Valve Prosthesis Patient Mismatch (VP PM)

Journal of the American College of Cardiology Vol. 60, No. 13, 2012 2012 by the American College of Cardiology Foundation ISSN 0735-1097/$36.00 Published by Elsevier Inc. http://dx.doi.org/10.1016/j.jacc.2012.05.035 STATE-OF-THE-ART PAPER AND COMMENTARY Valve Prosthesis Patient Mismatch (VP PM) A Long-Term Perspective Samuel A. Daneshvar, MD, Shahbudin H. Rahimtoola, MB, FRCP, DSC (HON) Los Angeles, California The concept/phenomenon of valve prosthesis/patient mismatch (VP PM), described in 1978, has stood the test of time. From that time to 2011, VP PM has received a great deal of attention but studies have come to varying conclusions. This is largely because of the determination of prosthetic heart valve area [called effective orifice area index (EOAi)] by projection rather than by actual measurement, variable criteria to assess severity of EOAi and the timing of determination of EOAi. All prosthetic heart valves have some degree of VP PM which must be placed in a proper clinical perspective. This can be done by determining its effects on function and outcomes. For mortality one needs to focus especially on severe/critical degree of VP PM and determine the cause of death was due to VP PM. For the period beyond 2011 a road map is suggested that will have uniformity of assessment of VP PM and a focusing on the important goals of VP PM. (J Am Coll Cardiol 2012;60:1123 35) 2012 by the American College of Cardiology Foundation Everything should be made as simple as possible, but not simpler. Albert Einstein (1) In 1978 The original paper that described valve prosthesis patient mismatch (VP PM) (2) stated that Mismatch can be considered to be present when the effective prosthetic heart valve area, after insertion into the patient, is less than that of a normal human valve ; all prosthetic heart valves (PHVs) are smaller than normal and thus are inherently stenotic. Two issues were emphasized. 1) See page 1136 The PHV mismatch was stated to be usually mild to moderate in severity and often of no immediate clinical significance. However, it could be severe (as was illustrated in Table 1 of that paper) when, after mitral valve replacement, the effective orifice area (EOA) (also previously referred to as PHV area) was severely reduced from 1.8 cm 2 to 1.1 cm 2 (range, 0.89 to 0.54 cm 2 /m 2 ); 14 months later, From the Griffith Center, Division of Cardiovascular Medicine, Department of Medicine, LAC USC Medical Center, Keck School of Medicine at University of Southern California, Los Angeles, California. Dr. Daneshvar has reported that he has no relationships relevant to the contents of this paper to disclose. Dr. Rahimtoola has received honoraria for educational lectures from American College of Cardiology Foundation, American College of Physicians, Society of Thoracic Surgeons, University of California Los Angeles, University of California Irvine, Cornell University, Creighton University, Thomas Jefferson University, Cedars-Sinai Medical Center, Harvard Medical School, University of Wisconsin, University of Hawaii, Cardiologists Association of Hong Kong, China, University of California, Fresno, ATS, St. Jude Medical, Carbomedics, Edwards LifeSciences, Merck, and Pfizer. Manuscript received April 5, 2012; accepted May 1, 2012. the patient had deteriorated symptomatically, and at rest, the mean left atrial pressure had increased from 16 to 37 mm Hg. 2) The EOA should be corrected for body size (EOA index [EOAi]), which, in the 1960s, was done for native aortic valve stenosis (AS) (3,4). This was illustrated in Figure 1 (of that publication and in this paper) where after aortic valve replacement (AVR) for severe isolated aortic regurgitation, the EOA averaged 1.7 cm 2 (range, 1.3 to 2.7 cm 2 ) but, when corrected for body size, averaged 1.0 cm 2 /m 2 and ranged from 0.7 to 1.6 cm 2 /m 2. Other issues that were discussed included the following: 1. The mismatch was related to 2 factors: a) the PHV that can be inserted in most patients has an effective EOAi that is less than that of the normal human valve and; b) the in vivo effective EOAi is further reduced because of endothelialization and tissue ingrowth. 2. The EOA is only 1 factor that has to be taken into account when selecting a valve replacement device for an individual patient. 3. The relationship between gradient and the EOA and PHV size is complex (Fig. 2). 4. In patients with moderate to severe PHV mismatch severe to very severe VP PM may develop with additional tissue ingrowth or thrombotic material, which could result in deterioration of cardiac function, which may be acute and could result in prosthetic thrombosis and sudden death. 5. When evaluating the poor late results of valve replacement on survival, functional class or left ventricular (LV) function it is important to evaluate PHV function and mismatch accurately.

1124 Daneshvar and Rahimtoola JACC Vol. 60, No. 13, 2012 Valve Prosthesis Patient Mismatch September 25, 2012:1123 35 Abbreviations and Acronyms AS aortic valve stenosis AVA aortic valve area AVR aortic valve replacement EOA effective orifice area EOAi effective orifice area index GOA geometric orifice area LV left ventricular PHV prosthetic heart valve VP PM valve prosthesis patient mismatch After 1978 VP PM: AVR. With the publication of VP PM, surgeons were more careful and inserted the largest PHV that could be safely inserted. PHVs with improved hemodynamic profile have also been developed (5 7). As a result severe VP PM has become a much less common clinical problem (5,8). The overwhelming majority of a large number of scientific publications have related to the aortic valve and not to mitral valve VP PM; therefore, the rest of this paper is restricted to aortic VP PM. In conclusion, VP PM has become an accepted part of the clinical aspect of PHV implantation. How should VP PM be measured? Different parameters have been evaluated to assess for the severity of VP PM. The most commonly used measures of valve size are the EOAi, the projected EOAi, and the geometric orifice area (GOA). MEASURED EOAi. The EOA is a physiological parameter analogous to the native AVA. The EOA can be measured both invasively (2 4,9) and noninvasively using echocardiography/doppler (10) or magnetic resonance imaging (11). The most readily and widely available method is echocardiography/doppler. The accuracy of EOA echocardiographic measurement in the bioprosthetic valve is limited by the same pitfalls that are present in the measurement of the AVA; in particular, the LV outflow tract diameter may be more difficult to measure because of reverberation artifact caused by the prosthetic heart valve, but in these instances, the sewing ring diameter may be a sufficient surrogate (12). In the bileaflet mechanical valve, the central orifice may produce a high-velocity jet, causing an underestimation of the EOA. Pressure recovery occurs with both bioprosthetic and mechanical heart valves, although the implications of pressure recovery in PHVs have not yet been clarified. The EOAi has a complex relationship with the mean gradient across the aortic valve and PHV (2,13 15) (Figs. 2 and 3). PROJECTED EOAI. The projected EOAi is obtained by using reference data to find the EOA for the labeled size and model of the PHV that was implanted. This is indexed to the patient s body surface area to obtain the projected EOAi. Reference data are derived primarily from in vivo measurements of the EOA in patients but also have been obtained using in vitro measurements under pulsatile flow conditions (16). Dumesnil et al. (15) demonstrated a strong correlation between the projected EOA and the in vivo EOA (r 0.86, SEE 0.16 cm 2,p 0.005). Pibarot et al. (17) also demonstrated strong correlation between the projected EOAi and the in vivo EOA (r 0.84, SEE 0.15 cm 2 ; EOA in vivo, r 0.95; EOA in vitro, r 0.03; p 0.0001). In a later study, Pibarot et al. (18) demonstrated that the projected EOA had a 73% sensitivity and 80% specificity for the prediction of combined moderate and severe (frequently referred to as moderate-severe) VP PM. Bleiziffer et al. (19) also investigated the correlation between the projected EOAi and the measured EOAi. They demonstrated a more modest correlation between the projected EOAi and the measured EOAi when using their own institutional data (r 0.62, p 0.001) as well as published reference data (r 0.53). Importantly, their institutional data came from echocardiography/doppler studies performed 6 months post-operatively. The projected EOAi had a sensitivity of 54% and specificity of 83% for predicting severe VP PM. In the study by Florath et al. Figure 1 EOAi in Absolute Values (cm 2 ) and Corrected for Body Size (cm 2 /m 2 ) Effective orifice area (EOA) (red circles) corrected for body size (EOAi) provides better knowledge of the hemodynamic consequences of the prosthetic heart valve. For example, in 2 patients, the EOA was identical (2.1 cm 2 ). Because the patients were of different sizes, the each EOAi was quite different (1.1 and 1.6 cm 2 /m 2 ). Reprint, with permission, from Rahimtoola (2).

