Court Filings

A pulsatile simulator for the in vitro analysis of the mitral valve with tri-axial papillary muscle displacement

Description
We developed a new pulsatile hydrodynamic simulator for the in vitro testing of mitral valve (MV) samples. The required specifications included a 3D positioning system for the papillary muscles (PMs) that is accurate and simple to manage; measurement
Categories
Published
of 9
All materials on our website are shared by users. If you have any questions about copyright issues, please report us to resolve them. We are always happy to assist you.
Related Documents
Share
Transcript
  © 2011 Wichtig Editore - ISSN 0391-3988 Int J Artif Organs( 2011;:4)383-39134 383 DOI: 10.5301/IJAO.2011.7729 INTRODUCTION Mitral valve (MV) diseases are among the most common pathologies of the heart (1). In recent decades valvular re-pair has become the preferred surgical approach to their treatment, with the introduction of novel surgical techni-ques (2-4), instruments, and annuloplasty devices (5, 6). Knowledgeable application of such new approaches relies  A pulsatile simulator for the  in vitro  analysis of the mitral valve with tri-axial papillary muscle displacement Riccardo Vismara 1  , Andrea Pavesi  1  , Emiliano Votta 1  , Maurizio Taramasso  2  , Francesco Maisano  2  , Gianfranco B. Fiore 1 1 Department of Bioengineering, Politecnico di Milano, Milan - Italy 2 Cardiac Surgery Division, IRCCS San Raffaele Hospital, Milan - Italy ABSTRACTPurpose: We developed a new pulsatile hydrodynamic simulator for the in vitro testing of mitral valve (MV) samples. The required specifications included a 3D positioning system for the papillary muscles (PMs) that is accurate and simple to manage; measurement of the force exerted by the chordae tendi- neae on the PMs; and the possibility to visually inspect the MV for kinematic analysis. Methods:  An atrial/ventricular chamber system was developed. The ventricular chamber housed a tri- axial actuator system that was aligned to a morphometric Cartesian frame, allowing for PM positioning even while tests are running. Each PM holder had an embedded load cell for force measurement. The atrial chamber was designed so as to permit MV visual inspection, maintaining a non-disturbed flow at the sample inlet. The setup was subjected to trials with fresh porcine MVs. Flow and pressure difference across the MVs and PM forces were measured in different MV configurations, with different PM spatial dislocations. High speed video recordings were acquired. Results:  The positioning accuracy was assessed. Tests with MVs showed good usability, even by the  non-engineering personnel. The effects of PM displacement on valve function (valve competence and PM forces) was consistent with previously published data, thus confirming the general soundness of the design principles. Conclusions:  The developed simulator is a promising instrument for performing MV in vitro tests in  a precise, well-repeatable manner. The ability to completely adjust the PM position while a test is  running boosts the simulator’s potential for detailed investigations of the pathological and surgically treated MV. KEY WORDS:  Mitral valve, In vitro, Papillary muscles  Accepted: January 17, 2011 ORIGINAL ARTICLE on accurate information about their effect on MV function and biomechanics. Given this scenario,  in vitro  and  in vivo  approaches are two complementary and reliable instru-ments for the study of new surgical procedures and/or devices from the proof-of-concept phase to the final op-timization step. The  in vitro  approach relies on the use of mock loop de-vices to replicate hemodynamic conditions in the labora-  © 2011 Wichtig Editore - ISSN 0391-3988 384 Pulsatile simulator for mitral valve test  logy (22-24). One of the most complex aspects to obtain  in vitro  is adequate control of papillary muscle (PM) posi-tioning, which is a determining factor for simulating both physiologic and pathologic conditions. Indeed, it is well known from the clinical literature that an important share of all possible MV pathologies, e.g., ischemic and functional mitral regurgitation, involve valve insufficiency secondary to ventricular wall dilation and associated PM dislocation, in absence of structural alterations of the MV per se.In the 1990s, a mock loop that allowed the operator to control the PM position in space and to measure the ten-sion force on the mitral chordae was developed and then refined (25-28). With that mock loop, investigations con-cerning the influence of PM dislocation on MV competen-ce, kinematics (thanks to video inspection capability), and chordal tension forces were carried out. A positioning sy-stem based on a cylindrical frame, aligned with the apical direction, allowed PM positioning and dislocation to be im-posed. However, the PM radial position in the plane paral-lel to the annular plane could be modified only by stopping the test and opening the ventricular chamber.In this paper we describe the design and development of a new mock loop for MV  in vitro  tests, and the preliminary experimental trials that were performed. The mock loop de-sign was devised so as to allow for individual dislocation of each PM in space according to a tri-axial Cartesian frame while the test is running. In addition, it was outfitted with capabilities to measure the relevant hydrodynamic