Interact CardioVasc Thorac Surg 2007;6:182-187. doi:10.1510/icvts.2006.142562
© 2007 European Association of Cardio-Thoracic Surgery
Institutional report - Valves |
Physiological function of stentless aortic valves is altered by trimming and removal of aortic wall components
Ralf-U. Kuehnela,*,
Ullrich A. Stocka,
Max O. Wendtb,
Ilka Degenkolbea,b,
Ute Jainskib,
Martin Hartrumpfa,
Manfred Pohlb and
Johannes M. Albesa
a Department of Cardiovascular Surgery, Heart Center Brandenburg, Bernau, Germany
b Institute of Medical Physics and Biophysics, Charite, Berlin, Germany
Received 31 August 2006;
received in revised form 22 December 2006;
accepted 23 December 2006
 Presented at the joint 20th Annual Meeting of the European Association for Cardio-thoracic Surgery and the 14th Annual Meeting of the European Society of Thoracic Surgeons, Stockholm, Sweden, September 10–13, 2006.
*Corresponding author. Herzchirurgie, Herzzentrum Brandenburg, Ladeburger Strasse 17, 16321 Bernau, Germany. Tel.: +49 3338 69 4500; fax: +49 3338 69 4545.
E-mail address: r.kuehnel{at}immanuel.de (R.-U. Kuehnel).
 |
Abstract
|
|---|
Various techniques of stentless aortic valve implantation with or without wall components exist. We investigated the in-vitro performance of stentless valves without or with aortic wall removal mimicking root versus subcoronary implantation. Glutaraldehyde-preserved stentless aortic valves (gpSVG), cryo-preserved human homografts (cpHG), cryo-preserved xenografts (cpXG), and fresh xenografts (fXG) were used. Valves were mounted as full roots or trimmed in a mock circuit. Mean transvalvular gradient (MTVG, mmHg) was measured. Distensibility was quantified using post-systolic backflow volume (BV, ml) – after valve closure. Function was visualized by means of a high-speed camera. Glutaraldehyde-preserved valves exhibited higher MTVG than cryo-preserved or fresh substitutes. After trimming, cpHG, cpXG, and fXG demonstrated marked reduction of MTVG (cpHG: 7.6–5.2 mmHg; cpXG: 6.7–4.9 mmHg; fXG: 8.4–5.2 mmHg). In contrast, after trimming gpSVG exhibited a significant increase of MTVG (7.1–9.2 mmHg). BV remained constant. Visualization indicated maintained distension of all valves and types of all sizes after trimming. In fresh and cryo-preserved grafts, aortic wall trimming resulted in significantly improved systolic performance while glutaraldehyde-preserved stentless valves demonstrated systolic impairment after wall resection. Subcoronary implantation of fresh or cryo-preserved aortic valves may therefore be preferred. In contrast, glutaraldehyde-preserved valves are dependent on wall suspension and may therefore be implanted as a root.
Key Words: Aortic valve; Biological valve replacement; Stentless bioprostheses hemodynamics; Hydrodynamic performance index
|
1. Introduction
|
|---|
Among biological substitutes, stentless valves have gained increasing interest because of their hemodynamic superiority even in small aortic sizes as compared to stented valves. Degeneration, however, is not significantly lower than in conventional stented valves [1]. The large variety of different implantation techniques existing nowadays can be roughly divided into root replacement techniques versus subcoronary implantation strategies [2]. It is, however, currently understood that these entirely different approaches are accompanied by a specific hemodynamic behavior eventually influencing extent and time-course of degeneration [3]. Furthermore, the preservation methods may also influence the valves initial as well as the chronic behavior [4]. In this experimental in vitro study we aimed at the hemodynamic differences of stentless valves after different preservation strategies and implantation technique utilizing the entire root versus subcoronary technique with trimmed wall components.
|
2. Materials and methods
|
|---|
2.1. Tested valves
Fourteen stentless valves were investigated in an in-vitro mock circuit unit as original roots with 40 mm height and after trimming of the wall components in a scalloped fashion thereby removing major parts of the aortic wall as well as the sinuses.
