FLEXIBLY DIVIDED DOUBLE LUMEN ENDOTRACHEAL TUBE

Viren Sunil Kalsekar, Joy Lee, Jenny Qiu, Nicholas Ragosta, Javier Ramos, David Taylor Massachusetts Institute of Technology Cambridge, MA, USA

ABSTRACT Endotracheal tubes (ETT), are used to maintain patient airway access during surgical procedures. The design presented here addresses a current need of the gold standard ETT by allowing for the passage of a larger bronchoscope through the inner lumen of the ETT while keeping outer dimensions constant. This design incorporates a flexible membrane with a wavy geometry to ensure a return to the neutral axis, thus allowing for even ventilation of both lungs. Flow tests simulating surgical ventilation conditions were conducted. These results confirmed that adequate air flow could be obtained independent of any pressure differences across the membrane. Tests of the prototype also confirmed that maximum instrumentation diameter was increased through this design when compared to a commercially available ETT with identical outer dimensions (35 Fr). Finite element analysis was conducted to determine the necessary thickness of the membrane for a safe design. One key design feature of the membrane is that its state of maximum deflection (2.1mm) does not significantly hinder airflow through the smaller lumen. Tests to further evaluate the functional requirements of safety and ease of use are currently under way. Methods of manufacturing this design are also under investigation. Keywords: endotracheal tube, flexible membrane, airway access, neutral axis INTRODUCTION Endotracheal tubes (ETT) are used to maintain patient airway access during surgical procedures. During thoracic surgery it is often necessary to ventilate the lungs separately. This allows one lung to be ventilated, while the other may be deflated. [1] Predicts the number of US thoracic surgeries in 2013 to be 38,500. Most of these surgeries use a double lumen endotracheal tube (DLT). The size of the DLT is selected for each patient based on their tracheal diameter; the outer diameter of the DLT must be small enough to fit in the patient, but large enough to have a small, easily sealed, gap between the tube and the trachea. During the procedure a bronchoscope is frequently used. It can serve as a guide along which the ETT is inserted down the trachea, it provides visual imaging to confirm correct positioning, and it can be used to remove large secretions. Larger bronchoscopes provide improved image resolution and suctioning.

Figure 1 shows a distribution of patients by tracheal size. A tracheal diameter of 14mm or less corresponds to a recommended DLT size of 35 French (Fr) [2]. While using a 35Fr DLT, the use of a full-sized (4.9mm) bronchoscope is not possible [3]. A tracheal size of 15-16mm corresponds to a recommended DLT size of 37Fr [2], which makes a full-sized bronchoscope very difficult to use [3]. Impossible Population Hard Easy

74%

8

12

16

20

24

Size of Trachea (mm)
Figure 1: Ability to use a full sized (4.9mm) bronchoscope, data based on [2, 3]

Using the patient statistics of [2] it was estimated that in 26% of patients require a 35Fr or 37Fr DLT and therefore during about 26% of thoracic surgeries full-sized bronchoscopes cannot be used. Using the surgery estimates of [1], this corresponds to an estimated 10,000 underserved patients per year. The modified double lumen ETT presented here aims to make the use of a full-sized bronchoscope possible even in patients who require a 35Fr sized-DLT. PROBLEM STATEMENT An estimated 10,000 patients per year require the use of a double lumen tube smaller than 39Fr, preventing the easy use of a full-sized bronchoscope. Current double lumen tubes are divided by a rigid wall which limits the size of instrument that can be inserted. To better serve these 10,000 patients per year, it was imperative to develop a new ETT that would allow the use of a full-sized bronchoscope in even the smallest patients. Such an ETT, would also need to be safe not alter current surgical procedures.

1

PRIOR ART Current designs of the ETT for ventilation and deflation mainly focus on two major concepts: the double lumen ETT and the single lumen ETT with bronchial blockers. The double lumen ETT involves the use of two separate tubes (lumens) to allow for independent ventilation and deflation of either lung (Figure 2: DLT Placement and Cross-Section). Because these two lumens together form a larger diameter tube that must fit within a patient’s trachea, each lumen is less than half the diameter of the full device. This constricts the use of larger, higher resolution bronchoscopes particularly in smaller female patients. Although the DLT is widely used due to the level of ventilation control its use allows, it continues to pose difficulty for high-resolution bronchoscopy.

