, the maximum value at each time step was used to collect data on shear strain rate (1 s1) and wall shear (Pa) for each waveform. After shear strain rate was calculated at each time step, the values were numerically integrated to produce the total strain. A similar process was carried out for the wall shear terms to produce the impulse per unit area (Pa*s) for each waveform.
R. M. Pidaparti and J. Swanson
J Med Eng Technol, 2015; 39(1): 1–8
J Med Eng Technol Downloaded from informahealthcare.com by Gazi Univ. on 01/10/15 For personal use only.
4
Figure 3. Comparison of results of wall shear stress for (a) original waveform with (b) original waveform with sine inhale mode of mechanical ventilation.
3. Results and discussion Several simulations were carried out by varying the mechanical ventilation waveforms and the results of wall shear stress and shear strain rate were obtained through computational analysis. Figure 3 shows the comparison of wall shear stress results between the OW and OSIW (see Figures 2(a) and (b)). High airway wall shear stress (WSS) areas were observed at the beginning of generation G2 and at the bifurcations (see
Figure 3). At the inhalation phase, higher WSS values were observed at the bifurcations for OW, whereas higher WSS values were observed at bifurcation 2 as well as at the other bifurcations. This trend was also observed in the exhalation phase. Figure 4 shows the wall shear strain rate comparisons between the OW and OWSI. The wall shear strain rate trends were similar to those of WSS values between the OW
Mechanical ventilation waveforms
5
J Med Eng Technol Downloaded from informahealthcare.com by Gazi Univ. on 01/10/15 For personal use only.
DOI: 10.3109/03091902.2014.968675
Figure 4. Comparison of results of shear strain rate for (a) original waveform with (b) original waveform with sine inhale mode of mechanical ventilation.
and OSIW. Figures 5–10 show the maximum wall shear stress vs time for each of the waveforms considered. It can be seen from Figures 5–10 that the OSIW waveform has a maximum WSS in comparison to other waveforms considered. This is followed by SSILSEW and OW, whereas other waveforms have an order lower WSS. It can be seen from Figures 5–10 that each of the waveforms greatly affects the WSS and, thus, the resulting cellular damage. The comparison of various fluid dynamics parameter results for various waveforms in
relation to the original waveform is summarized in Table 2. It can be seen from Table 2 that NTSW has the maximum strain rate, followed by OSEW and TW. However, the maximum wall shear is observed for the NTSW waveform, followed by OSIW, TW and OSEW waveforms. Overall the results presented in Table 2 reflect that the mechanical ventilation waveform greatly affects the fluid dynamics parameters. The results of maximum WSS for various mechanical ventilation waveforms are presented in Figure 11. It can be
6
R. M. Pidaparti and J. Swanson
J Med Eng Technol, 2015; 39(1): 1–8
Maximum Wall Shear Stress (Pa)
Maximum Wall Shear Stress (Pa)
OW WSS vs. Time
0.14 0.12 0.1 0.08 0.06 0.04 0.02 0 1
0
2
3
4
SSILSEW WWS vs. Time 0.16 0.14 0.12 0.1 0.08 0.06 0.04 0.02 0
5
0
1
2
Time (s)
0
1
2
3
4
5
Maximum Wall Shear Stress (Pa)
Maximum Wall Shear Stress (Pa)
OSIW WSS vs. Time
0.2 0.18 0.16 0.14 0.12 0.1 0.08 0.06 0.04 0.02 0
Maximum Wall Shear Stress (Pa)
OSEW WSS vs. Time
0
1
2
TW WSS vs. Time 0.1 0.09 0.08 0.07 0.06 0.05 0.04 0.03 0.02 0.01 0
0
1
3
4
NTSW WSS vs. Time 0.06 0.05 0.04 0.03 0.02 0.01 0 1
1.5
2
2.5
3
4
5
3
Figure 10. Variation of maximum wall shear stress vs time for TW.
3.5
Waveform
Max shear strain rate
Total shear strain
Max wall shear
Total wall shear
OSIW OSEW NTSW SSILSEW TW
15.82% 20.89% 36.12% 15.82% 17.02%
7.74% 0.79% 1.77% 4.79% 1.77%
36.02% 29.15% 58.18% 13.60% 31.27%
24.42% 14.14% 21.55% 5.80% 5.91%
5
Figure 7. Variation of maximum wall shear stress vs time for OSEW.
0.5
2
Table 2. Comparison of various fluid dynamics parameters between different waveforms with the original waveform.
Time (s)
0
5
Time (s)
Figure 6. Variation of maximum wall shear stress vs time for OSIW.
