Valvular Heart Disease

Transcription

1 Valvular Heart Disease Prosthesis-Patient Mismatch Predicts Structural Valve Degeneration in Bioprosthetic Heart Valves Willem Flameng, MD, PhD; Marie-Christine Herregods, MD, PhD; Monique Vercalsteren; Paul Herijgers, MD, PhD; Kris Bogaerts, PhD; Bart Meuris, MD, PhD Background Prosthesis-patient mismatch (P-PtM) after aortic valve replacement results in disturbed valve performance associated with increased pressure gradients. However, it is unknown whether this can be related to future structural valve deterioration (SVD) of the bioprosthesis. Methods and Results In 564 patients (mean age, 74 5 years) receiving an aortic valve bioprosthesis, clinical follow-up (median, 6.1 years; maximum, 16.4 years) was analyzed including echocardiography. SVD was diagnosed in 40 patients (7%) as substantially increased stenosis (n 24) or regurgitation (n 16) of the operated valve over time. When patients with P-PtM (effective orifice area index 0.85 cm 2 /m 2 ;n 285) developed SVD, it was preferentially of the stenosis type, whereas when patients without P-PtM (n 279) developed SVD, the majority was of the incompetence type (P 0.05). Multivariable analysis including patient- and valve-related variables revealed that P-PtM and label size 21 were independent predictors of SVD (P 0.04 and P 0.02, respectively). A nonparametric Turnbull estimate analysis showed that SVD is virtually nonexistent for up to 9 years in patients without P-PtM. Thereafter, SVD starts to occur and is mainly of the incompetence-type SVD (79% of cases). In patients with P-PtM, SVD starts to occur after 2 to 3 years after implantation and is mainly of the stenosis-type SVD (81% of cases). Conclusions These data suggest that stenosis-type SVD is an early, P-PtM related, and thus preventable phenomenon. Incompetence-type SVD is a time-dependent, nonspecific wear damage to bioprosthetic valves, which is not related to P-PtM. (Circulation. 2010;121: ) Key Words: bioprosthesis prosthesis-patient mismatch surgery valves As introduced by Rahimtoola, 1 prosthesis-patient mismatch (P-PtM) is present when the effective orifice area (EOA) of the inserted prosthetic valve is too small in relation to body size. As shown by Pibarot and Dumesnil, 2 P-PtM is common (20% to 70% of aortic valve replacements) and may be associated with less regression of left ventricular hypertrophy, more cardiac events, and lower survival. The effect of P-PtM on clinical outcome after aortic valve replacement is not accepted universally because several reports fail to show P-PtM as a risk factor for postoperative mortality and morbidity Nevertheless, the concept of P-PtM implies as the main hemodynamic consequence the generation of higher than expected gradients through normally functioning prosthetic valves. 1 Such a disturbance of hemodynamic flow patterns may have an influence on the structural integrity of the prosthetic valve tissue over time. We know from experimental work 11 that the origin of calcification of bioprosthetic valves is multifactorial. Besides the design-related stress distribution on the device, the age of the recipient, the type of agent used to cross-link the tissue, the anticalcification treatment used during the preparation of the device, and the kind of tissue itself have been shown to determine the calcification potential of bioprosthetic valves. Editorial see p 2083 Clinical Perspective on p 2129 However, not all bioprostheses showing structural valve degeneration (SVD) suffer from stenosis and calcification. Some valves show only rupture of the cusps, whereas others show the combination of leaflet calcification and rupture. Cusp ruptures were often associated with fatigue of the material, which was glutaraldehyde-fixed porcine aortic tissue or bovine pericardium. On the other hand, tears in the leaflets were also associated with microscopic or macroscopic calcification of the tissue. 12 Therefore, follow-up echocardiographic data, including pressure gradients across the valve, descriptive measures of leaflet calcification, and semiquantitative indications of regurgitation, become mandatory to identify the mode of valve failure, such as leaflet calcification and degeneration leading to valve stenosis or leaflet rupture leading to regurgitation. It is the objective of this study to classify bioprosthetic valve failure in this respect Received August 11, 2009; accepted March 17, From the Divisions of Cardiac Surgery (W.F., B.M., M.V., P.H.) and Cardiology (M.H.), Department of Cardiovascular Diseases, Katholieke Universiteit Leuven, Leuven, Belgium; and Biostatistical Center, Universiteit Hasselt, Diepenbeek, Belgium (K.B.). Correspondence to Willem Flameng, MD, PhD, Cardiac Surgery, University Clinic Gasthuisberg, Herestraat 49, B-3000 Leuven, Belgium. willem.flameng@med.kuleuven.be 2010 American Heart Association, Inc. Circulation is available at DOI: /CIRCULATIONAHA

