| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
|
|
|
|||||||
|
|||||||||
Ann Thorac Surg 2005;79:1284-1290
© 2005 The Society of Thoracic Surgeons
a Hospital Clínico Universitario, Valencia
b Hospital Clinic, Barcelona
c Hospital Universitario Marqués de Valdecilla, Santander
d Hospital Universitario 12 de Octubre, Madrid, Spain
Accepted for publication October 4, 2004.
* Address reprint requests to Dr Otero Coto, Servicio de Cirugía Cardiovascular, Hospital Clínico de Valencia, Avda Blasco Ibáñez 17, 46010 Valencia, Spain (E-mail: otero_edu{at}gva.es).
 |
Abstract |
|---|
 Top Abstract IntroductionMaterial and MethodsResultsCommentAcknowledgmentsReferences |
|---|
METHODS: Between May 1999 and March 2002, 80 patients (31 to 81 years of age; mean, 65 years) with a small aortic annulus were randomized to undergo aortic valve replacement with either the Sorin Slimline (n = 40) or St. Jude High Performance (n = 40) valve prosthesis. Clinical and echocardiographic Doppler follow-up was performed at 3 to 4 weeks, and 6 and 12 months postoperatively.
RESULTS: One patient died of non–valve-related causes 9 months after operation. Mean and peak pressure gradients at 6 and 12 months in the Sorin Slimline valve were lower than in the St. Jude High Performance valve for both size 19 and 21 mm. Effective orifice area and effective orifice area index were not significantly different. There was a significant (p = 0.0001) reduction in left ventricular mass and left ventricular mass index between preoperative measurements and at 12 months after surgery for both valves, but there was no difference (p = 0.27) between the Sorin Slimline and St. Jude High Performance valve prosthesis at any other follow-up period. Clinical results showed similarly good results with both valves.
CONCLUSIONS: No clinically significant difference in the hemodynamics of both valves was appreciated; patients with a Sorin Slimline valve exhibited statistically significantly lower pressure gradients, but the small differences in effective orifice area and effective orifice area index did not reach significance. A significant left ventricular mass regression was observed with both valve models. Both prostheses provided a satisfactory clinical outcome.
|
Introduction |
|---|
|
TopAbstractIntroductionMaterial and MethodsResultsCommentAcknowledgmentsReferences |
|---|
Introduced in 1998, the Sorin Slimline (SL) aortic mechanical valve prosthesis is a modification of the Sorin Bicarbon valve, with an increased geometric orifice area. The valve has a sewing ring made of polymeric fabric assembled to an acetal resin ring specially fitted for supraannular implantation, avoiding protrusions of the ring into the patient's annulus; this design may have a larger effective orifice area and other hemodynamic advantages when compared with previous prosthetic models, but thus far no randomized comparative study has been reported on the hemodynamic performance of this mechanical valve prosthesis.
We report the results of a randomized comparative study between a well-tested hemodynamically improved valve prosthesis, the St. Jude High Performance (HP) aortic valve, and the SL aortic valve prosthesis.
|
Material and Methods |
|---|
|
TopAbstractIntroductionMaterial and MethodsResultsCommentAcknowledgmentsReferences |
|---|
Patients with severe aortic valve stenosis (defined as having a mean transvalvular gradient greater than 50 mm Hg with a normal cardiac output or a calculated aortic valve area of less than 0.6 cm2) and a small aortic annulus (defined as an annulus accepting a St. Jude sizing device not greater than 21 mm) were eligible for the study. Patients who required associated coronary artery bypass grafting surgery were also included. Exclusion criteria were patients who refused participation in the study or a mechanical valve prosthesis, patients noncompliant or with contraindications to permanent anticoagulation, intravenous drug abusers, HIV-positive patients, patients who required concomitant valve procedures or who were implanted with other valve prosthesis, patients who had a major noncardiac progressive disease, patients who could not be followed at the outpatient clinic, and patients who underwent LV myectomy during the same operation or having documented hypertrophic obstructive cardiomyopathy.