JACC Vol. 60, No. 13, 2012 September 25, 2012:1123 35 Daneshvar and Rahimtoola Valve Prosthesis Patient Mismatch 1125 Figure 2 Relationship of Mean Systolic Gradient to AVA Diagrammatic representation of the relationship between the mean systolic gradient (MSG) to the aortic valve area (AVA) assuming constant cardiac output and velocity of flow area. There is no gradient until the valve area is reduced by 50%, the MSG increases gradually (red circles), and when the valve area is reduced by 65%, it increases markedly (red line). Reprinted, with permission, from Rahimtoola (2). (20), the correlation between the projected EOAi and the EOAi measured within 10 days post-operatively was r 0.49 (p value not specified). The projected EOAi had a discriminative power of 0.72 for predicting severe VP PM and sensitivity of 75% and specificity of 52% for predicting moderate-severe VP PM. In conclusion, using the projected EOAi in predicting severe VP PM has a low accuracy; therefore, the projected EOAi to determine prediction and outcomes of VP PM is of limited value. Other limitations of the projected EOAi are discussed in the section on prediction of VP PM. Geometric orifice area. The GOA is an ex vivo measurement of the valve area. It is obtained by measuring the diameter of the inner surface of the PHV and calculating the area, assuming the valve has a circular orifice. The GOA also does not take into account factors related to leaflet or disk resistance or to morphological changes in the valve after implantation. Furthermore, the valve may change shape as it is manipulated during implantation. GOA does not predict VP PM (19). Assessment of severity of VP PM. With aortic VP PM. the obstruction to the LV outflow tract is similar to that seen with native AS. Thus, the severity of aortic VP PM should be assessed by the same criteria as for severe AS. In the 1960s, severe AS was defined: 1) by Braunwald as the mean aortic valve gradient, measured after energy recovery (also called pressure recovery), of 50 mm Hg (3), which was corroborated in a large series of patients in 1988 (21); and 2) as an AVA index of 0.6 cm 2 /m 2 determined after energy recovery (4). In 1989, a review of the natural history of unoperated AS in the previous 2 decades provided criteria for assessment of severity of AS (22). Measured stroke volume in stable patients undergoing diagnostic cardiac catheterization on 2 consecutive days showed a mean change of 9%, and, thus, it was thought that the confidence interval of calculated AVA was 10% (23). A prospective natural history of AS showed poor outcomes in those with severe AS using the criteria of 0.7 cm 2 /m 2 (24); however, the valve area of all patients in that study was 0.63 cm 2 /m 2, as measured at cardiac catheterization. Other studies also assessed severe VP PM as an EOAi 0.6 cm 2 /m 2 (20,25 32). Braunwald (33) has recommended critical AS is an AVA index of 0.4 cm 2 /m 2 after energy recovery. In 2000, on the basis of the EOAi previously referred to as PHV index, by echocardiography/doppler the diagnosis of severe and moderate VP PM was stated to be 0.65 cm 2 /m 2 and 0.65 to 0.85 cm 2 /m 2, respectively (14). These values were picked by computer on the basis of the average of increase of mean gradient in relation to the reduction in valve area (Fig. 3). There was a spread of gradient at these levels of the EOAi, but the confidence intervals of the gradients and the EOAi were not presented. Moreover, none of the techniques used to measure the EOA allow for calculating the mean EOA to hundredth of a centimeter; the precision, reproducibility, and variability of values of an EOAi of 0.65 and 0.85 have not been presented. In conclusion, the severity of aortic VP PM should be assessed by the same criteria that are used for native AS (Table 1) (22,33 37). There is an increasing consensus that severe AS should be assessed by the criterion of an AVAi of 0.6 cm 2 /m 2 (36); it would be reasonable that this criterion should also be applied to severe VP PM. When should the severity of VP PM be determined? Virmani (38) has described that there are 4 phases of physiological healing of mechanical and bioprosthetic PHVs; they are platelet and fibrin deposition, inflammation, granulation tissue, and finally encapsulation. Longterm device fibrous encapsulation with extension to adjacent tissues add to structural stability. Bioprosthetic valves undergo morphological changes of both the tissue material as well as the supporting structures, which may contribute to VP PM. Valve leaflets become covered by fibrin, platelets, and other cellular material. The matrix of the leaflets undergoes microcalcification as well insulation with plasma materials, causing changes in the matrix structure. These changes may change the resistive properties of leaflet materials. In both mechanical and bioprosthetic valves, a fibrous sheath may also encapsulate the supporting structure of the valve, encroaching on the PHV orifice and also possibly causing valve leaflet or disk immobilization (39 41). In 1988, the Veterans Administration Randomized Trial of PHV presented the findings of repeat research cardiac catheterization that was performed approximately 6 months after PHV implantation. For PHVs of the same size, there

1126 Daneshvar and Rahimtoola JACC Vol. 60, No. 13, 2012 Valve Prosthesis Patient Mismatch September 25, 2012:1123 35 Figure 3 Mean Systolic Gradient Versus EOAi The relationship of the mean systolic gradient (MSG) with the effective orifice area index (EOAi) at rest is shown. Mismatch is said to occur at an EOAi of 0.85 cm 2 /m 2. However, there is a curvilinear relationship of MSG to EOAi, similar to that shown in Figure 2. SEE standard error of the estimate. Reprinted, with permission, from Pibarot and Dumesnil (14). was a wide range of EOAs (Fig. 4) (42). Two more recent echocardiographic/doppler studies from the Mayo Clinic of patients studied within 1 week of mitral porcine mitral bioprosthesis (43) (Fig. 5) and of tricuspid mechanical prostheses (44) showed a wide range of gradients and EOAs with the same size of PHV, even in the same brand of PHV. There are at least several explanations for these findings. 