Figure 3: Single Lumen ETT and Bronchial Blocker

Figure 2: DLT Placement and Cross-Section

There has been previous work in solving the problem of space limitations in double lumen tubes. A patent filed by Peter T. Walling in 1978 (US4233984) [5] describes a double lumen endotracheal tube that has a non-rigid division separating the two lumens of the tube. This non-rigid division would act as a flexible membrane, and be capable of flopping from one side of the main tube inner wall to the other (Figure 4: Motion of Flexible Division Described by US4233984). This movement would allow either lumen to use the full inner diameter of the main tube during ventilation and bronchoscopy. A limitation of this design is that it lacks a neutral state for its flexible division. This means that full occlusion of one lumen is possible and that even ventilation of both lungs may not always occur. Additionally, the manufacturing processes necessary to construct such a device were not discussed and the inventor was unable to produce a prototype.

Single lumen ETTs with bronchial blockers provide general ventilation of both lungs and single lung isolation. First, a single lumen ETT is inserted into the trachea. Next, a catheter balloon is extended beyond the lumen to inflate and seal off one of the lungs (Figure 3: Single Lumen ETT and Bronchial Blocker). Unfortunately this device is much less reliable than the DLT because the blocker is prone to dislodging when the patient coughs or is rolled onto his or her side. Additionally, it takes considerable skill to accurately place the bronchial blocker. On average anesthesiologists spend 3 minutes and 30 seconds placing a single tube and bronchial blocker; this is significantly longer than the 2 minutes taken to place a double lumen tube [4]. Because bronchial blockers are unreliable and difficult to use, they are employed almost exclusively in patients that are too small for the use of a double lumen tube [6].

Figure 4: Motion of Flexible Division Described by US4233984

REQUIREMENTS The functional requirements for a redesigned ETT were as follows, it must: 1. provide single lung isolation (the complete separation of one lung from the other for the purpose of thoracic surgeries); 2. allow the easy passage of a 4.9mm bronchoscope to either lung during surgery; 3. not alter current surgical procedure, it must therefore: a) be easy to use;

2

4.

b) allow air to escape from the deflating lung in a time that is comparable to a DLT of equivalent size; c) have approximately equal flow resistance in each lumen, independent of the pressure in the other lumen; & be safe to use.

(Note: We need the same outer diameter for the functional requirements) ������������������������������������ ������������ ������������������������������������������ = Where, b = (D-d)/2 and h=d. In case of the modified double lumen tube the term is 12 negligible because of the small thickness of the membrane. Using equation (1) for the moment of inertia and equating them for the two cross sections we get an expression for the inner diameter of the modified double lumen tube. We are targeting the population having trachea size 14-15 mm using the 35 Fr (OD=11.67mm) ETT. The smallest wall thickness of the DLT currently sold is 0.8 mm [8]. For a corresponding thickness of the new modified DLT we have ID= 8.77 mm with a thickness t=1.44 mm. Hence, we choose a thickness of 1.5 mm for our modified DLT with the same hardness (Shore A 85) as of the current DLT design. [9] MEMBRANE Balancing the three broad functional requirements of giving access to a full sized bronchoscope, integration into current surgical procedures, and patient safety, required careful optimization of the membrane’s stiffness and deflection. To best give access to the bronchoscope, the membrane was required to deflect away from the neutral axis. The larger the allowable deflection, the easier the bronchoscope would be able to pass. However if the allowable deflection was too large, it would interfere with the flow properties on the other side of the membrane. It was decided to limit the deflections of the membrane by having the membrane act as a spring that would resist deformation past the critical threshold. A simple flexible material has a resistance force-displacement relationship as depicted in Figure 6. While the magnitude of the resistance for membrane deformation is scalable with material and material thickness, the shape is not. There could be no jump in the required resistance force at the critical threshold. This was not desired as friction scales with the normal force, which equaled here the required membrane deflection force. It was therefore determined the optimal design would be a two-staged spring, as in Figure 6. In this case, low forces are required up to the critical threshold and then high forces are required past that point.
������ℎ ������(������ −������ ) 64