varia-bles and the forces acting on the PMs, plus visual inspec-tion windows for video recording of the MV in operation. Preliminary tests were run with fresh porcine MVs in order to evaluate the potential of the new apparatus, focusing on the immediacy, width and accuracy of its PM displacement capability. MATERIALS AND METHODS The mock loop consisted of three main elements (in Fig. 1, left, a schematic is provided; right, a photograph of the setup): • a pulsatile pump, consisting of a piston-cylinder system driven by a computer-controlled actuator (P), described previously (29, 30). • a MV test section (MVts), featuring an atrial and a ventri -cular chamber (A and V), a holder system for the mitral annulus, and a system for PM positioning and PM force tory, and its potential is well recognized by the scientific community (7, 8). Preliminary  in vitro  investigations allow one to precisely design subsequently targeted  in vivo  tests, thus limiting the amount of animal experiments needed. For these reasons  in vitro  tests are mandatory and regu-lated by international standards in the design/evaluation phase of devices such as prosthetic heart valves. In vitro  methods can also conveniently contribute to addressing specific issues related to novel surgical approaches. As re-gards the mitral valve, past  in vitro  investigations allowed for mechanistic insight into its function as well as into the effects of conservative/reparative surgical procedures. The information obtained included quantitative data about the force equilibrium within the mitral apparatus, and the force exchange between the latter and the ventricular myocar-dium at the annular level and through the papillary muscles (9-12). Undoubtedly,  in vitro  tests may only partially repli-cate the complexity of the natural organs, since patholo-gical models are generally very complex to replicate with an acceptable degree of accuracy. For this reason,  in vitro  tests are often preliminary to  in vivo  trials, which remain the sole tests able to replicate in a comprehensive fashion the final intended working conditions of the device or sur-gery (in terms of hemodynamics, and the biochemical and biomechanical environment) (13). To some extent, the  in vivo  experimental approach also provides the possibility of replicating pathological models (14, 15). In vivo  experi-ments, however, are quite costly and complex to manage, control and replicate, whereas  in vitro  testing may ensure controllable and repeatable experimental conditions at a reasonable cost, provided that a design trade-off between manageability and physiopathological likeness is found.In the literature, both steady-flow and pulsatile-flow mock loop designs for the  in vitro  analysis of the MV have been proposed. Generally speaking, steady-flow apparatuses allow insight to be gained into specific issues with a rela-tively simple in-lab management of the experimental trial (16), but poorly replicate the  in vivo  working condition of the valve. Pulsatile-flow mock loops are in general more complex to develop, maintain and manage, but, if properly designed, they are intrinsically superior in that they repro-duce the cyclic opening-closing behavior of the MV com-plex when realistic pulsatile flow dynamics are imposed. Starting from the 1970s, pulsatile-flow mock loops have evolved toward increasingly accurate replication, control, and repeatability of the local fluid dynamics (17-21), toge-ther with growing knowledge regarding MV physiopatho-  © 2011 Wichtig Editore - ISSN 0391-3988 385 Vismara et al  the hydraulic connections to the upstream atrial chamber and to the downstream afterload were placed. Mounting of the MV sample under testing at the upstream connection involved suturing the mitral annulus at a ring-shaped, inex-tensible, impermeable patch, which was gripped between the A and V chambers when the bayonet coupling was fa-measurement; • a hydraulic, adjustable afterload mimicking the hydrau -lic systemic input impedance (I), described previously (29, 30). Here we present the design and development of the ele-ments constituting the MVts, together with the subsequent trial tests. The whole process was driven by the following required specifications:  i) accurate PM positioning in spa-ce during tests;  ii)  real-time measurement of PM forces;  iii)  MV visual inspection from the atrial view to monitor its kinematics during tests;  iv)  flexibility with respect to tested MV size. Flow chambers and mitral annulus holder  system The MVts is composed of an atrial and a ventricular cham-ber, both made of polymethylmethacrylate (PMMA), con-nected through a half-turn bayonet coupling. In order to allow for image acquisition of the MV from the atrial view, the atrial chamber (Fig. 2, panel A) was provided with a tangential hydraulic connection for flow inlet. To minimi-ze the fluid-dynamic alterations at the valve inlet, a flow straightener made of a porous material was interposed between the upstream, circumferential-flow region and the downstream, axial flow region leading to the MV. The ventricular chamber (V, Fig. 1) included the housing for a complete excised MV sample (valve annulus and leaflets, chordae tendineae, and PMs) and for the positioning and force measurement subsystems (see below for details). It consisted of an PMMA box, connected laterally to the pumping system, P. The ventricular chamber was closed above by a dismountable top (  t  , Fig. 2, panel B), where Fig. 1 - Left : A schema-tic of the mock loop. A  and V=atrial and ventri-cular chambers respec-tively; P=computer-con-trolled pulsatile actuator; I=systemic input impedan-ce simulator; R=reservoir;  arrows indicate the flow direction. Right : a pho-tograph of the setup, with the external controllers for the PM displacement  marked. Fig. 2 - Schematics of the relevant subsystems of the mock circuit.  