Asian Cardiovasc Thorac Ann 2004;12:350-356
© 2004 Asia Publishing EXchange Ltd
Multiple Purpose Simulator Using a Natural Porcine Mitral Valve
Makoto Arita, MS1,
Sumihiro Tono, MS,
Hitoshi Kasegawa, MD2,
Mitsuo Umezu, PhD
1 Advanced Research Institute for Science and Engineering
Department of Mechanical Engineering, Waseda University
2 Department of Cardiovascular Surgery, The Sakakibara Heart Institute, Tokyo, Japan
For reprint information contact: Makota Arita, MS Tel: 81 3 5286 3256 Fax: 81 3 3200 2516 Email: arita{at}kurenai.waseda.jp Advanced Research Institute for Science and Engineering, Waseda University, 3-4-1, Ohkubo, Shinjuku-ku, Tokyo 169-85555, Japan.
 |
ABSTRACT
|
|---|
An in vitro pulsatile simulator with a porcine mitral valve was developed in order to simulate physiologic and diseased mitral valve conditions. Evaluation of these conditions was conducted from a hydrodynamic and annulus behavior point of view. We found it possible to simulate mild "mitral valve prolapse" and to obtain quantitative data related to the condition. The diseased condition produced a 40% greater regurgitant volume than that observed under the normal condition ( p < 0.0001). Regarding the leakage volume, the diseased condition exhibited about 2.6 times more leakage than the normal condition. The mitral valve simulator proposed in this study is considered fairly stable with respect to both hemodynamics and the behavior of the annulus, and it is an adequate simulator for modeling various types of normal and diseased mitral valve conditions.
|
INTRODUCTION
|
|---|
In 1971, Carpentier and associates introduced his concept of valvular remodeling1 as well as a rigid annuloplasty ring for mitral valve reconstruction. The rigid ring enhanced valve repair by avoiding recurrent annular dilation and decreased annulus size while maintaining sufficient orifice area. Duran and colleagues modified Carpentier’s annular frame concept in 1976 by using a flexible annuloplasty ring that could conform to the dynamic motion of the mitral annulus during the cardiac cycle.2 The development of prosthetic annular rings allowed more universal use of mitral valve repair techniques and an expansion of the indications for choosing repair over mitral valve replacement. In fact, several studies have reported that valvuloplasty for mitral valve disease, in comparison with prosthetic mitral valve replacement, has been associated with enhanced patient survival and a lower rate of complications, especially thromboembolism and anticoagulant-related hemorrhage in the late postoperative period.3–6
Advancements in cardiopulmonary bypass and cardioplegic techniques have also played an important role in establishing the current surgical techniques of mitral valve annuloplasty. However, surgical correction of ischemic mitral regurgitation (MR) depends not only onf the degree of ischemia or infarct producing valvular incompetence, but also on the clinical condition of the patient and the approach, which is based on the surgeons’ experience. Hence, there are no universally accepted guidelines to suggest the most appropriate methodology or surgical treatment for ischemic MR.
The aim of the present study was to develop an in vitro mitral valve simulator capable of reproducing normal and several diseased mitral valve conditions, including ischemic MR. It was hoped this would allow the evaluation of various surgical techniques for repairing mitral valve disease and provide surgeons a means of selecting the most appropriate surgical technique. First, we introduced the sophisticated mock circulatory system and replicated anomalous papillary muscle positioning as an example of a diseased model. In addition, we were then able to validate this new evaluation system from both a hydrodynamic and annulus behavior point of view.
|
MATERIALS AND METHODS
|
|---|
A freshly excised natural porcine mitral valve with intact chordae tendineae and papillary muscles was used in this experiment. Initially, the valve’s annulus configuration was measured using a digital microscope (Model VH-6300, KEYENCE Corp., Osaka, JAPAN). Next, an acrylic plate was fabricated upon which we could mount the valve with reference to the annulus measurements, and a 5 mm diameter nylon sewing band was glued to the cut surface. The trimmed valve was then sutured to the nylon band using 2-0 Ethibond polyester suture (ETHICON, Inc., Somerville, NJ, USA), followed by sealing the annular suture line with epoxy resin to prevent perivalvular leak.
All experiments were conducted in a pulsatile flow simulator developed in the Umezu BioMechanical Engineering Laboratory at Waseda University. The simulator is a modified Windkessel-type pulse duplicator with a pneumatically driven pump, into which the fresh porcine mitral valve fixture was installed in the mitral position (Figure 1
). To facilitate observation of the mitral annulus, the atrioventricular casing is made of transparent acrylic plate. The left ventricle is made from latex rubber tubing, which is compressed and relaxed with compressed air regulated by a microprocessor-based pulse duplicator. Pulse rate and systolic duration were controlled by this pneumatic driver. In order to achieve a proper aortic pressure waveform, the pulse pressure was controlled by adjusting the air regulator attached to the aortic compliance tank, thus increasing or decreasing the pressure. Additionally, systemic resistance was adjusted by using a throttling valve. An ATS mechanical bileaflet valve, nominal diameter of 25 mm, was used in the aortic valve position of the simulator.
View larger version (22K):
[in this window]
[in a new window]
|
Figure 1. Schematic drawing of the mitral valve simulator developed by Waseda University. This is based on a modified Windkessel-type pulse duplicator with a pneumatically driven pump and a freshly excised natural porcine valve with intact chordae tendineae and papillary muscles, which was attached in arterioventricular casing.
|
|
In order to simulate various mitral valve conditions, the fixation points of the papillary muscles could be changed from normal to several different "diseased" positions, using a guiding element. In this study two experimental conditions were examined, normal papillary muscle position (normal condition), and a 5 mm upwardly displaced position (diseased condition). The normal condition was defined by referring to the papillary muscle locations and orientations in the native porcine heart. Comparing the two conditions allowed us to understand whether this system could adequately simulate several diseased conditions.
With respect to acquisition of hemodynamic data, two electromagnetic flowmeters (MFV-2100, Nihon Kohden, Tokyo, JAPAN) and cannulation-type flow probes were used. These had inner diameters of 30 mm (FF-300TA, Nihon Kohden, Tokyo, JAPAN) and of 18 mm (FF-180TA, Nihon Kohden, Tokyo, JAPAN) for measuring mitral and aortic flows, respectively. Three strain gauge pressure transducers (UK-801(TW), Baxter, Irvine, CA, USA) were used (Figure 2) for continuous measurement of ventricular, aortic, and atrial pressures. The pressure signals were acquired digitally at a sampling frequency of 500 Hz and stored on a digital data recorder (DR-M3, TEAC Corp., Tokyo, JAPAN). Regarding observation of annular 3-D movement, a laser displacement sensor (LB-01/LB-60, KEYENCE Corp., Osaka, JAPAN) and a digital video camera (DCR-TRV900, SONY Corp., Tokyo, JAPAN), were installed above the left atrial chamber. Eight measurement points on the anterior and posterior mitral annulus were marked for measurement (A1 to A3 on the anterior annulus and P1 to P5 on the posterior annulus) as depicted in Figure 3.
View larger version (8K):
[in this window]
[in a new window]
|
Figure 2. Diagram of mitral valve simulator and measurement devices used in this study. Hemodynamic data were acquired using the electromagnetic flowmeters and the pressure transducers. White circles show the flowmeters and black circles indicate pressure transducers. Furthermore, a laser displacement sensor and a digital video camera were applied to observe the annular 3-D movement. LA: left atrium; LV: left ventricle.
|
|
View larger version (320K):
[in this window]
[in a new window]
|
Figure 3. Measurement points for investigating annulus behavior. Circles show the measurement points; A1 to A3 on the anterior annulus and P1 to P5 on the posterior. (A) end systole (B) end diastole.
|
|
In all of the experiments, a pulse rate of 75 beats•min–1 and a systolic fraction of 35% (ejection time of approximately 280 msec) were used. Systolic and diastolic pressures were set at 120 mm Hg and 80 mm Hg, respectively, by adjusting the peripheral resistance; atrial pressure was maintained at 7 mm Hg. The flow condition corresponded to a cardiac output of approximately 5 L•min–1. Backflow was divided into two components according to the method described by Dellsperger and Wieting.7 One component of backflow is caused by the closing action of the valve (regurgitant volume), and the other component is due to leakage after the valve is closed (leakage), as shown in Figure 4. Calculations and statistical analyses were conducted using 7 pulses because pressure and flow waveforms were unstable due to characteristics of the porcine valve. To preserve fresh tissue properties, St Thomas’s solution with a viscosity of 1.0 cP at a temperature of 22°C was used as the working fluid throughout this experiment.
View larger version (27K):
[in this window]
[in a new window]
|
Figure 4. Typical pressure and flow waveforms obtained from the simulator. The right upper circle on (A) is an enlargement of mitral backflow that shows the definition of regurgitant volume and leakage. (A) normal condition (B) diseased condition.
|
|
All data were statistically analyzed using a statistical software package (STATA ver.5.0, Stata Corp., TX, USA) and a p-value in the Welch’s test of less than 0.05 was considered statistically significant.
|
RESULTS
|
|---|
1. COMPARISON OF HYDRODYNAMICS
Figures 4A and 4B
show typical pressure and flow waveforms obtained from experiments in normal and diseased conditions, respectively. No significant difference was observed in the contour of the pressure waveforms. However, with respect to mitral flow, regurgitant volume at valve closing was significantly different between the two conditions. Figure 5 compares the regurgitant volume calculated from the mitral flow waves during both conditions. Mean flow rate derived from flow waveforms was 5.5 L•min–1 in normal condition and 5.3 L•min–1 in diseased condition. The difference of 0.2 L•min–1 was caused by the difference from regurgitant volume as presented in Figure 5, and thus, actual forward flow was almost the same. The diseased condition produced a regurgitant volume (5.6 mL/beat) 40% higher than that observed with the normal condition (4.0 mL/beat) (Welch’s t-test p <0.0001). Leakage volume, which represents the other index of mitral valve backflow, was also calculated from the mitral flow waveforms in the two conditions. As summarized in Figure 6, although the differences were small (statistically not significant), the diseased condition showed a minimal amount of leakage volume (0.3 mL/stroke), whereas the normal condition showed almost no leakage at all (0.1 mL/stroke).
View larger version (8K):
[in this window]
[in a new window]
|
Figure 5. Comparison of regurgitant volume between the normal and diseased models. Data is given by mean ± sd using 7 pulses.
|
|
View larger version (8K):
[in this window]
[in a new window]
|
Figure 6. Comparison of leakage volume between the normal and diseased models. Data is given by mean ± sd.
|
|
2. ANNULUS BEHAVIOR
Figures 7A and 7B show, respectively, an inflow view of the mitral valve with a normal papillary muscle position (normal condition) and with an upwardly displaced papillary muscle position (diseased condition). From these images, it can be observed that both anterior and posterior leaflets remained in the basal plane of the acrylic fixture under the normal condition. Under the diseased condition, both leaflets were observed to bulge upwards due to the abnormal papillary muscle position, resulting in a "mitral valve prolapse" configuration. In addition to the gross visual observations of mitral valve behavior, annulus displacements measured at 6 points from end systole to end diastole are depicted in Figure 8. By changing the papillary muscle position from the normal to the diseased condition, annular displacement increased at almost all points. The average incremental increase in displacement at each point was 28%. Displacement of the posterior side of the annulus was greater than that of the anterior side under both conditions. For instance, the posterior portion (P3) of the annulus exhibited a displacement about 2.2 times (normal condition) and 1.8 times (diseased condition) greater than that observed at the anterior part A2 of the annulus.