1) PHVs of the same labeled size are not necessarily of precisely the same size. Bioprostheses and other biological valves may also have differences in tissue materials. 2) There are variations from patient to patient with regard to healing changes, hemodynamic conditions, and pressure recovery. EOA Table and1 EOAi EOA in AS and and EOAi VP-PM in AS and VP-PM Grading AVA and EOA, cm 2 AVA Index and EOAi, cm 2 /m 2 Mild 1.5 0.9 Moderate 1.0 1.5 0.6 0.9 Severe 1.0 0.6 Very severe/critical* 0.7 0.4 Patients with borderline values of AVAs between moderate and severe (0.9 to 1.1 cm 2 ; 0.55 to 0.65 cm 2 /m 2 ) should be individually considered. *Critical is from Morrow et al. (33). Adapted, with permission, from Rahimtoola (22,34 37). AS aortic valve stenosis; AVA aortic valve area; EOA effective orifice area; EOAi effective orifice area index; VP-PM valve prosthesis patient mismatch. In conclusion, it is best to measure VP PM early (at 1 week after PHV implantation or at time of hospital discharge) to determine the variations in actual size of the PHV that was implanted in an individual patient. Importantly, VP PM should also be assessed at 6 to 12 months when the physiological and other morphological changes in the PHV are mostly complete. The severity of VP PM determined at this time can be expected to determine the long-term impact of VP PM on patients outcomes. Outcomes. Studies investigating the importance of moderate and severe VP PM have focused on hospital and long-term mortality, LV mass regression, and measures of patients functional status and quality of life. These studies have had mixed results, likely due to methodological differences in the studies and the patient population being studied. Studies varied in the parameter used to assess VP PM, the EOAi values at which moderate and severe VP PM were defined, and whether moderate and severe VP PM were evaluated separately or grouped together, that is, as moderatesevere VP PM. Furthermore, studies also have differed in patient s age, LV systolic function, LV mass index, and the presence of associated coronary artery disease. There is likely a complex interplay between different patient-related factors and VP PM on outcomes (45). Given these considerations, studies that used a measured EOAi, where measurement was done 6 to 2 months after

JACC Vol. 60, No. 13, 2012 September 25, 2012:1123 35 Daneshvar and Rahimtoola Valve Prosthesis Patient Mismatch 1127 Figure 4 EOA of Bjork-Shiley and Hancock Porcine Valves Data from cardiac catheterization studies performed 6 months after implantation of Bjork-Shiley (BS) (red circles) and Hancock (H) (blue triangles) valves in sizes 21 to 29. With both BS and H, the measured effective orifice area (EOA) in square centimeters shows a wide range. Error bars show mean and SD. Reprinted, with permission, from Khuri et al. (42). PHV implantation and separated the data of moderate from severe VP PM, provide the most reliable assessment of VP PM in assessing patient outcomes. Furthermore, given the large number of comorbid cardiac and noncardiac conditions in patients undergoing AVR, multivariate analysis should be performed for any conclusions regarding the influence of VP PM on outcomes. For assessing the role of VP PM on survival, deaths due to cardiac causes should be evaluated separately, and, again, multivariate analysis should be performed. Figure 5 LITERATURE SEARCH. In January 2012, a PubMed search was performed using the terms patient prosthesis mismatch, valve prosthesis patient mismatch, and a general search for patient prosthesis mismatch for all Englishlanguage articles published from 1998 to the current time. The search returned 402 articles, of which 159 were excluded (18 related to mitral valve replacement, 1 related to the Ross principle, 83 related to prosthesis other than heart valves, and 12 related to children, and 10 studies were not available at our institution). To ensure that no studies were missed, the references of the studies obtained were reviewed for any further studies that provided outcomes in relation to VP PM. Of the remaining articles, 37 provided data regarding outcomes in relation to VP PM. A. Studies using a measured EOAi. Given the considerations outlined previously, studies using direct measurement of the EOAi in individual patients after surgery will have the most sensitivity and specificity for VP PM. 1. Survival a. Operative or 30-day mortality. As discussed previously, the measured EOAi immediately postoperatively does not necessarily reflect the EOAi in the long term. Hanayama et al. (27) showed increased early mortality in patients with severe VP PM; however, this study did not correct for confounding variables. Subsequent studies have shown that VP PM using a measured EOAi does not have a significant impact on operative or 30-day mortality (28,46). b. Long-term mortality i. Moderate VP PM. Five studies measured the EOAi and examined the impact of moderate VP PM on mortality (Table 2) (20,25,28,32,46). None of them showed an increased mortality with moderate VP PM. ii. Severe VP PM. Seven studies measured the EOAi and examined the impact of severe VP PM on mortality (Table 3) (20,25,27,28,32,46,47). The incidence of severe VP PM ranged from 6.9% to 39.8%. Two studies showed an increased mortality with severe VP PM (20,25) and 5 other studies did not document an increase in mortality (27,28,32,46,47). Of the studies that did not show an increase in mortality, 1 study had a mean follow-up of 1.19 0.76 years, and in another study, only 4 of 58 patients had severe VP PM. iii. Moderate-severe VP PM. Studies that measured the EOAi and combined moderate and severe VP PM showed no statistically significant increase in mortality in this group (Table 3) (45,48 51). The incidence of moderate-severe VP PM ranged from 3.3% to 65.7%. Given that severe VP PM has a significant impact on mortality, whereas the influence of mod- EOA of Different Sizes of Mitral Bioprosthesis Carpentier-Edwards Duraflex normal porcine bioprosthesis sizes 27 to 35. Patient heart rates 100 beats/min, n 218. Mean effective area (EOA) by conventional equation. Error bars shown are 2 SD. There is a considerable range of EOAs for each prosthesis size. Reprinted, with permission, from Blauwet et al. (43).

