Flow Dynamics in Pediatric Rigid Bronchoscopes Using Computer-Aided Design Modeling Software

Transcription

1 The Laryngoscope VC 2015 The American Laryngological, Rhinological and Otological Society, Inc. Flow Dynamics in Pediatric Rigid Bronchoscopes Using Computer-Aided Design Modeling Software Mitchell D. Barneck, BS; J. Taylor Webb, BS; Ryan E. Robinson, BS; J. Fredrik Grimmer, MD Objectives/Hypothesis: Observed complications during rigid bronchoscopy, including hypercarbia and hypoxemia, prompted us to assess how well rigid bronchoscopes serve as an airway device. We performed computer-aided design flow analysis of pediatric rigid bronchoscopes to gain insight into flow dynamics. Study Design: We made accurate three-dimensional computer models of pediatric rigid bronchoscopes and endotracheal tubes. SOLIDWORKS (Dassault Systemes, Velizy-Villacoublay, France) flow analysis software was used to analyze fluid dynamics during pressure-controlled and volume-controlled ventilation. Methods: Flow analysis was performed on rigid bronchoscopes and similar outer diameter endotracheal tubes comparing resistance, flow, and turbulence during two ventilation modalities and in common surgical scenarios. Results: Increased turbulent flow was observed in bronchoscopes compared to more laminar flow in endotracheal tubes of similar outer diameter. Flow analysis displayed higher resistances in all pediatric bronchoscope sizes except one (3.0 bronchoscope) compared to similar-sized endotracheal tubes. Loss of adequate ventilation was observed if the bronchoscope was not assembled correctly or if increased peak inspiratory pressures were needed. Anesthesia flow to the patient was reduced by 63% during telescope insertion. Conclusions: Flow analysis illustrates increased turbulent flow and increased airflow resistance in all but one size of pediatric bronchoscopes compared to endotracheal tubes. This increased turbulence and resistance, along with the unanticipated gas distal exit pattern, may contribute to the documented hypercarbia and hypoxemia during procedures. These findings may explain why hypoxemia and hypercarbia are commonly observed during rigid bronchoscopy, especially when positive pressure ventilation is needed. Key Words: Bronchoscopy, airway, foreign body, rigid bronchoscope. Level of Evidence: NA Laryngoscope, 126: , 2016 INTRODUCTION The modern ventilating bronchoscope is used for evaluating the upper airway, removing foreign bodies, and performing minor surgical procedures. It had its beginnings with Gustav Killian in 1897, when he fastened a small electric light bulb to the distal end of a metal tube. 1 Over half a century later, the first ventilating bronchoscope was developed by Friedel. 1 A great advance in viewing came in 1963 with the introduction of the Hopkins rod lens (telescope). 1 The current Doesel-Huzly ventilating From the Department of Bioengineering (M.D.B., J.T.W., R.E.R.), and the Division of Otolaryngology (J.F.G.), University of Utah, Salt Lake City, Utah, U.S.A. Editor s Note: This Manuscript was accepted for publication August 31, Presented in part at the Combined Otolaryngology Spring Meetings for the American Society of Pediatric Otolaryngology, Las Vegas, Nevada, U.S.A., May 17, All authors contributed in some form in conception, generation and analysis of data, drafting of the manuscript, and/or manuscript revision. The authors have no funding, financial relationships, or conflicts of interest to disclose. Additional Supporting Information may be found in the online version of this article. Send correspondence to J. Fredrik Grimmer, MD, 100 North Mario Capecchi Drive, Suite 4500, Salt Lake City, UT j.grimmer@imail2.org DOI: /lary bronchoscope (Karl Storz, Tuttlingen, Germany) is based on these early innovations with slight modifications. 2 Aside from making telescopes thinner and modifying ports, 1 4 little innovation has been made to improve the rigid bronchoscope as an airway device for ventilation. Ventilation during bronchoscopy is of particular concern in patients with cardiopulmonary instability or foreign body aspiration. In fact, a recent comprehensive study reported a 4% risk of anoxic brain injury or death after foreign body aspiration. 5 Major complications associated with ventilation include pneumothorax, pneumomediastinum, cardiac arrest, and hypoxic brain injury. 6 During bronchoscopy for foreign body removal, hypoxemia, and hypercarbia occur in approximately one in four patients. 6,7 Conversion to tracheostomy or thoracotomy during endoscopic foreign body removal happens in nearly 1% of cases. 8 The estimated mortality associated with bronchoscopy during foreign body removal is between 0.5% and 1.1%. 6,9 Due to the high complication rate, some advocate flexible bronchoscopy prior to rigid bronchoscopy to confirm its necessity. 10 Complications seen during foreign body removal are related to a number of factors including airway obstruction, cardiopulmonary changes from anesthesia, and the ability to ventilate through a rigid bronchoscope. The studies cited above do not distinguish between these confounding variables, but poor flow dynamics through a