- gpSVG: commercially available stentless aortic valve root (Root Elan, Köhler Chemical, Alsbach, Germany)
- fXG: fresh porcine pulmonary xenograft valve
- cpXG: cryopreserved porcine pulmonary xenograft valve
- cpHG: cryopreserved homograft
Three valves were analyzed in group gpSVG and cpHG, four valves were investigated in group fXG and cpXG. The size of the valves were measured by a Hegar's dilator.
2.2. Mounting of the valves
The valves were initially used as full root. The basis and the supracommissural plane were sewn to rubber plates. These rubber plates were then fixed in the apparatus and measurements were performed. Thereafter, the valves were removed from the apparatus. The upper rubber plate was removed and the aortic wall was trimmed in a scalloped fashion by removal of aortic wall components of the three sinuses. The upper plate was then reattached by means of three commissural sutures while the mounting of the basis remained intact. The valves were again fixed and measurements were performed.
2.3. Heart valve testing device and set-up
An established extracorporeal mock circuit unit designed for highly precise in-vitro measurements of the hemodynamic performance of heart valves and heart valve substitutes was utilized (Fig. 1). The apparatus developed by Schichl and Affeld has been characterized in detail previously [5, 6]. Briefly, a piston-pump generated a constant stroke volume, which was repeatedly ejected through the test valve. This process was controlled by a computer-driven programmable disk armature motor (F12M4, MATTKE AG, Freiburg, Germany) simulating a physiological aortic flow profile. After completion of the forward flow phase, the controller switched from flow control to pressure control so that a physiological diastolic pressure gradient was maintained by a computer-controlled retrograde motion of the piston. Upon termination of diastole, a new cardiac cycle was initiated in order to generate several consecutive beats by the device. The test fluid was saline.
2.4. Parameter assessment and measuring protocol
Measurements during the first beat were discarded to attain a steady pulsatile state. In every experimental setting, measurements were conducted in ten consecutive beats. The pressure differences across the valve were measured with two separate pressure transducers (PR 10, KELLER Ges. für Druckmesstechnik mbH, Jestetten, Germany). The piston displacement was measured with a digital angular transducer (GiO 40, TWK Messelektronik GmbH, Düsseldorf, Germany). Data were recorded in an electronic database for further processing. The resulting flow and pressure curves were digitized with a resolution of 1 ms. This process results in an output diagram including the integrated data: Mean flow rate (MF; ml/s), root mean square flow (RMS; ml/s), stroke volume (SV; ml), simulated heart rate (HR; bpm), cardiac output (CO; l/min), mean transvalvular pressure gradient (MTVG; mmHg). From these basic parameters the following parameters were calculated: Effective Orifice Area (EOA; cm2), Power Transfer (PT; mW), Back Flow Volume (BV; ml), and Hydrodynamic Performance Index (HPI; %). Geometric orifice area (GOA; cm2) was assessed by means of Hegar-stick application. Back Flow Volume (BV) describes the post-systolic backflow volume after valve closure and is a measure of the distensibility of the root. It is noteworthy that BV does not include the backflow through the valve while closing and not the leakage volume through the closed valve. Power Transfer (PT) is the dissipated power representing the amount of power for bending and stretching valve components [7]. Hemodynamic Performance Index (HPI) describes the hydrodynamic capacity of the valve to utilize the respective geometric orifice area is defined by the ratio of EOA and GOA (see Appendix). Function of the valves was visualized by means of a high-speed camera (FAST GmbH, München, Germany) at 1000 frames/s. All frames were digitalized. The respective opening areas of all frames (1 ms) were electronically assessed by planimetry and presented as time curves.
2.5. Data acquisition and statistical analysis
All data were compiled by means of specifically designed software with a standard personal computer. Data were organized in a calculation sheet (Excel for Windows, Microsoft Inc, Seattle, WA). Descriptive and comparative statistical analysis was performed with standard software (SPSS for Windows, SPSS Inc., Chicago, Illinois, USA). Parametrical data were compared by means of ANOVA and post hoc t-tests. Non-parametrical data were analyzed by Mann-Whitney U-test. Data are presented as means±standard deviation (S.D.).
|
3. Results
|
|---|
3.1. Geometric orifice area (GOA; cm2)
Directly prior testing the geometric orifice areas (GOA) were evaluated by means of Hegar-dilators. The investigated commercially available valves demonstrated lower GOA than both xenografts and homografts. The respective sizes of these groups showed only minor variation (Table 1).