2 2124 Circulation May 18, 2010 Table 1. Preoperative Patient Characteristics Table 2. Valve Types, Implant Period, and Follow-Up Total No. of patients 564 Gender, % male 51 Age, mean SD, y Body surface area, mean SD, m Aortic valve stenosis regurgitation, % 93 Ejection fraction, mean SD, % Proportion of patients with ejection fraction 50%, % 19 Concomitant CABG, % 47 Rhythm, % Sinus rhythm 88 Atrial fibrillation 7 Pacemaker 5 NYHA functional class, % Class I 5 Class II 47 Class III 41 Class IV 7 Metabolic syndrome ( 3 of the following 5 factors present), % 10 Obesity (body mass index 30), % 15 Arterial hypertension, % 67 Hypercholesterolemia, % 31 Statin therapy, % 17 Diabetes mellitus (oral treatment or insulin), % 14 CABG indicates coronary artery bypass grafting; NYHA, New York Heart Association. and to elucidate a potential relationship between the mode of valve failure and P-PtM. Our hypothesis is that P-PtM, producing disturbed hemodynamic flow patterns across the bioprosthesis, can promote leaflet calcification and valve stenosis over time. Methods Patient Population A group of 564 consecutive patients underwent single aortic valve replacement with the use of a bioprosthesis. Exclusion criteria were double valve replacement, failed prosthesis replacement, Bentall operation, and acute endocarditis. Inclusion criteria for this cohort were age 70 years and the personal choice of the patient for a tissue valve. Patient characteristics are listed in Table 1. Surgical Technique Aortic valve replacement was performed with the use of standard techniques. Forty-seven percent of patients received concomitant coronary artery bypass grafting. Two prosthesis types were used: stented bovine pericardial or porcine bioprosthesis (n 365) or stentless xenograft bioprosthesis (n 199). Specific valve models and their manufacturers are listed in Table 2. Median label size was 23 (range, 19 to 29); for stentless valves, this value was 25 (range, 19 to 29), and for stented valves, it was 23 (range, 19 to 29). All stentless valves were implanted in a subcoronary fashion. Prosthesis-Patient Mismatch P-PtM was calculated with the use of patient s body surface area and the normal reference values of the EOA of the implanted prosthetic valves according to Pibarot and Dumesnil. 2 For the values of the Sorin Pericarbon valve, because they are not included in the list of No. of Implants Implant Period, Month/Year Follow-Up, Median (Maximum), y Toronto SPV Stentless 85 09/95 06/ (12.4) Bioprosthesis (St Jude Medical, Inc, St Paul, Minn) Medtronic Freestyle 64 01/97 10/ (11.5) Bioprosthesis (Medtronic, Inc, Minneapolis, Minn) Edwards Prima Stentless 50 09/91 04/ (16.4) Aortic Bioprosthesis (Edwards Lifesciences, Irvine, Calif) Mosaic Bioprosthesis /97 02/ (10.7) (Medtronic, Inc, Minneapolis, Minn) Carpentier-Edwards /95 12/ (12.8) Perimount (Edwards Lifesciences, Irvine, Calif) Pericarbon 48 08/92 12/ (15.4) (Sorin Biomedica, Sallugia, Italy) All /91 12/ (16.4) Pibarot and Dumesnil, we used the data presented by Seguin et al. 13 P-PtM was defined as an EOA index 0.85 cm 2 /m 2. Follow-Up and SVD Clinical and echocardiographic follow-up was performed shortly before hospital discharge and periodically by the referring cardiologist. Survival, reoperation, cerebrovascular accidents, bleeding complications, anticoagulation therapy, New York Heart Association class, and cardiac rhythm were recorded. Follow-up was 97.3% complete. Median follow-up period was 6.1 years, with a maximum of 16.4 years. Implant and follow-up periods for the different valve models are given in Table 2. A total of 3583 patient-years of data were available for analysis. Echocardiography was performed in 95.5% of the hospital survivors. In 58% of the patients, the last echocardiography was recorded within the last year of follow-up, and in 78%, the last echocardiography was recorded within