Author + information
- Received November 20, 2004
- Revision received December 24, 2004
- Accepted March 10, 2005
- Published online June 21, 2005.
- Florian Botzenhardt, MD⁎ (, )
- Walter B. Eichinger, MD,
- Sabine Bleiziffer, MD,
- Ralf Guenzinger, MD,
- Ina M. Wagner, MD,
- Robert Bauernschmitt, MD and
- Ruediger Lange, MD
- ↵⁎Reprint requests and correspondence:
Dr. Florian Botzenhardt, Department of Cardiology, Krankenhaus Muenchen Bogenhausen, Englschalkinger Str. 77, 81925 Munich, Germany
Objectives The present study evaluates complete supra-annular bioprostheses in patients with an aortic annulus of 18 to 23 mm in diameter.
Background Aortic valve replacement in patients with small aortic annulus using stented bioprostheses is often associated with unsatisfactory hemodynamic results and high incidence of patient-prosthesis mismatch.
Methods Between February 2000 and January 2004, 156 patients with aortic valve disease and an aortic annulus of 18 to 23 mm in diameter received the stented bovine Soprano (Sorin Biomedica Cardio, Saluggia, Italy) (n = 18), Perimount (Edwards Lifesciences, Irvine, California) (n = 52), Perimount Magna (Edwards Lifesciences) (n = 42), or the stented porcine Mosaic (Medtronic Inc., Minneapolis, Minnesota) (n = 44) bioprostheses. Intraoperatively, the surgeon measured the aortic annulus diameter by inserting a hegar dilator. Thus, postoperative hemodynamic results could be referred to the patient’s aortic annulus diameter instead of referring the results to the labeled valve size. This allows for objective comparisons between different valve types.
Results There was no significant difference in hemodynamic results between the different valve types in patients with an aortic annulus 18 to 20 mm. In patients with an annulus 21 to 23 mm, the Magna was significantly superior to the other investigated devices in mean pressure gradient, effective orifice area, and incidence of patient-prosthesis mismatch. There was no significant difference between the complete supra-annular bioprostheses Mosaic and Soprano and the intra-supra-annular Perimount valve.
Conclusions In patients with an aortic annulus of 18 to 20 mm in diameter, hemodynamic performance is independent of the implanted stented valve type and the annular position. Root enlargement or stentless valves may be beneficial alternatives. Patients with annulus diameter 21 to 23 mm benefit from the Magna in complete supra-annular position leading to superior hemodynamic results.
Although the aim of aortic valve replacement in patients with aortic stenosis is to relieve the stenosis and to approach physiologic transvalvular gradients, a residual gradient may remain, especially in small valve sizes. Obviously, parts of the sewing ring and the stent construction are positioned within the blood flow and cause a relative flow obstruction. Yoganathan et al. (1) reported that the area effectively used for blood flow only accounts for 40% to 70% of the total area occupied by the prosthesis, resulting in an incidence of patient-prosthesis mismatch of up to 52% among the patients with a stented aortic bioprosthesis (2).
Aortic root enlargement and, consecutively, the implantation of a larger valve or the implantation of stentless bioprostheses are strategies to achieve an optimal flow orifice and to minimize the transvalvular gradient in small aortic roots. Another concept is the construction of stented prostheses, which are suitable for a complete supra-annular position, so that the stent material is totally placed on top of the annulus and no stent material impairs the blood stream. This technique was realized in the Mosaic (Medtronic Inc., Minneapolis, Minnesota) bioprosthesis, followed by the Perimount Magna (Edwards Lifesciences, Irvine, California), and the Sorin Soprano (Sorin Biomedica Cardio, Saluggia, Italy) bioprostheses. The aim of the present study was to evaluate the hemodynamic performance of these bioprostheses in patients with small aortic annulus and to analyze the existence of hemodynamic benefits resulting from complete supra-annular placement.
Description of the implanted bioprostheses
The Carpentier-Edwards Perimount (Edwards Lifesciences) valve is a stented bovine pericardial bioprosthesis. It received Food and Drug Administration approval in 1991. Long-term studies have constantly shown low-pressure gradients and high freedom rates from valve-related adverse events, even in small valve sizes (3–6).