2 ventilating bronchoscope should also be considered as a contributing factor. Furthermore, patients who require more respiratory support (i.e., having a foreign body obstructing a bronchus) may expose the limits of ventilation with the current systems. Improvements in ventilation during rigid bronchoscopy have come from innovations in anesthesia, 11,12 but few studies have evaluated the rigid bronchoscope as an airway device. 13 The purpose of this study was to examine whether gas flow and resistance in rigid bronchoscopes is altered compared to endotracheal tubes. MATERIALS AND METHODS Flow simulations were performed using SOLIDWORKS (Dassault Systemes, Velizy-Villacoublay, France) flow simulation software. We first created computer-aided design (CAD) models identical to the reported physical specifications of Doesel-Huzly rigid bronchoscopes and Mallinckrodt endotracheal tubes (Covidien, Dublin, Ireland). A mesh of the hollow internal spaces of each model was then created, essentially subdividing the internal volume into smaller regions for individual calculations. In particular, a refined structured Cartesian immersed-body mesh was used. 14 We conducted a mesh independence study on the models to ensure convergence of the solutions. The flow simulation solves the Navier-Stokes equations, which are formulations of mass, momentum, and energy conservation laws for fluid flows. 15 In essence, once boundary and initial conditions are specified for a particular model, solving these equations elucidates the internal changes in volumetric flow rates, velocities, pressures, resistances, and laminar/turbulent flow characteristics throughout the mesh, which are the parameters of interest for this study. This is a simplification for clarity, and a more robust discussion of the software calculation is available (see Supporting Information, Appendix, in the online version of this article). The reported dimensions and measurements of Doesel- Huzly bronchoscopes and Mallinckrodt endotracheal tubes are reported in Table I. The sizes and lengths of bronchoscopes chosen were the most commonly used at Primary Children s Hospital, although various other lengths of bronchoscopes are available to purchase from Karl Storz. A depiction of the crosssectional area of various sizes of endotracheal tubes and bronchoscopes is provided as Supporting Figure 1 in the online version of this article. For the first part of the study, common pediatric Doesel- Huzly bronchoscopes and Mallinckrodt endotracheal tubes were analyzed based on pressure-controlled and volume-controlled ventilation. The bronchoscopes were tested in the most functional setup, with the telescope inserted and all appropriate ports closed. All studies used predefined material properties for air and an ambient temperature of 15.58C (608F). For pressure-controlled ventilation, each model s flow analysis study was set with an anesthesia inlet pressure of 20 cm H 2 O and a series of outlet pressures corresponding to an inflating pediatric lung (0.5 cm H 2 O increments ranging from cm H 2 O). These values were chosen based on the common surgical scenarios and studies regarding the internal pressure of pediatric lungs for various age groups. 16,17 The software then calculated the flow characteristics for each model, recalculating for each distal simulated lung pressure in the series. For volume-controlled ventilation, a tidal volume was calculated based on the average size of the patients for which a particular device is indicated. This tidal volume represents that of a patient with a weight in the 50th percentile for the age group. TABLE I. Physical Dimensions and Calculated Flow Necessary for Bronchoscopes and Endotracheal Tubes. Outer Diameter Inner diameter Length Telescope Diameter Calculated Necessary Flow Rate (L/Min) Doesel-Huzly bronchoscope conventional size N/A Mallinckrodt endotracheal tube conventional size N/A N/A N/A N/A N/A N/A N/A N/A N/A Date presented are specifications for common pediatric sizes of the Doesel-Huzly bronchoscope and Mallinckrodt endotracheal tube. Necessary flow rates were calculated based on the tidal volumes and respiratory rates for the 50th percentile weight of appropriate age groups. N/A 5 not