+

������ℎ 12

(1)

DESIGN Various strategies for achieving lung isolations were considered before a continuous flexible divider along the length of the tube was selected. It was determined that this design incorporating a complete physical separation of the two lungs and airway access minimized the risk of complications involving accidental reinflation of a deflated lung. The design of this flexible divider involves a wavy geometry that allows the membrane to naturally remain along a neutral axis but also expand to allow for instrument access down one side of the divided tube. There are several key interfaces of this design, namely at the ends of the tube with the wavy membrane cross section. These interfaces include that of the y-junction with the wavy membrane, the membrane with the bottom of the opening at the lower portion of the tube, and the tapered end which has a different cross sectional diameter than the rest of the tube. The overall dimensions of this design were selected to match the outer dimensions of a commercially available 35Fr DLT to address the segment of the population with the greatest need for larger instrument access while minimizing disruptions to current surgical procedure. OUTER WALL The current ETT uses PVC for the outer tube including Mallinckrodt, Portex and Sheridan [7]. As the material is approved by the FDA through mandatory tests and it is working with good results we decided to use PVC for the outer tube of the ETT. The Single Lumen Tube is the closest to our design of the modified ETT but we need to incorporate the same stiffness as in a double lumen tube in our design. The stiffness of the outer tube is dependent on the Young’s modulus and the geometry of the cross section. A first order analysis of the moment of inertia of the cross section: d

b

D

Figure 5: Cross section of a double lumen tube

3

Figure 7 show that as the lumen eclipses the log of the resistance value increases significantly. Normal Force Required Flexible Membrane Spring 2-Staged Spring
X

5
Log of Resistance Value [log(cm/(L/s)^2)]

4 3 2 1 0 0 -1 2 4

b

-6

-4

-2

0
Critical thresholds

2

4

6

Membrane Displacement (X) from Neutral Axis (mm)
Figure 6: Desired Force-Displacement Curve for DLT Divider

6

b, Cresent Width at Center (mm) Flow Experiment through an Eclipsing 9.5mm ID tube CFD of Flow Through Eclipsing Lumen Flow Through Tracheal Lumen of Portex 35F DLT

A loose membrane that would tighten at the critical threshold would meet the requirement of a two-staged spring. In fact, it would minimize friction because it would require no normal force before the critical threshold. However, in order to integrate into current surgical procedures, the flow resistance on each side of the membrane without a bronchoscope present should be approximately equal. To achieve this, the membrane was designed to return to a symmetric position when there was not a bronchoscope in either lumen. The first stage spring must therefore have a non-zero spring constant, eliminating the loose membrane option. This small but non-zero spring force in the first stage of membrane deflection can also be seen in Figure 6. CRITICAL THRESHOLDS FOR MEMBRANE DEFLECTION The smallest critical threshold for the membrane was set by how the bronchoscope interacts with the ETT. Using current double lumen tubes as a reference for how much clearance is required for an ‘easy’ fit, as in Table 1, shows that a clearance distance of 5.6mm is best. A 35F size with 1/16 inch wall thickness has an inner diameter of about 9.5mm, translating into a minimum deflection of X=0.85mm to achieve the same looseness as the 41Fr.
Table 1: Ease of Bronchoscopy by DLT Size