The Carpentier-Edwards Perimount Magna (Edwards Lifesciences) valve is a stented bovine pericardial bioprosthesis that corresponds to the standard Perimount valve in terms of tissue preparation and prosthetic material. However, the stent design is modified to allow for complete supra-annular placement. Thus, the implantation of a Magna of larger size, compared to the standard Perimount, can be considered. This prosthesis has been in clinical use since September 2002. Up to now there are no hemodynamic and clinical data available.
The Medtronic Mosaic (Medtronic Inc.) valve is a stented porcine bioprosthesis that has been in clinical use since 1994 (Europe) and 2000 (U.S.), respectively, and has proven low-pressure gradients and high freedom rates from adverse events in mid-term follow-up studies (7–9).
The Sorin Soprano (Sorin Biomedica Cardio) valve is a stented bovine bioprosthesis. It corresponds to the Sorin Pericarbon More, which has been in clinical use for over 15 years (10–12), in terms of tissue preparation and prosthetic material, but differs in stent design. It is constructed to allow for complete supra-annular implantation, intended to improve the hemodynamic performance. It has been in clinical use since August 2003.
The geometric dimensions of these prostheses are depicted in Table 1.
Aortic valve replacement was performed using standard cardiopulmonary bypass with moderate hypothermia and cold cristalloid cardioplegic cardiac arrest. After removal of the native aortic valve and after decalcification of the aortic annulus and root, the internal diameter of the aortic annulus was measured by inserting a gauge (hegar dilator) into the annulus (unit: 1 mm). Then the prosthetic valve size was determined by use of the original sizers provided by the manufacturer of the prosthesis. The final choice of the prosthetic size and type was based on the sizing of the patient’s annulus and the best fitting sizer at time of surgery. All bioprostheses were implanted with pledgeted interrupted noneverting mattress sutures in supra-annular position. The position of the Perimount after implantation is called intra-supra-annular, and the position of the Magna, Mosaic, and Soprano is called complete supra-annular. The difference is that the complete supra-annular prostheses are implanted on top of the aortic annulus, and neither stent nor sewing ring material reach into the blood stream and impair the physiologic flow pattern. In contrast, in intra-supra-annular prostheses, the valvular leaflets are located extra-annularly, but stent and sewing ring material reach into the annulus. In intra-annular prostheses, both the sewing ring and the stent reach into the annulus. The intra-annular, intra-supra-annular, and complete supra-annular positions are illustrated in Figure 1.
Between February 2000 and January 2004, 156 patients with aortic valve disease and an aortic annulus of 18 to 23 mm in diameter were included in this prospective, nonrandomized study; 52 patients received the Perimount, 42 patients received the Perimount Magna, 44 patients received the Mosaic, and 18 patients received the Soprano valve. All patients gave informed consent to participate in the study. The study was approved by the local ethics committee. The inclusion criteria were the requirement of aortic valve replacement due to isolated aortic stenosis and combined aortic lesion. Concomitant procedures were admitted, apart from valve replacement in another position. Preoperative and operative data are summarized in Tables 2 and 3.⇓⇓There was no significant difference between the valve type groups regarding preoperative and operative data. Patients were followed-up preoperatively and within 10 days postoperatively using transthoracic echocardiography to evaluate the hemodynamic performance of the prostheses.
Effective orifice area (EOA)was calculated according to the continuity equation: [EOA (cm2) = (left ventricular outflow tract [LVOT] diameter × 0.5)2× π × (velocity time integral [VTI]LVOT/VTIValve). Effective orifice area index (EOAI)(cm2/m2) was calculated by dividing EOA by body surface area. Peak pressure gradient (PG)was calculated by using the Bernoulli equation: [PG (mm Hg) = 4 × (VmaxValve)2]. Mean PGwas provided by a software application integrated in the ultrasound machine based on Bernoulli. Aortic annulus diameterwas measured intraoperatively by inserting a hegar dilator into the patient’s aortic annulus. Aortic annulus areawas calculated by assuming that the aortic annulus approaches the circle shape [aortic annulus area (cm2) = (annulus diameter × 0.5)2× π]. Moreover, we calculated the ratio of the prosthetic orifice area and the patient’s anatomic orifice area, represented by the ratio of EOA and aortic annulus area. It is called effective orifice fraction (EOF)(in %) and has been described earlier by our group (13).