applicable. The tidal volume was used to determine the theoretical volume flow rate during inspiration. The outlet (distal) pressures were defined in the same manner as for pressure-controlled ventilation, which were based on inflating lung pressures. We then compared the flow dynamics of equivalent outerdiameter bronchoscopes and endotracheal tubes. The calculated volume flow rate from the pressure-controlled ventilation and the calculated proximal pressures for the volume-controlled ventilation were used to calculate the flow resistance based on Poiseuille s law (see Supporting Information, Appendix, in the online version of this article). In the second part of the study, the rigid bronchoscope was modeled both with and without an inserted telescope. The 3.5 bronchoscope was additionally tested for various possible configurations including an open proximal end (immediately prior or subsequent to inserting the telescope), an open suction port, an open prism insertion port, and various combinations of these openings. The 3.5 size was chosen for analysis, as it tended to be the most commonly used size bronchoscope at our institution. RESULTS The results from these simulations provide visual and numerical information on turbulent intensity, volumetric flow, resistance, and pressure. This allows for a comparative investigation of airflow in Doesel-Huzly rigid bronchoscopes and Mallinckrodt endotracheal tubes. Pressure-Controlled and Volume-Controlled Ventilation Flow simulations of the bronchoscopes and endotracheal tubes using pressure-controlled and volumecontrolled initial conditions resulted in the flow profile 1941

3 Fig. 1. The rigid bronchoscope percent resistance increase from comparable endotracheal tube versus lung pressure. (Left) The comparative resistances of the five tested sizes of rigid bronchoscopes, defined by the outer diameter, during pressure-controlled ventilation. (Right) The comparative resistances of the same rigid bronchoscopes during volume-controlled ventilation. Resistances are reported as a percent increase from a similar-sized outer diameter endotracheal tube for contextual comparison. [Color figure can be viewed in the online issue, which is available at for each device size based on the simulated ventilation. When compared to a similar diameter endotracheal tube using pressure-controlled ventilation, bronchoscopes showed decreased volumetric flow rates across all internal lung pressures in all cases except one. When 20 cm H 2 O of peak inspiratory pressure (PIP) is applied at an inflated internal lung pressure of 5 cm H 2 O, the 2.5, 3.5, 3.7, and 4.0 bronchoscopes exhibited a 34.19%, 46.35%, 37.32%, and 39.77% reduction in volume flow rate, respectively, when compared to the endotracheal tube. This PIP and lung pressure were chosen as typical values in healthy children during mechanical ventilation. The only bronchoscope to exhibit improved flow when compared to an endotracheal tube was the 3.0 bronchoscope, which resulted in a 70.76% increase. For the volume-controlled system, the simulations used the predefined distal pressures and volume flow rates to calculate the pressure necessary to achieve adequate flow. In all cases, the bronchoscope required higher pressures to generate an equivalent flow in the similar outer diameter endotracheal tube. The 2.5, 3.0, 3.5, 3.7, and 4.0 bronchoscopes required a 38.75%, 3.96%, 56.43%, 20.32%, and 15.53% increase in pressure, respectively, compared to the endotracheal tubes (20 cm H 2 O PIP with an internal lung pressure of 5 cm H 2 O). The percent increase from the comparable endotracheal tube resistances of the tested bronchoscopes are indicated in Figure 1. Using volume-controlled ventilation, the bronchoscopes demonstrated an even greater increase in resistance compared to pressure-controlled ventilation (Fig. 1, right). Flow Characteristics and Common Surgical Scenarios Each flow simulation generated flow profiles unique to each instrument that included velocity, pressure, turbulence, and vorticity. When flow becomes turbulent, chaotic motion and eddies are formed that do not contribute to the volume flow rate. This turbulence increases the 1942 resistance substantially, requiring large increases in pressure to increase the volume flow rate. Compared to laminar flow in similar application, turbulent flow will exhibit increased resistance and decreased flow. In the endotracheal tubes, the internal flow characteristics indicated steady and more laminar flow, with minimal turbulence throughout all models as depicted in Figure 2. In contrast, several design features and peripheral devices of the bronchoscope contribute to an increased level of turbulent