Summary Table 2 of Summary Studies That of Studies Used Measured That UsedEOAi Measured to Assess EOAi VP-PM: to Assess DataVP-PM: of Moderate Data of and Moderate Severe VP-PM and Severe Analyzed VP-PM Separately Analyzed Separately Study population Milano et al. (28) Hanayama et al. (27) Mohty-Echahidi et al. (25) Florath et al. (20) Nozohoor et al. (46) Vicchio et al. (32) Okamura et al. (47) Patients who received 19- or 21-mm St. Jude prosthetic valve for aortic stenosis and had echocardiographic evaluation before discharge. Excluded patients with greater than mild AI, concomitant mitral or tricuspid valve surgery AVR. Patients with complete echocardiographic data Patients with AVR with 19- or 21-mm St. Jude Mechanical valve; excluded: 30-day mortality, patients without follow-up echocardiography within 1 yr All patients with AVR in whom EOAi was measured within 10 days postoperatively Patients undergoing AVR with Sorin Soprano or Medtronic Mosaic valves; excluded patients undergoing other valve surgery, aortic dissection, aortic arch surgery, surgery with deep hypothermia Patients 70 years of age with severe aortic stenosis who received bileaflet mechanical valve; hospital survivors 58/78 patients with echocardiography done at discharge after AVR No. of patients 205 788 388 533 372 345 58 Time of EOAi Measured before hospital Measured within 10 Measured by Measured at 1-yr follow-up assessment discharge days of AVR echocardiography Definition of VP-PM Moderate: EOAi 0.60 and 0.90 cm 2 /m 2 ; severe: 0.60 cm 2 /m 2 % with VP-PM Moderate: 149/205 (72.7%); severe: 24/205 (11.7%) Type of prosthetic valve 19- or 21-mm St. Jude Mechanical valve Lowest EOAi on serial post-operative echocardiograms Follow-up 8 5 yrs Not given for total population Short-term outcome Moderate: no impact on hospital mortality; severe: no impact on hospital mortality Measured EOAi; 91% were measured within 30 d post-operatively Severe: 0.60 cm 2 /m 2 Moderate: 0.85 cm 2 /m 2 ; Mild: 0.85 cm 2 /m 2 ; severe: 0.60 cm 2 /m 2 moderate: 0.85 cm 2 /m 2 ; severe: 0.60 cm 2 /m 2 76/768 (9.9%) Moderate: 168/388 (43.3%); severe: 66/388 (17.0%) Homograft: 7/1129 (0.6%); 19- and 21-mm St. Jude standard bioprosthesis: 500/1,129 or high-performance valves (44.3%); stentless: 211/1,129 (18.7%); mechanical 411/1129 (36.4%) (data for all patients with AVR during study period) Severe: early mortality was higher in patients (2.63% vs. 0.14%, p 0.001). No multivariate analysis done. Echocardiography performed at median 6 days. 331/388 (91%) occurred within 30 days. Clinical follow-up: 5.3 3.3 yrs Moderate: 277/533 (52%); severe: 150/533 (28%) Bioprosthetic: 85/533 (15.9%); stentless: 284/533 (53.3%); mechanical: 164/ 533 (30.8%) Moderate-severe: 0.85 cm 2 /m 2 ; severe: 0.65 cm 2 /m 2 Overall: severe: 39.8%. Sorin Soprano: moderate-severe: 84%; severe: 36%; Medtronic Mosaic moderate-severe: 74%; severe: 46% Sorin Soprano: 235/372 (63.1%); Medtronic Mosaic: 137/372 (26.9%) 4.7 2.2 yrs Clinical follow-up: 1.19 0.76 years; echocardiographic follow-up: 4.5 6.9 months Moderate: no impact on hospital mortality; severe: no impact on hospital mortality Measured by echocardiography at discharge Severe: 0.60 cm 2 /m 2 ; Moderate-severe: 0.85 cm 2 /m 2 ; moderate: 0.61 0.84 cm 2 /m 2 severe: 0.65 cm 2 /m 2 Moderate: 175/345 (50.7%); severe: 33/345 (9.6%) Bileaflet mechanical valves Moderate-severe: 18/58 (31%); severe: 4/58 (6.9%) 17-mm St. Jude valve 4.2 3.1 yrs 32.6 19.6 months (in entire cohort) Continued on next page 1128 Daneshvar and Rahimtoola JACC Vol. 60, No. 13, 2012 Valve Prosthesis Patient Mismatch September 25, 2012:1123 35

JACC Vol. 60, No. 13, 2012 September 25, 2012:1123 35 Daneshvar and Rahimtoola Valve Prosthesis Patient Mismatch 1129 Continued Table 2 Continued Milano et al. (28) Hanayama et al. (27) Mohty-Echahidi et al. (25) Florath et al. (20) Nozohoor et al. (46) Vicchio et al. (32) Okamura et al. (47) Moderate-severe: no difference in long-term survival; severe: no valve-related complications in 4 patients Moderate: no impact on 10-yr survival, 10-yr freedom of valverelated complications; Severe: no impact on 10-yr survival, 10-yr freedom of valverelated complications Moderate: no impact on overall mortality; severe: no impact on overall mortality Moderate: no impact on survival; severe: predictor of decreased survival by multivariate analysis (HR: 1.9; 95% CI: 1.08 to 2.68) Moderate: no impact on survival; severe: predicted decreased survival by multivariate analysis (HR: 2.18; 95% CI: 1.24 to 3.85). EOAi as a continuous variable: increased. EOAi as a continuous variable also was a predictor of improved survival on multivariate analysis. GOAi was not associated with survival on univariate or multivariate analysis Severe: no impact on longterm mortality Moderate: no impact of moderate on overall mortality, valve-related death; severe: no impact on overall mortality but increased number of cardiac events Long-term outcome Moderate and severe are analyzed separately. AI aortic insufficiency; AVR aortic valve replacement; CI confidence interval; GOAi, geometric orifice area index; HR hazard ratio; other abbreviations as in Table 1. erate VP PM is not likely to be significant, grouping patients with moderate and severe VP PM together, will likely only have an impact on mortality if there is a sufficient proportion of patients with severe VP PM. 2. LV mass regression Studies have also assessed the impact of VP PM on LV mass regression. Traditionally, the state of the myocardium has been referred to as LV hypertrophy when evaluated by electrocardiography and imaging techniques such as angiography, echocardiography, and magnetic resonance imaging. However, in reality, these are measures of increased LV mass because they are measuring cardiac myocytes, blood vessels, connective tissue, and fibrosis; in the normal heart, only about one half of the cells are cardiomyocytes, and fibroblasts are the predominant cell types in the remainder (52 54). This is important because increased LV mass may be due to changes in factors other than myocyte hypertrophy. Indeed, LV mass regression is complicated by multiple factors including female sex (29), the degree of pre-operative LV hypertrophy (55), the presence of systemic hypertension (56), renal dysfunction (57), and preoperative LV fibrosis (58). These should be kept in mind when considering the effect of VP PM on LV mass regression because LV mass regression is multifactorial. Several studies using a measured EOAi have attempted to elucidate the role of VP PM in LV mass regression. Tasca et al. (29,59) demonstrated that a larger increase from the pre-operative AVA to the post-operative EOAi was associated with increased LV mass regression at a mean follow-up of 1.5 0.31 years. Dalmau et al. (60) demonstrated that at 5-year follow-up, moderate-severe VP PM was associated with a higher incidence of increased LV mass (odds ratio: 12.6; 95% confidence interval: 2.6 to 44.1; p 0.001). On the other hand, several studies showed that there is no association between moderate-severe VP PM and long-term LV mass regression (49,50,61). Gelsomino et al. (51) demonstrated that up to 1 year post-operatively, patients with moderate-severe VP PM had a higher LV mass index compared with those patients without VP PM; however, at 3-year follow-up, patients with VP PM had a similar LV mass as those patients without VP PM. 3. Functional status and quality of life. Bleiziffer et al. (62) demonstrated that moderate-severe VP PM and the presence of coronary artery disease are predictors of decreased exercise capacity 6 months post-operatively. However, moderate-severe VP PM does not significantly affect long-term functional class (28,50) or quality of life based on the Short Form-36 questionnaire (49). The greater the amount

