| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
|
|
|
|||||||
|
|||||||||
Ann Thorac Surg 2002;73:767-778
© 2002 The Society of Thoracic Surgeons
a Division of Cardiovascular Surgery, Sunnybrook and Women’s College Health Sciences Centre, and the University of Toronto, Toronto, Ontario, Canada
* Address reprint requests to Dr Christakis, Sunnybrook and Women’s College Health Sciences Centre, H-406, 2075 Bayview Ave, Toronto, Ontario, Canada M4N 3M5
e-mail: george.christakis{at}swchsc.on.ca
Presented at the Thirty-seventh Annual Meeting of The Society of Thoracic Surgeons, New Orleans, LA, Jan 29–31, 2001.
 |
Abstract |
|---|
 Top Abstract IntroductionMaterial and methodsResultsCommentDiscussionReferences |
|---|
Methods. Fifty-three patients were randomized to receive the stented C-E pericardial valve (CE) and 46 patients the Toronto Stentless Porcine valve (SPV). Annuli were sized for the optimal insertion of both valve types, such that surgeons were required to commit to specific valve sizes before randomization. Echocardiographic measurements and functional status (Duke Activity Status Index) were assessed at 3 and 12 months postoperatively.
Results. Although cardiopulmonary bypass times (CE: 118.6 ± 36.3 minutes; SPV: 148.5 ± 30.9 minutes; p = 0.0001) and aortic cross-clamp times (CE: 95.4 ± 28.6 minutes; SPV: 123.6 ± 24.1 minutes; p = 0.0001) were significantly prolonged in the SPV group, perioperative morbidity and mortality was similar between groups. Neither valve offered a superior internal diameter for any given annular diameter (mean decrease in left ventricular outflow tract diameter after valvular implantation: SPV: 3.4 ± 1.11 mm versus CE: 3.7 ± 1.33 mm;E p = 0.25). Although labeled mean valve size was significantly larger in the SPV group, the actual mean valve size based on internal valvular diameter was no different between groups (CE: 21.9 ± 2.0 mm; SPV: 22.3 ± 2.0 mm; p = 0.286). Although effective orifice areas increased, and mean and peak transvalvular gradients decreased in both groups over time, no differences were demonstrated between groups at 12 months. Similarly, although significant regression of left ventricular mass was accomplished in both groups over time, no differences were demonstrated between groups. Finally, Duke Activity Status Index scores of functional status improved in both groups over time; however, no differences were noted between groups at 12 months postoperatively.
Conclusions. Although offering excellent outcomes, stentless valves did not demonstrate superior hemodynamic indices in comparison to stented valves up to 12 months after implantation.
|
Introduction |
|---|
|
TopAbstractIntroductionMaterial and methodsResultsCommentDiscussionReferences |
|---|
The Toronto Stentless Porcine Valve (SPV) was designed to preclude such limitations by eliminating the valvular sewing ring and stent, thereby maximizing overall area available for flow. The following study attempts to define the possible hemodynamic and clinical benefits of stentless versus stented valves by way of a multicenter prospective randomized trial undertaken in patients undergoing elective bioprosthetic AVR.
|
Material and methods |
|---|
|
TopAbstractIntroductionMaterial and methodsResultsCommentDiscussionReferences |
|---|
|
Due to the proposed benefit of the SPV in small annuli, and due to the possible effects of coronary artery disease on mass regression, patients were stratified for annulus size (
23 mm versus > 23 mm), as well as for the presence or absence of coronary artery disease (
50% stenosis of cross-sectional area of any of the left anterior descending, obtuse marginal, or right coronary arteries) for each surgeon.
Randomization was performed in the operating theater using the sealed envelope technique after patient eligibility was confirmed. Patients were randomized to receive either the Carpentier-Edwards (stented) Pericardial valve (CE; Edwards Life Sciences Inc, Irvine, CA) or the Toronto Stentless Porcine Valve (SPV; St. Jude Medical Inc, Minneapolis, MN). In each patient, sizing for both valve types was undertaken using the appropriate valve sizers (CE and SPV) before randomization. As such, each surgeon was required to commit to a specific valve size before valve selection. This unique protocol was designed to prevent surgeon-specific selection bias. Moreover, to account for the variable influence of each surgeon’s ability to crowd a large stented aortic valve into a small annulus, stratification was undertaken for each surgeon. Finally, annular diameters were recorded for each patient based on manual measurements using standardized metric sizers and calipers.
The study coordinator for each center was instructed on randomization procedures and was primarily responsible for ensuring the integrity of the system. Regular audits were undertaken to ensure adherence to randomization. The analyst charged with assessing randomization was blinded to patient outcome.
Eligibility criteria
All eligible patients who agreed to participate in the study provided written informed consent before enrollment. The consent form was approved by each center’s Institutional Review Board. Eligible patients included those with aortic stenosis who were referred for aortic valve replacement and who, for medical or lifestyle reasons, required a bioprosthetic valve. Patients with mixed aortic valve disease were eligible if the peak aortic transvalvular gradient exceeded 50 mm Hg or if the aortic valve area was less than or equal to 0.80 cm2. Finally, candidates would have to meet the surgical and technical requirements for implantation of both the stented and stentless valve types before randomization.
Preoperative exclusion criteria
Patients specifically referred for the Ross procedure were deemed ineligible for trial enrollment. Similarly, patients were excluded if the surgeon anticipated preoperatively the requirement for concomitant cardiac procedures other than coronary bypass surgery (ie, mitral or tricuspid valve surgery, ventricular myotomy or myectomy, replacement of ascending aorta, or operation for congenital defects). Other exclusion criteria included emergent or reoperative operation, as well as patient unwillingness or inability to return for echocardiographic follow-up.
All patients meeting the above eligibility criteria were investigated by transthoracic echocardiography. Preoperative echocardiography was necessary to identify those patients in whom the SPV could be safely implanted. The diameter of the annulus and the size of the native aorta at the level of the sinotubular junction were measured accordingly. If the diameter of the annulus was greater than that of the sinotubular junction, or if the diameter of the annulus was more than 10% smaller than that of the sinotubular junction, the SPV was deemed unsafe for implantation, and the patient was automatically excluded from enrollment in the trial.