A:  exploded view of the atrial chamber, showing the position of the camera for the atrial view of the MV (arrows refer to flow direction towards the MV sample).  B:  3D sketch of the PM positioning sub- system, embedded into the dismountable top (t) of the ventricular chamber (x=CC direction – b cylinder; y=SL direction – c cylinder;  z=apical direction – a linear guide). C:  a detailed sketch of the PM  holder with an embedded load cell located between elements d and e (arrows refer to the direction of applied forces).  D:  photograph of a MV sample housed in the mock loop.  © 2011 Wichtig Editore - ISSN 0391-3988 386 Pulsatile simulator for mitral valve test  Two mini button load cells (Model 13; Honeywell Interna-tional Inc., Morristown, NJ, USA) were inserted between the elements d   and e  of the holder system and silicone-coated to make them suitable for submerged operation (Sylgard ®  184 kit; Dow Corning, Midlands, MI, USA). Each cell was then re-calibrated by applying calibrated weights (0.98-10.8 N) to the load cell and recording the correspon-ding voltage outputs. PM anchoring involved placing the PM itself in the element d   and suturing it to the element e  with an inextensible e-PTFE thread, thus pre-compressing the load cell. The output of the load cell in these conditions was zeroed before each test. The traction force exerted by the chordae tendineae was thus converted into a compres-sion force that could be recorded by the calibrated load cell. Panel D in Figure 2 shows a photograph of an MV sample housed in the mock loop. Hydrodynamic mock loop instrumentation The mock loop was equipped with three piezoelectric pressure transducers (140 PC series; Honeywell Interna-tional Inc., Morristown, NJ, USA), located in the atrial and ventricular chambers, and downstream from the one-way valve in the aortic position, for recording the pressure drop across the MV and the pressure signal at the inlet of the systemic impedance simulator. Transit time flowmeters (HT 110R; Transonic Systems Inc., Ithaca, NY, USA) equipped with 1 inch probes were placed downstream of the valve in the aortic position and upstream from the MV, at the inlet connection of the atrial chamber. The atrial view of valve kinematics was recorded with a high speed CCD camera (up to 1200 fps) (Phantom ®  Miro2; Vision Research, Way-ne, NJ, USA). Experimental protocol and setup for mock loop trial  Preliminary tests with porcine MV samples were aimed at evaluating the conformity of the setup to its design speci-fications. The use of the tri-axial positioning system during the setup phase was evaluated in collaboration with expe-rienced surgeons to bring to light any limitations possibly arising when the desired geometrical configuration for the MV complex was sought. Example hydrodynamic tests were then run with the MV samples, during which the in-fluence exerted on MV performances by PM dislocations was studied. Given the aim of the present work, namely, to stened. At the downstream connection, in turn, a jelly-sh one-way valve, developed in house, acted as the aortic valve. The PM positioning system was also embedded into the top, thus simplifying MV mounting into the system and the anchoring of PMs to the corresponding holders. PMs positioning sub-system This sub-system of the MVts was conceived to allow for in-dependent, repeatable tri-axial positioning of the PMs with sub-millimetric accuracy, even with the mock loop running. The reason for this specification was to avoid having to dismount the setup between successive tests, thus enhan-cing test repeatability. All of the system’s elements were water-resistant and easily replaceable for maintenance purposes. The system was designed with parametric 3D CAD (ProEngineer, PTC, Needham, MA, USA), as shown in the sketch in Figure 2, panel B.The 3D positioning of each PM is made possible by three actuators that translate along mutually orthogonal axes, corresponding respectively to the apical, commissure-commissure (CC) and septo-lateral (SL) directions of the valve. A linear guide driven by a worm screw (IgusDryLin ® ; Igus, Providence, RI, USA) with 1.25 mm pitch (Fig. 2, panel B, element  a  ) enabled apical motion. Two double-effect mini-cylinders (SMC CUJ ®  ) (   b  and c , respectively) allowed for movements in the other two directions. In order to guarantee adjustments during tests, the drivers of the actuators were placed outside the ventricular chamber. For each PM, the external drivers consisted in a counting dial, fixed on the top, driving the worm screw (Fig. 1, right) and two twin cylinders connected to the inner mini-cylinders via a flexible hydraulic line passing through the top (Fig. 1, right). Vacuum-degassed, deionized water was used as the fluid for mini-cylinder actuation. The accuracy of the coupling between the slave mini-cylinders and their master cylinders was experimentally assessed by comparing the imposed displacement versus the obtained displacement, both measured via a vernier caliper. PM holders and force measurement system Figure 2, panel C, shows a detail of each PM holder and force measurement system. Each such system was hosted at the tip of each SL mini-cylinder (Fig. 2, panel B, element c  ), fixed to the piston rod with a hinge coupling, so as to al-low the holder system to rotate around the piston rod axis.  © 2011 Wichtig Editore - ISSN 0391-3988 387 Vismara et al  a 70 mL stroke volume. Under each set of experimental conditions, the force on the PMs, the pressure drop across the valve, and the transvalvular flow rate were acquired.  Valve competence was evaluated by calculating the re-gurgitant volume (RV) through the closed valve (static lea-kage), without considering the dynamic leakage that took place during the valve closure phase (30). Data were ac-quired at a sampling rate of 200 Hz via an A/D acquisition board (6036E; National Instruments, Austin, TX, USA). In addition, the atrial view of the valve kinematics was recor-ded with the high speed camera. The resulting data for each sample were expressed as the mean value ± the standard deviation, calculated over five subsequent beat cycles. Statistical significance, where re-levant, was defined as p < 0.05. RESULTS The behavior of the hydraulic mini-cylinder positioning sy-stem is exemplified in Figure 3, in which the displacement of the slave piston is plotted with respect to the displace-ment of the master piston. Provided that the driving circuit was accurately debubbled, the remaining slight impreci-sion was mainly caused by the free play of the piston ga-skets. A maximum positioning error of 0.75 mm was found in the characterization of the system. The procedures for MV sample mounting and realization of the ev   configuration were carried out by at least one experienced surgeon each, mimicking the geometrical po-sition of the PM as quantified in the ex vivo  heart sample, and were deemed satisfying by the clinicians in terms of immediacy, manageability, and control of PM positioning.  A complete MV sample mounting procedure took 30 to 50 minutes for an experienced user.The effect of the lateral displacement of the PMs on val-ve competence was found to be statistically significant verify the reliability of the mock loop, we focused on apical and lateral (along the CC axis) PM dislocations, for which it was particularly straightforward to verify whether the out-come of our  in vitro  tests was consistent with the previous experimental evidence. According to previous studies (27), apical dislocations have the greatest impact on MV com-petence, whereas lateral dislocations barely affect it.Fresh swine hearts were obtained from the local abattoir and three MVs were harvested by experienced surgeons. Only samples with physiologic anatomies were considered for tests. During the excision procedures, annular size was characterized through the CC dimension, evaluated using a commercial go/no-go sizer; other main anatomical data were measured with a vernier caliper. Table I shows the main measurements taken for samples 1-3. After excision of each valve, its PMs alone were dipped in a 5% glutaral-dehyde solution for 1 minute to increase their stiffness and resistance to stitching.Each MV sample underwent a pulsatile test. The sample was sutured to the impermeable patch by an experienced surgeon so as to avoid any undersizing or oversizing of the valve. The sample was housed in the test section on the mock loop and each PM positioning system was adjusted to reach an initial geometrical configuration that matched the anatomical measurements taken ex vivo  (  ev   configu-ration), then the pulsatile pump was turned on. With the mock loop in operation, the PMs were first laterally dislo-cated from the ev   configuration up to maximal displace-ments of 20 mm (  L  configuration). Then, the influence of the apical dislocation was evaluated by moving the PMs from the ev   configuration in the apical direction at 2.5 mm steps until effects on valve performance were found to be statistically significant (   A  configuration). From the  A  confi-guration, the effect of 3 mm, 6 mm, and 9 mm additional apical dislocations was evaluated (   A+3 ,  A+6 ,  A+9  configu-rations, respectively). Tests were run with the pump set at a 75 bpm beat rate and TABLE I - RELEVANT MORPHOMETRIC DATA OF THE THREE PORCINE MV SAMPLES USED FOR TESTS Sample Heart weight [g]MV diameter [mm]APM-AP distance [mm]PPM-AP distance [mm]APM-PPM distance [mm]144032172013248240162717349436172616  AP = annular plane; APM = anterior papillary muscle; PPM = posterior papillary muscle.   Sample Heart weight [g]MV diameter [mm]APM-AP distance [mm]PPM-AP distance [mm]APM-PPM distance [mm]
Search
Similar documents
View more...
Related Search
We Need Your Support
Thank you for visiting our website and your interest in our free products and services. We are nonprofit website to share and download documents. To the running of this website, we need your help to support us.

Thanks to everyone for your continued support.

No, Thanks
SAVE OUR EARTH

We need your sign to support Project to invent "SMART AND CONTROLLABLE REFLECTIVE BALLOONS" to cover the Sun and Save Our Earth.

More details...

Sign Now!

We are very appreciated for your Prompt Action!

x