Figure 7: Resistance Increases Dramatically as a Lumen Narrows

The maximum critical point was set before the jump in resistance values observed in the experimental data at b=2.7mm which for an infinitely thin membrane is a displacement value of X= 2.1mm. To summarize therefore the membrane was required to: 1. return to the neutral axis, 2. displace a minimum of 0.85mm for ease of bronchoscopy, & 3. displace of maximum of 2.1mm to minimize flow resistance. Based on structural analysis below it was determined that the membrane equivalent thickness needed to be 1.4mm peak-topeak. Therefore the membrane’s deflection would ideally be 0.85mm+1.4mm. Unfortunately this would violate the flow requirements. Since the membrane was neutrally stable, it would not deform past the contact point with the bronchoscope and so the extra room to create an easy fit was not necessary. The minimum fit distance would therefore be 1.65mm of deflection. To allow for tolerances, a target deflection of 1.9mm was chosen. This deflection corresponds to a b value of 2.9mm and an X value of 0.5mm (note that the membrane no longer has infinite thinness). MEMBRANE DESIGN A membrane curved-wave design (Figure 8) was developed to comply with the three geometrical requirements established from the flow test. The membrane was analyzed in the three most relevant failure mode configurations: deformation due to bronchoscope passage, deformation due to ventilation air pressure, and bronchoscope puncture. FEA simulations were performed to analyze membrane failure modes. First order models were developed to verify FEA results.

DLT Size (French) 35 37 39 41

ID of each lumen (mm) [10] 4.9 5.2 5.5 5.6

Ease of use with 4.9mm Bronchoscope [3] Impossible Difficult, requires lubrication Moderately easy Easy

The largest critical threshold for the membrane was set by the flow resistance that the other side of the membrane would experience. To determine the resistance to air that and eclipsing lumen provides, CFD simulations and experiments were conducted for a 33cm long tube with an un-eclipsed inner diameter of 9.5mm. The flow experiment was conducted using the apparatus described and pictured in Annex A. The results in

4

1 2
Figure 8: Cross Section of ETT with curve-wave membrane design

CURVED-WAVE MEMBRANE DESIGN A curved-wave configuration was chosen due to its inherent ability to comply with deformation forces while still maintaining the ability to spring back to the neutral position. Symmetry and overall simplicity allowed for analysis using first order models. Three main parameters determined the compliance mechanics of the membrane: material properties, thickness, and number of waves. Flexible, medical grade, Polyvinyl Chloride of Shore hardness 85A was assumed for modeling purposes due to its wide availability for medical applications, low cost, and adaptability to various manufacturing processes.
Table 2: Properties and Dimensions of Flexible Membrane

MEMBRANE DEFORMATION DUE TO

Figure 9: Membrane deformed due to bronchoscope passage. Contact points 1 and 2 are shown

VENTILATOR PRESSURE Typical range of ventilator pressures during lung isolation procedures is 5 − 25 ������������������2 ������. The membrane was designed to sustain a pressure of 50 ������������������2 ������.

Membrane Properties
Material PVC (Shore 85A) Wave Periods 4 Thickness (mm) 0.5 Length (mm) 9.4 Corner Fillets (mm) 0.5

MEMBRANE DEFORMATION DUE TO BRONCHOSCOPE PASSAGE A reference bronchoscope of diameter 5mm was used to virtually deform the membrane. An FEA simulation showed that the membrane will have two contact points with the bronchoscope (Figure 9). The FEA showed that the force at contact point 1 is ������1 ≈ 0.012 ������/������ and that the force at contact point 2 is ������2 ≈ 0.002 ������/������������. So, ������1 ≈ 6 × ������2. Maximum membrane stress was found to be 0.4 ������������������, well below the 10 ������������������ yield strength [11] of typical flexible PVC. The FEA predicted a maximum membrane deflection of 1.1 ������������. Results show that the passage of a 5 mm bronchoscope through the presented cross-section will not cause failure of the membrane with a predicted safety factor of 25.

Figure 10: FEA simulation results of membrane deformed to ventilator pressure of 50 cmH2O

A maximum stress of 0.6 ������������������ was predicted by the FEA simulation. A first order analysis modeling the deformed shape of the membrane as a 1D planar deformation of a curved beam gives a maximum stress of 0.3 ������������������. The first order estimate confirms the order of magnitude of the membrane deformation calculated by the FEA simulation. The FEA predicts a maximum membrane deflection of 1 ������������. At 200 200 ������������������2 ������ FEA simulations predict that the membrane will still have a safety factor of ~2. The results show that the ventilator pressure will not cause failure of the membrane with a safety factor of 17.