flow for the rigid bronchoscope. The flow was laminar near the inlet port in all models, but quickly became turbulent with increased vorticity throughout the length of the bronchoscope shaft. The insertion of the telescope dramatically increased the turbulence within each bronchoscope model. The volume flow rate for bronchoscopes was reduced to between 54.6% and 73.2% of the original volume flow rate when a telescope was inserted. In pressure-controlled ventilation, the resistance in the main shaft was high enough in the 3.5 rigid bronchoscope to force 98.66% to 99.04% of the flow out of the side vents (Fig. 2C). Only about 1% of the ventilation actually exited the distal end of the device. The differences in flow patterns between endotracheal tubes and bronchoscopes are summarized in Figure 2. In another flow simulation study, we analyzed the proportion of airflow that was received by the patient in several surgical scenarios with and without telescopes inserted. These scenarios are comprised of various configurations of open and closed ports including the anesthesia inlet, instrument opening, suction port, and prism insertion port of the 3.0, 3.5, and 4.0 Doesel-Huzly bronchoscope. By opening one or more of these ports, a leak is created that results in decreased resistance (overall increased flow). In all cases, the gas administration to the patient with an open proximal end port was less than half of the total gas volume entering the device. The volume delivered to the patient is preserved until a critical point when the leak becomes so substantial (such as the all

4 Fig. 2. Images of SOLIDWORKS computer-aided design models of a 3.5 bronchoscope and 4.0 endotracheal tube with flow analysis. (A, B) Flow patterns and analysis on proximal and distal ends, respectively, of a 3.5 bronchoscope with a 2.9-mm telescope inserted. Traces illustrate particle paths and colors indicate turbulence intensity. (C, D) Proximal and distal ends of a 4.0 endotracheal tube, respectively. The turbulence intensity map is to the same scale for all figures. ports open case in Table II) that the flow to the patient is reduced. For example, in the 3.5 bronchoscope, the open ports caused a decrease in gas administered to the patient s lungs of nearly 40% when compared to the normal closed-port scenario. These results are summarized in Table II (see Supporting Table 1 and Supporting Table 2 in the online version of this article). DISCUSSION Nearly all sizes of bronchoscopes in our study demonstrated reduced volume flow, increased resistance, and increased turbulence compared to similar outer-diameter TABLE II. Results of Computer-Aided Design Flow Analysis of 3.5 Rigid Bronchoscope With Inserted Telescope. Bronchoscope Operating Details Inlet Flow Rate (ml/s) Flow Rate to Patient (ml/s) Proximal End Leakage (ml/s) % Anesthesia Into Environment Normal setup % Open proximal end % Open proximal end % and suction port All available ports open % Data presented are volume flow rates for pressure-controlled ventilation in a 3.5 rigid bronchoscope during common surgical scenarios at sea level. These values are provided at an applied pressure of 20 cm H 2 O and lung pressure equal to 5 cm H 2 O positive end-expiratory pressure. This corresponds to a partially inflated pediatric lung. The values illustrate substantial leakage of anesthesia with any combination of open ports, possibly affecting the surgical staff. endotracheal tubes. We chose to compare rigid bronchoscopes to endotracheal tubes because endotracheal tubes are the gold standard device used during mechanical ventilation, providing smooth laminar flow. A bronchoscope with its other intended functions may never achieve this degree of laminar flow and reduced resistance, but it is helpful to make these comparisons to identify areas for future improvement for device development. One surprising finding is that the 3.5 rigid bronchoscope compares much less favorably than the 3.0. However, the 3.5 bronchoscope is longer and utilizes a larger diameter telescope, which increases airflow resistance. The short 3.5 bronchoscope (18.5 cm), which was not tested in our current study, would have better flow characteristics and would be a better choice in a child difficult to ventilate, although the length may be inadequate for the distal airway. The short 3.0 bronchoscope, with the 1.9-mm telescope, provided the least airway resistance and highest volume flow supporting Poiseuille s law (see Supporting Information, Appendix, in the online version of this article). This illustrates that the length of the device is linearly correlated with the resistance and flow rate. These studies indicate that using the shortest length and largest diameter bronchoscope appropriate for the child provides the least airway resistance. Another surprising finding was that ventilation modality variably affected the performance of identical devices. For instance, as seen in Figure 1, the 3.5 bronchoscope experienced the largest percent resistance increase compared to endotracheal tubes in pressurecontrolled ventilation. However, the 2.5 bronchoscope experienced the greatest percentage resistance increase 1943