1130 Daneshvar and Rahimtoola JACC Vol. 60, No. 13, 2012 Valve Prosthesis Patient Mismatch September 25, 2012:1123 35 Summary Table 3 of Summary Studies That of Studies Use Measured That UseEOAi Measured to Assess EOAi VP-PM: to Assess DataVP-PM: of Moderate Data of and Moderate Severe VP-PM and Severe Are Combined VP-PM Are Combined Study population Gelsomino et al. (51) Sakamoto et al. (48) Vicchio et al. (49) Bleiziffer et al. (45) Garatti et al. (50) AVR with Cryolife O Brien bioprosthesis; hospital survivors AVR in patients 65 yrs of age. Patients with a root too small to accommodate a 19-mm valve underwent a root enlargement by the Nicks method. Patients 70 yrs of age with severe aortic stenosis who received a 19-mm bileaflet mechanical valve; hospital survivors All patients with bioprosthetic valves for whom echocardiography was performed at 6 months 17-mm AVR with mechanical valve n 62 51 134 survivors of AVR 645 78 Time of EOAi Measured at time of Mean echocardiographic Not provided Measured at 6.5 1.5 months Not provided assessment discharge follow-up: 2.2 1.5 yrs Definition of VP-PM EOAi: 0.85 cm 2 /m 2 Moderate: 0.85 cm 2 /m 2 ; EOAi: 0.85 cm 2 /m 2 EOAi: 0.85 cm 2 /m 2 EOAi 0.85 cm 2 /m 2 severe: 0.60 cm 2 /m 2 % with VP-PM 12/62 (19.4%) at discharge; at 6 months, 2/61 (3.3%) 9/51 (17.6%) with moderate VP-PM; no patients with severe VP-PM 88/134 (65.7%) 248/622 (39.9%) 29/78 (37%) Type of prosthetic valve 21- or 23-mm Cryolife O Brien valve 19-mm Carpentier- Edwards Perimount 19-mm mechanical valves Bioprosthetic 583/645 (90.4%); stentless: 62/645 (9.6%) Follow-up 37 12 months 2.4 1.8 yrs 3.9 2.7 yrs Clinical: mean 2.66 (range, 0.45 to 7.19) yrs; echocardiography at 6.5 1.5 mo Short-term outcome There were 2 early deaths that were not valve related. However, there was a statistically significant difference in 30-day mortality between the group with and without VP- PM Long-term outcome No statistically significant difference in long-term mortality Data for moderate and severe are combined. LV left ventricular; other abbreviations as in Table 2. No overall or valve-related mortality with moderate VP-PM; no difference in risk of thromboembolism with moderate VP-PM No difference in overall survival, valve-related complications 1. Increased cardiac mortality if logarithmic EOAi used as a continuous variable, but not if VP-PM taken as a category. 2. Increased cardiac mortality with residual left ventricular hypertrophy with VP-PM compared with residual LV hypertrophy without VP-PM St. Jude bileaflet, Sorin Bicarbon Slim Discharge, annual: 83 45 months No difference in gradients or mortality for patients with VP-PM of myocardial fibrosis before AVR results in a worse New York Heart Association functional class postoperatively (58) (Fig. 6). B. Studies using a predicted EOAi. A large number of studies used the projected EOAi to assess the severity of VP PM (Online Appendix). Given the limitations of a projected EOAi discussed previously, this presents a serious methodological limitation, and results should be interpreted very cautiously. It is of little surprise that these studies present conflicting results regarding the impact of VP PM on longterm mortality. Several of these studies have shown an interaction between patient characteristics and VP PM. C. Studies using the GOA index. The problems related to using GOA as a measure of VP PM were discussed previously. Several large studies have also demonstrated that the GOA index and the z-score of the valve internal diameter do not predict long-term mortality (63,64) and quality of life (65). Therefore, studies that used the GOA to evaluate outcomes are not discussed further. D. Using the EOAi as a continuous variable. The influence of VP PM on mortality may be more carefully assessed by using the EOAi as a continuous variable. In doing so, studies avoid creating categorical variables that may have additional confounding factors that are difficult to assess. Studies using the EOAi as a continuous variable have shown that a lower EOAi is associated with increased mortality (25,45). This is compatible with the knowledge that the more the narrowing of the LV outflow tract, the worse the outcome; this is similar to valvular AS. Interestingly, the study by Bleiziffer et al. (62), which demonstrated that the EOAi as a continuous variable was associated with increased mortality, was unable

JACC Vol. 60, No. 13, 2012 September 25, 2012:1123 35 Daneshvar and Rahimtoola Valve Prosthesis Patient Mismatch 1131 Figure 6 Myocardial Fibrosis and Functional Class The relationship of New York Heart Association (NYHA) functional classes I to IV at baseline (blue boxes) and 9 months (red boxes) post-operatively to baseline myocardial fibrosis in patients with before and after aortic valve replacement for aortic stenosis. Those with severe fibrosis died or were in NYHA functional classes III and IV. Reprinted, with permission, from Weidemann et al. (58). to demonstrate increased mortality using a cutoff of 0.85 cm 2 /m 2 to define VP PM. This makes evident the difficulty in documenting the increased mortality of VP PM when combining the data of moderate and severe VP PM, that is, in moderate-severe VP PM. However, Bleiziffer et al. did not demonstrate a specific value of an EOAi that increases mortality, for example, at critical VP PM ( 0.40 cm 2 /m 2 ) versus severe VP PM ( 0.60 cm 2 /m 2 )(Fig. 7). Prediction and prevention of severe VP PM. The importance of the need for a technique which will permit accurate pre-operative prediction of the valve size the surgeon will realistically be able to insert in an individual patient was emphasized in the original description of Figure 7 Severe VP-PM Diagram emphasizing that severe valve prosthesis patient mismatch (VP-PM) rather than moderate or mild