Valve size distribution was as follows: 3 patients received size 17 (1 SL, 2 HP), 29 patients received size 19 (14 SL, 15 HP), and 48 patients received size 21 valve prostheses (25 SL, 23 HP; Table 1). The three hospitals operated on 24, 27, and 29 patients.
|
All patients underwent routine preoperative investigations according to each hospital's protocol. Echocardiographic Doppler examination and complete clinical information for all patients were required. Postoperative clinical examination was performed before patient discharge, and at 3 to 4 weeks and at 6 and 12 months after surgery.
Operative Techniques
Standard cardiopulmonary bypass with mild hypothermia and cold blood antegrade, retrograde, or combined cardioplegia were used. After valve excision, the aortic annulus was completely debrided of calcium, and then measured to determine the prosthesis size to be implanted. Both prostheses were implanted in a supraannular position with noneverting mattress U-shaped sutures with small pledgets. The septum-perpendicular orientation was routinely used for implantation of both prostheses.
Echocardiography
Echocardiography Doppler examinations were performed in most patients before discharge or at 1 month, and routinely at 6 and 12 months postoperatively. Follow-up was completed at 6 and 12 months in 80 and 79 patients. Good quality records were available at 3 to 4 weeks for 40 patients only. Echocardiographers were blinded as to type of prosthesis used in each patient. Two-dimensional and M-mode echocardiographic data were obtained using apical and parasternal views, following standardized protocols. Data on valve hemodynamics, LV geometry, LV mass and function, and mitral valve function were obtained in all patients. Measurements were averaged for three beats in patients in sinus rhythm and five beats in patients in atrial fibrillation. Effective orifice area (EOA) was calculated by the continuity equation, LV outflow tract area as LV outflow tract velocity time integral times the stroke volume divided by the aortic valve velocity time integral, mean and peak transvalvular pressure drop (in millimeters of mercury) by the expanded Bernoulli equation (with correction for subvalvular velocity), ejection fraction by the Teichholz formula, and LV mass by the Devereux formula.
Statistical Analysis
Data are presented as mean ± standard deviation or absolute numbers. A repeated-measures analysis of variance was used to compare changes in hemodynamic and geometric variables (mean and peak gradients, EOA and EOA index, stroke volume, cardiac output, LV ejection fraction, fractional shortening, LV mass, and LV mass index). This analysis was used to adjust the within-subject variation occurring when multiple measurements were made on the same patient (as in the case of follow-up echocardiographic examinations). The analyses were performed by using the GLM (General Linear Models) Procedure in SAS version 6.12 (SAS Institute Inc, Cary, NC). Statistical significance was determined at p less than 0.05. The analysis of variance was performed on each of the following variables: mean and peak gradients, EOA, EOA index, stroke volume, cardiac output, LV ejection fraction, fractional shortening, LV mass, and LV mass index.
A full analysis of variance model was used to simultaneously test significant differences between valves (SL versus HP), valve size (19 mm versus 21 mm), and timing (preoperative, 6 months, and 12 months). If a significant difference was observed between valve types, then statistical comparisons were made at each follow-up interval. Probability values for comparison of SL and HP (sizes 19 and 21, 6 and 12 months combined) were obtained from the full model split-plot analysis of variance.
The analysis of LV mass and LV mass index was performed across all periods (preoperative, 1 , 6, and 12 months) to detect sequential improvement in LV hypertrophy from the preoperative to the postoperative period. The remaining analysis of echocardiographic variables was made at 6-month and 12-month follow-up periods.
No statistical analysis was performed for the data on size 17 valves because of the small number of patients receiving this size.
|
Results |
|---|
|
TopAbstractIntroductionMaterial and MethodsResultsCommentAcknowledgmentsReferences |
|---|
|
Mean echocardiographic measurements are summarized in Tables 3 and 4.
|
Analysis of EOA showed an increase of 10% (p = 0.32) at 6 months and of 20% (p = 0.11) at 12 months for patients with a size 19 SL prosthesis when compared with those with a size 19 HP prosthesis; patients with a size 21 SL prosthesis showed a 5% (p = 0.43) increase of EOA at 6 months and a 4% (p = 0.33) increase at 12 months. A comparison of all sizes (19 and 21) and all periods (6 and 12 months) showed a possibly significant difference (p < 0.05), but all other comparisons were not significant. The EOA index showed a small increase in patients with an SL prostheses; however, this was not statistically significant.
The low number of patients with a size 17 valve prosthesis precluded any meaningful comparison of the findings. The patient with an SL valve had a body surface area of 1.62 m2 whereas the mean body surface area of the HP patients was 1.38 m2.
Data on stroke volume and cardiac output were significantly lower in patients with a size 19 mm SL prosthesis at 6 months, but not at 12 months; instead they were significantly lower for patients with a size 21 mm SL prosthesis at 12 months, but not at 6 months. Combining both sizes together, patients with an SL prosthesis had significantly lower stroke volume and cardiac output values than patients with an HP prosthesis.
There was no difference between valve types for LV ejection fraction or shortening fraction.
There was a statistically significant (p = 0.0001) reduction in LV mass and LV mass index between preoperative measurements and at 12 months after surgery (Table 4). There was no difference in LV mass between 6 and 12 months (p > 0.05), suggesting that the larger changes in LV mass occurred in the first 6 months after surgery. However, LV mass regression was not significantly different (p = 0.27) between the SL and HP valves. Early (3 to 4 weeks after surgery) LV mass regression was detected, being larger in patients with the SL valve, but these data were derived from a smaller population and the difference did not reach statistical significance.
|
|
Comment |
|---|
|
TopAbstractIntroductionMaterial and MethodsResultsCommentAcknowledgmentsReferences |
|---|
Hemodynamically improved valve prostheses offer the possibility of better results, particularly for patients with a small aortic annulus. In comparison with the standard valve prostheses, previously cited, the HP valve has an annular diameter of 19 and 21 mm and a geometric orifice area of 2.06 and 2.55 cm2, respectively, for sizes 19 and 21 (with an area increase of 26% and 23% when compared with the same sizes of the standard St. Jude valve), whereas the SL valve measures 19.2 and 21.3 mm of annular diameter, and has 2.27 and 2.83 cm2 in geometric orifice area for the same sizes (with an increase of 10% and 11% when compared with the HP valve). With these premises it would be expected that the SL valve prosthesis showed improved hemodynamics in comparison with the standard models and the HP valve (as would also be expected for the new St. Jude Regent valve prostheses, not available at the start of this study).
Our study shows reduced peak and mean transvalvular gradients in the patients with SL valves when compared with patients with the same size HP valves, but differences in stroke volume and cardiac output may have amplified this reduction. There was a small, although not statistically significant, difference in EOA between both types of prostheses, although there was a borderline difference in a combined evaluation of all sizes and both periods of follow-up, perhaps as a result of greater power of the test owing to the increased number of observations. However, 8 patients with a 19-mm HP valve had a calculated EOA index of less than 0.8 cm2 at 12 months of follow-up, whereas only 3 patients in the 19-mm SL subgroup had values below this limit. The values obtained for the size 17 valves seem suboptimal, particularly in the patient with the SL prosthesis; this valve could be acceptable for low-weight patients but significant mismatch may be present in large patients.
Vitale and colleagues [13], in a clinical evaluation of St. Jude HP versus standard St. Jude aortic valve prostheses, reported significantly decreased peak and mean gradients, and a slight increase in EOA and EOA indices, not reaching statistical significance, in patients with the HP valves. Inasmuch as the difference in geometric area between the HP and standard St. Jude valve is greater than the difference between the SL and HP valves, the small differences in EOA and EOA index seen in our patients are in agreement with the findings of the study by Vitale and coworkers [13]. Although theoretically a much larger number of patients could have allowed detection of a statistically significant difference in EOA and EOA index between HP and SL prostheses, such a large study is probably impossible to perform in a selected population of patients with a small aortic annulus; moreover, the clinical influence of such a small difference is doubtful.
Adequacy of echocardiographic Doppler measurements for evaluation of valve prosthesis has been disputed, particularly for some types of prostheses. Baumgartner and associates [14] reported that localized high velocities in bileaflet valves do not reflect the mean velocity distribution across the orifice, and this factor causes an underestimation of the valve area. Laske and colleagues [15] and Bech-Hanssen and coworkers [16] have shown that echocardiography overestimates gradients across bileaflet valves and underestimates EOAs, and noted that high gradients may not be caused by valve dysfunction.
Whether differences in design can affect the results of echocardiographic Doppler evaluation of different bileaflet valves is open to discussion, but some data suggest a positive answer. Shipkowitz and associates [17] showed differences in the angle of opening of the leaflets and in the pattern of flows and velocities with different bileaflet prostheses.
Not only valve-related design factors but also patient-related and technical factors (such as protruding pledgets) may influence EOA and gradient values. Shipkowitz and colleagues [17] showed that leaflet opening angle was significantly affected by the outflow configuration, suggesting that pressure distribution across the leaflets was affected by the amount of leaflet exposed downstream of the orifice. Estimation of EOAs could be influenced by individual anatomic and functional factors, even though mechanical valves are orientated in the same position after implantation. Subvalvular flow velocity is considered important because the standard method leads to overestimation of the gradient obtained as a result of flow acceleration in the LV outflow tract when the proximal velocity exceeds 1 m/s [18]; for this reason the expanded Bernoulli equation, with correction for subvalvular velocity, was used to calculate transvalvular gradients in our series. Pressure recovery may produce an overestimation of the gradients, being different for every bileaflet valve [18]. Changes of measurements with time have been noticed by Vitale and coworkers [13] (in the 23-mm HP valves only); our patients with a size 21 prosthesis, both HP and SL, but not those with a size 19 valve, also showed small increases of average EOA and EOA index; whether this reduction is related to changes in outflow tract anatomy is speculative, but variations in 6-month and 12-month measurements (sometimes relatively large in individual patients) were frequent in our series, thus suggesting technical reasons and possible limitations of individual measurements for evaluation. On the other side, general data may not be applicable to the individual patient for anatomic and functional reasons.
Gillinov and associates [19] proposed that inconsistencies in the determination of the true EOA in clinical practice might support the use of other alternative concepts for bileaflet valves, such as in vitro determined EOA (in spite of its limitations), a theoretical calculation of internal orifice area minus area of obstructing elements, or other alternatives [19]. At the moment no clear recommendations can be established.
In most patients LV mass was found to be within the normal limits at 12 months' follow-up. Other authors have reported that LV mass regression is a late finding after valve replacement [20], but Christakis and coworkers [21] have reported that some changes in LV mass were found early after surgery, a finding also seen in our patients, but in those with a size 17 prosthesis. The observed difference versus time of the LV mass regression could be the result of improved hemodynamics with the new-generation prostheses; Sim and associates [22] and González-Juanatey and colleagues [23] have shown the influence of prosthesis size on LV mass regression after aortic valve replacement, and De Carlo and coworkers [4] have described that a larger prosthesis size appeared to be the only significant predictor of a greater regression of LV hypertrophy.
No clinical differences between both groups of patients were appreciated during follow-up. Therefore, the small hemodynamic differences found in this study seem not to influence significantly the early evolution of these patients.
We conclude that the improved hemodynamic characteristics of these valves make them suitable for use in the small aortic annulus with a low incidence of early complications, low gradients, and adequate LV mass regression, but long-term results have yet to be reported.
|
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 |