the last 3 years. The diagnosis of SVD was based on echocardiographic findings. Two types of SVD were diagnosed. A stenosis-type SVD was diagnosed when echocardiographic follow-up showed calcification of the leaflets. When the leaflets became thickened and less pliable, but the pressure gradient across the valve increased above the value at baseline to values 55 mm Hg, the valve was also diagnosed as stenotic. Baseline was defined as the first echocardiogram obtained during the first postoperative year. When these criteria were met but some degree of regurgitation was also found, the valve remained classified as stenosis-type SVD. The incompetence-type SVD was diagnosed when the valve showed no leaflet abnormalities on the echocardiogram, but transprosthetic regurgitation worsened. Worsening was defined as an increase of at least 1 degree of severity of regurgitation during follow-up compared with the value at baseline. Transprosthetic regurgitation was graded semiquantitatively by color Doppler echocardiography according to the following scale: 0, none; 1, trivial; 2, mild; 3, moderate; and 4, severe. Only transvalvular regurgitation was taken into account. Perivalvular regurgitation was scored but was considered nonstructural valve dysfunction.

3 Flameng et al P-PtM Predicts SVD in Bioprosthetic Valves 2125 Table 3. Single-Predictor Analysis of Late Survival (Cox) and SVD (Cox With Multiple Imputation Method) Late Survival SVD Hazard Ratio 95% CI P Hazard Ratio 95% CI P Age, y Female gender Body surface area NYHA class Ejection fraction AS AI Rhythm (SR/non-SR) Concomitant CABG Design (stented/stentless) Size Tissue (porcine aortic/pericardial) Anticalcification treatment EOA index Discharge peak gradient P-PtM CI indicates confidence interval; NYHA, New York Heart Association; AS, aortic stenosis or mixed aortic valve disease; AI, pure aortic insufficiency; SR, sinus rhythm; and CABG, coronary artery bypass grafting. Statistical Analysis For the formulation of valve-related complications, standard guidelines and definitions of terms were used according to recently published recommendations. 14 The Fisher exact test was used to analyze categorical data. The standard t test was used to analyze transvalvular gradients. Early survival (3-month mortality) was analyzed by logistic regression. Overall survival was analyzed by standard single-predictor and multivariable (P 0.1 threshold to enter the model) Cox proportional hazards models. SVD was analyzed with an interval-censored technique, with the time interval between the last echocardiographic follow-up demonstrating normal valve function and the first echocardiographic follow-up demonstrating SVD. For SVD, we used single-predictor and multivariable (P 0.1 threshold to enter the model) Cox proportional hazards models, using the poor man s data augmentation multiple imputation method for interval-censored data as described by Pan. 15 A total of 20 imputations per model have been used. In addition, the nonparametric Turnbull estimate was used to create a graphic representation of the time to SVD depending on P-PtM status. P 0.05 was considered statistically significant for the study. Results Survival Hospital mortality was 5%. Overall survival at 10 years was %, and freedom from cardiac death was % at 10 years. Including hospital mortality, a total of 233 deaths were recorded, of which 71 were definitely cardiac related. Single-predictor analysis of possible predictors of mortality revealed age (P 0.002) for early mortality and age (P ), coronary artery bypass graft surgery (P 0.03), and valve size 21 (P 0.01) as significant factors for late survival (Table 3). Multivariable analysis revealed only age as significant for early (P 0.01) as well as for late mortality (P ). Clinical Outcome At 10 years, freedom from hospital readmission for cardiac reasons was % at 10 years, freedom from thromboembolic events and/or major anticoagulation-related bleeding was %, and freedom from reoperation was %. Fourteen patients (2%) developed acute bacterial endocarditis during the follow-up course. These patients were not considered to have SVD and were censored at the time of onset of endocarditis. Prosthesis-Patient Mismatch Fifty percent of the patients had an EOA index 0.85 cm 2 /m 2, 46% had an EOA index between 0.85 and 0.65 cm 2 /m 2 (moderate P-PtM), and 4% had an EOA index 0.65 cm 2 /m 2 (severe P-PtM). For