View this table:
[in this window]
[in a new window]
|
Table 1 Geometric orifice area and hemodynamic parameters at standard perfusion (Heart rate: 70 bpm/Cardiac output: 4.9 l/min, 35% systole, and physiologic aortic flow profile)
|
|
3.2. Mean transvalvular gradient (MTVG; mmHg)
In the commercially available stentless valves (gpSVG) the group mean value of MVTG increased from 7.13 mmHg to 9.16 mmHg after trimming, whereas it decreased in all other groups from 7.78 mmHg to 5.21 mmHg. This effect of trimming is significant (P=0.00519) (Table 1; Fig. 2). This different behavior was even more pronounced when looking at EOA. While it decreased in gpSVG from 1.86 cm2 to 1.65 cm2 (group mean) after trimming it increased from 1.79 cm2 to 2.19 cm2 in all other valves (P=0.0000427).
3.3. Power Transfer (PT; mW)
PT of gpSVG valves increased from 53 to 98 mW after trimming while it decreased from 93 to 32 mW in all other valves (P=0.0023). In three instances, negative values appeared, as a consequence of statistical variability and are not significantly different from zero (Table 1).
3.4. Back flow volume (BV; ml)
Trimming did not influence BV significantly in all investigated valves (P=0.178). However, BV of gpSVG before and after trimming (3.22±1.04 ml vs. 2.83±0.03 ml) were significantly lower than the values of all other valves (6.96±1.38 ml vs. 5.58±0.89 ml) (P=0.0061 before trimming, P=0.0000037 after trimming) (Table 1).
3.5. Hydrodynamic performance index (HPI)
gpSVG showed higher HPI both in the untrimmend as well as the trimmed state (70.6±4.1% vs. 62.4±0.4%) compared to all other groups (39.4±3.6% vs. 48.4±5.6%) (P=0.0016). HPI of gpSVG decreased after trimming non-significantly (P=0.066). In contrast, HPI increased significantly in all other groups after trimming (P=0.00021) (Table 1, Fig. 3).
3.6. High speed visualization
The commercially available gpSVG showed a lower initially opening than all other valves and less variation of the orifice area during systole indicating a higher stiffness than the fresh or cryopreserved valves (Fig. 4).
View larger version (13K):
[in this window]
[in a new window]
|
Fig. 4. Opening area (mm2) during three systolic cycles in 1-ms intervals of a fresh xenograft (fXG) and a stentless valve (gpSVG) retrieved from highspeed visualization (1000 frames/s).
|
|
|
4. Discussion
|
|---|
It is currently understood from a variety of clinical and experimental studies that the aortic valve must be seen as a functional unit consisting of the left ventricular outflow tract, annulus, cusps, commissures, and aortic wall [8]. Particularly the role of the sinuses must not be underestimated. Such insights have been revealed already 500 years before by Leonardo da Vinci but remained long forgotten. Robicsek and Thubrikar [9] could show in a mock circuit that fresh and cryopreserved aortic roots did not exhibit differences regarding their hemodynamic properties. Following artificial stiffening of the root pressure-volume parameters did not change. However, movement pattern of the cusps was markedly altered. Different movement behavior of the cusps was also seen by Fujimoto et al. [10]. Subcoronary implantation was compared with full-root replacement in stentless bioprosthesis. In the subcoronary technique a larger effective orifice area and a higher forward flow was registered in contrast to root replacement. The authors interpreted this finding as sequelae of different elasticity resulting from the particular implantation technique. Erasmi et al. [11] looked at the influence of different techniques of aortic valve reconstruction on the distensibility of the aortic root. By means of analysis of a variety of functional parameters it was demonstrated that fixation of the native valve within rather stiff prosthetic material, as typically seen in the David-reimplantation technique, had a marked impact on the dynamic movement of the aortic root thereby resulting in hemodynamic impairment in contrast to a more physiological appearance of a valve preserved by using the remodelling technique according to Yacoub [8]. In a clinical study, Leyh et al. [12] could verify the influence of the aortic root on proper cusp movement after valve preserving procedures. Reimplantation of the valve within a tubular graft without prefabricated sinuses resulted in a sharply increased and turbulent flow velocity adversely affecting opening and closure time. As a consequence, contact of the cusps with the prosthetic wall appeared indicating the danger of early degeneration. Similar findings were seen by De Paulis [13] in clinical investigations utilizing the reimplantation technique. He compared a standard prosthesis with a modified elastic prosthesis. Valves reimplanted into the modified prosthesis indeed demonstrated improved physiological movement of the cusps. Lansac et al. [14] investigated the movement pattern of cusps in a sheep model in-vivo. Utilizing sonometric probes at the cusps he could demonstrate that all three cusps enlarged simultaneously but to a different extent resulting in a rotation of the root and a skewed valve plane.