Data are expressed as mean values ± 1 SD. Comparisons between groups were made using the ttest for independent samples in case of normal data distribution, and the Mann-Whitney Utest in case of non-normal data distribution. Comparisons within one group were performed using the ttest for dependent samples or the Wilcoxon matched pairs test, respectively. Normal data distribution was assessed by using Q-Q plots. Differences were considered significant with p < 0.05. Correlations were tested with the Pearson product-moment method.
Mean systolic pressure gradient, EOA, EOAI, and EOF at discharge, grouped by aortic annulus diameter (18 to 20 mm and 21 to 23 mm), are depicted in Table 4.
The hemodynamic results did not differ significantly between the various valve types implanted in patients with an aortic annulus of 18 to 20 mm in diameter using the ttest for independent samples. In patients with an aortic annulus of 21 to 23 mm in diameter, the Magna prosthesis showed significantly lower mean pressure gradients (p = 0.023 vs. Perimount, p < 0.001 vs. Mosaic, p = 0.009 vs. Soprano), larger EOAs (p = 0.015 vs. Perimount, p < 0.001 vs. Mosaic, p = 0.014 vs. Soprano), larger EOAIs (p = 0.010 vs. Perimount, p < 0.001 vs. Mosaic, p = 0.004 vs. Soprano), and larger EOFs (p = 0.028 vs. Perimount, p < 0.001 vs. Mosaic, p = 0.017 vs. Soprano) compared with the other devices using a ttest for independent samples. There was no significant difference between the complete supra-annular valves Soprano and Mosaic and the intra-supra-annular Perimount valve regarding the hemodynamic results using the ttest for independent samples.
Patient-prosthesis mismatch is rated in three grades according to the results in EOAI: no mismatch (EOAI >0.85 cm2/m2), moderate mismatch (EOAI ≤0.85 cm2/m2), and severe mismatch (EOAI ≤0.65 cm2/m2) (14). The incidence of patient-prosthesis mismatch after aortic valve replacement grouped by aortic annulus diameter is shown in Table 5.
There was no difference in the incidence of patient-prosthesis mismatch between the various prostheses in patients with an aortic annulus of 18 to 20 mm in diameter. In patients with an aortic annulus of 21 to 23 mm in diameter, the incidence of moderate and severe mismatch was 8.7% in the Magna group, 41.5% in the Perimount group, 40.0% in the Mosaic group, and 50.0% in the Soprano group.
In Table 6,the hemodynamic results are shown grouped by aortic annulus diameter and by the presence or absence of severe patient-prosthesis mismatch.
There is a significant hemodynamic inferiority of the patients with severe patient-prosthesis mismatch compared to those with moderate or no patient-prosthesis mismatch in all hemodynamic parameters and in both annulus groups using the ttest for independent samples.
Aortic valve replacement in patients with small aortic annulus and aortic root only allows the implantation of a prosthesis small in size, which is often associated with high transprosthetic gradients, low EOA, and low left ventricular mass regression. To warrant satisfactory hemodynamic results, even for patients with small aortic annulus, new technologies in heart valve replacement were introduced.
The aortic root can be enlarged by a patch after the techniques described by Nicks et al. (15) or Manougian and Seybold-Epting (16) to realize the implantation of a larger prosthesis. Castro et al. (17) reported that aortic root enlargement can be performed with minimal added risk relative to standard aortic valve replacement as a preventive strategy to minimize patient-prosthesis mismatch.
Stentless prostheses implanted in the subcoronary position or as full root replacement are associated with superior hemodynamic results. However, the implantation of a stentless valve requires surgical experience to achieve morbidity and mortality rates as low as with stented prostheses.
The concept of the so-called complete supra-annular bioprostheses is to maintain the advantage of the easy and safe implantation of stented valves, but to place them on top of the aortic annulus. The aortic root orifice is naturally narrowed by the aortic annulus ring. The seating of the complete supra-annular prosthesis is upon this annulus ring. Thus, there should be no narrowing of the flow area by the prosthetic stent in addition to the narrowing by the annulus ring.
The Mosaic, Magna, and Soprano bioprostheses belong to the latter category of prostheses. The aim of this study was to evaluate their hemodynamic performance in patients with small aortic annulus. The intra-supra-annular Perimount prosthesis is considered as a reference valve, which has proven low-pressure gradients even in small valve sizes (4,18).