further analysis, we considered a value 0.85 cm 2 /m 2 as P-PtM (without further distinction between moderate and severe P-PtM). Stentless valves had significantly less P-PtM (44/199 patients; 22%) than stented valves (240/365 patients; 65%; P 0.05). Patients with P-PtM had significantly higher peak gradients at discharge than patients without P-PtM ( versus mm Hg; P 0.001). Structural Valve Degeneration The diagnosis of SVD was made in 40 patients (7%). A flow chart representing the distribution of the 2 types of SVD in relation to P-PtM is presented in Figure 1. Twenty-four patients had a stenotic valve, 15 were calcified on echocardiography, and 9 showed degenerated and thickened leaflets. Four were explanted and showed massive calcification on gross macroscopic examination. Ten patients had concomitant regurgitation. Sixteen had no stenosis but a purely incompetent valve. On the basis of these echocardiographic criteria, freedom from SVD was significantly lower than that of reoperation at 10 years: % versus %. Changes in hemodynamics over time, in terms of peak pressure gradients across the valves, are shown in Figure 2 for 3 groups of patients: those not experiencing SVD, those

4 2126 Circulation May 18, 2010 Figure 1. Distribution of the 2 types of SVD in relation to P-PtM. developing incompetence-type SVD, and those developing stenosis-type SVD. In patients with stenosis-type SVD, peak gradients rose from to mm Hg (P ; Figure 2). Single-predictor analysis of possible predictors of SVD (Table 3) showed that the patient-related factors gender and body surface area and the valve-related factors anticalcification treatment, valve size 21, EOA index, and P-PtM were significant. Multivariable analysis revealed that anticalcification treatment, valve size 21, and P-PtM were independent predictors of SVD (Table 4). Furthermore, when patients with P-PtM developed SVD, it was predominantly the stenosis type (21/26 patients; 81%), whereas when patients without P-PtM developed SVD, it was mainly the incompetence type (11/14; 79%; P 0.001; Figure 1). Freedom from SVD in relation to time after surgery in patients with or without P-PtM is presented in Figure 3. In patients with P-PtM, SVD occurs much earlier than in patients without P-PtM. Patients without P-PtM remain virtually free from SVD during the first 9 years postoperatively, and when it occurs rather rapidly thereafter, it is mainly (79% of cases) of the incompetence type. In patients with P-PtM, the incidence of SVD starts early ( 2 to 3 years) after implantation, its occurrence rate is rather slow, and the majority (81% of cases) are of the stenosis-type SVD. Discussion SVD and Mode of Valve Failure In general, SVD related to bioprostheses is considered 1 entity because it refers to changes intrinsic to the valve, but it covers a wide range of valve abnormalities such as calcification, leaflet tear, stent creep, and suture line disruption of components of the bioprosthesis. 14 Most studies analyzing factors influencing SVD do not take the mode of failure of the valve into account and conclude that SVD as such is promoted by the age at implantation (younger age), site of implantation (mitral position), gender (male), and valve type (porcine) Cardiovascular risk factors were not universally found to be predictive of SVD In a recent study, including a data set of 3934 patients, it was shown that the lifetime risk of reoperation because of SVD is 22% for a 60-year-old man in whom a bioprosthesis is implanted. 22 As stated previously, all of these studies refer to SVD as 1 entity and have reoperation for SVD as the end point. By doing so, the problem of SVD is underestimated and not properly characterized. In a study population of primarily older patients, we also found a low reoperation rate at 10 years (4%), which was comparable to the rate found by others. 18,23 26 The discrepancy between the reoperation rate and the incidence of Figure 2. Evolution in peak pressure gradients across the aortic bioprosthesis. Shown are mean value, SE (box), and SD (whiskers). Values were obtained at discharge and at the most recent echocardiogram in patients not experiencing SVD (left), in patients with incompetence-type SVD (middle), and in those with stenosis-type SVD (right). For patients with SVD, the gradient at the moment of echocardiographic diagnosis of SVD was used. *P vs discharge peak gradient.