5

MEMBRANE FAILURE DUE TO BRONCHOSCOPE PUNCTURE Bronchoscope puncture is the most critical failure mode given the localized nature of the forces in action and their unpredictability due to the possible degrees of freedom of these forces. Puncture and tear phenomena are also strongly dependent on material properties. For analyzing the distribution of puncture stresses on the membrane, a bronchoscope diameter of 5 ������������ and a maximum force of 5 ������ were assumed. In terms of stress distribution, the worst possible configuration is a bronchoscope pushed normal to the surface of the membrane. An FEA simulation with the aforementioned configuration and a flat membrane was performed. The simulation showed a maximum stress concentration of 2 ������������������. A linear, first order estimate, of a flat plate showed a maximum stress of ~9 ������������������, which confirms the order of magnitude of the FEA simulation. As the analysis demonstrates, membrane puncture is the most critical failure mode. The FEA simulation shows that our design will have a safety factor of 5 but experimental results could further confirm the accuracy of the FEA simulation.

modified from a wavy profile to a flat profile in order to better accommodate assembly. MEMBRANE AND Y-JUNCTION INTERFACE The first critical interface was located at the base of the Yjunction piece and at the beginning of the membrane extrusion. The main requirement in designing the first interface was to create a fluid continuation between the surfaces in order to maintain lung separation and prevent unintentional air pathways. The membrane was modified so that there could be a conversion between the wavy component in the main assembly and the flat portion required in the y-junction. In order to ensure a secure connection between the membrane and the y-junction, there were two suggestions to the interface behavior. The first suggestion was that the wavy membrane in the partition conforms from the wave to a flat orientation as shown in Figure 12.

Figure 12: Connection interface between membrane and Yjunction

Figure 11: Membrane deformed due to 5 mm diameter bronchoscope force of 5 N. Table 3: FEA Results on Membrane Mechanics
Safety Factor (Yield Strength: 10 MPa)

Failure Mode

Maximum Displacement (mm)

Maximum Stress (MPa)

Ventilator Pressure Bronchoscope Passage Deformation Bronchoscope Puncture

1 1.1 2.8

0.6 0.4 2

17 25 5

If the membrane’s profile converted from a wavy form to a straight form, then the flattened portion could be compressed by the Y connector allowing for a secure and airtight connection between the two surfaces. The same could be achieved with the alternate situation in which the Y-junction had a profile that converts from a straight to a wavy form that matches the wavy membrane profile. However, in both cases, the two surfaces need a joining method in order to securely attach the extrusion to the injection molded y-junction. Several applicable joining techniques are airtight joining with an adhesive, solvent joining, and heat joining. With all considerations, the interface should adhere to all FDA approved methods. MEMBRANE AND OPENING INTERFACE The second critical interface is located at opening of the trachea exit close to the end of the ETT. The requirement for designing the second interface is similar to the first interface: creating a continuation between the surfaces in order to maintain lung separation and prevent unintentional air pathways. However, the tracheal membrane interface requires a more complicated solution than the first interface.

CRITICAL INTERFACES To incorporate the design of the flexible membrane partition with the rest of the assembly, two critical locations required extensive: the area between the y-junction near the top of the double lumen ETT and the main assembly as well as the area between the opening tracheal interface and the flexible membrane assembly. In both areas, the cross section must be

6

of manufacturing technique equivalents were used to create a model of the key design features. The first prototyping technique explored as shown in Annex B was inserting heat shrink tubing within soft PVC tubing to artificially create the membrane partition. The first prototype was able to simulate the possibility of incorporating two separate extrusions in the same assembly. The first extrusion would become the outer casing of the ETT and the second extrusion would contain an alternate material and the membrane partition. The second extrusion would be made to adhere to the side of the first extrusion as shown in Figure 15