Intraoperative exclusion criteria
Because preoperative echocardiography is not always capable of identifying the discordance between annular sizing and sinotubular junction sizing, measurements were also performed manually (using metric sizers and calibrated calipers) at the time of operation. In the event of inconsistency between echocardiographic and manual measurements, the intraoperative manual measurements were deemed conclusive.
Patients who exhibited extensive calcification of the aortic annulus, aortic sinus, or ascending aorta that was not amenable to safe debridement were also excluded intraoperatively due to the contraindication to SPV implantation.
Finally, to avoid potential contamination of study outcomes, patients requiring concomitant aortic root enlargement or additional cardiac procedures (other than coronary artery bypass grafting) as determined intraoperatively, were excluded before randomization.
Surgical technique
To maximize external validity, each surgeon was allowed to use his or her standard cardiopulmonary bypass and cardioplegic techniques. Once the native aortic valve was excised and the aortic annulus debrided, the surgeon accurately sized the sinotubular junction and the aortic annulus using sizers for both the SPV and CE valves. At this point, the surgeon was able to determine the need for additional cardiac procedures. Randomization proceeded after the requirement for additional procedures was ruled out, and only after the surgeon had definitively committed to the size of each valve.
For CE pericardial valve implantation, interrupted mattressed, pledgeted 2-0 Ticron sutures were placed circumferentially from below the annulus. The valve was implanted in the supraannular position, with the stent positioned so as not to interfere with the coronary ostia. The Toronto SPV was positioned with the muscular shelf corresponding to the noncoronary sinus and commissures positioned approximately 120 degrees apart. The base of the SPV was sutured in the annular position with interrupted 4-0 Ticron sutures, and the rims of the valve commissures were sutured to the native aorta using running 4-0 Prolene sutures.
Outcomes
The primary outcome of this trial was regression of left ventricular mass indexed on body surface area (LVMI). The LVMI has been used to illustrate the regression of ventricular hypertrophy in both experimental [9, 10] and clinical [11, 12] studies involving hypertension treatment. Evidence from the hypertension literature suggests a strong correlation between LVMI and sudden death [13], congestive heart failure [14], and long-term cardiac mortality [15]. The LVMI has also been used to document the incomplete regression of hypertrophy after aortic valve replacement using catheterization techniques [16], echocardiography [17], and computed tomography [18].
The secondary outcome in this trial was the Duke Activity Status Index (DASI). Although a primary goal of AVR is improvement of functional status, exercise capacity may be limited postoperatively, particularly in cases of patient-prosthesis mismatch. The DASI is a disease-specific quality of life index that has been shown to be highly correlated (p < 0.0001) with peak oxygen uptake (Spearman correlation coefficient = 0.58) [19]. As a clinical outcome proxy for exercise capacity, DASI scores enabled evaluation of functional status and quality of life both before and after surgery.
Measurements and calculations
Echocardiographic measurements and calculations are provided in Appendix 1. M-mode echocardiography has been shown to correlate well with contrast ventriculography for left ventricular mass measurement [20]. Transthoracic echocardiographic measurements were performed preoperatively, as well as 3 and 12 months postoperatively. Examinations included two-dimensional, two-dimensional-derived M-mode, and color Doppler analyses. Left parasternal, apical, right parasternal, subcostal, and suprasternal standard views were used in a successive pattern of interrogation. All measurements were averaged over three cardiac cycles in cases of sinus rhythm, and six cardiac cycles in patients with atrial fibrillation. Only two sonographers from each institution (previously assessed for < 5% interobserver variability) were used for the study. Hewlett-Packard Sonos 1000E echocardiographic machines (Hewlett Packard, Palo Alto, CA) were used with a 2.5-MHz transducer. Each examination was recorded and documented on a labeled VHS format videotape, and was submitted to the central institution (Sunnybrook and Women’s College Health Sciences Center). Each videotape included all tracings (M-mode, two-dimensional, continuous wave, and pulsed wave Doppler, color flow). To avoid bias by recognition of valve type on two-dimensional echocardiography, high resolution acetate copies of M-mode information (for calculation of left ventricular mass) were coded for patient identification and read separately. All readings were performed at the core site by two experienced echocardiographers who were blinded to patient and prosthesis. Effective valve orifice area, peak and mean pressure gradients, velocity of circumferential shortening, cardiac output, and LVMI were calculated using standard formulae (Appendix). The LVMI was estimated using criteria set forth by the Canadian Cardiovascular Society [21].
The DASI was similarly administered preoperatively, as well as during echocardiographic follow-up at 3 and 12 months postoperatively. Various aspects of daily activity were assessed, with possible scores ranging from 0 to 58.2 points.
Statistical methods
The SAS for PC (SAS Language Guide for Personal Computers. SAS Institute Inc, Cary, NC; 1988) and BMDP/PLR (Dixon WJ, ed. BMDP Statistical Software Manual. Berkeley, CA. University of California Press; 1992) programs were used for statistical analyses.
The baseline characteristics and hospital outcomes for the two groups were compared using 
2 or Fisher’s exact test for categorical data and t tests for continuous variables.
As previously noted, the regression of LVMI after aortic valve replacement was the primary outcome in this study. We hypothesized that, in comparison to conventional stented valves, implantation of the SPV would result in significantly more mass regression postoperatively. As such, analysis of the primary outcome was based on a treatment by time interaction in a two-way repeated measures analysis of variance.
Analysis of DASI scores was similarly based on a treatment by time interaction in a two-way repeated measures analysis of variance. Furthermore, as it was anticipated that any advantage of SPV implantation would be enhanced in smaller annuli, a separate analysis was undertaken using a three-way repeated measures analysis of variance with a treatment by annulus size by time interaction.
Results are reported as the mean ± standard deviation in text and tables, and as the mean ± standard error in figures, unless otherwise noted. Statistical significance was defined as a p value of less than 0.05. Power calculations were undertaken on an ad-hoc and posthoc basis using both GLM and Proc Mixed methodologies.
|
Results |
|---|
|
TopAbstractIntroductionMaterial and methodsResultsCommentDiscussionReferences |
|---|
Preoperative clinical characteristics including age, body surface area, New York Heart Association functional class, hypertension, the presence of coronary artery disease, and left ventricular grade were similar in the two groups (Table 1).