5 in volume-controlled ventilation. Additionally, the resistance increases illustrated by volume-controlled ventilation were substantially higher than pressure-controlled ventilation. This information may be valuable for anesthesiologists when determining ventilation modalities based on particular surgical scenarios. One limitation of our study is we did not make direct measurements of dimensions but rather used the manufacturer s specifications. Lockhart and Elliot measured smaller inner diameters of bronchoscopes than reported values by the manufacturer. This would account for the increased airway resistance in bronchoscopes they reported. 13 Our calculated values of endotracheal tube airway resistance were less but comparable to direct measurements using manometry. 18 Additionally, using Poiseuille s law to calculate the resistance in the bronchoscope may have slightly under-represented the actual resistances (see Supporting Information, Appendix, in the online version of this article). Finally, small air leaks from any of the three proximal ports would further increase turbulence and device resistance. These limitations would mean that in physical implementation, there may be even less flow and higher turbulence than reported in this study. The advantage of CAD modeling software over direct manometry is the ability to identify airflow turbulence. This is an important consideration because laminar flow provides more efficient ventilation than turbulent flow. We demonstrate that rigid bronchoscopes exhibit increased turbulent flow compared to endotracheal tubes due to several design features and peripheral devices of the bronchoscope. In some cases airflow was mainly directed through the vents toward the tracheal wall rather distally toward the lungs. To our knowledge, there are no published studies evaluating airway turbulence in pediatric rigid bronchoscopes. Although not specifically examined, during spontaneous ventilation it appears that resistance is too great for gases to flow through the lumen of the bronchoscope and will follow the path of least resistance. In this case, the bronchoscope may serve solely as an oxygen and anesthetic delivery device, with the patient breathing around the bronchoscope. When respiratory effort ceases, inspiration can be achieved by applying positive pressure, but expiration will largely continue passively around the bronchoscope. A challenge presented with positive pressure ventilation during rigid bronchoscopy is the large air leak escaping between the outer wall of the bronchoscope and the trachea. This is further exacerbated by turbulent airflow being delivered through the side vents toward the tracheal wall rather than laminar flow being delivered toward the lungs. Increasing proximal pressure by manually bagging would tend to further increase turbulent flow. These findings suggest that a rigid bronchoscope is not well suited for positive pressure ventilation. In comparison, endotracheal tubes typically have a smaller air leak, due to conventionally using a slightly larger outer-diameter size, and deliver appropriately directed laminar flow. When necessary, a balloon cuff can eliminate an air leak altogether. Using a balloon 1944 cuff, similar to an endotracheal tube, on a bronchoscope may not be appropriate as this would limit gas flow on expiration, increasing the risk of hyperinflation and pneumothorax. A balloon cuff would also occlude the side vents on the bronchoscope where the majority of airflow is delivered. A concerning finding is the anesthetic agent released to the surrounding environment when the proximal ports are left unoccluded. This occurs frequently during endoscopic foreign body removal, when cleaning the telescope, and during instrumentation. Our findings are consistent with findings by Westphal et al., who found a high incidence of anesthetic side effects in surgeons performing rigid bronchoscopy. 