VP-PM will have serious adverse outcome but that the level of severe VP-PM at which this occurs is not known at this time (see text). EOAi effective orifice area index; Mod. moderate. VP PM (2). Important in this statement is that valve size refers to the EOAi rather than the labeled valve size. This was elaborated by Pibarot and Dumesnil (14) using a 3-step protocol: 1) calculate the BSA using the Dubois method; 2) determine the minimum EOA required to ensure an EOAi of 0.85, 0.8, or 0.75 cm 2 /m 2 based on the minimum required EOAi for a given patient; and 3) select the type and size of the valve greater or equal to the minimal EOA value obtained in step 2. This method of Pibarot and Dumesnil has been demonstrated to reduce the incidence of moderate-severe VP PM when using reference data obtained by in vivo measurement performed 6 to 12 months after PHV implantation (19). However, there are several difficulties to recognize when using methods to prevent VP PM: A. Projected EOAi 1. The projected EOAi has limited value in predicting the presence or absence of VP PM (18 20,66) (also see above). The EOA of a given model and size of valve has a range of values and may vary between patients (Table 4) (19,42). Thus, even with the best values of VP PM, there is an EOAi range that may occur in any 1 patient (Fig. 8). 2. Reference data for EOA of different valves is available in the literature and also provided by PHV manufacturers in the form of tables. Manufacturerprovided tables have several flaws. The data provided in these tables comes from a multitude of sources and includes in vitro data, the GOA, and in vivo data obtained at different time intervals after PHV implantation (67,68). Manufacturer-provided tables may use favorable EOA data (67,68). Furthermore,

1132 Daneshvar and Rahimtoola JACC Vol. 60, No. 13, 2012 Valve Prosthesis Patient Mismatch September 25, 2012:1123 35 EOA AfterDetermined AVR for EOA PHVat Determined Labeled 6 Months Size at 23 6 Months Table 4 After AVR for PHV Labeled Size 23 n Mean EOA, cm 2 Range Edwards Perimount 113 1.82 1.7 1.9 Edwards Perimount Magna 38 1.82 1.7 2.0 Medtronic Mosaic 23 1.53 1.3 1.8 Sorin Mitroflow 19 1.53 1.4 1.7 Sorin Freedom Solo 7 2.00 1.6 2.3 St. Jude Epic Supra 35 1.81 1.6 2.0 St. Jude Toronto Root 6 1.60 1.4 1.8 For several valves, there are 10 patients. Reprinted, with permission, from Bleiziffer et al. (45). PHV prosthetic heart valve; other abbreviations as in Table 2. they provide only the mean EOA without range or confidence intervals. Ideally, given these considerations, EOA values should be based on in vivo measurements made 6 to 12 months after implantation (Table 5). Values provided should include the mean SD and range, as well as median and interquartile range. 3. There is no standard for the labeling of valve sizes, and valves with the same labeled valve size differ in external diameter and consequently may not fit into the same aortic root (69). Tables generally refer to the labeled valve size. This makes comparison between different models and sizes difficult. Figure 8 EOAi of 1 Prosthesis Type Histogram distribution of EOAi at 6 months after aortic valve replacement in 113 patients of the same type and size (Edwards Perimount size 23). The mean value of 1.82 cm 2 /m 2 may be the only parameter inserted into a reference table. Most patients would have moderate VP-PM, many would have mild VP-PM, and few would have severe VP-PM. Abbreviations as in Figure 7. Adapted and modified, with permission, from Bleiziffer et al. (45). PHV Table Areas 5 From PHV Randomized Areas From Randomized In Vivo TrialsIn Vivo Trials PHV Size, mm PHV Area Index, cm 2 /m 2 p Value A Medtronic Mosaic: 23 0.74 0.06 C-E Perimount: 23 0.79 0.08 0.05 Medtronic Mosaic: 25 0.80 0.09 C-E Perimount: 25 0.85 0.09 0.05 B C-E Perimount Magna 1.17 0.27 C-E Perimount 0.94 0.36 0.01 Medtronic Mosaic porcine 0.83 0.23 0.001 Soprano valve 0.87 0.20 0.004 C C-E Perimount: 22.9 2.0 1.9 0.09 Toronto stentless porcine: 26.3 2.0 (p 0.0001) 1.74 0.66 NS D C-E Perimount: 24.8 2.17 0.9 0.2 Toronto stentless porcine: 24.8 2.51 0.88 0.22 NS E C-E Perimount Magna: 23.4 2.1 1.07 0.4 C-E Perimount: 22.4 1.8 0.80 0.2 C-E Prima Plus: 24.3 1.7 0.87 0.3 0.028 F C-E Perimount Magna: 23 mm 1.12 0.28 0.007 Medtronic Mosaic: 23 mm 0.79 0.20 Data for A through E are from Rahimtoola (35) with permission. Data for F from Dalmau et al. (60). C-E Carpentier-Edwards. B. Severity of VP PM 1. A projected EOAi that falsely predicts the presence of VP PM may cause the patient to undergo a series of unnecessary steps intended to prevent VP PM. The surgical techniques available to prevent VP PM are less than ideal. Data are needed to determine whether the increased risk of techniques used to prevent VP PM outweigh the risk of VP PM and improve patient outcomes (70). 2. Prevention of all instances of VP PM is, at present, not possible. There is a need for a change of emphasis in the prevention of VP PM. Many studies have discussed prevention of moderate-severe, which includes both moderate and severe VP PM. There is need to focus on prevention of severe VP PM. This occurs with an EOAi 0.6 cm 2 /m 2, and, allowing for a 10% variability in determining the severity of VP PM, the goal should be an EOAi 0.7 cm 2 /m 2. C. Patient-related factors 1. It is unclear what an acceptable EOAi is for an individual patient. This likely depends on several patient-related factors such as age, left ventricular systolic function, severity of LV hypertrophy, and patient expectation of the level of activity after valve replacement. 2. Although avoidance of VP PM is preferable, several factors must be taken into account when deciding the choice of PHV (Table 6) (71). Choice of a prosthetic heart valve with an EOA that avoids mismatch must be considered in light of other competing factors such as structural valve deterioration and other complications. In conclusion, at the present time, one cannot accurately and reliably predict severe, or even critical, VP PM. Obvi-

