| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
|
|
|
|||||||
|
|||||||||
RESEARCH-ARTICLE
1Department of Biomedical Engineering and 2Department of Physiology, University of Wisconsin, Madison, Wisconsin
Address for reprint requests and other correspondence: A. J. Lokuta, Dept. of Physiology, Univ. of Wisconsin School of Medicine and Public Health, 1300 Univ. Ave., Madison, WI 53706 (e-mail: ajlokuta{at}physiology.wisc.edu).
Abstract
Respiratory mechanics is a difficult topic for instructors and students alike. Existing respiratory mechanics models are limited in their abilities to demonstrate any effects of rib cage movement on alveolar and intrapleural pressures. We developed a model that can be used in both large and small classroom settings. This model contains digital pressure displays and computer integration for real-time demonstration of pressure changes that correspond to the different phases of breathing. Moving the simulated diaphragm and rib cage causes a volume change that results in pressure changes visible on the digital sensors and computer display. Device testing confirmed the model's ability to accurately demonstrate pressure changes in proportion to physiological values. Classroom testing in 369 surveyed students showed improved understanding of respiratory concepts (P < 0.05). We conclude that our respiratory mechanics model is a valuable instructional tool and provide detailed instructions for those who would like to create their own.
Key words: respiratory physiology; physiology demonstration; pressure changes; classroom tool; student survey
OUR GOAL was to design and build a mechanical model that would improve student understanding of human respiratory mechanics. In particular, we sought to develop a model that demonstrates pressure changes in alveolar and intrapleural spaces with breathing as well as the three-dimensional expansion of the thoracic cavity by the rib cage and diaphragm. Although simple homemade models and basic commercial Plexiglas lung models are available (9), they have short life spans and parts that are difficult to replace. Also, most models do not display pulmonary pressures, making it difficult for students to visualize the forces driving gas exchange between the lungs and atmosphere (1, 5, 6). Other models do visualize the pressure changes using analog means, but the models are not interfaced with a computer or are not visually representative of human anatomy (2, 4). Furthermore, no currently available physical models illustrate the expansion of the rib cage. Although most of the lung's volume change is due to contractions of the diaphragm, rib cage movement may contribute between 5% and 42% of the lung's total volume change (3).
Design, Fabrication, and Cost
Design considerations. A model for teaching respiratory mechanics should contain both intrapleural and alveolar pressure displays to demonstrate pressure relationships during inspiration and expiration. To accommodate different classroom settings, the model should be functional in a small classroom as well as a large lecture hall. Because document cameras are frequently used in lecture halls to present information to students, the device must fit under a typical 13 × 17-in. document camera. To maximize its usefulness, the device should be compatible with computer display software and be operable by a single user. The container housing the lungs should be transparent so that the inner components of the model are visible. To allow for transport, the device should weigh no more than 20 lb. One of the major concerns with previous models is the difficulty of replacing components. Therefore, components under frequent stress should be durable and easily replaceable.
Mechanical design. The respiratory mechanics model we designed consists of a sealed transparent chamber in which diaphragm movement can be simulated using a piston, and rib cage movements can be simulated with elastic membranes, to inflate and deflate balloons representing the lungs (Fig. 1). The container, which corresponds to the thoracic cavity, was constructed of transparent polycarbonate to allow a clear view of the lungs. Polycarbonate was chosen over acrylic and other transparent materials for ease of machining. The container was designed as a rectangular box (7.25 × 7.25 × 10 in.) with a curved front panel. The box provides a flat back so that the model can be laid flat on a document camera or overhead projector while the curved front panel allows a wider viewing angle. To mimic the intrapleural space, a constant negative pressure must be maintained within the container. A plug in one side panel of the model can be removed to apply a residual negative pressure to reflect functional residual capacity in vivo. In addition, the plug can be removed to demonstrate a pneumothorax and subsequent atelectasis.