19 The authors recommend using total intravenous anesthesia, thereby avoiding inhaled agents altogether. If an inhaled agent is used, rapid exchange of instruments, occlusion of all proximal ports, and holding ventilation when ports are open may reduce but not totally eliminate surgeon exposure. Use of the rubber stoppers during instrumentation may also direct flow toward the patient rather than the surrounding environment. Attaching the anesthesia tubing directly to the instrument port may improve flow when ventilation becomes challenging. These studies also demonstrate areas for design improvement including maximizing laminar flow, reducing resistance to flow, and reducing gas escape during instrumentation. A primary limitation of the rigid bronchoscope is that ventilation, instrumentation, and viewing cannot be done simultaneously. During viewing with the telescope in place, airway resistance increases and volumetric flow decreases substantially compared to endotracheal tubes. When the surgeon proceeds to instrumentation, the proximal port must be opened, at which point ventilation cannot be assisted for the patient. To improve ventilation, the Hopkins telescope must be removed and all ports occluded, which does not allow for adequate viewing of the airway. Apart from concerns raised regarding airflow dynamics, one questions the wisdom in designing a life-saving device that requires excessive assembly. CONCLUSION Most sizes of pediatric rigid bronchoscopes have increased airway resistance and turbulence compared to similar-sized outer-diameter endotracheal tubes. In many instances, airflow was directed mainly through the vents rather than down the distal end toward the lungs. These characteristics of the rigid bronchoscope may explain why some patients are difficult to ventilate during rigid bronchoscopy. Further technological innovation is necessary to improve the ventilation, safety, and effectiveness of rigid bronchoscopes. BIBLIOGRAPHY 1. Becker HD. Bronchoscopy: the past, the present, and the future. Clin Chest Med 2010;31:1 18, Table of Contents. 2. Gallagher TQ, Hartnick CJ. Direct laryngoscopy and rigid bronchoscopy. Adv Otorhinolaryngol 2012;73: Yang CC, Lee KS. Comparison of direct vision and video imaging during bronchoscopy for pediatric airway foreign bodies. Ear Nose Throat J 2003;82:

6 4. Ayers ML, Beamis JF Jr. Rigid bronchoscopy in the twenty-first century. Clin Chest Med 2001;22: Kim IA, Shapiro N, Bhattacharyya N. The national cost burden of bronchial foreign body aspiration in children. Laryngoscope 2015;125: Fidkowski CW, Zheng H, Firth PG. The anesthetic considerations of tracheobronchial foreign bodies in children: a literature review of 12,979 cases. Anesth Analg 2010;111: Chen LH, Zhang X, Li SQ, Liu YQ, Zhang TY, Wu JZ. The risk factors for hypoxemia in children younger than 5 years old undergoing rigid bronchoscopy for foreign body removal. Anesth Analg 2009;109: Aydogan LB, Tuncer U, Soylu L, Kiroglu M, Ozsahinoglu C. Rigid bronchoscopy for the suspicion of foreign body in the airway. Int J Pediatr Otorhinolaryngol 2006;70: Latifi X, Mustafa A, Hysenaj Q. Rigid tracheobronchoscopy in the management of airway foreign bodies: 10 years experience in Kosovo. Int J Pediatr Otorhinolaryngol 2006;70: Cavel O, Bergeron M, Garel L, Arcand P, Froehlich P. Questioning the legitimacy of rigid bronchoscopy as a tool for establishing the diagnosis of a bronchial foreign body. Int J Pediatr Otorhinolaryngol 2012;76: Litman RS, Ponnuri J, Trogan I. Anesthesia for tracheal or bronchial foreign body removal in children: an analysis of ninety-four cases. Anesth Analg 2000;91: , TOC. 12. Shen X, Hu CB, Ye M, Chen YZ. Propofol-remifentanil intravenous anesthesia and spontaneous ventilation for airway foreign body removal in children with preoperative respiratory impairment. Paediatr Anaesth 2012;22: Lockhart CH, Elliot JL. Potential hazards of pediatric rigid bronchoscopy. J Pediatr Surg 1984;19: Fluid flow simulation: advanced boundary Cartesian meshing technology in SOLIDWORKS flow simulation. Dassault Systemes Corp. website Available at: Accessed February Fluid flow simulation: numerical basis of CAD-embedded CFD. Dassault Systemes Corp. website Available at: sw/docs/flow_basis_of_cad_embedded_cfd_whitepaper.pdf. Accessed February Centers for Disease Control and Prevention. Growth charts. Available at: Updated September 9, Accessed September Thorsteinsson A, Larsson A, Jonmarker C, Werner O. Pressure-volume relations of the respiratory system in healthy children. Am J Respir Crit Care Med 1994;150: Hentschel R, Buntzel J, Guttmann J, Schumann S. Endotracheal tube resistance and inertance in a model of mechanical ventilation of newborns and small infants-the impact of ventilator settings on tracheal pressure swings. Physiol Meas 2011;32: Westphal K, Lischke V, Aybeck T, Kessler P. Exposure of the pediatric surgeon to inhalation-anesthetic during pediatric bronchoscopy procedures [in German]. Pneumologie 1997;51: