| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
|
|
|
|||||||
|
|||||||||
|
Interact CardioVasc Thorac Surg 2009;8:182-186. doi:10.1510/icvts.2008.194134 © 2009 European Association of Cardio-Thoracic Surgery
Rapid prototyping of compliant human aortic roots for assessment of valved stentsDepartment of Cardio-Vascular Surgery, Centre Hospitalier Universitaire Vaudois, CHUV, Rue du Bugnon 46, CH-1011 Lausanne, Lausanne, Switzerland Received 15 September 2008; accepted 15 October 2008
*Corresponding author. Center of Cardiac Surgery, Pauls Stradins Clinical University Hospital, Pilsonu str. 13, Riga, LV-1002, Latvia.
Adequate in-vitro training in valved stents deployment as well as testing of the latter devices requires compliant real-size models of the human aortic root. The casting methods utilized up to now are multi-step, time consuming and complicated. We pursued a goal of building a flexible 3D model in a single-step procedure. We created a precise 3D CAD model of a human aortic root using previously published anatomical and geometrical data and printed it using a novel rapid prototyping system developed by the Fab@Home project. As a material for 3D fabrication we used common house-hold silicone and afterwards dip-coated several models with dispersion silicone one or two times. To assess the production precision we compared the size of the final product with the CAD model. Compliance of the models was measured and compared with native porcine aortic root. Total fabrication time was 3 h and 20 min. Dip-coating one or two times with dispersion silicone if applied took one or two extra days, respectively. The error in dimensions of non-coated aortic root model compared to the CAD design was <3.0% along X, Y-axes and 4.1% along Z-axis. Compliance of a non-coated model as judged by the changes of radius values in the radial direction by 16.39% is significantly different (P<0.001) from native aortic tissue – 23.54% at the pressure of 80–100 mmHg. Rapid prototyping of compliant, life-size anatomical models with the Fab@Home 3D printer is feasible – it is very quick compared to previous casting methods.
Key Words: Aortic root; 3D print; Stereolitography; Stent valve; Trancatheter valve replacement
With advancement of diagnostic imaging techniques capable of producing three-dimensional virtual reconstructions of human body parts from stacks of multi-planar images, 3D imaging has rapidly entered the clinical world. It is used routinely not just for diagnostic purposes but also for intervention planning and guidance in fields of radiology, neurosurgery and cranio-facial surgery [1–3]. The before mentioned specialties were pioneering this field, but very quickly the benefits provided by 3D visualization were appreciated and accepted also by other medical specialties including cardiologists and cardiovascular surgeons. It is being used for diagnostic purposes, for operation planning as well as for education and training [4–7]. More recently with decreasing prices and increasing overall popularity and availability of rapid prototyping devices also called 3D printers, the virtual 3D representation on the screen has found its realization in solid replica of anatomical structures – life-size, rigid physical models [8–11] which contribute to the realism, facilitate perception and recognition of the 3D structures [10], hence improving clinical performance [9]. In cardiovascular surgery, considering the rapid advancement of transcatheter procedures [12], there is a certain need for flexible hollow models e.g. a model of aorta or aortic root for endovascular and valved-stents procedures training and new device testing. There are several methods to create such models but all of them include several steps and are time consuming. The most detailed and precise flexible models can be made by combining rapid prototyping for creating molds and traditional casting of silicone [5, 13]. Recently, a new rapid prototyping system has become available, created by the Fab@Home project (Fig. 1) which to our best knowledge is the only 3D printer capable of using any viscous substance as a building material.
Aiming to reduce the complexity and time needed for creating a compliant physical model of aortic root we decided to pursue a goal of building a 3D model in a single-step procedure without employing the somewhat time consuming and rather unpredictable casting techniques.
We used Solidworks 2008 SE (SolidWorks Corporation, Concord, MA, USA) for creating a precise 3D model (Fig. 2a and Video 1) of the aortic root in silico using previously published anatomical and geometrical data with minimal modifications [14, 15]. The dimensions used in creating the CAD model can be seen in Fig. 2b. The geometry of the sinuses of Valsalva was modeled partially following the description of Reul and colleagues [15] and was treated as an epitrochoid. ‘An epitrochoid is a roulette traced by a point attached to a circle of radius r rolling around the outside of a fixed circle of radius R, where the point is a distance d from the center of the exterior circle’ – definition of an epitrochoid from www.wikipedia.org. In brief the sinuses were described as an epitrochoid with an R=13.2 mm, r=4.4 mm and d=2.8 mm (Fig. 3). It has to be noted that in order to have three sinuses R has to be equal to 3r. The X, Y coordinates of the respective curve were calculated in Microsoft Excel software using the equations which can be seen in Fig. 3. Later the calculated points were saved in an ASCII text file as X, Y, Z coordinates and imported into SolidWorks.
The model geometry was saved in STL format (Stereolithography) and printed using an open-source rapid prototyping system developed by the Fab@Home project (http://fabathome.org/) and distributed by Koba Industries (Albuquerque, NM, USA) which main components are a deposition tool with a syringe moved by several stepper motors. For printing we used a syringe tip with a diameter of 0.84 mm. The printing setup can be seen on Fig. 1. As a material for 3D fabrication we used common house-hold (sanitary) silicone (Forbo international, Schoenenwerd, Switzerland) which took around 4 h to dry and is semitransparent. Afterwards we dip-coated several models with dispersion silicone (Nusil Technology, Carpinteria, CA, USA) one or two times making a pause of at least 16 h between dipping to allow the previous layer to dry completely. We compared the elastic properties of the constructed models with a fresh porcine aortic root harvested within 6 h post mortem. We performed a test using a roller pump to pressurize the models with three sonomicrometry probes (Sonometrics, London, Canada) fixed on the outer surface of the models at the level of commissures. Mathematical analysis of the data was performed off-line with a software package for cardiovascular analysis (Sonosoft version 3.1.3., Sonometrics, London, Canada). We used Student's t-test for comparing means and chose P-value below 0.05 as the threshold for statistical significance. To estimate the cross-sectional compliance, changes in the radius and area of a circle circumscribed around the three piezoelectric probes were compared at four different pressure levels: 0, 80–100, 100–150 and 200–240 mmHg. Similarly the compliance of the fresh porcine aortic root was estimated.