|
58% of the lung's volume change, with rib expansion contributing the rest (3). Similarly, our model's diaphragm piston motion provides a larger volume change than the motion of the rib membranes. The 5-in.-diameter diaphragm piston is located on the bottom of the model and mimics the function and location of the diaphragm muscle in the human body. By pulling out the piston, the volume in the container increases, causing the pressure inside to decrease and the lungs to expand. The piston can be removed to provide access to the interior of the container for part replacement when needed. The rib membranes represent chest wall expansion and are located on both side panels of the model. Sections of gum rubber, selected for its durability and elasticity, are stretched over holes in the side panels that increase the internal volume when pulled outward. The gum rubber is attached to the panels by flanges, which are screwed on to create a leak-proof seal while allowing easy replacement of the membrane material. The small holes in the container beneath the membrane side panels allow volume change when the rib membrane is stretched but keep the rib membranes from collapsing inward when negative pressure is created inside the container. Handles are attached to both the piston and rib membranes for easy manipulation by the user. Elastic lungs are located within the model chamber and inflate or deflate according to the internal volume and pressure changes. Standard latex balloons were selected for the lungs because they are easy to replace, readily available, and have minimal leakage due to their seamless design. Two balloons are clamped onto a Y-tube fitting, with the third port passing through the container top via a rubber stopper and exposing the balloons to atmospheric pressure.
Two digital pressure sensors are attached to the top of the model for simultaneous real-time measurements of intrapleural and alveolar pressures. The intrapleural pressure sensor is exposed to the internal space of the container by threading it directly into the top panel. The alveolar pressure sensor threads through the top panel as well and is attached to a tube that passes through the Y-tube fitting and into one of the balloons. Although expensive, electronic compound pressure gauges with displays were selected for several reasons. First, a compound gauge is needed to measure negative and positive pressures created during simulated inspiratory and expiratory movements. Second, sensitive gauges are required to measure the small (0.2 psi) pressure changes. Third, electronic sensors are necessary so the model can interface with computer software to provide real-time graphs. Finally, digital displays are required so that the model can stand alone and function without the computer interface. The sensors are powered independently by a power adaptor that can be plugged into any 110- to 120-V wall outlet.
Electrical design.
Two digital compound pressure sensors (PSA-C01, Autonics), set to range of ±102 cmH2O, were used to measure pressure changes occurring within the alveolar and intrapleural spaces. Each sensor was interfaced with the BioPac MP30 or MP35 analog-to-digital converter by a 9-pin female D-sub connector (Fig. 2). For each transducer (alveolar and intrapleural), pin 2 on the D-sub 9 connector was connected to the signal output of the transducer and pins 3 and 4 were connected to ground (Fig. 3). Each pressure transducer was powered by a 15-V power supply. As a consequence, a 12-
resistor was soldered in parallel between pin 2 on the D-sub connector and ground to reduce the voltage input to the analog-to-digital converter, which was designed to handle a maximum 130-mV input. Pressure transducer digital outputs 1 and 2 were not used. After all connections were soldered, the D-sub housing units were secured; both connectors are shown connected to the MP30 in Fig. 4.
|
|
|
Set Data Acquisition Time) was set to 5 min to ensure a long enough period of time for demonstration purposes. Both channels were scaled (under MP30 Set Up Channels wrench icon Scaling) such that 59.3 mV = +102 cmH2O and 11.9 mV = –102 cmH2O according to calibration tests described below. Gain was set to 100, and the offset was changed as necessary to make sure a 0-cmH2O reading on the digital pressure sensor corresponded to a visual display of 0 cmH2O on the BioPac system. Cost. Project costs totaled $499.08 for two models: an initial prototype and the final product. This includes the costs of some unused materials and does not include the cost of donated items. This price also includes initial prototype costs, which would not be necessary in constructing a single respiratory model. The cost of a computer and the BioPac hardware are not included in the total cost because they are not required to use the model.
The projected costs for constructing one model with no donated materials would be $430.41. The primary expenses are the pressure sensors and the acrylic and polycarbonate stock materials, which comprise 56% and 31% of the total price, respectively. Reducing material waste and using bulk quantities would reduce costs somewhat further.
Testing
Physical testing. The most important aspect of ensuring the physical viability of our device was the strength of the seals around the cut polycarbonate pieces. We tested the efficacy of the seal in a variety of different ways. Submerging the prototype in water allowed us to assess the overall effectiveness of the seals. Water leaking into the prototype indicated a problem area. Small amounts of water were poured into the prototype, and the device was rotated to run the water along the sealed edges. Holes in the seals were indicated by water leaking out.
While submersion testing was efficient for large-scale leak testing, we also tested the seals using dry ice. A weigh boat containing dry ice was placed within the model and sprayed with water to produce a cloudy vapor. The piston was then replaced and moved inward to determine if and where the vapor was leaking out of the model. To examine the effectiveness of the seals on a still smaller scale, we rubbed soap along the seams and wet them slightly. When using the piston, bubbles appeared in areas where leakage occurred.