0.1 0.09 0.08 0.07 0.06 0.05 0.04 0.03 0.02 0.01 0
4
Figure 9. Variation of maximum wall shear stress vs time for SSILSEW.
Time (s)
Maxmum Wall Shear Stress (Pa)
J Med Eng Technol Downloaded from informahealthcare.com by Gazi Univ. on 01/10/15 For personal use only.
Figure 5. Variation of maximum wall shear stress vs time for OW.
3 Time (s)
4
4.5
Time (s)
Figure 8. Variation of maximum wall shear stress vs time for NTSW.
seen that OSIW produced the highest WSS value and the NTSW produced the lowest WSS value. In comparison to the OW, the OWSI and the SSILSEW produced higher values of WSS by 36.02% and 13.60%, respectively. However, other waveforms OSEW, TW, and NTSW reduced the WSS values by 29%, 31% and 58%, respectively, in comparison to the OW. It is interesting to note that a higher value of WSS seems to cause more airway inflammation as well as epithelial erosion [9,22]. These findings of WSS values both quantitatively and qualitatively are similar to those obtained by other computational studies, especially for sinusoidal and square waveforms [14,18]. Also, the location of maximum shear stress may vary for different waveforms and this may have implications for the cellular damage location. Since we are relating the WSS as a means for causing cellular damage, the waveform which gives less WSS may reduce lung injury. More research and further studies are required to estimate the
Mechanical ventilation waveforms
J Med Eng Technol Downloaded from informahealthcare.com by Gazi Univ. on 01/10/15 For personal use only.
DOI: 10.3109/03091902.2014.968675
7
Figure 11. Effect of mechanical ventilation waveform on maximum wall shear stress.
overall cellular damage resulting from the waveform and the patient condition. The OSIW develops much more strain and WSS, primarily because the peak velocity at inhalation is much higher than for the OW. This is the same reason as to why SSILSEW is also much higher for both of these quantities, and why they both share the same maximum value for shear strain rate. Overall, the results obtained from the present study indicate that mechanical ventilation waveforms play an important role in airway wall shear stress and strain rates. These results, combined with optimization studies, suggest that it is possible to develop mechanical ventilation strategies to avoid lung injuries in patients.
4. Conclusions An investigation of the role of mechanical ventilation waveform on airway wall shear stress and strain rate was carried out using fluid dynamics analysis. An airway geometry model of 2–3 bifurcations was considered. Six different waveforms were analysed. The simulation results showed that OSIW produced the highest WSS value and the NTSW produced the lowest WSS value. Also, the OSIW and the SSILSEW produced a higher shear strain rate in comparison to the OW. The results obtained suggest that it may be possible to design a waveform to minimize epithelium cell inflammation and reduce lung injury. The investigation showed that varying the waveform can successfully reduce both the shear strain and the wall shear stress. The NTSW had the greatest reduction and this is likely because of its slow changing nature. However, the OSEW also had reductions in both shear strain and wall shear stress; this shows that one component of these is the sudden discontinuous jump made from inhalation to exhalation on traditional mechanical ventilators. Future investigations should further optimize these waveforms so that the negative consequences of mechanical ventilation are minimized. The ultimate goal would be to conduct clinical trials to physically validate what has been shown computationally.
Acknowledgement The authors thank the US National Science Foundation for sponsoring the research through a grant CMMI-1430379. The authors thank Dr Guoguang Su for help with the airway bifurcation model.
Declaration of interest The authors report no conflicts of interest. The authors alone are responsible for the content and writing of this article.
References 1. Chang, W.C., 2006, Clinical Applications of Mechanical Ventilation, 3rd edn. (United States: Thomson). 2. Pierce, L.N., 2007, Management of the Mechanically Ventilated Patient, 2nd edn. (MO: Elsevier) ISBN: 9780721603971. 3. Sather, J.E., and Schuur, J.D., 2007, Epidemiology of acute respiratory failure and mechanical ventilation in U.S. emergency departments, 1993–2004. Annals of Emergency Medicine, 50, S110. 4. Rubenfield, G.D., Caldwell, E., Peabody, E., Weaver, J., Martin, D.P., Neff, M., Stern, E.J., and Hudson, L.D., 2005, Incidence and outcomes of acute lung injury. New England Journal of Medicine, 353, 1685–1693. 5. Imai, Y., and Slutsky, A.S., 2006, Systemic effects of mechanical ventilation. In: Dreyfuss, D., Saumon, G., and Hubmayr, R.D., editors Ventilator-Induced Lung Injury (New York, NY: Taylor & Francis Group). pp. 267–285. 6. Khadaroo, R.G., and Marshall, J.C., 2002, ARDS and the multiple organ dysfunction syndrome: Common mechanisms of a common systemic process. Critical Care Clinics, 18, 127–141. 7. Tobin, M.J., 2001, Advances in mechanical ventilation. The New England Journal of Medicine, 344, 1986–1996. 8. Fan, E., Needham, D.M., and Stewart, T.E., 2005, Ventilatory management of acute lung injury and Acute Respiratory Distress Syndrome. Journal of the American Medical Association, 294, 2889–2896. 9. Thammanomai, A., Hamakawa, H., Bartolak-Suki, E., and Suki, B., 2013, Combined effects of ventilation mode and positive end-expiratory pressure on mechanics, gas exchange and the epithelium in mice with acute lung injury. PLOS One, 8, 1–10. 10. Muscedere, J.G., Mullen, J.B., Gan, K., and Slutsky, A.S., 1994, Tidal ventilation at low airway pressures can augment lung injury. American Journal of Respiratory & Critical Care Medicine, 149, 1327–1334.
