| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
|
|
|
|||||||
|
|||||||||
a Laboratory of Experimental Cardiac Surgery, Department of Cardiovascular Diseases, Katholieke Universiteit (K. U.) Leuven, Leuven, Belgium
b Arbor Surgical Technologies, Irvine, California
Accepted for publication September 20, 2007.
* Address correspondence to Prof Dr Flameng, Cardiac Surgery, University Clinic Gasthuisberg, Herestraat 49, B-3000, Leuven, Belgium (Email: willem.flameng{at}med.kuleuven.be).
| Ms Cunanan and Mr Lane disclose that they have a financial relationship with Arbor Surgical Technologies.
|
 |
Abstract |
|---|
 Top Abstract IntroductionMaterial and MethodsResultsCommentAcknowledgmentsReferences |
|---|
Methods: Nine Trilogy valves (size 21) and six Perimount valves (size 23 or 25) were implanted in the mitral position in adolescent sheep and studied during five months. Hemodynamic measurements were performed at one week, three months, and five months using transthoracic echocardiography. Valve calcification was assessed by X-ray and calcium content was measured by atomic absorption spectrometry after five months implantation in sheep. Tissues were also evaluated histologically (Von Kossa staining).
Results: The nine Trilogy valves had lower peak velocity, peak gradient, and mean gradient compared with the six Perimount valves. These 21-mm Trilogy valves had similar deceleration time and effective orifice area compared with the 23- and 25-mm Perimount valves. Calcification of the Trilogy valves was significantly lower than Perimount valves (p < 0.01), particularly in the commissural (p < 0.01) and free margin regions (p < 0.03). In all parameters assessed, the Trilogy valves exhibited less variation valve-to-valve compared with Perimount valves.
Conclusions: These findings demonstrate that a valve designed to reduce stress in the tissue, improve leaflet kinematics, with advanced antimineralization treatment, can exhibit superior calcification resistance in the mitral position of adolescent sheep. The trilobal geometry and independent leaflet suspension design, combined with an advanced tissue treatment, appears to be a promising breakthrough in the effort to develop a more durable and hemodynamically efficient bioprosthetic valve.
Although long-term results of bioprosthetic heart valves are satisfactory in terms of clinical outcome [1], durability in younger patients is still questionable because of premature valve calcification [2, 3]. In a recent study we were able to show that the origin of heart valve calcification is multifactorial, but design-related stress distribution might be one of the major determinants [4, 5]. Therefore, we were especially interested in the calcification characteristics of the newly designed Trilogy valve (Arbor Surgical Technologies, Irvine, CA). This valve has a trilobal geometry and independent leaflet suspension design, which might potentially optimize stress distribution on the leaflets and reduce leaflet calcification. In adolescent sheep, we tested this valve against a commercially available bovine pericardial valve, the Carpentier-Edwards Perimount valve (Edwards Lifesciences, Irvine, CA), a model known for its accelerated calcification properties [5]. We previously summarized our results of over 120 implants, comparing the juvenile to the adolescent sheep model for a variety of commercially available valves. While calcification is more aggressive in the juvenile compared with the adolescent model, the adolescent model still demonstrated aggressive calcification that was powerful enough to distinguish between the effects of different antimineralization treatments. This is likely due to the still-enhanced calcium metabolism in these young animals. Using slightly older sheep avoids issues with the rapid growth of the animal during the study period.
Specific study endpoints examined were the following: leaflet biostability as assessed by macroscopic analysis, leaflet calcification as assessed by radiographic examination and quantitative calcium analysis of leaflet, pathological analysis of the valve at the conclusion of the study, any pathological consequences to the major organs, and morbidity and mortality of the study animals. During the study, the valves were assessed periodically (at one week, three months, and five months postoperatively) for hemodynamic performance, including measurements of the pressure difference across the heart valve substitute, and an assessment of regurgitation.