Finally, leak testing of our device was done by ensuring that the measured intrapleural pressure was always 
0 cmH2O. If the calibrated signal for the intrapleural space rose above 0 cmH2O, a leak existed somewhere in the device.
Periodically throughout construction, we tested the alveolar and intrapleural pressures generated by the piston and rib membranes separately. The device was connected to a high sensitivity pressure transducer (MPX 399/2, Hugo Sachs Elektronik) through the pressure sensor attachments at the top of the device. The measured pressures were recorded and graphed using LabView software. The alveolar and intrapleural pressures were tested separately in triplicate. The pressures generated by the piston and rib membranes were also tested separately and together, again in triplicate.
Educational testing. The most important aspect of this project was determining whether our model improved student understanding of respiratory mechanics concepts. To measure the instructional efficacy of the prototype, a method of surveying was developed for students in Human Physiology 335 at the University of Wisconsin (Madison, WI). Because students in undergraduate physiology classes will be the primary beneficiaries of the finished device, it was important to determine if their learning improved with use of the prototype in a classroom setting. Before any students were surveyed, a protocol was submitted to the Social and Behavioral Sciences (SBS) Institutional Review Board (IRB). This protocol was approved for SBS IRB exemption because the proposed study only involved surveying college students and posed no physical risk to the participants. Students participated voluntarily and anonymously with no incentive or risk to their grades, and each student was provided with a written consent form. The class had 427 registered students, but, for each question on the surveys, we encountered a different number of respondents and that number was always <427. Additionally, some questions were randomly skipped by the participating students, and fewer students overall chose to participate in the postsurveys, although there were equal losses from both control and experimental groups. Both the pre- and postinstructional surveys were developed in house after consultation with the SBS IRB, the University of Wisconsin Department of Physiology, and the University of Wisconsin Department of Biomedical Engineering.
Physiology students were randomly divided into control and experimental groups based on their laboratory sections All students were given a presurvey during their regular laboratory period to test their knowledge of respiratory physiology concepts before the material had been covered in lecture or laboratory workshops. Two weeks after the presurveys, the students then received postsurveys containing the same questions as the presurvey. After giving the presurvey, the laboratory instructor noted that answers 2 and 3 for question 2 are both valid for different points in the inspiration process. To compensate, answers 2 and 3 were both counted as correct answers, and the wording of the question was changed in the postsurvey to include all points of inspiration. All prelaboratory introductory material was presented to both groups by the same laboratory instructor. In the control group, the laboratory instructor gave a short introduction to the respiratory laboratory, explaining basic respiratory pressures and volumes. All of the material tested in the surveys was mentioned during the introduction. After the introduction, students in the control group were given the postsurvey on respiratory physiology concepts. In the experimental group, the laboratory instructor gave the same laboratory introduction but added a breathing demonstration using our model. The pressure changes in the intrapleural and alveolar spaces due to the diaphragm and rib membranes were graphed in real time using BioPac software and displayed on a projector. The response of the lungs after a puncture wound to the thoracic cavity (i.e., a pneumothorax) was also demonstrated. All of the material that was tested in the surveys was either mentioned during the introduction or shown with our model or both. After the introduction and demonstration, students in the experimental group were given the same postsurvey as the control group with additional questions specific to our prototype. The results of the pre- and postsurveys were tabulated and compared. A perfect survey score, 6/6, would indicate thorough understanding of the material. One-way ANOVA was used to compare experimental to control survey scores and pre- and postsurvey scores. P values of <0.05 were considered significant.
Results
Physical results. After the final construction of the prototype, the device was tested using the high-sensitivity pressure transducer noted above to determine the pressures generated by moving the rib membranes, piston, or both. The pressure sensors were removed from the prototype for the test. The transducer was connected in place of the intrapleural sensor connection at the top of the prototype, and pressures from the movement of the piston, rib membranes, and both were recorded (Fig. 5). The device generated negative pressures, which accurately represents the required negative pressure in the intrapleural space. The pressure testing trials shown in Figs. 5 and 6 were performed at a higher frequency than normal breathing to compensate for small leaks in the device and to maximize generated pressure.
|
|
Measurements of pressure in the intrapleural and alveolar spaces were taken in separate trials. Note that each graph in both Figs. 5 and 6 was generated by testing a single component of the device during separate trials using a single pressure port. The alveolar pressure was slightly more negative than the intrapleural pressure because the intrapleural space of the model has a greater volume than the alveolar space. This does not reflect pressure behavior when the diaphragm and ribs are operated concurrently, as in the human respiratory system. When operated concurrently, the intrapleural pressure generated by the piston will be more negative than the alveolar pressure to keep the lungs inflated. These generated pressures are large enough for the prototype to show the differences between pressures in the intrapleural and alveolar spaces as well as the differences between the contributions of the ribs and diaphragm.