Figure 13: Membrane and Tracheal exit interface

As shown in Figure 13, the membrane needed to be able to shift from a sinusoidal curve to a flattened plane. However, the location the tracheal exit required a reorientation of the natural position of the ETT. The pathway between the Y connection and the tracheal exit should be as straight as possible in order to deter any snagging between the bronchoscope and the inside of the ETT. The recommendation was that the ETT should be manufactured through extrusion as much as possible. As shown in Figure 14, the wavy membrane was straightened toward the end of the tracheal exit. In order to achieve the end geometry, the ETT was extruded in the proposed cross section geometry. At the end of the ETT, the tracheal exit was created by removing half of the extrusion. The wavy membrane in the half that was left behind was heated and reformed into a smooth geometry. The risk associated with the first recommendation is reformation end geometry failures.

Figure 15: The cross section of attached extrusions

The first prototype proved the robustness of the design because it allowed bronchoscope access without tearing. The second prototyping technique consisted of thermoforming the wavy partition. In order to demonstrate the possibility of creating a wavy partition, extensive testing on the heating properties of the material was conducted. It was shown that with PET, the material had to have even heating that slowly ramped up to a steady temperature. The PET wavy partitions were then subsequently heated in a controlled oven as shown in Annex B. The resultant product was that the creation of a permanent wavy form was entirely feasible.

Figure 14: Removal of half of the extrusion. Both membrane and tube are made of the PVC

A second recommendation was for a case in which the ETT’s membrane would be made from alternative materials as shown in Figure 13. The membrane ended at the tracheal exit. However, the problem concerned with modifying the membrane so that it could orientate from a sinusoidal geometry to a planar geometry to ensure easy attachment. In the area beyond the tracheal exit, the rest of the tube would be formed by alternative manufacturing techniques such as plastic dipping formation. PROTOTYPE CONSTRUCTION AND MANUFACTURING In order to fully ascertain the validity of the design, a prototype was manufactured for testing purposes. A multitude

The third prototyping technique used was to create a fully 3D printed model of the assembly. The 3D printed model was for visual purposes only. However, the model presented that the design of the ETT was feasible and the incorporation of different materials was possible. Several of the various manufacturing techniques that have been investigated for obtaining the cross section of a wavy membrane within an outer tube include multiple methods of extrusion. One option of extrusion is to use a single material and extrude the entire cross section at once. Limitations of this method include the material used since with a single extrusion, a material that not only allows for proper movement of the membrane, but enough stiffness to prevent collapse of the outer wall must be selected. A further limitation of this process is the range of thickness of the geometry. The thinnest part of the extrusion must be at least 1/5th the thickness of the thickest part. The other option available with extrusions is to use multiple

7

extrusions to produce the separate components of the crosssectional geometry. With this option, it is possible to obtain thinner membranes formed with more flexibility as relates to material selection. The membrane can be extruded in either a sheet form or in a tubular form. However, issues of joining the membrane to the outer wall exist for either option. Several options for joining were investigated. These forms of plastic welding each have their advantages and disadvantages, allowing for different capabilities with joining. One potential welding process is solvent welding, which uses chemicals to weld two parts together. While the two parts are joined without flash, one potential disadvantage is that the area surrounding the joint can be softer if any of the solvent remains [14]. Another issue is that many different materials are not compatible and sometimes cannot be joined together [14]. While this is not the case for PVC and acrylics, the materials selection does become more limited and is a potential disadvantage [14]. Another possible welding option is heat welding. Though this method is easy to use with PVC, joining dissimilar materials is difficult with this technique as well [13]. Ultrasonic welding uses ultrasound to heat the plastic to create the weld, and while it is a fast and well tested technique in most plastics, one difficulty is that with this process, joining dissimilar materials is not easily accomplished [14]. With laser welding, one advantage is that additional material does not have to be added to join the parts, so biocompatibility issues are diminished. Additional benefits include the precision with which the joints can be welded together; one disadvantage is that the parts must be positioned precisely to ensure a good weld [12]. With laser welding, internal geometries and curvatures can also be welded [13]. One additional type of joining process that was investigated is adhesives. With the selection of the correct adhesive, different materials can be joined together flexibly and with little structural deterioration. However, one significant disadvantage to using adhesives is the biocompatibility of many of the adhesives available. TESTING AND VALIDATION Several different tests were conducted to analyze the performance of the design. Flow testing at the hospital was conducted with the works-like prototype to analyze the ability to ventilate both lungs with and without a pressure difference across the membrane. Flow tests using two ventilators indicated that the ability to adequately ventilate both lungs was independent of any pressure differentials across the membrane. As shown in Figure 17 airflow through each side of the tube was maintained to within 0.1L volumetric flow difference for each scenario. Membrane testing was conducted through finite element analysis to determine if the designed material, geometry, and thickness would be able to withstand the necessary applied forces. All membrane testing discussed in this paper was done through FEA and is more fully discussed in the membrane design section. Key results showed that an A85 shore PVC