Table 2 demonstrates the perioperative and postoperative outcomes by group. Cardiopulmonary bypass times (CE: 118.6 ± 36.3 minutes; SPV: 148.5 ± 30.9 minutes; p = 0.0001) and aortic cross-clamp times (CE: 95.4 ± 28.6 minutes; SPV: 123.6 ± 24.1 minutes; p = 0.0001) were significantly prolonged in the SPV group. However, postoperative outcomes including early mortality, myocardial infarction, low output syndrome, pacing requirement, cerebrovascular accident, reoperation for bleeding, prolonged ventilation, sternal wound infection, and prosthetic valve endocarditis were similar between groups, as was the total duration of hospital stay.
|
0.85 cm2/m2) was not significantly different between groups (CE, 11%; SPV, 14%; p = 0.4858). Although the number of patients in this group was too small to permit meaningful analysis, the presence of patient-prosthesis mismatch did not appear to adversely affect the rate or extent of ventricular mass regression, and did not impact functional outcomes.
|
|
Left ventricular mass indexed on body surface area regressed significantly in both groups over time, with most of the effect occurring during the first 3 postoperative months (effect of time, p = 0.0001 in both groups). However, no differences in mass regression were shown between groups at either 3 months (p = 0.6248) or 12 months (p = 0.5671) postoperatively, and no significant effects were demonstrated for the treatment by time interaction (group x time effect, p = 0.5071) (Fig 2).
Overall left ventricular mass regression (LVMR) at 12 months was 22.3 g/m2 in the CE group and 23.8 g/m2 in the SPV group (p = 0.3883). Neither valve had a significant size-related advantage in patients with small ( 23 mm) aortic annuli (LVMR for annuli 23 mm = CE: 23.3 g/m2; SPV: 24.2 g/m2; p = 0.6381).
|
The DASI scores improved significantly over time in both groups (p = 0.0001). Although there was a moderate group effect at 3 months in favor of CE patients (p = 0.0349), no interactive effect of group and time was found in either group (group x time, p = 0.3272) (Fig 3).
|
|
Comment |
|---|
|
TopAbstractIntroductionMaterial and methodsResultsCommentDiscussionReferences |
|---|
Despite the excellent perioperative results of tissue valve replacement for aortic stenosis, 10-year survival rates range from 50% to 66% [1–4, 22, 23]. Compelling evidence suggests that such poor long-term results may be related to the incomplete regression of left ventricular hypertrophy caused by persistently elevated transvalvular gradients [3]. Tissue aortic valves are prone to high transvalvular gradients due to suboptimal leaflet opening at low flow rates, due to the obstructive nature of tissue valve sewing rings and stents, and due to the potential for insertion of small prostheses in large patients [3–8].
The SPV was designed to provide superior hemodynamic performance, thereby enabling more complete regression of left ventricular hypertrophy and improving overall clinical outcome. By eliminating the valvular sewing ring and stent, the manufacturers sought to enable implantation of prosthesis sizes that were 1 to 3 sizes larger than would otherwise be possible with a conventional stented bioprosthesis [24]. Studies from our institution as well as others have demonstrated low gradients and large valve areas after implantation of the SPV in the aortic position [24–29].
Nonetheless, despite such promising findings, few comparative trials exist with which to confirm or refute the favorable results associated with SPV implantation. In this trial, we compared the SPV to a popular stented bioprosthesis using a prospective randomized methodology. Left ventricular mass regression was chosen as a practical, yet clinically relevant surrogate outcome. Left ventricular mass indexed on body surface area is a noninvasive and easily reproducible estimate of the extent of left ventricular hypertrophy. It is a reflection of the severity of aortic stenosis and has been well correlated to peak aortic valve gradients [9–18, 30].
Our study demonstrated a reduction in peak and mean transvalvular gradients over time in both groups. However, no difference was detected between groups. The reduction in transvalvular gradients resulted in a regression of left ventricular mass over time in both groups. However, once again, no difference was noted between groups, despite stratification of randomization by surgeon, the presence of coronary artery disease, and small versus large annular diameters. Stratification for gender and hypertension, two additional factors that may affect LVMR, was not undertaken in this study. Although the upper limit of normal LVMI is lower in women and thus LVMI after operation may be lower in women than in men, the high prevalence of women undergoing AVR (
40%) and the increased complexity of stratifying for four groups made stratification by gender unnecessary and undesirable. Moreover, stratification for annulus size likely allowed for adequate randomization of women, as the female gender can be considered a proxy for small annulus. With respect to hypertension, preoperative identification of hypertension in patients with critical aortic stenosis is difficult if not impossible. Elevated blood pressure in patients with aortic stenosis may occur as a physiologic reflex to compensate for low flow states and heart failure. Moreover, aggressive afterload reduction and vasodilator therapy is contraindicated in this population. As such, preoperative stratification and randomization of patients with hypertension was not possible. The relatively high incidence of hypertension in the general population and the randomized nature of this study likely permitted adequate allocation of patients into each treatment group. Indeed, post hoc analyses confirmed similar allocation of female and hypertensive patients to both groups.
Oversizing of valves may have provided yet another possible source of bias in this trial. In an in vitro model developed by Nagy and colleagues [31], SPV oversizing was found to be hemodynamically detrimental, with a resultant elevation in transvalvular gradients. Accordingly, every effort was taken to ensure appropriate sizing. Furthermore, oversizing as a possible confounder was limited by surgeon stratification and by the multicenter (and multisurgeon) nature of the trial. Moreover, as is evident from the study findings, postoperative transvalvular gradients were extremely low for both valves, such that oversizing was unlikely.
The negative findings of this trial were somewhat unexpected considering the apparent differences in implanted valve sizes encountered. On the basis of manufacturer labeling, the average implanted SPV size was almost two valve sizes larger than the average implanted CE size. However, when based on the actual valvular internal diameters (as measured ex-vivo), no statistically or clinically significant differences in implanted valve sizes were noted between groups. Such misleading valve sizing has previously been reported by our group [32]. Although the external diameter of the CE valve was consistently larger than the external diameter of the SPV, such differences were effectively accommodated by the supraannular design of the former. In keeping with the sizing similarities, the decrease in left ventricular outflow tract diameter after valvular implantation was not significantly different between groups. Moreover, neither valve demonstrated a clear advantage in patients with small annular diameters. Finally, the absence of any size difference was also reflected in the measured effective orifice areas, which improved over time in both groups, but were not different between groups.