8
R. M. Pidaparti and J. Swanson
J Med Eng Technol Downloaded from informahealthcare.com by Gazi Univ. on 01/10/15 For personal use only.
11. Blick, A.M., Dee, K.C., and Gaver, D.P., 2003, Mechanisms of surface-tension induced epithelium cell damage in a model of pulmonary airway reopening. Journal of Applied Physiology, 94, 770–783. 12. Comeford, A., Forster, C., and Wall, W.A., 2010, Structured tree imnpedance outflow boundary conditions for 3D lung simulations. ASME Journal of Biomechanical Engineering, 132, 081002-1-10. 13. Gattinoni, L., Carlesso, E., Cadringher, P., Valenza, F., Vagginelli, F., and Chiumello, D., 2003, Physical and biological triggers of ventilator-induced lung injury and its prevention. European Respiratory Journal, 47(Suppl), 15S–25S. 14. Alzahrany, M., Banerjee, A., and Salzman, G., 2014, Flow transport and gas mixing during invasive high frequency oscillatory ventilation. Medical Engineering & Physics, 36, 647–658. 15. Pidaparti, R.M., Koombua, K., and Ward, K.R., 2011, Assessment of mechanical ventilation parameters on respiratory mechanics. Journal of Medical Engineering & Technology, 36, 34–41. 16. Muller, W.J., Gerjarusek, S., and Scherer, P.W., 1990, Studies of wall shear stress and mass transfer in a large scale model of
J Med Eng Technol, 2015; 39(1): 1–8
17. 18.
19. 20. 21. 22.
neonatal high-frequency jet ventilation. Annals of Biomedical Engineering, 18, 69–88. Green, A.S., 2004, Modeling of peak flow wall shear stress in major airways of the lung. Journal of Biomechanics, 37, 661–667. Koombua, K., Pidaparti, R.M., Worth Longest, P., and Ward, K., 2009, Biomechanical aspects of compliant airways due to mechanical ventilation. Molecular and Cellular Biomechanics Journal, 6, 203–216. Weibel, E.R., 1963, Morphometry of the Human Lung (Berlin: Springer Verlag). Hlastala, M.P., and Berger, A.J., 2001, Physiology of Respiration (New York: Oxford University Press). ANSYS, 2005, Ansys 10.0 user guide (Canonsburg: ANSYS Inc.). Dhanireddy, S., Altemeier, W.A., Matute-Bello, G., O’Mahony, D.S., Glenny, R.W., Martin, T.R., and Liles, W.C., 2006, Mechanical ventilation induces inflammation, lung injury, and extra-pulmonary organ dysfunction in experimental pneumonia. Laboratory Investigation, 86, 790–799.
INNOVATION
Effect of mechanical ventilation waveforms on airway wall shear Ramana M. Pidaparti*1 and John Swanson2 College of Engineering, University of Georgia, Athens, GA, USA and 2Department of Mechanical and Nuclear Engineering, Virginia Commonwealth University, Richmond, VA, USA
J Med Eng Technol Downloaded from informahealthcare.com by Gazi Univ. on 01/10/15 For personal use only.
1
Abstract
Keywords
Better understanding of airway wall shear stress/strain rate is very important in order to prevent inflammation in patients undergoing mechanical ventilation due to respiratory problems in intensive-care medicine. The objective of this study was to investigate the role of mechanical ventilation waveforms on airway wall shear/strain rate using computational fluid dynamics analysis. Six different waveforms were considered to investigate the airway wall shear stress (WSS) from fluid dynamics analysis for the airway geometry of two-to-three generations. The simulation results showed that Original with Sine Inhale Waveform (OSIW) produced the highest WSS value and the Near True Sine Waveform produced the lowest WSS value. Also, the Original with Sine Inhale Waveform and the Short Sine Inhale with Long Sine Exhale Waveform (SSILSEW) produced a higher shear strain rate in comparison to the Original Waveform (OW). These results, combined with optimization, suggest that it is possible to develop a set of mechanical ventilation waveform strategies to avoid inflammation in the lung.