extruded membrane can withstand maximum ventilation pressures of 50 cm H2O (factor of safety of 20), bronchoscope insertion forces (factor of safety 20), and puncture forces of 5 N (factor of safety 5).

Normal Force Required

Flexible Membrane Spring 2-Staged Spring Wavy Membrane

-6

-4

-2

0

2

4

6

Membrane Displacement (dX) from Neutral Axis (mm)

Figure 16: Curved-Wave Membrane's Performance vs. Desired Performance

The deflection performance of the curved-wave membrane was simulated by FEA as a spring in tension. The linear stretching was converted to an equivalent dX and normal force value with basic geometric relationships. Figure 16 compares these results to the desired performance outlined by the flow analysis. While the results do improve on the flat membrane, most of the difference is in the scale of the force, not the shape of the curve. Further work needs to be done to either optimize the shape of the curved membrane or move back to a flat and stretchy membrane divider. Future membrane performance tests will be conducted on a prototype constructed using the final selected material and manufacturing process. These tests are expected to be a key component of the membrane design validation.

Figure 17: Ventilator Flow Data through Prototype

Bronchoscope insertion testing was conducted to confirm the ability to pass a larger diameter instrument through a 1X prototype. A 4.9 mm bronchoscope model was successfully

8

inserted through both lumens of the 1X prototype, as shown in Figure 18 and Figure 19.

functional requirements of safety and ease of use are currently underway. Once a scale prototype of our device is fabricated using the manufacturing processes outlined previously, we will further evaluate the behavior of the flexible membrane. Tests to be conducted include characterizing frictional forces acting on the membrane during bronchoscope insertion and puncture testing of the membrane. These tests will allow us to finalize our membrane material and geometry and evaluate the overall safety of our device. A recent patent application filed by representatives from Mallinckrodt Medical Inc, a respiratory pharmaceutical manufacturer, was found regarding the possibility of having a flexible membrane within a dual lumen endotracheal tube. The claims filed within the application were strictly concerned with the idea of the various configurations the membrane could have. Although the application includes a variation of the team design, the patent was limited in its description of manufacturing processes which should permit future work on the device [JQ2]. Continuing research into manufacturing practices should be a direction that the design shall follow in the future. ACKNOWLEDGMENTS The authors would like to thank the 2.750 staff for their advice and support throughout the design and fabrication stages of the project. They would also like to thank members of the PERG lab who provided insight into various aspects of the fabrication processes. They would also like to thank Barbara Hughey for her help with obtaining equipment for testing. The team would also like to thank Dave Dow and Pat McAtamney in the LMP for their insight into and support with thermoforming.

Figure 18: Portex 35F DLT with a simulation 4.9mm bronchoscope maximally inserted

Figure 19: 35F prototype shows easy passage for 4.9mm bronchoscope

As Figure 18 and Figure 19 indicate, the presented design allows for a larger diameter instrument to pass through than can be passed through a commercially available ETT with identical outer dimensions. By allowing for larger instrument access while keeping outer dimensions constant, the presented design better serves the patient population and increases the percentage of patients who can be intubated with the use of a 4.9 mm bronchoscope as illustrated in Figure 20.

Figure 20: Increased access of 4.9 mm bronchoscope using presented design.