|
Material and Methods |
|---|
|
TopAbstractIntroductionMaterial and MethodsResultsCommentAcknowledgmentsReferences |
|---|
Design Specifications and Tissue Treatment of the Trilogy Valve
The Trilogy valve was designed to relieve stress and reduce wear-and-tear on the tissue leaflets, and therefore it has more support structures for the tissue compared with other valve designs [7]. As shown in Figure 1, the finished valve has a distinctive appearance with the trilobal design at the base of the leaflets. The computer-assisted design schematic shows the support elements, including the multilayered frames for each leaflet (called laminates), which suspend each leaflet independently and help absorb the energy of valve opening and closing. The leaflet-laminate assemblies are attached to the valve frame (called the valve crown), which places the structural support elements into the sinuses and out of the blood flow. This design enables the valve to have a larger orifice area compared with other valves, while still providing more structural support and reducing stress to the tissue [7].
|
Design Specifications and Tissue Process for the Perimount Valve
The Perimount valve is a traditional circular valve comprised of an Elgiloy wire frame and three individual leaflets [10]. This valve is also treated with Tween-80 as a calcification mitigant [11, 12]. According to the website of Edwards Lifesciences (http://www.edwards.com), the tissue is cross-linked using a Neutralogic fixation, which occurs in a stress-free bath of glutaraldehyde. Further details are not disclosed [13].
Surgery and Follow-Up
Adolescent ewes accepted for the study were less than 12 months of age. The animals were bred at the Zootechnical Center of the K.U. Leuven and procured for research through the Animalium K.U. Leuven.
A total of 18 animals were studied: 11 were implanted in the mitral position with Trilogy valves and 7 with Perimount valves. The study design was based on International Standards Organization 13485, which specifies six test and two control valves. We desired to achieve six survivors for each valve type at five months, but due to our inexperience with the Trilogy valve we implanted a greater number of animals with the test valve to ensure an adequate number of survivors at five months. The operative technique is described previously [5] and included placing the animal on cardiopulmonary bypass. One Trilogy animal died from endocarditis on postoperative day 11 and one Trilogy animal suffered accidental strangulation and death in its cage at day 65. Both animals were excluded from further analysis. Six Perimount animals survived to the five-month endpoint and one Perimount animal was electively sacrificed at three months and removed from the study (this was the only mitral type valve; all other valves were aortic type valves). Blood screening was performed at regular intervals to ensure overall animal health.
Echocardiographic Follow-Up
At day 7 postoperative, two-dimensional echo, color flow, and pulsed Doppler studies were performed by an experienced echocardiographer (M.C.H.) by transthoracic echocardiography (TTE), repeated at 90 days, and immediately prior to sacrifice at 150 days. The following parameters were scored: leaflet mobility, number of functional leaflets (0/3 to 3/3); flow turbulence, obstruction (0 = absent; 1 = mild; 2 = moderate; 3 = severe); regurgitation (0 to 4); coaptation area, complete coaptation – slight malcoaptation – severe malcoaptation; attachments (vegetation, thrombus) to the leaflets (Y/N); laminar flow, presence of laminar flow through the implant (Y/N); pressure gradients (peak and mean), deceleration time, and valve surface area; and flow velocities.
Valve Explantation
At five months the animals were anesthetized and euthanized with an overdose of a potassium chloride solution (intravenously) and the heart was removed. The implanted bioprosthesis was excised and, after careful rinsing, evaluated. Macroscopic photographs of the valves were taken from both the inflow and outflow surfaces.
X-Ray Assessment
X-ray examination was performed under mammography conditions (Faxitron, Wheeling, IL) to demonstrate and localize gross calcification. When considering the possible sites of calcification in each valve, scoring each leaflet, each commissure, and three corresponding wall portions, there were nine possible sites for calcification on each valve. Due to the unique structural features of each valve, which are evident upon X-ray, it was not possible to score the X-ray images in a masked fashion.
Histologic Examination
For each specimen, 5-µm-thick cross-sections were prepared from the leaflet of the bioprosthesis. Sections were embedded in paraffin. All sections were stained with hematoxylin and eosin, Masson’s trichrome stain for collagen, Von Giesson stain for elastin, phosphotungstic-acid-hematoxylin stain for fibrin, and Von Kossa for calcium staining. By means of light microscopy the histologic integrity of the tissue was evaluated, including localization and extent of calcification, presence of an inflammatory response in the tissue, and extent of fibrous sheathing over the valve tissue.