In our study it became apparent that both fresh and cryopreserved grafts showed a maintained elasticity in contrast to glutaraldehyde-preserved valves. Significant differences between fresh and cryopreserved valves were absent indicating that cryopreservation is an adequate tool to provide preserved grafts of high quality. Interestingly though, trimming of the wall components mimicking subcoronary implantation did improve elastic properties as seen by a reduction in backflow volume as well as systolic performance as indicated by a reduction of mean transvalvular gradient. These findings were supported by the high-speed camera analysis showing most physiological opening and closure behavior of the cusps of both fresh and cryopreserved valves.
Glutaraldehyde preservation obviously did impair hemodynamic properties. The commercially available stentless valves investigated in our study demonstrated a markedly higher stiffness of the aortic wall and pliability of the cusps indicated by a rather small backflow volume and the movement pattern assessed by the high-speed camera in contrast to the fresh- and cryopreserved valves (Fig. 4). Though the mean transvalvular gradients were not significantly different in all groups, the Hemodynamic Performance Index of gpSVG was better than in the other groups. The reason is the smaller geometric orifice area of the gpSVG. We hypothesize that these valves obviously made full use of the given geometric orifice area particularly in the native setting.
4.1. Limitation of the study
Artificial models offer a wide spectrum of experimental settings while providing accurate data. Furthermore, they can reduce the number of animal experiments. However, the transfer to the in-vivo situation is limited. Aspects of the central and peripheral circulation, such as the ‘Windkessel’ effect play a major role in valve behavior. This is only simulated by the implemented software in our model. The outflow tract of our model is straight and not naturally curved. As a consequence, the in-vivo hydraulic behavior of a valve may differ from our idealized assumptions. Isotonic saline was used in our model. In contrast, blood is a non-Newtonian fluid exhibiting velocity dependent viscosity. However, Pohl et al. have already demonstrated that the fluid viscosity has little influence on systolic valve parameters [15]. By using physiologic saline, however, pressure gradients may be slightly underestimated. Most certainly, our in vitro study can provide but first insights into the assumed potential and drawbacks of different stentless biological valves regarding a recommendation for a particular implantation technique.
4.2. Conclusions
Glutaraldehyde preservation reduces elasticity of stentless valves while cryopreservation does obviously not adversely affect the performance of stentless substitutes in contrast to fresh valves. Most interestingly, trimming of wall components impaired hydrodynamic performance of glutaraldehyde xenografts while a positive hydrodynamic effect was observed in fresh- or cryopreserved grafts. As a clinical consequence, glutaraldehyde-preserved grafts may be implanted as a root while cryopreserved valves perform better as a subcoronary implant. However, in the clinical scenario, the positive effect of removal of the wall components of cyropreserved valves may disappear if the recipient's aortic root is markedly calcified. In vivo studies are required to verify our initial findings.
|
Appendix
|
|---|
List of abbreviations and symbols
|
Conference discussion
|
|---|
Dr A. Moritz (Frankfurt, Germany): I have a question about the model you made. When you cut out the aortic wall of the sinuses of your prosthesis, how did you replace this? At least in the pulse duplicator model we saw, it looked like you replaced this with a Dacron prosthesis. If you did this, this is not resembling a subcoronary replacement. It is more a Yacoub-like replacement.
Dr Kuehnel: You raised an interesting point. This study is an in vitro study. We don't used Dacron tubes but instead a very elastic tube to reimplant the scalloped valve and we fixed it only with sutures. So we think this model is a good way to make this problem clear.