The follow-up period in this trial did not extend beyond 12 months as the majority of mass regression has been shown to occur early after AVR [18]. Indeed, the majority of mass regression in this trial occurred within the first 3 months postoperatively.
Exercise capacity may be limited after AVR due to high exercise gradients, especially in patients with small aortic prostheses. Tatineni and colleagues [33] demonstrated that prosthesis size is an independent predictor of exercise tolerance after AVR. Graded bicycle ergometric stress testing or dobutamine stress echocardiography would be the best method for evaluating exercise capacity after AVR. However, the risk of cardiac morbidity and mortality associated with stress testing in patients with critical aortic stenosis precludes its preoperative use. As such, in this trial we elected to use the DASI to evaluate functional capacity. As a clinical outcome proxy for exercise capacity, DASI scoring permitted a reliable evaluation of functional status and quality of life both before and after operation. Although DASI scores improved over time in both groups, no difference was found between groups up to 1 year postoperatively. Moreover, late New York Heart Association functional status, while improving in both groups, was similar at 12 months postoperatively, regardless of implanted valve type.
As expected, due to a somewhat more demanding implantation technique, cardiopulmonary bypass and aortic cross-clamp times were both prolonged in patients receiving the SPV. Nonetheless, such increased times were relatively well tolerated, and did not result in increased perioperative morbidity or mortality. Moreover, the operative procedure was easily manageable by experienced surgeons, and no valvular malfunction (aortic insufficiency) was reported in any patient postimplantation.
Power calculations were performed on a post hoc basis using both GLN and Proc Mixed methodologies. Post hoc estimates of actual sample effect size included F = 0.02 for the group by time interaction, F = 0.04 for the group effect, and F = 0.38 for the time effect. Therefore, for a medium effect size in the true population, this study demonstrated 79% power to detect such a difference (ie, a difference of 15 g/m2 in LVMI, and 10 points in DASI score). Data from the hypertension literature suggests that for each 20 g/m2 increase in LVMI, there is a 1.5 x increase in the relative risk of cardiac morbidity. Thus, a 20 g/m2 regression in ventricular mass index is considered to be clinically significant, and would likely have been detected by this trial. Similarly, an increase of 10 points in DASI scores correlates with a clinically significant improvement in exercise performance, a difference that would likely have been detected by this trial [34]. As such, we believe that this trial was sufficiently powered to enable the above conclusions.
In conclusion, both the stented and stentless valve varieties provided excellent options for bioprosthetic AVR. Although offering outstanding outcomes, the stentless valve did not demonstrate superior hemodynamic indices in comparison to the stented valve up to 12 months after implantation. In view of the above findings, we believe that the advantages of stentless valves, if any, are unlikely to be related to sizing, and will only be determined through long-term follow-up to assess late patient outcome and valvular durability.
|
Appendix |
|---|
R2/4), in square centimeters, obtained from two-dimensional measurement of LVOT diameter; TVILVOT = time velocity integral of forward blood flow, in centimeters, derived from pulse wave (PW) Doppler in the LVOT; TVIAO = time velocity integral of forward blood flow, in centimeters, derived from transvalvular continuous wave (CW) Doppler.
Cardiac output
Cardiac output (CO) is calculated using the following formula: CO = (TVILVOT x CSALVOT) x HR, in liters per minute, where TVILVOT = time velocity integral of forward blood flow, in centimeters, derived from PW Doppler in the LVOT; CSALVOT = LVOT cross-sectional area (R2/4), in square centimeters, obtained from two-dimensional measurement of LVOT diameter; HR = heart rate, in beats per minute.
Peak pressure gradient
Peak velocities obtained from PW and CW Doppler are converted into pressure gradients using Bernoulli’s equation: 
PPEAK = 4(V22 -V12), for an aortic valve, peak systolic pressure gradient in mm Hg, where V2 = peak transvalvular velocity, in meters per second, as measured with CW; V1 = peak velocity in the LVOT, in meters per second, as measured with PW.
Mean pressure gradient
Mean transvalvular pressure gradient is calculated by subtraction of the mean pressure proximal to the aortic valve from the mean distal pressure. Mean pressures are obtained by planimetry of the Doppler spectral envelope: PMEAN = (P2 -P1), mean transvalvular pressure gradient in mm Hg, where P2 = mean distal pressure, in mm Hg, as measured with CW; P1 = mean pressure in the LVOT, in mm Hg, as measured with PW.
Left ventricular mass
Left ventricular mass = 0.8 x [1.04 x (LVIDd + IVSd + PWTd)3 -(LVIDd)3] + 0.6 g, where LVIDd = left ventricular internal dimension at end-diastole, in centimeters; IVSd = interventricular septal thickness at end-diastole, in centimeters; PWTd = posterior wall thickness at end-diastole, in centimeters. This formula for left ventricular mass is based on the volume-corrected ASE-cube method [11].
Left ventricular function
Percent fractional shortening, when derived from M-mode measurements, is based on minor axis shortening and assumes the ventricle contracts symmetrically: %D = (LVIDd - LVIDs)/LVIDd x 100%, where LVIDd = left ventricular internal dimension at end-diastole, in centimeters; LVIDs = left ventricular internal dimension at end-systole, in centimeters.
Velocity of circumferential fiber shortening, when derived from M-mode measurements, represents the velocity of fiber shortening in the minor axis rather than in the whole circumference: Vcf = (LVIDd - LVIDs)/LVIDd x LVET), in circumferences per second, where LVIDd = left ventricular internal dimension at end-diastole, in centimeters; LVIDs = left ventricular internal dimension at end-systole, in centimeters; LVET = left ventricular ejection time, in milliseconds, measured from the onset to the end of systolic flow.
|
Discussion |
|---|
|
TopAbstractIntroductionMaterial and methodsResultsCommentDiscussionReferences |
|---|
This slide shows a comparison of a stentless and a stented 23 mm valve, and I think it is obvious before insertion that the stentless valve has a larger effective orifice area. To end up with two valves of similar size after implantation, one must either make the stented valve larger, which seems impossible to me, or make the flexible stentless valve smaller after implantation, which seems a lot more likely to me. We have used a different stentless valve and have implanted it as a total root replacement. This eliminates the possibility of stuffing one valve inside a fixed container, the native aortic root, and thereby possibly producing some obstruction or altering the size at the time of implantation. The consistent result that we have seen by echo pictures with stentless valves is a valve flow pattern with no turbulence, as shown on the bottom left. The stented valves that we have implanted have typically shown a somewhat turbulent pattern, as is shown there on the right, which demonstrates some degree of obstruction.