JACC Vol. 60, No. 13, 2012 September 25, 2012:1123 35 Daneshvar and Rahimtoola Valve Prosthesis Patient Mismatch 1133 Factors Table 6to Consider Factors When to Consider Choosing When a PHV Choosing a PHV Age of the patient Comorbid conditions: cardiac and noncardiac Expected life span of patient Choose a PHV That does not require root replacement for isolated aortic valve disease With long-term follow-up outcomes that are at least as good as the best of the available PHV With which individual physicians and medical centers have the necessary skill and experience Probability of adherence and compliance with warfarin therapy Patient s wishes and expectations Other extenuating circumstances Reprinted, with permission, from Rahimtoola (35). PHV prosthetic heart valve. ously a small PHV (for example, 17 or 19 mm) is not suitable for the average-sized adult in many parts of the world. Thus, we have to rely on the skilled and experienced cardiac surgeon to implant the largest PHV that can be safely implanted. When the data described in the section Beyond 2011 are available, we will have reached another major milestone. VP PM in transcatheter valve therapy. VP PM has been described in percutaneously implanted aortic valves. The incidence of severe VP PM in the Edwards Sapien valve has been reported to be 11% to 18.2% when measured 6 to 12 months after implantation (72,73). In the Medtronic Corevalve, the incidence of severe VP PM has been reported to be 2% to 16% when measured at the time of hospital discharge (74 76). The study of the Corevalve by Giannini et al. (76) also evaluated VP PM 6 months after transcatheter valve therapy and found 12% incidence of severe VP PM. Propensity-matched data suggest that the incidence of VP PM is less than in surgical PHV implantation (72). In a study using the Cribier-Edwards and Edwards SAPIEN valves to perform transcatheter valve therapy, patients who had surgical AVR had smaller EOAi despite starting with a larger aortic valve annulus (77). However, more data are needed before conclusions can be reached. Beyond 2011 The concept/phenomenon of VP PM has stood the test of time. Several of the expectation and predictions that have not been fulfilled are reviewed here. Thus, it is reasonable to review changes that need to be made in the assessment of VP PM. 1. EOAi should be measured at 1 to 4 weeks or at hospital discharge to evaluate the actual valve size that was implanted. This should also be done at 6 to 12 months to evaluate the severity of VP PM that will affect long-term outcomes. 2. The grading of severity of VP PM should be similar to another common LV outflow tract obstruction, namely, valvular AS. VP PM can be mild (EOAi 0.9 cm 2 /m 2 ), moderate (EOAi 0.6 to 0.9 cm 2 /m 2 ), or severe (EOAi 0.6 cm 2 /m 2 ). 3. Mild VP PM, like mild AS, is unlikely to have clinically significant untoward effects. Outcomes with moderate and severe VP PM should be evaluated separately. Moderate VP PM is unlikely to reduce survival unless there is progression of valve obstruction, for example, with pannus formation. Severe VP PM has negative effects on outcomes, but its effect on mortality is still unproven, and more focused study is needed. To assess the effects of VP PM on mortality, the goal should be to determine by multivariate analysis the role of VP PM on mortality due to cardiac causes (78). 4. Prediction of severity of VP PM is problematic. The primary goal should be not to prevent VP PM but rather to prevent severe VP PM. 5. Use of the EOAi as a continuous variable may help to define the level of severe VP PM that results in increased mortality, and this may occur at a critical level of obstruction ( 0.4 cm 2 /m 2 )(Fig. 7). Reprint requests and correspondence: Dr. Shahbudin H. Rahimtoola, University of Southern California, Old General Hospital Room 3221, 1200 North State Street, Los Angeles, California 90033. E-mail: rahimtoo@usc.edu. REFERENCES 1. Albert Einstein. Available at: http://en.wikiquote.org/wiki/ Albert_Einstein. Accessed June 6, 2012. 2. Rahimtoola SH. The problem of valve prosthesis-patient mismatch. Circulation 1978;58:20 4. 3. Braunwald E, Goldblatt A, Aygen MM, Rockoff SD, Morrow AG. Congenital aortic stenosis. I. Clinical and hemodynamic findings in 100 patients. II. Surgical treatment and the results of operation. Circulation 1963;27:426 62. 4. Tobin JR, Rahimtoola SH, Blundell PE, Swan HJ. Percentage of left ventricular stroke work loss. A simple hemodynamic concept for estimation of severity in valvular aortic stenosis. Circulation 1967;35: 868 79. 5. Garcia Fuster R, Estevez V, Rodriguez I, et al. Prosthesis-patient mismatch with latest generation supra-annular prostheses. The beginning of the end? Interact Cardiovasc Thorac Surg 2007;6:462 9. 6. Botzenhardt F, Eichinger WB, Guenzinger R, et al. Hemodynamic performance and incidence of patient-prosthesis mismatch of the complete supraannular perimount magna bioprosthesis in the aortic position. Thorac Cardiovasc Surg 2005;53:226 30. 7. Dalmau MJ, Mariagonzalez-Santos J, Lopez-Rodriguez J, Bueno M, Arribas A. The Carpentier-Edwards Perimount Magna aortic xenograft: a new design with an improved hemodynamic performance. Interact Cardiovasc Thorac Surg 2006;5:263 7. 8. Moon MR, Lawton JS, Moazami N, Munfakh NA, Pasque MK, Damiano RJ Jr. POINT: prosthesis-patient mismatch does not affect survival for patients greater than 70 years of age undergoing bioprosthetic aortic valve replacement. J Thorac Cardiovasc Surg 2009;137: 278 83. 9. Shavelle D. Evaluation of valvular heart disease by cardiac catheterization and angiocardiography. In: Otto C BR, editor. Valvular Heart Disease: A Companion to Braunwald s Heart Disease. Philadelphia, PA: Saunders/Elsevier, 2009:86 100.

1134 Daneshvar and Rahimtoola JACC Vol. 60, No. 13, 2012 Valve Prosthesis Patient Mismatch September 25, 2012:1123 35 10. Zoghbi WA, Chambers JB, Dumesnil JG, et al. Recommendations for evaluation of prosthetic valves with echocardiography and doppler ultrasound. J Am Soc Echocardiogr 2009;22:975 1014; quiz 1082 4. 11. Karamitsos TD, Myerson SG. The role of cardiovascular magnetic resonance in the evaluation of valve disease. Prog Cardiovasc Dis 2011;54:276 86. 12. Chafizadeh ER, Zoghbi WA. Doppler echocardiographic assessment of the St. Jude Medical prosthetic valve in the aortic position using the continuity equation. Circulation 1991;83:213 23. 13. Gorlin R, Gorlin SG. Hydraulic formula for calculation of the area of the stenotic mitral valve, other cardiac valves, and central circulatory shunts. I. Am Heart J 1951;41:1 29. 14. Pibarot P, Dumesnil JG. Hemodynamic and clinical impact of prosthesis-patient mismatch in the aortic valve position and its prevention. J Am Coll Cardiol 2000;36:1131 41. 15. Dumesnil JG, Honos GN, Lemieux M, Beauchemin J. Validation and applications of indexed aortic prosthetic valve areas calculated by Doppler echocardiography. J Am Coll Cardiol 1990;16:637 43. 16. Yoganathan AP, Woo YR, Sung HW, Williams FP, Franch RH, Jones M. In vitro hemodynamic characteristics of tissue bioprostheses in the aortic position. J Thorac Cardiovasc Surg 1986;92:198 209. 17. Pibarot P, Dumesnil JG, Lemieux M, Cartier P, Metras J, Durand LG. Impact of prosthesis-patient mismatch on hemodynamic and symptomatic status, morbidity and mortality after aortic valve replacement with a bioprosthetic heart valve. J Heart Valve Dis 1998;7: 211 8. 18. Pibarot P, Dumesnil JG, Cartier PC, Metras J, Lemieux MD. Patient-prosthesis mismatch can be predicted at the time of operation. Ann Thorac Surg 2001;71:S265 8. 