Airway wall shear, fluid dynamics analysis, mechanical ventilation, strain rate
1. Introduction Patients with acute lung injury (ALI), including acute respiratory distress syndrome (ARDS), airway and other pulmonary diseases, often require mechanical ventilation [1,2]. In the US alone, the incidence of respiratory failure and acute lung injury is 137–253 per 100,000 in the general population [3,4]. Ventilator Assisted Lung Injury (VALI) can be fatal, especially for ARDS patients, and might contribute to multiple organ dysfunction syndrome (MODS) from volutrauma, atelectrauma or biotrauma mechanisms [5,6]. Many techniques have been suggested to prevent VALI and mainly centre around reductions in tidal volumes to reduce airway pressure. However, decreasing tidal volume can have drawbacks such as causing hypercapnia, decreasing aerated lung volume, increase shunting and worsening oxygenation [7]. Increasing positive end-expiratory pressure (PEEP) to compensate for reductions in tidal volume can occasionally lead to transient oxygen desaturation, hypotension, barotrauma, arrhythmia and bacterial translocation [8]. Thammanomai et al. [9] investigated the combined effects of ventilation mode and PEEP and concluded that higher PEEP improved the mechanics and gas exchange, but also increased the epithelial cell injury in mice. Also, inadequate PEEP can *Corresponding author. Email: [email protected]
History Received 30 April 2014 Revised 4 September 2014 Accepted 19 September 2014
result in epithelial lung injury due to mechanical ventilation induced shear and normal stresses [10,11]. Developing new modes of mechanical ventilation is challenging, in part because of the difficulty in predicting the effects. Recently, computational models representing airflow in patient-specific human lungs under mechanical ventilation have been developed [12]. A greater appreciation now exists for the effects of distending forces on the lung from mechanical ventilation and how these forces are distributed via the lung’s fibrous skeleton [13]. The airway wall shear stress due to mechanical ventilation is likely to play a central role in initiating, maintaining and/or exacerbating new or existing inflammatory responses in the lung and the subsequent development of VALI [9]. Computational studies also suggest that high airway wall shear stresses lead to inflammation [14,15]. Many experimental studies have suggested the potential for using computational methods to investigate the effects of mechanical ventilation on lung injury in hopes of developing means to reduce injury in the future [14–18]. However, the role of mechanical ventilation waveforms, especially with non-sinusoidal inhale/exhale shapes on airway wall shear stress and shear strain rate has not been studied quantitatively. The objective of this study was to investigate the role of mechanical ventilation waveform on airway wall shear/strain rate using computational fluid dynamics analysis. Using
2
R. M. Pidaparti and J. Swanson Input: Airflow Velocity from Mechanical Venlaon
J Med Eng Technol, 2015; 39(1): 1–8
the compressibility effects are very small. Also, the diameters of the lung bifurcations are sufficiently small to safely assume that the density variation of air throughout the lung is minimal. In light of this, incompressible flow field is assumed in the simulation. Additionally, the flow was modelled as laminar based on the Reynolds number calculation, as shown below. d at 37 C 1:138 mkg3 4:54 ms ð0:005mÞ Re ¼ 1:9 105 Pa s Re ¼ 1360
J Med Eng Technol Downloaded from informahealthcare.com by Gazi Univ. on 01/10/15 For personal use only.
Re ¼
Figure 1. An overview of the airway bifurcation geometry (generations 2–3) with input of airflow due to mechanical ventilation and pressure outlets.
airway geometry of two-to-three generations, the results of airway wall shear stress obtained from computational analysis for six different waveforms are compared and discussed.