Total fabrication time of the life-size aortic root model was 3 h and 20 min (Video 2). Dip-coating with dispersion silicone was performed one or two times to increase the mechanical strength of model and it took one or two extra days, respectively – see Fig. 2c for a final look of the built model after double dipping.
The dimensions of non-coated aortic root model compared to the dimensions of the CAD design can be seen in Fig. 2b. A maximum error of 2.94% along X, Y-axes and 4.07% along Z-axis has been produced during 3D printing (for detailed results see Fig. 2b). The surface of the printed object was not ideally smooth, with visible layered structure. To prove the appropriateness of the built mockup for close to real-life simulations in artificial circulatory systems we conducted several tests to evaluate the cross-sectional compliance of the uncoated and dip-coated models and compared these data with the parameters of a native porcine aortic root. The results are summarized in Table 1. The table is missing extensibility results of the uncoated model for pressures above 80–100 mmHg for the reasons that higher pressure introduced defects resulting in massive leakage, which prohibited further testing. As judged by the changes of the circumscribed circle radius values by 16.39% compliance of a non-coated model is significantly different (P<0.001) from native aortic tissue – 23.54% at the pressure of 80–100 mmHg, but still comparable. Unfortunately the non-coated model lacks mechanical strength to withstand testing with the full range of physiological blood pressure occurring in the aortic root. Both coated models are significantly more rigid compared to the non-coated at the respective tested pressures. Even single dipping decreases model compliance dramatically to 3.25% at 80–100 mmHg pressure. The model is almost rigid after a second dipping, showing only 3.12% increase in radius values when raising pressure from 0 to above 200 mmHg.
Up to now the only option for making a custom flexible anatomical model was to use a casting method, which is very time consuming and complicated. It is even more complicated to produce a hollow flexible object – this requires not only a mold but also a breakable or melting insert. We have demonstrated that Fab@Home – a new type rapid prototyping system – can be used for production of compliant, life-size replica of anatomical structures in a very short time. For comparison, a recent paper by Sulaiman and colleagues [13] describes a procedure for creating anatomical models in silicone with similar properties as ours by casting technique using 3D printed molds which takes around five days from a CAD design to a finalized model. The method we have used reduces the production time dramatically to hours instead of days. At its current development stage the Fab@Home system has limitations in production detail and precision, which stem in part from the very principle of the machine – using any viscous substance as a building material. It is impossible to have full control of a deposited viscous material (like silicone) before it hardens. The only possibility to increase the precision is to use the smallest diameter nozzle as possible for the syringe of the deposition system, which would dispense silicone in very thin threadlike paths preventing it from running to undesirable places. Using a very small in diameter syringe tip, although it increases precision it also increases production time dramatically. A compromise is required – the setup we used allows for roughly 0.8 mm resolution along X, Y axes and 0.5 mm along Z with a nozzle diameter of 0.84 mm. Another factor which limits the precision of the system is shrinkage of silicone during the drying time. The latter factor can be eliminated by utilizing industry-grade silicone with near zero shrinkage or with a certain well-defined ratio. A strictly technical limitation of this 3D printer is inability to build near-horizontal hanging parts with a single deposition tool, but this has been already solved by creating custom systems with two deposition tools, where one of them is used to deposit an easy to remove support material. We did not manage to adjust the mechanical properties of the models to native aortic tissue with silicone readily available to us. Though knowing the enormous versatility of available silicone types this should not be a problem for future works. Most likely it would require working with combinations of two or three silicone types to model almost any physiological or pathological state of the aortic wall. We conclude that rapid prototyping of compliant, life-size anatomical models with the Fab@Home 3D printer is feasible in a single step – it is very rapid compared to previous casting methods, relatively simple and, with a complete 3D printing setup below USD 5000, relatively inexpensive. The fabrication system has an acceptable production error of <3.0% along the horizontal and 4.1% along the vertical axis. Common house-hold silicone can be used for creating flexible anatomical models for training and educational purposes, but it lacks the mechanical strength and sturdiness for high pressure testing, which can be added to the model by dip-coating it with dispersion silicone. The lack of compliance with the current production may well mimic the stiffer vascular walls in elderly patients, the main group of patients undergoing trans-catheter valve replacement. We have successfully started to use this type of model in in-vitro valved stents testing integrating the aortic root in an artificial circulatory loop. Currently work on a simplified whole heart model for training in trans-apical valved-stents delivery is under way.
Dr Martins Kalejs stay and work at the Centre Hospitalier Universitaire Vaudois Lausanne was supported by the Zarins-Knight Traveling Fellowship and by the Latvian National Research Programme in Medicine 2006–2009. We would like to thank Mr Saad Abdel Sayed for valuable help with dip-coating of the models.
This article has been cited by other articles:
|
||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
| ANN THORAC SURG | ASIAN CARDIOVASC THORAC ANN | EUR J CARDIOTHORAC SURG |
| J THORAC CARDIOVASC SURG | ICVTS | ALL CTSNet JOURNALS |
Top
Abstract
1. Introduction