Software.
Graphs of the intrapleural and alveolar pressures generated by the model, as displayed when interfaced with BioPac software, are shown in Fig. 7. Alveolar pressure correctly demonstrated a decrease in pressure and a return to atmospheric pressure when the piston or rib membranes were pulled outward (demonstrating inhalation) and an increase in pressure and again a return to atmospheric pressure as the piston or rib membranes were pushed back to their initial starting positions (demonstrating exhalation). The intrapleural space was always 0 cmH2O, demonstrating that the final device was leak free.
|
|
|
|
|
One-way ANOVA was used to compare experimental to control survey scores and pre- and postsurvey scores. As expected, presurvey scores from the control group were not statistically different from presurvey scores from the experimental group. Furthermore, in the control group, no improvement in survey test scores was evident with instruction without the demonstration. In contrast, postsurvey scores were significantly higher than presurvey scores in the experimental group (P < 0.0001), and postsurvey scores were significantly higher in the experimental group than control group (P < 0.0001). These findings demonstrate that the model had a significant and positive impact on survey test performance.
Discussion
We constructed a device for representing human respiratory mechanics, with the option of using a computer to display the pressures. While the estimated cost of manufacture of this device may be high, it offers improved and expanded functionality over other available models. Most currently available models do not display pulmonary pressures, making it difficult for students to visualize the forces driving gas exchange between the lungs and atmosphere (1, 5, 6). Other models do visualize the pressure changes using analog means, but the models are not interfaced with a computer or are not visually representative of human anatomy (2, 4). Furthermore, no currently available physical models illustrate the expansion of the rib cage. Our model demonstrates the effects of the rib cage and/or diaphragm on internal pressures, the biphasic nature of alveolar pressure changes, and the physiological effects of pneumothorax. All of these can be visualized in real time, further enhancing a student's ability to understand the functional interaction of the components of the human respiratory system.
It should be noted that no model is perfect, and ours suffers from some of the same shortcomings as others. For example, the pleural space is, in reality, a virtual space filled with a small volume of fluid and not an air-filled cavity. Since gases readily compress and expand (following Boyle's law), the presence of an air-filled pleural space in our model blunts the effect of diaphragm and rib movements on lung (balloon) volume. Also, our model cannot represent the full range of flexibility of the actual thoracic cavity or how it interacts mechanically with the diaphragm. Contraction of the diaphragm in the context of relaxed intercostal muscles can actually pull the rib cage inward, as in the "retractions" that occur in babies with respiratory distress syndrome. Shortcomings such as these need to be explicitly addressed with students whenever a mechanical model is used for teaching such complex topics.
Analysis of the survey data indicated that the students exposed to the model scored 1.0 point higher on a 6-point scale than students given a lecture only. Furthermore, students exposed to the model demonstrated greater improvement in understanding than those only exposed to a lecture via analysis of the pre- and postsurveys. Thus, these data support our conclusion that the instructor's use of the model improved student understanding of respiratory physiology.
Because the postsurveys were administered immediately after the concepts were explained and the model was demonstrated, the data may not capture whether the device helps improve long-term retention of the key concepts tested. In the future, administation of the postdemonstration surveys after a longer duration might better capture long-term improvements in learning.
When using this model in a classroom setting, we recommend familiarizing oneself with the functionality of the device and practicing the demonstration beforehand. As previously mentioned, there are numerous concepts that can be demonstrated with this device, so prelesson experimentation with the device and lesson planning are recommended. We welcome questions and inquiries to be directed to us at jjanderson1{at}wisc.edu.
Acknowledgments
The authors give special thanks to Karen Seashore, Donna Jahnke, Lillian Larson, Wendy Crone, Kristyn Masters, Kynan Shook, Petro Extrusion Technologies, Kraton, and Autonics for the support, advice, and donations of time and materials that helped make this model possible.
Received for publication August 22, 2008. Accepted for publication January 19, 2009.
REFERENCES
lecturedemonstrations/Newdemos/LungModel.htm [2612009].
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
| Visit Other APS Journals Online |