Quantitative Calcium Examination
Half of every leaflet in every valve was used for quantitative calcium determination. The leaflet half was divided into a basal portion, a commissural region, and a free edge. After lyophilization, the tissue was pulverized and desiccated to constant weight in an oven. Hydrolysates were made in 6N hydrogen chloride. Calcium content was measured by flame atomic absorption spectrometry and expressed as microgram per milligram dry weight.
Statistical Analysis
Because nonparametric statistics were used the data are represented as median and range. Different statistical techniques were used based on the data properties. Calcium score distributions from X-ray assessment were analyzed with a nonparametric Kruskal Wallis exact test for singly ordered R x C tables (StatXact 4.0.1; Cytel Software Corp, Cambridge, MA). Absolute calcium levels of the different groups were compared with a nonparametric Wilcoxon-Mann Whitney test (SPSS 14.0 for Windows; SPSS Inc, Chicago, IL).
Variables followed in time (hematology and TTE) were analyzed within the groups using a nonparametric paired Friedman test. To determine valve differences and time effect the data were first analyzed with a full factorial mixed linear model (time, valve and time valve interaction) but if the interaction term was not significant it was omitted from the model (SPSS. 14.0 for Windows, SPSS Inc.). To achieve a random intercept model the sheep were considered individual subjects. Variables were considered significantly different if p was less than 0.05.
|
Results |
|---|
|
TopAbstractIntroductionMaterial and MethodsResultsCommentAcknowledgmentsReferences |
|---|
Valve Hemodynamics
The TTE measurements performed at one week, three months, and five months were evaluated and transvalvular peak velocity, peak gradient, mean gradient, deceleration time, and valve surface area were determined. The results are listed in Table 1.
|
Leaflet coaptation area was also assessed during the TTE exams at one week, three months, and five months. In general, leaflet coaptation was considered complete for all valves at all time points, with the exception of one Perimount valve, which exhibited slight malcoaptation at five months.
Valve attachments were also assessed during the echocardiographic examination. Three of six Perimount valves (50%) exhibited thickened leaflets and (or) calcification by echo assessment at five months. Only one Trilogy animal demonstrated thickened leaflets by echo and this animal was found to have developed endocarditis.
Color Doppler techniques were used to evaluate the degree of valvular regurgitation. Overall, the average regurgitation score for the Trilogy and the Perimount valve groups did not differ significantly nor change with time (p = 0.497, stratified R x C tables, Mann-Whitney test). Leaflet mobility was acceptable in all leaflets of all valves at all time points. Flow was observed to be laminar in all valves at all time points. No evidence of paravalvular leak was observed in any valve at any time point.
Gross Examination
All leaflets were observed to have flexibility at the time of explant. All sewing rings were covered with a glistening white pannus, with re-endothelialized suture tails protruding above the sewing ring on the outflow side. Moderate to heavy pannus growth onto the leaflets was observed on most valves, particularly on the outflow side of the valve. There were no signs of thrombosis or excess platelet deposition on any of the valves. Typical examples of explants are shown in Figure 2.
|
When calcifications were seen on X-ray, they were preferentially located at the commissures. Commissural calcifications were found in five of six Perimount valves and four of nine Trilogy valves. Leaflet calcifications were less frequent: they were found in two of six Perimount valves and in one of nine Trilogy valves. Wall portion calcifications were rare; in one of six Perimount valves and in none of Trilogy valves. Figure 2 represents typical Faxitron results from both valve types.
When considering the possible sites of calcification in each valve, scoring each leaflet, each commissure, and three corresponding wall portions, there were nine possible sites for calcification on each valve. Out of the possible sites in the Trilogy valves, only 11% had any evidence of calcification by radiologic techniques; the Perimount valves exhibited positive signs of calcification in 37% of the sites. This difference in calcification incidence between Trilogy and Perimount valves is statistically significant (p < 0.02, Kruskal Wallis exact-singly ordered R x C tables).