View larger version (233K):
[in this window]
[in a new window]
|
Figure 7. Images of mitral valve during systolic phase. This shows an example of a diseased "mitral valve prolapse" model simulated by our sophisticated mock circulatory system. (A) normal condition (B) diseased condition produced by upward displacment of the papillary muscle by 5 mm.
|
|
View larger version (12K):
[in this window]
[in a new window]
|
Figure 8. Displacement of annulus measurement points under normal and diseased conditions. Measurement points are depicted in Figure 3.
|
|
|
DISCUSSION
|
|---|
1. CONTOUR OF PRESSURE/FLOW WAVEFORMS AND ANNULUS BEHAVIOR
Pressure and flow waveforms obtained from this mitral valve simulator were considered acceptable simulations of physiologic conditions. Although only one test set was applied in this experiment, the simulator is capable of producing a wide variety of test conditions, such as flow rates ranging from 2 to 7 L•min–1, pulse rates from 50 to 180 beats•min–1, and aortic pressures of up to 160 mm Hg. With respect to hemodynamics, therefore, both normal physiologic models and many diseased models (low cardiac output, hypertension, ventricular tachycardia, or combinations of these) could be simulated.
In addition to hemodynamic control, the simulator used in these experiments employed a natural mitral valve so that actual mitral valve disease conditions could also be simulated. In this study, a mild mitral valve prolapse model (Figure 7B) was achieved by changing the relative position between the papillary muscles and the basal plane of the mitral valve. In this case, the results of regurgitant and leakage volume measurements demonstrated that the regurgitation in the diseased condition was statistically significantly higher compared to that of the normal model, whereas there was no significant difference in the leakage volume. Although no significant difference in leakage volume was observed, if the papillary muscle position was changed more dramatically, leakage volume as well as regurgitant volume would probably be increased, and thus, a severe mitral regurgitation model could also be simulated.
Annulus behavior observed in these experiments indicated that the posterior portion of the annulus displaced to a greater degree than the anterior annulus, which corresponds in similar fashion to investigations regarding native mitral valve annular flexibility by Komoda and associates.8 In our diseased condition model, the distance between the mitral annulus and the papillary muscle was slightly shortened experimentally compared to the normal condition, and therefore, it was inferred that annular displacement was greater than normal due to the resultant slack in the chordae tendineae. Because of this, valve closing time in the diseased condition would be longer than in the normal condition, and thus, the regurgitant volume was greater. However, although there was slack in the chordae, displacement of the anterior annulus remained significantly less than that observed for the posterior annulus, which is similar to the normal condition. This finding suggests that the natural mitral valve has sufficient adaptation to or tolerance for mild changes in papillary muscle positioning and chordae tension.
Consequently, the mitral valve simulator introduced in this study was considered to be fairly reliable with respect to simulating the hemodynamics and annulus behavior of the natural native mitral valve. We believe the simulator to be an adequate device for modeling various types of natural and diseased mitral valve conditions. Hence, we believe that our initial purpose was achieved in developing a versatile mitral valve simulator.
2. APPLICATIONS OF THE MITRAL VALVE SIMULATOR
We produced the diseased condition by changing the relative position of the papillary muscle with respect to the plane of the mitral valve annulus. This concept focuses on the clinical situation wherein transient ischemia or left ventricular dilation changes the position of the papillary muscle and causes mitral regurgitation. Currently, surgeons cannot directly observe native normal or diseased mitral valve performance. Also, in the postoperative period, there are no means except by angiography or echocardiography to evaluate whether a surgical correction was adequate. To facilitate observation of the mitral annulus, the simulator is made of transparent acrylic plates and thus, mitral valve behavior can be monitored from various directions. Therefore, if a diseased condition can be produced and a surgeon can perform a mitral valve repair using this mitral valve simulator, the surgeon can visually confirm the appropriateness of the technique.
In addition to confirmation of their repair techniques, this simulator also allows the surgeon to use it as a training simulator for enhancing surgical techniques. The simulator can model various degrees of mitral regurgitation severity caused by chordae tendineae rupture or by papillary muscle displacement due to ischemia or left ventricular dilation. With respect to chordae tendineae rupture, mitral regurgitation can be reproduced by cutting one or more of the chordae. Reconstruction or substitution of the chordae can be investigated using this system with the aim of obtaining objective and quantitative data on the hemodynamic effects of these repairs.