2. Modelling and simulations Geometric representation of the airway generations 2–3, which is the 2nd and 3rd bifurcation within the lungs, in three-dimensions based on anatomical Weibel adult human tracheobronchial airway descriptions [19] was considered in this study. These airway generations have significantly fewer cartilage plates and rings [20] and, hence, the presence of cartilage plates in the respiratory generations 2–3 was ignored in this study. The geometry of the airway bifurcation model developed in the current study is shown in Figure 1. The computational model involves modelling the airways bifurcation for fluid dynamics analysis. The airways were considered as rigid (no deformation) in the computational fluid dynamics (CFD) model from which the airway pressures and wall shear stresses were obtained. Numerical solution of the airflow through airway bifurcation during mechanical ventilation was investigated using ANSYS CFX software [21]. The ANSYS workbench was used to conduct the FSI simulations using CFX and ANSYS Mechanical. To obtain the input flow waveform for the airway generation 2, airflow at the level of the trachea from mechanical ventilation was assumed to divide equally at each bifurcation. The input velocity profile is assumed to be uniform and air was modelled as an incompressible fluid in this study. Using the maximum velocity of the air in lung bifurcations, the Mach number is estimated to be much less than 0.2, in which case
where Re is the Reynolds number, is the air density at 37 C, v is the maximum velocity of the air, d is the inlet diameter of the model, and m is the viscosity of air at 37 C. Since the Reynolds number is less than the transition region of Re ¼ 2300 for a circular pipe, laminar flow can be assumed. A zero constant-distribution pressure was applied at the outlets of the fluid domain. This boundary condition was used in our earlier study [18] for fluid dynamics analysis. However, for fluid-solid analysis, we imposed a zero-displacement boundary condition at the outlets. The pressure boundary condition applied at the outlets of the model in relation to the inlet will greatly affect the resulting results of wall shear stresses which are related to pressure gradients between inlet and outlets. Also, at different bifurcations the pressure boundary condition will be different. It is noteworthy to point out that the sensitivity of model results to pressure boundary conditions needs to be investigated for a realistic application. This is one of the limitations of the present study. A mesh-convergence study was performed to confirm that a fine enough element had been used to represent the fluid domain for analysis. Changes in maximum pressure and velocity were used as convergence criteria. A converged model was obtained when changes in those criteria were less than 5%. The converged finite element model consisted of 107,520 hexahedra elements. Six different waveforms of mechanical ventilation, as shown in Figure 2, were chosen for this study. The first waveform was the original waveform (OW), which is the waveform currently used during mechanical ventilation. All other waveforms were constructed by numerically integrating the area underneath the original waveform and adjusting the waveforms so that the area underneath the inhale and underneath the exhale remained constant for each waveform. The second waveform was modified so that the inhale followed a sinusoidal rise and fall, yet the exhale was the same as the original waveform and was designated as the Original with Sine Inhale Waveform (OSIW). The third waveform was modified so that the exhale followed a sinusoidal fall and rise, yet the inhale followed the original waveform and was designated as the Original with Sine Exhale Waveform (OSEW). The fourth waveform, Near True Sine Waveform (NTSW), attempted to follow a true sine wave as closely as possible by having half the time period for inhalation and half for exhalation but, because of the slight difference in areas of inhalation and exhalation, is not quite a true sine wave. The fifth waveform used the sinusoidal inhale
Mechanical ventilation waveforms
DOI: 10.3109/03091902.2014.968675
Original Waveform
1
2
3
Original Waveform with Sine Inhale (b) 5 4 3 2 1 0 5 -1 0 -2 -3 -4 -5 -6 Velocity (m/s)
Velocity (m/s)
(a) 4 3 2 1 0 -1 0 -2 -3 -4 -5 -6
3
4
1
2
Time (s)
3
4
5
3
4
5
3
4
5
Time (s) Near True Sine Waveform
Original Waveform with Sine Exhale
(d) 3
(c) 3
Velocity (m/s)
0 0
1
2
3
4
5
-1
1 0 0
1
2
-1 -2
-2
-3 Time (s)
Time (s) Trapezoid (f) 4
Short Sine Inhale with Long Sine Exhale Waveform
3
Velocity (m/s)
(e) 5 4 3 2 1 0 −1 0 −2 −3
Velocity (m/s)
J Med Eng Technol Downloaded from informahealthcare.com by Gazi Univ. on 01/10/15 For personal use only.
1
Velocity (m/s)
2 2
1
2
3
4
5
2 1 0 −1
0
1
2
−2 Time (s)
Time (s)
Figure 2. Six different types of mechanical ventilation waveforms considered. (a) Original Waveform; (b) Original Waveform with Sine Inhale; (c) Original Waveform with Sine Exhale; (d) Near True Sine Waveform; (e) Short Sine Inhale with Long Sine Exhale; and (f) Trapezoid.