The subcoronary technique used to implant the Toronto stentless porcine (SPV) valve, as mentioned in Dr Cohen’s paper, describes oversizing. This is done to maximize leaflet coaptation after implantation and thereby insure valvular competence. The more one oversizes, the more tissue one puts in the same size container and the more likely you are to produce some obstruction. So I wonder whether oversizing is a partial explanation for your early results.
This next slide shows some of your previous data from your institution and also shows the Food and Drug Administration trial data using the Toronto SPV valve. It really shows that you may be simply too early in your randomized trial to draw any comparative conclusions with regard to these two valves, as the Toronto SPV valve demonstrates continual increase in size during at least 3 years. I know of no data that show this increase in size over this length of time in the orifice area of a stented tissue valve. I also wonder whether echo studies at rest are too insensitive to demonstrate the significant difference in effective orifice area in the two groups of patients you studied.
So my questions for you are this. What role do you think technique and oversizing may play in your results, and what results would you anticipate if you compare these groups of patients after 3 years’ time or include exercise data?
Thank you for the privilege of discussing this most interesting paper.
DR TIRONE E. DAVID (Toronto, Ontario, Canada): I congratulate you for a beautiful presentation. I find myself in a difficult position because the authors were my residents who I cultivated their inquiring minds and now they are challenging my own.
The findings of this randomized trial are unique in multiple aspects. Let me mention a few. The aortic valve orifices that you obtained with stented valves in this trial are unique. Those of us who implant pericardial valves every day know that very few patients ended up getting a valve with an effective orifice area of 2 cm2 postoperatively. We do an echo on every patient at Toronto General Hospital just before they are discharged after valve grafting. We do not see valve areas as large as you reported with your patients with stented valves. If you take a look at the Baxter database for the PMA on this valve, the average size implanted was 21, the second most commonly implanted valve was 19, the third was 23, making a total of 76% of all valves. In your study the most commonly implanted valves were 25 and 27. Your patients had very large aortic annuluses to accommodate such large valves. The surgeon may have introduced bias in the randomization process. You said that the diseased aortic valve was excised, the annulus measured and only then were the patients randomized. It looks like most randomized patients had large aortic annuluses.
Finally, the hemodynamic benefit of aortic valve replacement does not end at 1 year. We have published a paper showing that left ventricular mass and left ventricular remodeling continues up to 5 years postoperatively, particularly in patients who had aortic insufficiency.
It is difficult to argue against a randomized trial, but I do not believe your patient population is representative of patients who undergo aortic valve replacement.
DR FRIEDRICH MOHR(Leipzig, Germany): I also would like to congratulate the friends from Toronto on this excellent presentation; however, I have to address my concerns about their results. As many of you may know, we did a randomized trial with both the Toronto and the Freestyle valve, which was published last year in Circulation, focusing on regression of left ventricular hypertrophy. In contrast to the data from Toronto, there seemed to be a major difference.
If you look at the left ventricular hypertrophy demonstrated in this patient cohort, it is only mild hypertrophy with a body index of 110. If you look into our paper, starting off with patients with severe left ventricular hypertrophy, starting off with 140, there is a significant advantage for both stentless valves within 6 and 12 months, and even in the follow-up in 24 months. And I just wonder whether the differences were so little because these patients did not have massive left ventricular hypertrophy. I think we all agree that left ventricular hypertrophy is a high risk factor and the early regression and the mortality in the first year after aortic valve replacement is more significant. I wonder whether you could address that in knowledge of our paper.
DR PAUL KURLANSKY (Miami Beach, FL): I would like to congratulate the presenters on a very well-designed and beautifully presented trial. There are certain questions, however, that arise in addition to those very serious questions that have been raised by the other discussants.
There were approximately 50 patients in each group, and although I am not a statistician, as I understand it, it would take a difference, upward of 70%, to achieve a statistical significance in a group this size. And so I wonder why the presenters are presenting the paper at this time rather than waiting until there are more patients in this study so that the data could be a little bit more meaningful.
Now, the second issue is that I think that by the means of analyzing the data, the presenters have robbed themselves of the beauty of the design of the study. The study design permits comparison of patients based on their actual annular diameter, not the diameter associated with any particular manufacturer or valve size. However, in the presentation of the data, they lumped all of the patients together, rather than telling us for a given actual size internal diameter what was the valve implanted and what was the result. There clearly may be a difference in results that are seen in the smaller size annulus, which is not seen in the larger size and which certainly, with such a small group of patients, gets washed out in a statistical analysis.
And the final question is, I seem to recall an abstract presented to the American College of Chest Physicians out of the same group, which was presented a month ago, which came to exactly the opposite conclusion with a slightly larger number of patients. So I was just wondering how that could be reconciled.
Thank you so much.
DR COHEN: Thank you, Dr Matloff, and I’d like to thank all the discussants for their comments and questions.
With regards to Dr Kon’s comments, I do appreciate the differences demonstrated in your discussion slides. However, such hemodynamic comparisons are based on in vivo findings performed in a pulse duplicator, which as you know is not necessarily generalizeable to the in vivo scenario.
With respect to the possibility of inappropriate valve sizing, I do not believe that our stentless valves were oversized. First, the multicenter nature of this trial and the process of stratification by surgeon served to effectively minimize the impact of such surgeon-specific technical biases. Although you may argue that one surgeon may have had a tendency to oversize, it is unlikely that all of the participating surgeons routinely oversized their SPV valves. Moreover, in a study by Nagy and colleagues, SPV oversizing was shown to result in increased transvalvular gradients. As you have seen from our data, the transvalvular gradients achieved after SPV implantation were remarkably low; certainly as low as, or even lower than most published accounts.
With respect to the previous data you quoted, it is very difficult to compare such results to our current findings as these studies are based not on randomized trials, but on retrospective data, and are thus subject to significant biases. To this day, the SPV is not implanted routinely in all comers, and is reserved mainly for highly selected patients. Thus, such nonrandomized results lack generalizeability.