5 Flameng et al P-PtM Predicts SVD in Bioprosthetic Valves 2127 Table 4. Multivariable Analysis of SVD (Cox With Multiple Imputation Method) Hazard Ratio 95% CI P Size Anticalcification treatment P-PtM CI indicates confidence interval. SVD as diagnosed echocardiographically indicates that not all patients with SVD undergo reoperation within the time frame of the 15-year follow-up. Furthermore, it shows that data collection on the mode of failure of SVD is poor in most studies. Nevertheless, some pathology studies refer to it and distinguish between valve calcification and cusp rupture. Calcification intrinsic to the cusps is the major pathological process necessitating bioprosthetic valve reoperation. 12 This is mainly so for stented bovine pericardial and porcine aortic valves In our study, the diagnosis of SVD was based on echocardiographic findings allowing imaging of structural changes in the valve leaflets over time. However, only gross calcifications can be seen in the leaflets with this technique, and leaflet thickening is never recorded quantitatively. Nevertheless, the combination of substantially increasing pressure gradients associated with such descriptive measures allows the diagnosis of stenosis of the prosthesis. In our study, the majority of stenosed prostheses were calcified, and all of the valves explanted for stenosis were calcified. On the other hand, for stentless porcine aortic valves, the mode of failure leading to reoperation is predominantly cusp rupture. 23,31 Cusp ruptures, inducing relatively sudden but significant valve regurgitation, can be seen as fatigue damage due to buckling and tension of the glutaraldehyde cross-linked collagen of the leaflets. 32 Our finding that all valve models are prone to this kind of wear after 10 years of implantation, irrespective of the valve model, valve size, or presence of P-PtM, is not surprising because all of these different valve types, although carefully designed, are glutaraldehyde crosslinked. When these valves finally will wear and rupture in ex vivo accelerated fatigue cycling tests, why should they not do this after 30 to 40 million cycles in vivo? However, it is remarkable in our study that the stenosis mode of failure is also not related to any kind of valve design, whether stented or stentless, pericardial or bovine; 5 of the 6 valve types studied are represented in the stenosis- as well as in the incompetence-type SVD group. Because it is obvious that only patients prone to P-PtM develop stenosis-type SVD over time, there must be a link between disturbed hemodynamics and this kind of pathology. It is known that the predominant mechanism of calcification in bioprostheses is cell mediated, being nucleated at the membranes and in organelles of the transplanted cells. 12 On the other hand, predilection of certain sites of bioprosthetic valve leaflets to calcification has provoked suggestions that design factors and mechanical stresses may be major reasons leading to degeneration. 33 Because bioprostheses are intrinsically obstructive, especially in small sizes, a relative P-PtM will induce disturbances in valve performance, abnormal flow profiles, and changes in stress distribution. Besides mechanical stress, flow turbulences within the transprosthetic flow jet and downstream of the prosthesis may also predispose to SVD by enhancing degenerative and/or atherosclerotic-like processes at the surface of the cusps. 20,21,34 36 Another potential mechanism to explain the findings of the present study is the smaller EOA reserve of patients with P-PtM. Patients with P-PtM, by definition, have smaller indexed EOA at the outset Figure 3. Nonparametric Turnbull estimate plotting freedom from SVD since time of operation for patients with and without P-PtM and assuming a constant hazard within each region of support.