from the second waveform and the sinusoidal exhale from the third waveform and was designated as the Short Sine Inhale with Long Sine Exhale Waveform (SSILSEW). The sixth waveform was a trapezoid with 0.25 second rises and falls for the inhale and 0.9 s rises and falls for the exhale and was designated as the Trapezoid Waveform (TW). These waveforms were constructed and chosen based on their similarity to the original waveform. They are idealized representations utilized to investigate the influence of the shape of the velocity waveform on the WSS. A time step of 0.1 s was used for the numerical solution for each waveform. This computational speed was chosen after conducting a time step study with the Original Waveform in which the maximum wall shear stress only changed 0.71%, despite decreasing the step size by a factor of 50. The results of this study are shown in Table 1. Both the wall shear stress (WSS) and shear strain rate were calculated using computational analysis. The WSS is a tangential stress at an airway wall
Table 1. Effect of time step size on computational simulation. Time step size (s) Time steps Maximum wall shear (Pa) Percentage change
0.1 42 0.1286
0.05 84 0.1311 1.98%
0.01 420 0.1345 4.63%
0.002 2100 0.1295 0.71%
which is obtained as fluid (air) viscosity multipled by the transverse velocity gradient. In contrast, airway pressure acting in the normal direction to the airways creates stresses across the thickness of the airways. After the simulations were run, the maximum value at each time step was used to collect data on shear strain rate (1 s1) and wall shear (Pa) for each waveform. After shear strain rate was calculated at each time step, the values were numerically integrated to produce the total strain. A similar process was carried out for the wall shear terms to produce the impulse per unit area (Pa*s) for each waveform.
R. M. Pidaparti and J. Swanson
J Med Eng Technol, 2015; 39(1): 1–8
J Med Eng Technol Downloaded from informahealthcare.com by Gazi Univ. on 01/10/15 For personal use only.
4
Figure 3. Comparison of results of wall shear stress for (a) original waveform with (b) original waveform with sine inhale mode of mechanical ventilation.
3. Results and discussion Several simulations were carried out by varying the mechanical ventilation waveforms and the results of wall shear stress and shear strain rate were obtained through computational analysis. Figure 3 shows the comparison of wall shear stress results between the OW and OSIW (see Figures 2(a) and (b)). High airway wall shear stress (WSS) areas were observed at the beginning of generation G2 and at the bifurcations (see
Figure 3). At the inhalation phase, higher WSS values were observed at the bifurcations for OW, whereas higher WSS values were observed at bifurcation 2 as well as at the other bifurcations. This trend was also observed in the exhalation phase. Figure 4 shows the wall shear strain rate comparisons between the OW and OWSI. The wall shear strain rate trends were similar to those of WSS values between the OW
Mechanical ventilation waveforms
5
J Med Eng Technol Downloaded from informahealthcare.com by Gazi Univ. on 01/10/15 For personal use only.
DOI: 10.3109/03091902.2014.968675
Figure 4. Comparison of results of shear strain rate for (a) original waveform with (b) original waveform with sine inhale mode of mechanical ventilation.
and OSIW. Figures 5–10 show the maximum wall shear stress vs time for each of the waveforms considered. It can be seen from Figures 5–10 that the OSIW waveform has a maximum WSS in comparison to other waveforms considered. This is followed by SSILSEW and OW, whereas other waveforms have an order lower WSS. It can be seen from Figures 5–10 that each of the waveforms greatly affects the WSS and, thus, the resulting cellular damage. The comparison of various fluid dynamics parameter results for various waveforms in
relation to the original waveform is summarized in Table 2. It can be seen from Table 2 that NTSW has the maximum strain rate, followed by OSEW and TW. However, the maximum wall shear is observed for the NTSW waveform, followed by OSIW, TW and OSEW waveforms. Overall the results presented in Table 2 reflect that the mechanical ventilation waveform greatly affects the fluid dynamics parameters. The results of maximum WSS for various mechanical ventilation waveforms are presented in Figure 11. It can be
6
R. M. Pidaparti and J. Swanson
J Med Eng Technol, 2015; 39(1): 1–8
Maximum Wall Shear Stress (Pa)
Maximum Wall Shear Stress (Pa)
OW WSS vs. Time
0.14 0.12 0.1 0.08 0.06 0.04 0.02 0 1
0
2
3
4
SSILSEW WWS vs. Time 0.16 0.14 0.12 0.1 0.08 0.06 0.04 0.02 0
5
0
1
2
Time (s)
0
1
2
3
4
5
Maximum Wall Shear Stress (Pa)
Maximum Wall Shear Stress (Pa)
OSIW WSS vs. Time
0.2 0.18 0.16 0.14 0.12 0.1 0.08 0.06 0.04 0.02 0
Maximum Wall Shear Stress (Pa)
OSEW WSS vs. Time
0
1
2
TW WSS vs. Time 0.1 0.09 0.08 0.07 0.06 0.05 0.04 0.03 0.02 0.01 0
0
1
3
4
NTSW WSS vs. Time 0.06 0.05 0.04 0.03 0.02 0.01 0 1
1.5
2
2.5
3
4
5
3
Figure 10. Variation of maximum wall shear stress vs time for TW.