With respect to your question regarding exercise testing, although this would have been the most appropriate method to assess hemodynamics, such testing is quite risky and often contraindicated in patients with critical aortic stenosis. This being the case, we would not have been able to evaluate our patients preoperatively with such methods. Unfortunately, the absence of preoperative values would have precluded a paired comparison, thus introducing significant variability into the trial outcomes. In view of such limitations, a decision was made to use the Duke Activity Status Index (DASI) as a clinical outcome proxy for exercise capacity. As I mentioned, the DASI is highly correlated with peak oxygen uptake, and is both cost effective and safe. And as we demonstrated in our paired comparison of DASI scores, no differences were observed between groups, regardless of valve type. Perhaps in the future, we’ll continue to assess the progress of our patients using more traditional methods such as dobutamine stress echo or graded bicycle ergometric stress testing.
In response to Dr David’s comments, I believe that the reason for the apparent smaller implanted valve sizes observed in the Baxter data is related to the fact that when the Carpentier-Edwards valve was introduced, it was intended for implantation specifically in patients with small annuli. This would have accounted for the tendency toward smaller mean implanted valve sizes. With regard to the possibility of continued regression of left ventricular mass after 1 year, this is certainly possible, and may be brought to bear with continued follow-up of our patients. Our decision to carry our end point out only to 1 year was based on numerous studies in the aortic valve literature demonstrating that the majority of mass regression occurs within the first year after operation. In addition, if you look at the Food and Drug Administration database, continued regression of left ventricular mass between the first and eighth postoperative years after SPV implantation amounted to no more than 3%. I am doubtful that such differences are statistically or clinically significant. Nonetheless, I can assure you that we will continue to follow these patients long-term to rule out or rule in any ongoing effects.
Regarding Dr Mohr’s comments, I am familiar with the excellent study that your group published recently involving a comparison of stentless and stented valves. Although your study demonstrated significant benefits of stentless valves, the design and statistical analyses were quite different from our trial. First, you included both SPV and Freestyle valves in your trial, which may have introduced significant variability. Although both valves can be implanted using a subcoronary technique, the design features of the two valves are entirely different, which may have affected overall hemodynamics. In view of such differences, I do not believe that the two valves should be considered a single entity in a trial of this nature. Moreover, your group used a comparison of means to compare hemodynamic outcomes between stentless and stented valves. In our trial, we used a paired comparison, such that each patient acted as his or her own control, based on values obtained preoperatively, as well as at 3 and 12 months postoperatively. Methodologically, I believe that this is the most optimal approach for comparison of such data. Finally, you suggested that the left ventricular hypertrophy demonstrated in our cohort was uncharacteristically low, and possibly not representative of the general population of patients presenting with aortic stenosis. However, I call to your attention the results of the Food and Drug Administration database, which includes hemodynamic follow-up on more than 400 patients who have received the SPV since, I believe, 1992. This is by far the largest known SPV database of its kind. And, if you’ll allow me, I’d like to read you the findings with regard to left ventricular mass: The mean discharge LV mass index in the Food and Drug Administration database was 151 g/m2, which decreased down to 124 at 1 year. By comparison, the mean preoperative left ventricular mass index in our trial was 130 g/m2 which decreased down to 106 at 1 year. The overall mass regression was 27 g/m2 in the Food and Drug Administration trial versus 24 g/m2 in our trial. So, I’m sure you’ll agree that these numbers are not different, especially when you take into consideration the standard deviations of 49 and 33 g/m2, respectively.
With regard to Dr Kurlansky’s question, our trial was sufficiently powered for both the primary and secondary outcomes presented. Certainly, if we were to be comparing clinical outcomes, much larger numbers of patients would be required. However, this trial was not designed to compare clinical outcomes. As I mentioned in the presentation, the primary outcome was regression of left ventricular mass indexed on body surface area, whereas the secondary outcome was DASI scores. Assuming a medium effect size in the true population, this study had 79% power to detect a difference as low as 15 g/m2 in ventricular mass index, and 10 points in the DASI score. Because we know from the hypertension literature that a clinically significant difference in mass regression is 20 g/m2 and that a 10-point increase in DASI scores correlates with a clinically significant improvement, the trial was more than sufficiently powered for the conclusions offered. Moreover, to ensure adequate power, our target enrollment was based on highly conservative estimates of the two main outcomes, such that it is more than likely that in actuality, differences greater than 20 g/m2 in mass index or 10 points in the DASI score would be necessary to demonstrate clinical benefit. This being the case, I do not believe that these data were submitted prematurely.
Regarding your second comment, the patients were indeed compared based on annular diameters, and if you recall, I showed you a slide demonstrating the valve sizes implanted for each annular diameter by group, and the remarkable consistency among surgeons in choosing a particular valve size for any given annular diameter. Moreover, the design of the trial was such that annular size was effectively taken out of the equation by virtue of the fact that each annulus was sized for the optimal insertion of both valve types before randomization. Furthermore, as I mentioned, randomization was stratified by annular diameter, such that any differences in small versus large annuli were effectively ruled out.
Finally, regarding the abstract you referred to, you are correct in identifying the obvious inconsistencies. However, as you know, due to its abridged nature, an abstract by definition does not provide sufficient information for proper evaluation. The study you are referring to includes both randomized patients from this trial, as well as nonrandomized patients; hence, the larger numbers. Our trial includes only randomized subjects. Moreover, what the abstract fails to mention is that attrition rates throughout the follow-up period were significant, such that although 127 patients were entered at initiation, only 40 patients were available for early follow-up, and only about 20 patients were available for late follow-up. So I’m sure you’ll agree that any conclusions based on such numbers are presumptuous at best.
Thank you very much.
|
References |
|---|
|
TopAbstractIntroductionMaterial and methodsResultsCommentDiscussionReferences |
|---|
21 mm) aortic valve prostheses assessed by Doppler echocardiography. Amer Heart J 1993;125:138-146.[Medline]
Related Article
This article has been cited by other articles:
|
|
 |
|
  2006 WRITING COMMITTEE MEMBERS, R. O. Bonow, B. A. Carabello, K. Chatterjee, A. C. de Leon Jr, D. P. Faxon, M. D. Freed, W. H. Gaasch, B. W. Lytle, R. A. Nishimura, et al. 2008 Focused Update Incorporated Into the ACC/AHA 2006 Guidelines for the Management of Patients With Valvular Heart Disease: A Report of the American College of Cardiology/American Heart Association Task Force on Practice Guidelines (Writing Committee to Revise the 1998 Guidelines for the Management of Patients With Valvular Heart Disease): Endorsed by the Society of Cardiovascular Anesthesiologists, Society for Cardiovascular Angiography and Interventions, and Society of Thoracic Surgeons Circulation, October 7, 2008; 118(15): e523 - e661. [Full Text] [PDF]
|
|
|
|
 |
|
 S. Narang, D. K. Satsangi, A. Banerjee, and M. A. Geelani Stentless valves versus stented bioprostheses at the aortic position: midterm results. J. Thorac. Cardiovasc. Surg., October 1, 2008; 136(4): 943 - 947. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 R. O. Bonow, B. A. Carabello, K. Chatterjee, A. C. de Leon Jr, D. P. Faxon, M. D. Freed, W. H. Gaasch, B. W. Lytle, R. A. Nishimura, P. T. O'Gara, et al. 2008 Focused Update Incorporated Into the ACC/AHA 2006 Guidelines for the Management of Patients With Valvular Heart Disease: A Report of the American College of Cardiology/American Heart Association Task Force on Practice Guidelines (Writing Committee to Revise the 1998 Guidelines for the Management of Patients With Valvular Heart Disease) Endorsed by the Society of Cardiovascular Anesthesiologists, Society for Cardiovascular Angiography and Interventions, and Society of Thoracic Surgeons J. Am. Coll. Cardiol., September 23, 2008; 52(13): e1 - e142. [Full Text] [PDF]
|
|
|
|
 |
|
 D. Wendt, M. Thielmann, T. Buck, R.-A. Janosi, T. Bossert, N. Pizanis, M. Kamler, and H. Jakob First clinical experience and 1-year follow-up with the sutureless 3F-Enable aortic valve prosthesis Eur. J. Cardiothorac. Surg., April 1, 2008; 33(4): 542 - 547. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
T. E. David, C. M. Feindel, J. Bos, J. Ivanov, and S. Armstrong Aortic valve replacement with Toronto SPV bioprosthesis: Optimal patient survival but suboptimal valve durability J. Thorac. Cardiovasc. Surg., January 1, 2008; 135(1): 19 - 24. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 N. D. Desai and G. T. Christakis Bioprosthetic Aortic Valve Replacement: Stented Pericardial and Porcine Valves Card. Surg. Adult, January 1, 2008; 3(2008): 857 - 894. [Full Text]
|
|
|
|
 |
|
 A. Kallikourdis and S. Jacob Is a stentless aortic valve superior to conventional bioprosthetic valves for aortic valve replacement? Interactive CardioVascular and Thoracic Surgery, October 1, 2007; 6(5): 665 - 672. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 J. Dunning, R.J. Graham, J. Thambyrajah, M.J. Stewart, S.W.H. Kendall, and S. Hunter Stentless vs. stented aortic valve bioprostheses: a prospective randomized controlled trial Eur. Heart J., October 1, 2007; 28(19): 2369 - 2374. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 S. Lehmann, T. Walther, J. Kempfert, S. Leontjev, A. Rastan, V. Falk, and F. W. Mohr Stentless Versus Conventional Xenograft Aortic Valve Replacement: Midterm Results of a Prospectively Randomized Trial Ann. Thorac. Surg., August 1, 2007; 84(2): 467 - 472. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
B. Kunadian, K. Vijayalakshmi, A. R. Thornley, M. A. de Belder, S. Hunter, S. Kendall, R. Graham, M. Stewart, J. Thambyrajah, and J. Dunning Meta-Analysis of Valve Hemodynamics and Left Ventricular Mass Regression for Stentless Versus Stented Aortic Valves Ann. Thorac. Surg., July 1, 2007; 84(1): 73 - 78. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
A. Ali, J. C. Halstead, F. Cafferty, L. Sharples, F. Rose, E. Lee, R. Rusk, J. Dunning, V. Argano, and S. Tsui Early Clinical and Hemodynamic Outcomes After Stented and Stentless Aortic Valve Replacement: Results From a Randomized Controlled Trial Ann. Thorac. Surg., June 1, 2007; 83(6): 2162 - 2168. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
C. R. Bridges, S. M. O'Brien, J. C. Cleveland, E. B. Savage, J. S. Gammie, F. H. Edwards, E. D. Peterson, and F. L. Grover Association between indices of prosthesis internal orifice size and operative mortality after isolated aortic valve replacement J. Thorac. Cardiovasc. Surg., April 1, 2007; 133(4): 1012 - 1021. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
J. B. Chambers, H. M. Rimington, R. Rajani, F. Hodson, and F. Shabbo A randomized comparison of the Cryolife O'Brien and Toronto stentless replacement aortic valves J. Thorac. Cardiovasc. Surg., April 1, 2007; 133(4): 1045 - 1050. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
S. Yadav, A. J. Hodge, A. D. Hilless, and P. D. Skillington Outcomes with Toronto stentless porcine aortic valve: the Australian experience Interactive CardioVascular and Thoracic Surgery, December 1, 2006; 5(6): 709 - 715. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
D. Chatel, C. Mica, D. Blanchard, F. Baud, D. Bruere, A. Maudiere, and P. Peycher Short-term hemodynamic advantages of stentless CryoLife-O'Brien valve over stented bioprostheses for aortic valve replacement Interactive CardioVascular and Thoracic Surgery, October 1, 2006; 5(5): 578 - 580. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
L. H. Cohn Evidence-based cardiac surgery J. Thorac. Cardiovasc. Surg., September 1, 2006; 132(3): 457 - 458. [Full Text] [PDF]
|
|
|
|
|
|