CONCLUSION AND FUTURE WORK In summary, the design presented improves on the current gold standard of intubation and lung isolation by employing a flexible division to separate the two lumens. This flexible membrane was designed such that it would easily allow for the passage of a bronchoscope down either lumen (membrane deflections of 0.86mm). At the same time, this membrane was designed so that its state of maximum deflection (2.1mm) would not significantly hinder airflow through the smaller lumen. Additionally, the partition maintains a neutral position at the center of the ETT when a bronchoscope is not being used, this central neutral position ensures even ventilation to both lungs when necessary. Our design achieved the functional requirements previously outlined of: single lung isolation, bronchoscope access to both lungs and ventilation capabilities similar to existing double lumen ETTs. Testing to evaluate the

9

REFERENCES [1] US Surgical Procedures Volumes #A606. Medtech Insight. February 2007. [2] J. B. Brodsky, A. Macario, and J. B. D. Mark, “Tracheal diameter predicts double-lumen tube size: a method for selecting left double-lumen tubes,” Anesthesia and analgesia, vol. 82, no. 4, pp. 861–864, 1996. [3] J. Benumof and C. A. H. M.D, Benumof’s Airway Management: Principles and Practice. Elsevier Health Sciences, 2007. [4] Campos, J. H. (2003). An update on bronchial blockers during lung seperation techniques in adults. Anesth Analg 2003; 97:1266-1274. [5] Walling, P.T. (1978). “Respiratory Ventilating Device.” US. Patent No. 4233984. Dallas, TX: U.S. [6] Hammer GB, Fitzmaurice BG, Brodsky JB. Methods for single-lung ventilation in pediatric patients. Anesth Analg 1999; 89: 1426 –1429 [7] Sheridan Endotracheal tube brochure: www.tradekey.com/brochure/196407-7241573-3/hudsonrci.pdf

[8] Portex endotracheal tubes sizing and dimensions http://www.smithsmedical.com/userfiles/trachealtubechart.pdf [9] Viva Sight control for Lung Isolation http://www.etview.com/sites/ETView/UserContent/files/pdf/vivasight_brochure_for-net.pdf [10] D. Amar, D. P. Desiderio, P. M. Heerdt, A. C. Kolker, H. Zhang, and H. T. Thaler, “Practice Patterns in Choice of Left Double-Lumen Tube Size for Thoracic Surgery,” Anesth Analg, vol. 106, no. 2, pp. 379–383, Feb. 2008. [11] Titow, W.V. 1984. PVC Technology. Elsevier Applied Science. London. Pp. 1186. [12] Dawes, C., 1992, Laser Welding: A Practical Guide. Abington Publishing Cambridge, England 1992. [13] Grewells, D., Benatar, A., “Welding of Plastics: Fundamentals and New Developments” Intern. Polymer Processing, 22 (1), pp 43-60. [14] Troughton, M., 2008, Handbook of Plastics Joining: A Practical Guide. William Andrew Inc., Norwich, New York.

10

ANNEX A DETAILS OF FLOWS-LIKE EXPERIMENTAL SETUP The experimental setup attempted to use volume measurements of water to correlate volume of air passed through the flow testing device. This assumed the incompressibility of air. Since the pressure variation during the experiment were small (225 Pa), this assumption was reasonable. Observed flow velocities ranged from 3.9 to 18.6m/s. Using a hydraulic diameter definition of D=4*Area/Perimeter, the equivalent Reynolds number was found to range from 758 to 5723. The ‘constant’ pressure ranged from 1.1kPa to 1.4kPa. The resistance was calculated using the flow equation ������ = ������������2 .

Pressure level of water= pressure of air

Direction of Air Flow

H2O

b AIR

Volume of water = volume of air displaced
Figure 22: Experimental Layout Figure 21: Apparatus for controlling the eclipsing of a tube

11

Constant Pressure Source

Water

33cm 9.5mmID lumen with controlled eclipse

Air

Volume Passed Through Device

Figure 23: Experimental Setup

12

ANNEX B PROTOTYPING A CURVED-WAVE

13