3.5
Waveform
Max shear strain rate
Total shear strain
Max wall shear
Total wall shear
OSIW OSEW NTSW SSILSEW TW
15.82% 20.89% 36.12% 15.82% 17.02%
7.74% 0.79% 1.77% 4.79% 1.77%
36.02% 29.15% 58.18% 13.60% 31.27%
24.42% 14.14% 21.55% 5.80% 5.91%
5
Figure 7. Variation of maximum wall shear stress vs time for OSEW.
0.5
2
Table 2. Comparison of various fluid dynamics parameters between different waveforms with the original waveform.
Time (s)
0
5
Time (s)
Figure 6. Variation of maximum wall shear stress vs time for OSIW.
0.1 0.09 0.08 0.07 0.06 0.05 0.04 0.03 0.02 0.01 0
4
Figure 9. Variation of maximum wall shear stress vs time for SSILSEW.
Time (s)
Maxmum Wall Shear Stress (Pa)
J Med Eng Technol Downloaded from informahealthcare.com by Gazi Univ. on 01/10/15 For personal use only.
Figure 5. Variation of maximum wall shear stress vs time for OW.
3 Time (s)
4
4.5
Time (s)
Figure 8. Variation of maximum wall shear stress vs time for NTSW.
seen that OSIW produced the highest WSS value and the NTSW produced the lowest WSS value. In comparison to the OW, the OWSI and the SSILSEW produced higher values of WSS by 36.02% and 13.60%, respectively. However, other waveforms OSEW, TW, and NTSW reduced the WSS values by 29%, 31% and 58%, respectively, in comparison to the OW. It is interesting to note that a higher value of WSS seems to cause more airway inflammation as well as epithelial erosion [9,22]. These findings of WSS values both quantitatively and qualitatively are similar to those obtained by other computational studies, especially for sinusoidal and square waveforms [14,18]. Also, the location of maximum shear stress may vary for different waveforms and this may have implications for the cellular damage location. Since we are relating the WSS as a means for causing cellular damage, the waveform which gives less WSS may reduce lung injury. More research and further studies are required to estimate the
Mechanical ventilation waveforms
J Med Eng Technol Downloaded from informahealthcare.com by Gazi Univ. on 01/10/15 For personal use only.
DOI: 10.3109/03091902.2014.968675
7
Figure 11. Effect of mechanical ventilation waveform on maximum wall shear stress.
overall cellular damage resulting from the waveform and the patient condition. The OSIW develops much more strain and WSS, primarily because the peak velocity at inhalation is much higher than for the OW. This is the same reason as to why SSILSEW is also much higher for both of these quantities, and why they both share the same maximum value for shear strain rate. Overall, the results obtained from the present study indicate that mechanical ventilation waveforms play an important role in airway wall shear stress and strain rates. These results, combined with optimization studies, suggest that it is possible to develop mechanical ventilation strategies to avoid lung injuries in patients.
4. Conclusions An investigation of the role of mechanical ventilation waveform on airway wall shear stress and strain rate was carried out using fluid dynamics analysis. An airway geometry model of 2–3 bifurcations was considered. Six different waveforms were analysed. The simulation results showed that OSIW produced the highest WSS value and the NTSW produced the lowest WSS value. Also, the OSIW and the SSILSEW produced a higher shear strain rate in comparison to the OW. The results obtained suggest that it may be possible to design a waveform to minimize epithelium cell inflammation and reduce lung injury. The investigation showed that varying the waveform can successfully reduce both the shear strain and the wall shear stress. The NTSW had the greatest reduction and this is likely because of its slow changing nature. However, the OSEW also had reductions in both shear strain and wall shear stress; this shows that one component of these is the sudden discontinuous jump made from inhalation to exhalation on traditional mechanical ventilators. Future investigations should further optimize these waveforms so that the negative consequences of mechanical ventilation are minimized. The ultimate goal would be to conduct clinical trials to physically validate what has been shown computationally.
Acknowledgement The authors thank the US National Science Foundation for sponsoring the research through a grant CMMI-1430379. The authors thank Dr Guoguang Su for help with the airway bifurcation model.
Declaration of interest The authors report no conflicts of interest. The authors alone are responsible for the content and writing of this article.