6 2128 Circulation May 18, 2010 of operation, and they will thus develop critical stenosis more rapidly than patients with no P-PtM. For example, a 30% reduction in EOA due to SVD will result in a much higher increase in gradient in a patient with P-PtM than in one without P-PtM because the patient is on the steep portion of the gradient EOA curve. Although this does not explain the process of valve calcification, it can help to explain that a significant portion of stenotic-type SVD patients show only thickened leaflets without visible calcifications that are associated with high transvalvular gradients. However, the fundamental mechanism of the phenomenon of prosthetic valve calcification must be elucidated. 37 Limitations of the Study After hospital discharge, postoperative echocardiography was performed by the referring cardiologists. This may have influenced the quality of postoperative evaluation because echocardiographic assessment is observer dependent. However, this observer-dependent variation will be small because only the in-hospital discharge echocardiogram can be done by a cardiologist other than the one doing the follow-up. The number of patients developing SVD is relatively small in our series, which can limit the statistical power of the study. This may be related to the older age of the population studied because it is possible that the impact of P-PtM on SVD becomes even more pronounced in a younger population. To separate the effect of P-PtM in different age strata, further studies will be needed. Clinical Implications These findings have definite clinical implications because, as demonstrated by several studies, P-PtM can be predicted at the time of operation, and, on the basis of this information, the surgical technique and/or the selection of the prosthesis can be modified to prevent P-PtM or reduce its severity By avoiding P-PtM, the incidence of SVD should be reduced by 50%, which, in turn, can influence the choice for a bioprosthetic valve instead of a mechanical one. 20 Furthermore, because intrinsically stentless bioprostheses are less prone to P-PtM, this type of bioprosthesis should be preferred over stented valves. None. Disclosures References 1. Rahimtoola SH. The problem of valve prosthesis-patient mismatch. Circulation. 1978;58: Pibarot P, Dumesnil J. Hemodynamic and clinical impact of patient-prosthesis mismatch in the aortic position and its prevention. J Am Coll Cardiol. 2000;36: Blais C, Dumesmil JG, Baillot R, Simard S, Doyle D, Pibarot P. Impact of prosthesis-patient mismatch on short-term mortality after aortic valve replacement. Circulation. 2003;108: Howell NJ, Keogh BE, Barnet V, Bonser RS, Graham TR, Rooney SJ, Wilson IC, Pagano D. Patient-prosthesis mismatch does not affect survival following aortic valve replacement. Eur J Cardiothorac Surg. 2006;30: Flameng W, Meuris B, Herijgers P, Herregods MC. Prosthesis-patient mismatch is not clinically relevant in aortic valve replacement using the Carpentier-Edwards Perimount valve. Ann Thorac Surg. 2006;82: Mohty D, Malouf JF, Girard SE, Schaff HV, Grill DE, Enrique-Sarano ME, Miller FA Jr. Impact of prosthesis-patient mismatch on long-term survival in patients with small St Jude medical mechanical prostheses in aortic position. Circulation. 2006;113: Mohty D, Dumesmil JG, Echahidi N, Mathieu P, Dagenaie F, Voisine P, Pibarot P. Impact of prosthesis-patient mismatch on long-term survival after aortic valve replacement: influence of age, obesity, and left ventricular dysfunction. J Am Coll Cardiol. 2009;53: Walther T, Rastan A, Falk V, Lehman S, Garbade J, Funkat AK, Mohr FW, Gummert JF. Patient prosthesis mismatch affects short- and long-term outcomes after aortic valve replacement. Eur J Cardiothorac Surg. 2006;30: Kohsaka S, Mohan S, Virani S, Lee VV, Contreras A, Reul GJ, Coulter SA. Prosthesis-patient mismatch affects long-term survival after mechanical valve replacement. J Thorac Cardiovasc Surg. 2008;135: Mihaljevic T, Nowicki ER, Rajeswaran J, Blackstone EH, Lagazzi L, Thomas J, Lytle BW, Cosgrove DM. Survival after valve replacement for aortic stenosis :implications for decision making. J Thorac Cardiovasc Surg. 2008;135: Flameng W, Meuris B, Yperman J, De Visscher G, Herijgers P, Verbeken E. Factors influencing calcification of cardiac bioprosthesis in adolescent sheep. J Thorac Cardiovasc Surg. 2006;132: Schoen FJ, Levy RJ. Pathology of substitute heart valves: new concept and developments. J Card Surg. 1994;9: Seguin JR, Grandmougin D, Folliguet T, Warembourg H, Laborde F, Chaptal PA. Long-term results with the Sorin Pericarbon valve in the aortic position: a multicenter study. J Heart Valve Dis. 1998;7: Cary W, Akins D, Miller C, Turina MI, Kouchoukos NT, Blackstone EH, Grunkemeier GL, Takkenberg JJM, David TE, Butchart EG, Adams DH, Shahian DM, Hagl S, Mayer JE, Lytle BW. Guidelines for reporting mortality and morbidity after cardiac valve interventions. J Thorac and Cardiovasc Surg. 2008;134: Pan W. A multiple imputation approach to Cox regression with intervalcensored data. Biometrics. 2000;56: Rahimtoola SH. Choice of prosthetic heart valve for adult patients. JAm Coll Cardiol. 2003;41: Jamieson WR, Germann E, Aupart MR, Neville PH, Marchand MA, Fradet GJ. 15-year comparison of supra-annular porcine and Perimount aortic bioprosthesis. Asian Cardiovasc Thorac Ann. 2006;14: Gao G, Wu Y, Grunkemeier GL, Furnary AP, Starr A. Durability of pericardial versus porcine aortic valves. J Am Coll Cardiol. 2004;44: Gring CN, Houghtaling P, Novaro GM, Roselli E, Smedira N, Banburry M, Blackstone E, Griffin BP. Preoperative cholesterol levels do not predict explant for structural valve deterioration in patients undergoing bioprosthetic aortic valve replacement. J Heart Valve Dis. 2006;15: Briand M, Pibarot P, Despres JP, Voisin P, Dumesmil JG, Dagenais F, Mathieu P. Metabolic syndrome is associated with faster degeneration of bioprosthetic valves. Circulation. 2006;114: Farivar RS, Cohn LH. Hypercholesterolemia is a risk factor for bioprosthetic valve calcification and explantation. J Thorac Cardiovasc Surg. 2003;126: Van Geldorp MWA, Jamieson E, Kappetein P, Ye J, Fradet GJ, Eijkemans MJC, Grunkemeier GL, Bogers A, Takkenberg J. Patient outcome after aortic valve replacement with a mechanical or biological prosthesis: weighing lifetime anticoagulant-related event risk against reoperation risk. J Thorac Cardiovasc Surg. 2009;137: Mohammadi S, Baillot R, Voisine P, Mathieu P, Dagenais F. Structural deterioration of the Freestyle aortic valve: mode of presentation and mechanisms. J Thorac Cardiovasc Surg. 2006;132: Bach DS, Metras J, Doty JR, Yun KL, Dumesnil JG, Kon ND. Freedom from structural valve deterioration among patients aged or 60 years undergoing Freestyle stentless aortic valve replacement. J Heart Valve Dis. 2007;16: Desai ND, Merin O, Cohen GN, Herman J, Mobilos S, Sever JY, Fremes SE, Goldman BS, Christakis GT. Long-term results of aortic valve replacement with the St. Jude Toronto stentless porcine valve. Ann Thorac Surg. 2004;78: Fann JI, Miller DC. Porcine valves: Hancock and Carpentier-Edwards aortic prosthesis. Semin Thorac Cardiovasc Surg. 1996;8: Suzigan S, Braile DM, Souza DR, Rossi MA. Pathologic findings in 74 dysfunctional Braile-Biomédica bovine pericardial heart valves recovered at reoperation, 1978 to J Cardiovasc Surg. 1994;35:

7 Flameng et al P-PtM Predicts SVD in Bioprosthetic Valves Butany J, Yu W, Silver MD, David TE. Morphologic findings in explanted Hancock II porcine bioprosthesis. J Heart Valve Dis. 1999; 8: Roselli EE, Smedira NG, Blackstone EH. Failure modes of the Carpentier-Edwards pericardial bioprosthesis in the aortic position. J Heart Valve Dis. 2006;15: Valente M, Ius P, Bortolotti U, Talenti E, Bottio T, Thiene G. Pathology of the Pericarbon bovine pericardial xenograft implanted in humans. J Heart Valve Dis. 1998;7: Butany J, Collins MJ, Nair V, Leask RL, Scully HE, Williams WG, David TE. Morphological findings in explanted Toronto stentless porcine valves. Cardiovasc Pathol. 2006;15: Krucinski S, Vesely I, Dokainish MA, Campbell G. Numerical simulation of leaflet flexure in bioprosthetic valves mounted on rigid and expansile stents. J Biomech. 1993;26: Sabbah HN, Hamid MS, Stein PD. Mechanical factors in the degeneration of porcine bioprosthetic valves: an overview. J Card Surg. 1989;4: Nollert G, Miksch J, Kreuzer E, Reichart B. Risk factors for atherosclerosis and the degeneration of pericardial valves after aortic valve replacement. J Thorac Cardiovasc Surg. 2003;126: Shetty R, Pibarot P, Charest A, Janvier R, Dagenais F, Perron J, Couture C, Voisine P, Després JP, Mathieu P. Lipid-mediated inflammation and degeneration of bioprosthetic heart valves. Eur J Clin Invest. 2009;39: Pibarot P, Dumesnil JG. Prosthetic heart valves: selection of the optimal prosthesis and long-term management. Circulation. 2009;119: Yoganathan AP, Chandran KB, Sotiropoulos F. Flow in prosthetic heart valves: state-of-the-art and future directions. Ann Biomed Eng. 2005;33: Pibarot P, Dumesnil JG, Cartier PC, Métras J, Lemieux M. Patient-prosthesis mismatch can be predicted at the time of operation. Ann Thorac Surg. 2001;71:S265 S Bleiziffer S, Eichinger WB, Hettich I, Guenzinger R, Ruzicka D, Bauernschmitt R, Lange R. Prediction of valve prosthesis-patient mismatch prior to aortic valve replacement: which is the best method? Heart. 2007;93: Castro LJ, Arcidi JMJ, Fisher AL, Gaudiani VA. Routine enlargement of the small aortic root: a preventive strategy to minimize mismatch. Ann Thorac Surg. 2002;74: Botzenhardt F, Eichinger WB, Bleiziffer S, Guenzinger R, Wagner IM, Baurnschmitt R, Lange R. Hemodynamic comparison of bioprostheses for complete supra-annular position in patients with small aortic annulus. J Am Coll Cardiol. 2005;45: Dalmau MJ, Gonzalez-Santos JM, Lopez-Rodriguez J, Bueno M, Arribas A, Nieto F. One year hemodynamic performance of the Perimount Magna pericardial xenograft and the Medtronic Mosaic bioprosthesis in the aortic position: a prospective randomized study. Interact Cardiovasc Thorac Surg. 2007;6: Perez de Arenaza D, Lees B, Flather M, Nugara F, Husebye T, Jasinski M, Cisowski M, Khan M, Henein M, Gaer J, Guvendik L, Bochenek A, Wos S, Lie M, Van Nooten G, Pennell D, Pepper J. Randomized comparison of stentless versus stented valves for aortic stenosis: effects on left ventricular mass. Circulation. 2005;112: CLINICAL PERSPECTIVE Prosthesis-patient mismatch can be predicted at the time of operation. On the basis of this information, the surgical technique and/or the selection of the prosthesis can be modified to prevent prosthesis-patient mismatch or reduce its severity. By avoiding prosthesis-patient mismatch, the incidence of structural valve deterioration should be reduced by 50%, which in turn can influence the choice for a bioprosthetic valve instead of a mechanical one. Furthermore, because intrinsically stentless bioprostheses are less prone to prosthesis-patient mismatch, this type of bioprosthesis should then be preferred over stented valves.