In the past, the number or length of artificial chordae tendineae used has largely been based on surgeons’ empirical experience, and not by detailed studies conducted in both in vitro and in vivo experiments. The only in vitro study known to the authors concerns a comparative investigation of the mechanical properties of surgical sutures and natural porcine mitral valve chordae reported by Cochran and associates.9 However, this study focused on the application of suture materials for chordal substitution, not on surgical repair techniques. Hence, we regard our mitral valve simulator as an evaluation system for studying the effects of chordal reconstruction techniques from valvular hemodynamic and valve behavior points of view.
Concerning mitral regurgitation secondary to ventricular dilation, lateral displacement of the papillary muscles causes restriction of the motion of the leaflets, preventing coaptation.10 In our mitral valve simulator, this situation can easily be simulated by shifting papillary muscle position laterally using a guiding element. We intend to describe such an experiment and the results there of in a future report.
Another application for the mitral simulator would be to study the characteristics of various types of annuloplasty rings. Many types of annuloplasty rings have recently been developed and introduced into the clinical market,11–14 however, detailed investigations regarding their functional characteristics have not been carried out sufficiently. By using our simulator, it would be possible to conduct a comparison of various types of these devices under more controlled conditions. Our laboratory has previously developed an in vitro evaluation system for annuloplasty rings, and its validity was demonstrated in previous studies.15,16 However, these studies were carried out under static conditions, and the dynamic characteristics of the annuloplasty rings were not evaluated.
Clinically, the functional characteristics of annuloplasty devices have only been discussed subjectively, based primarily on clinical data from transesophageal echocardiography or other methods.17–20 By use of this mitral valve simulator, more quantitative data about annuloplasty ring performance could be obtained under controlled and reproducible conditions that closely simulate the clinical situation. In particular, application of an annuloplasty ring for mitral valve insufficiency caused by ventricular dilation or an apically displaced papillary muscle as previously mentioned would be one of the more practical investigations for providing surgeons a better understanding of how these devices function.
Consequently, some clinical significance of this research was found in the ability of our in vitro model to simulate physiologic and diseased mitral valve parameters. As the data described above were from preliminary simulation results, the authors have been trying to verify fundamental characteristics based on a comparative study between in vitro and clinical data.
|
CONCLUSION
|
|---|
The authors have confirmed that mechanical engineering studies such as those described in this report can provide cardiac surgeons with useful tools for establishing improved methods of surgical technique. Although it is a simple evaluative simulation system and further tests are needed, we believe our test apparatus is capable of determining annulus behavior following performing various annuloplasty techniques. Not only can our proposed simulator demonstrate the suitability of surgical corrections for mitral incompetence, such as chordal rupture or displacement of papillary muscles, it can also be used to demonstrate and compare the dynamic characteristics of different annuloplasty devices.
|
ACKNOWLEDGMENTS
|
|---|
This research was carried out with the aid of the following research funds: The Program for Promotion of Fundamental Studies in Health Science of the Organization for Drug ADR Relief, R&D Promotion and Product Review of Japan (No.96–12); Grant-in-aid for Scientific Research of Japan (No.09557112, No.09470288); and Health Sciences Research Grants (No.H-11-Drug-006). The authors would like to express their gratitude to Mr. Dave Myers of Medtronic Heart Valves, for reviewing the manuscript. Also, we are very thankful to Mr. Tomoya Mizunuma, undergraduate students in Prof. Umezu’s laboratory, who provided much valuable assistance in the performance of these experiments.
|
REFERENCES
|
|---|
- Carpentier A, Deloche A, Dauptain J, Soyer R, Blondeau P, Piwnica A, et al. A new reconstructive operation for correction of mitral and tricuspid insufficiency. J Thorac Cardiovasc Surg
1971;61:1–13.[Medline]
- Duran CG, Ubago JL. Clinical and hemodynamic performance of a totally flexible prosthetic ring for atrioventricular valve reconstruction. Ann Thorac Surg
1976;22:458–63.[Abstract]
- Oury JH, Peterson KL, Folkerth TL, Daily PO. Mitral valve replacement versus reconstruction. An analysis of indications and results of mitral valve procedures in a consecutive series of 80 patients. J Thorac Cardiovasc Surg
1977;73:825–35.[Abstract]
- Rankin JS, Feneley MP, Hickey M, Muhlbaier LH, Wechsler AS, Floyd RD, et al. A clinical comparison of mitral valve repair versus valve replacement in ischemic mitral regurgitation. J Thorac Cardiovasc Surg
1988;95:167–77.