Dr H-H. Sievers (Luebeck, Germany): I have a question to your model. I saw that your peripheral resistance is a water column. We found that this does not resemble very exactly the peripheral resistance of the arterial system. You can get different motion patterns of the valve. So do you think that your resistance with a water column does represent the three part total sum of the total resistance in the arterial system of humans.
Dr Kuehnel: We used a rather complicated apparatus. In diastole the mock circuit switched from pressure control to flow control. Therefore, you can modulate a constant pressure under diastolic conditions. Does that answer your question?
Dr Sievers: Yes, but it is not only the diastolic pressure, it is indeed the very complex peripheral resistance which also dictates the movement of the leaflets, but I think you have probably considered it.
|
References
|
|---|
- David TE, Feindel CM, Scully HE, Bos J, Rakowski H. Aortic valve replacement with stentless porcine aortic valves: a ten-year experience. J Heart Valve Dis May 1998; 7:250–254.[Medline]
- Bach DS, Cartier PC, Kon ND, Johnson KG, Deeb GM, Doty DB. Impact of implant technique following freestyle stentless aortic valve replacement. Ann Thorac Surg Oct 2002; 74:1107–1113. discussion 1113–1114.[Abstract/Free Full Text]
- Athanasiou T, Jones C, Jin R, Grunkemeier GL, Ross DN. Homograft implantation techniques in the aortic position: to preserve or replace the aortic root. Ann Thorac Surg 2006; 81:1578–1585.[Abstract/Free Full Text]
- Sung HW, Chang Y, Chiu CT, Chen CN, Liang HC. Mechanical properties of a porcine aortic valve fixed with a naturally occurring crosslinking agent. Biomaterials 1999; 20:1759–1772.[CrossRef][Medline]
- Schichl K, Affeld K. A computer controlled versatile pulse duplicator for precision testing of artificial heart valves. Int J Artif Organs 1993; 16:722–728.[Medline]
- Werner S, Wendt MO, Schichl K, Pohl M, Koch B. Testung hydrodynamischer Eigenschaften von Herzklappenprothesen mit einem neuen Prüfstand. Biomed Technik 1994; 39:204–210.[Medline]
- Kuehnel R-U, Pohl A, Puchner R, Wendt MO, Hartrumpf M, Pohl M, Albes JM. Opening and closure characteristics of different types of stented biological valves. Thorac Cardiov Surg 2006; 54:85–90.
- Albes JM, Stock UA, Hartrumpf M. Restitution of the aortic valve: what is new, what is proven, and what is obsolete? Ann Thorac Surg Oct 2005; 80:1540–1549.[Abstract/Free Full Text]
- Robicsek F, Thubrikar MJ. Role of sinus wall compliance in aortic leaflet function. Am J Cardiol 1999; 84:944–946.[CrossRef][Medline]
- Fujimoto T, Kawaguchi D, Shiraishi Y, Iwasaki K, Arira M, Tanaka T, Umezu M. Effect of the elastic conditions around a stentless valvular bioprosthesis on opening behavior. J Artif Organs 2004; 7:133–136.[CrossRef][Medline]
- Erasmi A, Sievers HH, Scharfschwerdt M, Eckel T, Misfeld M. In vitro hydrodynamics, cusp-bending deformation, and root distensibility for different types of aortic valve-sparing operations: remodeling, sinus prosthesis, and reimplantation. J Thorac Cardiovasc Surg 2005; 130:1044–1049.[Abstract/Free Full Text]
- Leyh RG, Schmidtke C, Sievers HH, Yacoub MH. Opening and closing characteristics of the aortic valve after different types of valve-preserving surgery. Circulation 1999; 100:2153–2160.
- DePaulis R. Opening and closing characteristics of the aortic valve after valve-sparing procedures using a new aortic root conduit. Ann Thorac Surg 2001; 72:487–494.[Abstract/Free Full Text]
- Lansac E, Lim HS, Shomura Y, Lim KH. Aortic root dynamics are asymmetric. J Heart Valve Dis 2005; 14:400–407.[Medline]
- Pohl M, Wendt MO, Werner S, Koch B, Lerche D. In Vitro testing of artificial heart valves: comparison between Newtonian and non-Newtonian fluids. Artif Organs 1996; 20:37–46.[Medline]
|
|
Top
Abstract
1. Introduction