References 1. Chang, W.C., 2006, Clinical Applications of Mechanical Ventilation, 3rd edn. (United States: Thomson). 2. Pierce, L.N., 2007, Management of the Mechanically Ventilated Patient, 2nd edn. (MO: Elsevier) ISBN: 9780721603971. 3. Sather, J.E., and Schuur, J.D., 2007, Epidemiology of acute respiratory failure and mechanical ventilation in U.S. emergency departments, 1993–2004. Annals of Emergency Medicine, 50, S110. 4. Rubenfield, G.D., Caldwell, E., Peabody, E., Weaver, J., Martin, D.P., Neff, M., Stern, E.J., and Hudson, L.D., 2005, Incidence and outcomes of acute lung injury. New England Journal of Medicine, 353, 1685–1693. 5. Imai, Y., and Slutsky, A.S., 2006, Systemic effects of mechanical ventilation. In: Dreyfuss, D., Saumon, G., and Hubmayr, R.D., editors Ventilator-Induced Lung Injury (New York, NY: Taylor & Francis Group). pp. 267–285. 6. Khadaroo, R.G., and Marshall, J.C., 2002, ARDS and the multiple organ dysfunction syndrome: Common mechanisms of a common systemic process. Critical Care Clinics, 18, 127–141. 7. Tobin, M.J., 2001, Advances in mechanical ventilation. The New England Journal of Medicine, 344, 1986–1996. 8. Fan, E., Needham, D.M., and Stewart, T.E., 2005, Ventilatory management of acute lung injury and Acute Respiratory Distress Syndrome. Journal of the American Medical Association, 294, 2889–2896. 9. Thammanomai, A., Hamakawa, H., Bartolak-Suki, E., and Suki, B., 2013, Combined effects of ventilation mode and positive end-expiratory pressure on mechanics, gas exchange and the epithelium in mice with acute lung injury. PLOS One, 8, 1–10. 10. Muscedere, J.G., Mullen, J.B., Gan, K., and Slutsky, A.S., 1994, Tidal ventilation at low airway pressures can augment lung injury. American Journal of Respiratory & Critical Care Medicine, 149, 1327–1334.
8
R. M. Pidaparti and J. Swanson
J Med Eng Technol Downloaded from informahealthcare.com by Gazi Univ. on 01/10/15 For personal use only.
11. Blick, A.M., Dee, K.C., and Gaver, D.P., 2003, Mechanisms of surface-tension induced epithelium cell damage in a model of pulmonary airway reopening. Journal of Applied Physiology, 94, 770–783. 12. Comeford, A., Forster, C., and Wall, W.A., 2010, Structured tree imnpedance outflow boundary conditions for 3D lung simulations. ASME Journal of Biomechanical Engineering, 132, 081002-1-10. 13. Gattinoni, L., Carlesso, E., Cadringher, P., Valenza, F., Vagginelli, F., and Chiumello, D., 2003, Physical and biological triggers of ventilator-induced lung injury and its prevention. European Respiratory Journal, 47(Suppl), 15S–25S. 14. Alzahrany, M., Banerjee, A., and Salzman, G., 2014, Flow transport and gas mixing during invasive high frequency oscillatory ventilation. Medical Engineering & Physics, 36, 647–658. 15. Pidaparti, R.M., Koombua, K., and Ward, K.R., 2011, Assessment of mechanical ventilation parameters on respiratory mechanics. Journal of Medical Engineering & Technology, 36, 34–41. 16. Muller, W.J., Gerjarusek, S., and Scherer, P.W., 1990, Studies of wall shear stress and mass transfer in a large scale model of
J Med Eng Technol, 2015; 39(1): 1–8
17. 18.
19. 20. 21. 22.
neonatal high-frequency jet ventilation. Annals of Biomedical Engineering, 18, 69–88. Green, A.S., 2004, Modeling of peak flow wall shear stress in major airways of the lung. Journal of Biomechanics, 37, 661–667. Koombua, K., Pidaparti, R.M., Worth Longest, P., and Ward, K., 2009, Biomechanical aspects of compliant airways due to mechanical ventilation. Molecular and Cellular Biomechanics Journal, 6, 203–216. Weibel, E.R., 1963, Morphometry of the Human Lung (Berlin: Springer Verlag). Hlastala, M.P., and Berger, A.J., 2001, Physiology of Respiration (New York: Oxford University Press). ANSYS, 2005, Ansys 10.0 user guide (Canonsburg: ANSYS Inc.). Dhanireddy, S., Altemeier, W.A., Matute-Bello, G., O’Mahony, D.S., Glenny, R.W., Martin, T.R., and Liles, W.C., 2006, Mechanical ventilation induces inflammation, lung injury, and extra-pulmonary organ dysfunction in experimental pneumonia. Laboratory Investigation, 86, 790–799.