To allow representative comparisons between different bioprostheses of various manufacturers, we introduced a new concept of grouping the prostheses and the obtained results. As shown in Table 1, all prostheses differ in their geometric dimensions. These differences do not allow the comparison of their hemodynamic performance based on the same valve size label. Moreover, our interest is not focused on the performance of a certain valve type in a certain size. Emphasizing our patient-centered point of view, we are more interested in the following question: what hemodynamic performance can a patient with an aortic annulus of 20 mm expect after aortic valve replacement with certain prosthesis? This point of view is also the content of the proposals of the International Organization for Standardization (ISO) concerning the valve size labeling of heart valve prostheses (ISO/CD 5840): the ISO defines “valve size” as “manufacturer’s designation of a valve that indicates the tissue annulus diameter of the patient into which the valve is intended to be implanted.” The tissue annulus diameter is, in aortic valve replacement, the patient’s aortic annulus diameter.
To investigate the prosthetic performance in certain aortic annulus diameters, we started to refer our results to the aortic annulus diameter: patients with an aortic annulus of 18 to 20 mm in diameter and patients with an aortic annulus of 21 to 23 mm in diameter.
In patients with an aortic annulus of 18 to 20 mm in diameter, the implanted valve type and the implantation position (intra-supra-annular, complete supra-annular) did not influence the hemodynamic outcome after aortic valve replacement. This result may be explained by the adversarial ratio of the prosthetic stent to the small blood flow area in this patient group, causing an artificial flow obstruction, independently of the implanted valve type. These patients may benefit form aortic root enlargement and implantation of a larger valve.
In contrast, in patients with an aortic annulus of 21 to 23 mm in diameter, the Magna prosthesis is significantly superior regarding mean pressure gradient, EOA, EOAI, and EOF compared with the other devices. The improvement of the Magna can be attributed to the renewed stent design with smaller sewing ring and lower profile in the Magna compared with the standard Perimount, whereas there is no difference in the internal diameter between both prostheses. Practically, a larger Magna than a standard Perimount can be implanted in the same patient.
Effective orifice area index is the decisive parameter to describe the incidence of patient-prosthesis mismatch. In case of mismatch, there can be a high residual transvalvular gradient, although the valve is absolutely normal. It has been demonstrated that, to avoid any significant gradient at rest and during exercise, EOAI of an aortic valve prosthesis should not be <0.85 cm2/m2(19–21). We rated the extent of patient-prosthesis mismatch as not present in EOAI >0.85 cm2/m2, moderate in EOAI ≤0.85 cm2/m2, and severe in EOAI ≤0.65 cm2/m2(14). This graduation corresponds to the general concept that moderate aortic stenosis of a native valve is present with EOAI <0.90 cm2/m2(22).
Obviously, the stent is positioned within the blood flow and causes a relative flow obstruction. The smaller the aortic annulus, the smaller the implanted prostheses, and, thus, the ratio of stent to effectively used blood flow area worsens.
However, the incidence of patient-prosthesis mismatch of various valve types should not be compared based on the labeled valve size. This method would not consider the benefit of the complete supra-annular implantation technique: the implantation of a larger valve compared with conventional prostheses in the same patient. Therefore, we referred the incidence of patient-prosthesis mismatch to the aortic annulus diameter as an invariant variable.
Our data show that there is no significant difference in the incidence of patient-prosthesis mismatch among the various prostheses in patients with an aortic annulus of 18 to 20 mm in diameter. Regarding the patients with an aortic annulus of 21 to 23 mm, there is a clear superiority of the Magna prosthesis concerning the incidence of moderate and severe mismatch, when compared with the Perimount, Mosaic, and Soprano. This result shows that the complete supra-annular position alone is not decisive for a low incidence of patient-prosthesis mismatch. It is the combination of the complete supra-annular position and a stent design that is characterized by a narrow sewing ring and a low profile, realized in the Magna prosthesis.