19. Bleiziffer S, Eichinger WB, Hettich I, et al. Prediction of valve prosthesis-patient mismatch prior to aortic valve replacement: which is the best method? Heart 2007;93:615 20. 20. Florath I, Albert A, Rosendahl U, Ennker IC, Ennker J. Impact of valve prosthesis-patient mismatch estimated by echocardiographicdetermined effective orifice area on long-term outcome after aortic valve replacement. Am Heart J 2008;155:1135 2. 21. Griffith MJ, Carey C, Coltart DJ, Jenkins BS, Webb-Peploe MM. Inaccuracies in using aortic valve gradients alone to grade severity of aortic stenosis. Br Heart J 1989;62:372 8. 22. Rahimtoola SH. Perspective on valvular heart disease: an update. J Am Coll Cardiol 1989;14:1 23. 23. McAnulty JH, Kremkau EL, Rosch J, Hattenhauer MT, Rahimtoola SH. Spontaneous changes in left ventricular function between sequential studies. Am J Cardiol 1974;34:23 8. 24. Frank S, Johnson A, Ross J. Natural history of valvular aortic stenosis. Br Heart J 1973;35:41 6. 25. Mohty-Echahidi D, Malouf JF, Girard SE, et al. Impact of prosthesispatient mismatch on long-term survival in patients with small St Jude Medical mechanical prostheses in the aortic position. Circulation 2006;113:420 6. 26. Tasca G, Mhagna Z, Perotti S, et al. Impact of prosthesis-patient mismatch on cardiac events and midterm mortality after aortic valve replacement in patients with pure aortic stenosis. Circulation 2006; 113:570 6. 27. Hanayama N, Christakis GT, Mallidi HR, et al. Patient prosthesis mismatch is rare after aortic valve replacement: valve size may be irrelevant. Ann Thorac Surg 2002;73:1822 9; discussion 1829. 28. Milano AD, De Carlo M, Mecozzi G, et al. Clinical outcome in patients with 19-mm and 21-mm St. Jude aortic prostheses: comparison at long-term follow-up. Ann Thorac Surg 2002;73:37 43. 29. Tasca G, Brunelli F, Cirillo M, et al. Impact of valve prosthesispatient mismatch on left ventricular mass regression following aortic valve replacement. Ann Thorac Surg 2005;79:505 10. 30. Howell NJ, Keogh BE, Barnet V, et al. Patient-prosthesis mismatch does not affect survival following aortic valve replacement. Eur J Cardiothorac Surg 2006;30:10 4. 31. Mascherbauer J, Rosenhek R, Fuchs C, et al. Moderate patientprosthesis mismatch after valve replacement for severe aortic stenosis has no impact on short-term and long-term mortality. Heart 2008;94: 1639 45. 32. Vicchio M, Della Corte A, De Santo LS, et al. Prosthesis-patient mismatch in the elderly: survival, ventricular mass regression, and quality of life. Ann Thorac Surg 2008;86:1791 7. 33. Morrow AG, Roberts WC, Ross J, et al. Obstruction to left ventricular outflow. Current concepts of management and operative treatment. Ann Intern Med 1968;69:1255 86. 34. Rahimtoola SH. Choice of prosthetic heart valve for adult patients. J Am Coll Cardiol 2003;41:893 904. 35. Rahimtoola SH. Choice of prosthetic heart valve in adults: an update. J Am Coll Cardiol 2010;55:2413 26. 36. Rahimtoola SH. Determining that aortic valve stenosis is severe: back-to-the-future: physical examination and aortic valve area index/ energy loss index 0.6 cm 2 /m 2. J Am Coll Cardiol Img 2010;3:563 6. 37. Rahimtoola SH. Valvular heart disease: a perspective on the asymptomatic patient with severe valvular aortic stenosis. Eur Heart J 2008;29:1783 90. 38. Ladich E, Michaels MB, Jones RM, et al. Pathological healing response of explanted MitraClip devices. Circulation 2011;123:1418 27. 39. Schoen FJ, Levy RJ. Founder s Award, 25th Annual Meeting of the Society for Biomaterials, perspectives. Providence, RI, April 28-May 2, 1999. Tissue heart valves: current challenges and future research perspectives. J Biomed Mater Res 1999;47:439 65. 40. Flameng W, Herregods MC, Vercalsteren M, Herijgers P, Bogaerts K, Meuris B. Prosthesis-patient mismatch predicts structural valve degeneration in bioprosthetic heart valves. Circulation 2010;121: 2123 9. 41. Carpentier A. Hemodynamic factors affecting the fate of valvular bioprosthesis. Circulation 2010;121:2083 4. 42. Khuri SF, Folland ED, Sethi GK, et al. Six month postoperative hemodynamics of the Hancock heterograft and the Björk-Shiley prosthesis: results of a Veterans Administration cooperative prospective randomized trial. J Am Coll Cardiol 1988;12:8 18. 43. Blauwet LA, Malouf JF, Connolly HM, et al. Doppler echocardiography of 240 normal Carpentier-Edwards Duraflex porcine mitral bioprostheses: a comprehensive assessment including time velocity integral ratio and prosthesis performance index. J Am Soc Echocardiogr 2009;22:388 93. 44. Blauwet LA, Burkhart HM, Dearani JA, et al. Comprehensive echocardiographic assessment of mechanical tricuspid valve prostheses based on early post-implantation echocardiographic studies. J Am Soc Echocardiogr 2011;24:414 24. 45. Bleiziffer S, Ali A, Hettich IM, et al. Impact of the indexed effective orifice area on mid-term cardiac-related mortality after aortic valve replacement. Heart 2010;96:865 71. 46. Nozohoor S, Nilsson J, Luhrs C, Roijer A, Sjogren J. Influence of prosthesis-patient mismatch on diastolic heart failure after aortic valve replacement. Ann Thorac Surg 2008;85:1310 7. 47. Okamura H, Yamaguchi A, Nagano H, et al. Mid-term outcomes after aortic valve replacement with the 17-mm St. Jude Medical Regent Valve. Circ J 2012;76:365 71. 48. Sakamoto Y, Hashimoto K, Okuyama H, Sasaki T, Takakura H, Onoguchi K. Patient-prosthesis mismatch may be irrelevant after aortic valve replacement with the 19-mm Perimount pericardial bioprosthesis in patients aged 65 years or older. J Artif Organs 2007;10:206 11. 49. Vicchio M, De Santo LS, Della Corte, A et al. Aortic valve replacement with 19-mm bileaflet prostheses in the elderly: left ventricular mass regression and quality of life. J Heart Valve Dis 2008;17:216 21. 50. Garatti A, Mori F, Innocente F, et al. Aortic valve replacement with 17-mm mechanical prostheses: is patient-prosthesis mismatch a relevant phenomenon? Ann Thorac Surg 2011;91:71 7. 51. Gelsomino S, Morocutti G, Masullo G, et al. Patient-prosthesis mismatch after small-size stentless aortic valve replacement. J Card Surg 2004;19:91 7. 52. Manabe I, Shindo T, Nagai R. Gene expression in fibroblasts and fibrosis: involvement in cardiac hypertrophy. Circ Res 2002;91:1103 13. 53. Rudy Y. Conductive bridges in cardiac tissue: a beneficial role or an arrhythmogenic substrate? Circ Res 2004;94:709 11. 54. Nag AC. Study of non-muscle cells of the adult mammalian heart: a fine structural analysis and distribution. Cytobios 1980;28:41 61. 55. Penta de Peppo A, Zeitani J, Nardi P, et al. Small functional size after mechanical aortic valve replacement: no risk in young to middleage patients. Ann Thorac Surg 2005;79:1915 20. 56. Bove T, Van Belleghem Y, Francois K, Caes F, Van Overbeke H, Van Nooten G. Stentless and stented aortic valve replacement in elderly