- Perier P, Deloche A, Chauvaud S, Fabiani J, Rossant P, Bessou J, et al. Comparative evaluation of mitral valve repair and replacement with Starr, Bjork, and porcine valve prostheses. Circulation
1984;70(supple I):187–92.
- Sand ME, Naftel DC, Blackstone EF, Kirklin JW, Karp RB. A comparison of repair and replacement for mitral valve incompetence. J Thorac Cardiovasc Surg
1987;94:208–19.[Abstract]
- Dellsperger KC, Wieting DW. Backflow considerations in the design of prosthetic heart valves. Biomech Symposium AMD
1981;43:3–5.
- Komoda T, Hetzer R, Oellinger J, Siniawski H, Hofmeister J, Hubler M, et al. Mitral annular flexibility. J Card Surg
1997;12:102–9.[Medline]
- Cochran RP, Kunzelman KS. Comparison of viscoelastic properties of suture versus porcine mitral valve chordae tendineae. J Caridiac Surg
1991;6:508–13.
- Cosgrove DM, Stewart WJ. Mitral valuvuloplasty. Curr Probl Cardiol
1989;14(7):353–416.
- Cosgrove DM, Arcidi JM, Rodriguez L, Stewart WJ, Powell K, Thomas JD. Initial experience with the Cosgrove-Edwards Annuloplasty System. Ann Thorac Surg
1995;60:499–504.[Abstract/Free Full Text]
- Lee EM, Shapiro LM, Wells FC. Midterm results of mitral valve repair with the Sculptor Annuloplasty rings. Ann Thorac Surg
1997;63:1340–5.[Abstract/Free Full Text]
- Seguin JR. Preservation of three-dimensional annular movement with the SJMTM-Seguin mitral annuloplasty ring. J Heart Valve Disease
1996;5:641–6.[Medline]
- Carpentier AF, Lessana A, Relland JY, Belli E, Mihaileanu S, Berrebi AJ, et al. The "physio-ring": an advanced concept in mitral valve annuloplasty. Ann Thorac Surg
1995;60:1177–85.[Abstract/Free Full Text]
- Arita M, Kasegawa H, Umezu M. Development of in vitro evaluation system for annuloplasty rings. Asian Cardiovasc Thorac Ann
2001;9:14–8.[Abstract/Free Full Text]
- Arita M, Kasegawa H, Umezu M. Mechanical characteristics of a flexible mitral annuloplasty band. Proc of 10th Int Conf on Biomed Eng
2000:289–90.
- Odell JA, Schaff HV, Orszulak TA. Early results of a simplified method of mitral valve annuloplasty. Circulation
1995;92(suppl II):150–154.[Abstract/Free Full Text]
- Okada Y, Shomura T, Yamaura Y, Yoshikawa J. Comparison of the Carpentier and Duran prosthetic rings used in mitral reconstruction. Ann Thorac Surg
1995;59:658–63.[Abstract/Free Full Text]
- Gorton ME, Piehler JM, Killen DA, Hoskins ML, Borkon AM. Mitral valve repair using a flexible and adjustable annuloplasty ring. Ann Thorac Surg
1993;55(4):860–3.[Abstract]
- Yamaura Y, Yoshida K, Hozumi T, Akasaka T, Okada Y, Yoshikawa J. Three-dimensional echocardiographic evaluation of configuration and dynamics of the mitral annulus in patients fitted with an annuloplasty ring. J Heart Valve Disease
1997;6:43–7.[Medline]
TOP
ABSTRACT
INTRODUCTION