When analyzing the effect of patient-prosthesis mismatch on the hemodynamic results, the transprosthetic pressure gradient is expected to increase with decreasing EOAI. In our series, mean pressure gradient in patients with EOAI ≤0.65 cm2/m2was 16.7 ± 4.6 mm Hg vs. 10.8 ± 5.0 mm Hg (p < 0.001) in patients with EOAI >0.65 cm2/m2. The consequence of elevated pressure gradients is a reduced or missing left ventricular mass regression after aortic valve replacement (23,24), which may negatively influence late survival, systolic and diastolic cardiac function, and physical capacity (25,26). The Magna bioprosthesis significantly minimizes the risk of patient-prosthesis mismatch in patients with an aortic annulus of 21 to 23 mm. The impact of this reduced incidence of mismatch on the transvalvular pressure gradients could have been shown in this study. The effect of this benefit on left ventricular mass regression and late clinical outcome will have to be studied in long-term investigational series (27).
Patients’ mean age at operation date was 75.6 years (range 55 to 89 years), representing the increasing number of elderly patients undergoing aortic valve replacement. In our institution, patients from the age of 65 receive a bioprosthesis, so that anticoagulation can be limited to three postoperative months to prevent embolic events caused by prosthetic thrombogenity and episodic atrial fibrillation. Then the risk of major bleeding significantly exceeds the risk of embolic events. Especially older women often present with narrow LVOT, small aortic annulus, and heavy calcification. According to the recommendations by the American College of Cardiology, the surgical procedure in this patient group, according to our results properly requiring root enlargement, must consider the balance between the potential for improved symptoms and survival and the morbidity and mortality of surgery (28). In our institution, we do not always perform aortic root enlargement in case of patient-prosthesis mismatch in this patient group, setting priority to achieve low rates of perioperative adverse events. However, we sometimes have to accept suboptimal hemodynamic performance.
After aortic valve replacement in patients with an aortic annulus of 18 to 20 mm in diameter, there is no statistical difference in the hemodynamic outcome between intra-supra-annular and complete supra-annular stented bioprostheses, and there is no statistical difference among the various complete supra-annular valves. All stented prostheses in this patient group are associated with high incidence of patient-prosthesis mismatch. Hence, there is no benefit of the complete supra-annular prostheses in these patients. Therefore, we recommend for patients with an aortic annulus up to 20 mm the use of aortic root enlargement and, consecutively, the implantation of a larger stented bioprostheses.
Patients with an aortic annulus of 21 to 23 mm in diameter benefit from the implantation of the complete supra-annular Magna bioprosthesis, resulting in a significantly superior hemodynamic performance and a minimized risk of patient-prosthesis mismatch. This benefit could not be demonstrated for the other complete supra-annular valves, Mosaic and Soprano. There was no significant difference between those valves and the intra-supra-annular Perimount prosthesis.
The echocardiographic examinations of the prostheses were financially supported by the manufacturers of the heart valve prostheses: Medtronic Inc., Edwards Lifesciences, Sorin Biomedica Cardio. The authors had full control of study design, methods used, outcome parameters, analysis of data, and production of the written report.
- Abbreviations and Acronyms
- effective orifice area
- effective orifice area index
- effective orifice fraction
- International Organization for Standardization
- left ventricular outflow tract
- Received November 20, 2004.
- Revision received December 24, 2004.
- Accepted March 10, 2005.
- American College of Cardiology Foundation
- Neville P.H.,
- Aupart M.R.,
- Diemont F.F.,
- Sirinelli A.L.,
- Lemione E.M.,
- Marchand M.A.
- Eichinger W.B.,
- Botzenhardt F.,
- Günzinger R.,
- et al.
- Blais C.,
- Dumesnil J.G.,
- Baillot R.,
- Simard S.,
- Doyle D.,
- Pibarot P.
- Nicks R.,
- Cartmill T.,
- Bernstein L.
- Dumesnil J.G.,
- Honos G.N.,
- Lemieux M.,
- Beauchemin J.
- Dumesnil J.G.,
- Yoganathan A.P.
- Sim E.K.W.,
- Orszulak T.A.,
- Schaff H.V.,
- Shub C.
- Sullivan J.M.,
- Zwaag R.V.,
- El-Zeky F.,
- Ramanathan K.B.,
- Mirvis D.M.
- Pibarot P.,
- Dumesnil J.G.
- Bonow R.O.,
- Carabello B.,
- de Leon A.C. Jr..,
- et al.