R. O. Bonow, B. A. Carabello, K. Chatterjee, A. C. de Leon Jr, D. P. Faxon, M. D. Freed, W. H. Gaasch, B. W. Lytle, R. A. Nishimura, P. T. O'Gara, et al. ACC/AHA 2006 Guidelines for the Management of Patients With Valvular Heart Disease: A Report of the American College of Cardiology/American Heart Association Task Force on Practice Guidelines (Writing Committee to Revise the 1998 Guidelines for the Management of Patients With Valvular Heart Disease) Developed in Collaboration With the Society of Cardiovascular Anesthesiologists Endorsed by the Society for Cardiovascular Angiography and Interventions and the Society of Thoracic Surgeons J. Am. Coll. Cardiol., August 1, 2006; 48(3): e1 - e148. [Full Text] [PDF]
|
|
|
|
|
|
A. Ali, J. C. Halstead, F. Cafferty, L. Sharples, F. Rose, R. Coulden, E. Lee, J. Dunning, V. Argano, and S. Tsui Are Stentless Valves Superior to Modern Stented Valves?: A Prospective Randomized Trial Circulation, July 4, 2006; 114(1_suppl): I-535 - I-540. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
M. J. Dalmau, J. MariaGonzalez-Santos, J. Lopez-Rodriguez, M. Bueno, and A. Arribas The Carpentier-Edwards Perimount Magna aortic xenograft: a new design with an improved hemodynamic performance Interactive CardioVascular and Thoracic Surgery, June 1, 2006; 5(3): 263 - 267. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
J. B. Chambers, H. M. Rimington, F. Hodson, R. Rajani, and C. I. Blauth The subcoronary Toronto stentless versus supra-annular Perimount stented replacement aortic valve: Early clinical and hemodynamic results of a randomized comparison in 160 patients J. Thorac. Cardiovasc. Surg., April 1, 2006; 131(4): 878 - 882. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
M. A. Borger, S. M. Carson, J. Ivanov, V. Rao, H. E. Scully, C. M. Feindel, and T. E. David Stentless Aortic Valves are Hemodynamically Superior to Stented Valves During Mid-Term Follow-Up: A Large Retrospective Study Ann. Thorac. Surg., December 1, 2005; 80(6): 2180 - 2185. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
D. Perez de Arenaza, B. Lees, M. Flather, F. Nugara, T. Husebye, M. Jasinski, M. Cisowski, M. Khan, M. Henein, J. Gaer, et al. Randomized Comparison of Stentless Versus Stented Valves for Aortic Stenosis: Effects on Left Ventricular Mass Circulation, October 25, 2005; 112(17): 2696 - 2702. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 M. Tamim, T. Bove, Y. Van Belleghem, K. Francois, Y. Taeymans, and G. J Van Nooten Stentless vs. Stented Aortic Valve Replacement: Left Ventricular Mass Regression Asian Cardiovasc Thorac Ann, June 1, 2005; 13(2): 112 - 118. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
C. Amarelli, A. Della Corte, G. Romano, G. Iasevoli, G. Dialetto, L. S. De Santo, M. De Feo, M. Torella, M. Scardone, and M. Cotrufo Left ventricular mass regression after aortic valve replacement with 17-mm St Jude Medical mechanical prostheses in isolated aortic stenosis J. Thorac. Cardiovasc. Surg., March 1, 2005; 129(3): 512 - 517. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
M. Doss, S. Martens, J. P. Wood, A. Miskovic, T. Christodoulou, G. Wimmer-Greinecker, and A. Moritz Aortic Leaflet Replacement With the New 3F Stentless Aortic Bioprosthesis Ann. Thorac. Surg., February 1, 2005; 79(2): 682 - 685. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
S. H. Rahimtoola The year in valvular heart disease J. Am. Coll. Cardiol., January 4, 2005; 45(1): 111 - 122. [Full Text] [PDF]
|
|
|
|
|
|
N. D. Desai, O. Merin, G. N. Cohen, J. Herman, S. Mobilos, J. Y. Sever, S. E. Fremes, B. S. Goldman, and G. T. Christakis Long-Term Results of Aortic Valve Replacement With the St. Jude Toronto Stentless Porcine Valve Ann. Thorac. Surg., December 1, 2004; 78(6): 2076 - 2083. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
J. C. Halstead and S. S. Tsui Randomized trial of stentless versus stented bioprostheses for aortic valve replacement Ann. Thorac. Surg., October 1, 2003; 76(4): 1338 - 1339. [Full Text] [PDF]
|
|
|
|
|
|
S. C. Gauthier, J. G. Barton, M. M. Lane, and R. C. Elkins Pulmonary autografts in patients with severe left ventricular dysfunction Ann. Thorac. Surg., September 1, 2003; 76(3): 689 - 693. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
P. Pibarot and J. G. Dumesnil Is the hemodynamic performance of the carpentier-edwards perimount valve really equivalent to that of stentless valves? Ann. Thorac. Surg., August 1, 2003; 76(2): 656 - 657. [Full Text] [PDF]
|
|
|
|
|
|
G. Cohen Reply Ann. Thorac. Surg., August 1, 2003; 76(2): 657 - 658. [Full Text] [PDF]
|
|
|
|
|
|
O. Lund, K. Emmertsen, I. Dorup, F. T. Jensen, and C. Flo Regression of left ventricular hypertrophy during 10 years after valve replacement for aortic stenosis is related to the preoperative risk profile Eur. Heart J., August 1, 2003; 24(15): 1437 - 1446. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
M. Doss Reply to Walther and Falk Eur. J. Cardiothorac. Surg., August 1, 2003; 24(2): 335 - 336. [Full Text] [PDF]
|
|
|
|
|
|
M. Doss, S. Martens, J.P. Wood, T. Aybek, P. Kleine, G. Wimmer Greinecker, and A. Moritz Performance of stentless versus stented aortic valve bioprostheses in the elderly patient: a prospective randomized trial Eur. J. Cardiothorac. Surg., March 1, 2003; 23(3): 299 - 304. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
N. D. Desai and G. T. Christakis Stented Mechanical/Bioprosthetic Aortic Valve Replacement Card. Surg. Adult, January 1, 2003; 2(2003): 825 - 856. [Full Text]
|
|
|
|
|
|
S. Bevilacqua, J. Gianetti, A. Ripoli, U. Paradossi, A. Giuseppe Cerillo, M. Glauber, M. L. Sacha Matteucci, M. Senni, A. Gamba, E. Quaini, et al. Aortic valve disease with severe ventricular dysfunction: stentless valve for better recovery Ann. Thorac. Surg., December 1, 2002; 74(6): 2016 - 2021. [Abstract] [Full Text] [PDF]
|
|
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
| 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 |