Continuing to examine this cohort, we see that the difference in calcification incidence at the commissures is 25% in the Trilogy valves, while 77% of the commissural sites in the Perimount valves calcified. Statistical significance in this study cohort, taking into consideration the distribution of the calcification in the individual valves, is only borderline (p = 0.06; Pearson 
2 exact test).
Histologic Evaluation
Tissues in all valve leaflets appeared well-preserved, with substantial pannus overgrowth on the leaflets, particularly on the inflow side of the valves. Trace calcification of the valve leaflets was detected histologically in two of the nine Trilogy animals (22%) and in three of the six Perimount animals (50%).
Calcium Content by Atomic Absorption Spectroscopy
Overall, the average calcium content of tissue leaflets in the Trilogy group was 1.05 (range 0.65 to 2.58) µg Ca++/mg dry weight, compared with an average content of 3.23 (range 1.52 to 23.8) µg Ca++/mg dry weight in the Perimount group (p = 0.01, see Table 2).
|
|
Comment |
|---|
|
TopAbstractIntroductionMaterial and MethodsResultsCommentAcknowledgmentsReferences |
|---|
The Trilogy valve demonstrated both low overall calcification as well as no position dependence for calcification. This is in sharp contrast to the Perimount valve, which not only exhibited higher levels of overall calcification, but also demonstrated a strong positional dependence, where calcification at the commissures was particularly high compared to the Trilogy valve commissure calcification. These data support the basic design premise of reducing stress at the commissures due to the unique independent suspension used in the leaflets of the Trilogy valve.
The free margin region of the Trilogy valve also calcified less than the free margin region of the Perimount valve, which indicates that the tissue preparation method used in the manufacturing of the Trilogy valve provides an edge, which calcifies less than the edge of the Perimount valve leaflets. This supports the design intent of better retention of tissue architecture, which would by its nature result in less protein insudation and calcification.
Finally, the base region of the Perimount valve contained twice the average calcium content of the base region of the Trilogy valve. This comparison demonstrates the difference in tissue treatments, because the base region of the leaflets is not under stress. These differences in processing include the addition of enzymatic inhibitors and controlling tissue swelling in the Trilogy valve tissue, as well as cross-linking the tissue while it is under an applied load, which produces more consistent biomechanical behavior across different tissues [8]. Both valves are treated with Tween-80, an effective anticalcification treatment [9].
Therefore, we can determine that there is an improvement in the tissue process in the Trilogy valve that reduces its propensity to calcify compared with the Perimount valve. The interpretation of the observed superiority of the Trilogy valve in terms of hemodynamic performance is more complex. The manufacturer sized the valves used in this study as "size 21 mm." This was done because this size represents the size of a tube fitting into the valve having a trilobal configuration. However, our measurements of the geometric orifice area assessed from the Faxitron pictures reveal that the Trilogy valve and the 23-mm Perimount valve have nearly the same geometric orifice area. The majority of our control valves (four of six) were Perimount size 23 valves. This may explain why in vivo echocardiography reveals similar deceleration times, reflecting similar effective orifice areas between both Trilogy and Perimount 23-mm valves. The fact that we found, predominantly at the early postimplant echocardiograms, lower peak and mean velocities and gradients, must be related to the differences in valve design and leaflet suspension between both valve types. Then, data indicate that the so-called "size 21" Trilogy valve performs like a 23 or 25 Perimount valve.
Taken together, the data support the improvements in the design, processing, and manufacturing of the Trilogy valve, which translate to measurable and statistically significant differences in outcomes compared with the Perimount control valve. Based on this comparison, the Trilogy valve is suitable for study in humans.
|
Acknowledgments |
|---|
|
TopAbstractIntroductionMaterial and MethodsResultsCommentAcknowledgmentsReferences |
|---|
|
References |
|---|
|
TopAbstractIntroductionMaterial and MethodsResultsCommentAcknowledgmentsReferences |
|---|
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
| 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 |
