The Last Generation of Pericardial Valves in the Aortic Position: Ten-Year Follow-up in 589 Patients

Transcription

1 The Last Generation of Pericardial Valves in the Aortic Position: Ten-Year Follow-up in 589 Patients Michel R. Aupart, MD, Agnes L. Sirinelli, MD, Frank F. Diemont, MD, Yvon A. Meurisse, MD, Xavier B. Dreyfus, MD, and Michel A. Marchand, MD Department of Cardiac Surgery, Trousseau University Hospital, Tours, France Background. The first generation of pericardial valves has been withdrawn from the market because of excessively high rates of premature failure. With its original design, the Carpentier-Edwards pericardial valve has promised improved results. Methods. In our institution, 589 patients underwent an isolated aortic valve replacement with a Carpentier- Edwards pericardial bioprosthesis between July 1984 and December The patients' mean age was years, and 49% of the patients were in New York Heart Association clinical class III or IV. The operative mortality rate was 2.3% (14 of 595). All patients but 4 were followed up for an average of 4.1 years after their operation, and total follow-up was 2,408 patient-years. Results. At the time of the study, more than 85% of the patients were in New York Heart Association class I or II. There were 79 late deaths. After 10 years, the actuarial survival rate was 71% + 7%. Nineteen patients died of valve-related causes (3 endocarditis, 7 thromboembolic complications, 1 structural failure, and 8 sudden deaths). The actuarial rate of freedom from valve-related death was 94% + 3% at 10 years. Valve-related complications included 23 thromboembolic episodes (0.9% per patient-year), 14 endocarditis (0.5% per patient-year), 9 reoperations (0.4% per patient-year), and 4 structural valve failures with calcification and stenosis (0.2% per patient-year). After 10 years, freedom from valve-related complications was 84% --. 6%, from reoperation 97% - 2%, and from valve failure 96% + 4%. Conclusions. Because of its low rate of valve-related events at 10 years and low rate of structural deterioration with no leaflet tears, this prosthesis is an outstanding choice for patients who need tissue valves and for patients aged 60 years or older. (Ann Thorac Surg 1996;61:615-20) he first generation of pericardial valves has been T abandoned by physicians because of poor clinical results and excessively high rates of deterioration, characterized by leaflet tears. With its original design, the Carpentier-Edwards valve has promised improved results. Intermediate follow-up has shown low rates of valve-related events, in particular a low rate of deterioration with no leaflet tears [1-3]. This study reports our experience at 10 years with 595 Carpentier-Edwards pericardial aortic valves (model 2900; Baxter Healthcare Corp., Santa Ana, CA) in 589 patients. Material and Methods The Valve The Carpentier-Edwards pericardial valve is a threeleaflet bioprosthesis made of glutaraldehyde-preserved bovine pericardium. This valve presents two conceptual improvements: Complete strut flexibility is achieved with an Elgiloy wire, and the design of the stent allows the pericardial cusp to be mounted by drawing it into the stent from below rather than suturing. Accepted for publication Sep 21, Address reprint requests to Dr Aupart, Unite de Chirurgie Cardiaque, Hopital Trousseau, ~fours Cedex, France. Patients From July 1984 to December 1993, 595 Carpentier- Edwards pericardial bioprostheses were used in 589 consecutive patients for isolated aortic valve replacement in our hospital. Patients undergoing multiple valve replacement were excluded from this study, but there were no exclusions for other concomitant operations. There were 426 men (71.5%) and 169 women (28.5%). The age ranged from 16 to 88 years, with a mean of 67.5 _ years; 46 patients were younger than 50 years old (Fig 1). The main indications for valve replacement were idiopathic calcific stenosis in 361 patients, dystrophic insufficiency in 157, rheumatic fever in 23, prosthetic valve dysfunction in 21, endocarditis in 16, congenital anomaly in 12, and acute aortic dissection in 5. Clinical presentation was aortic stenosis in 308 patients, aortic insufficiency in 217, and mixed disease in 70. The mean New York Heart Association clinical status was 2.4. Ten percent of patients had chronic atrial fibrillation, and 12% had coronary artery bypass grafting. There were 125 associated procedures (21%). These included aortocoronary bypass in 72 patients, mitral valvuloplasty in 21, tricuspid annuloplasty in 11, ascending aortic operations in 8, myomectomy in 10, and aneurysm repair in 3. The patients were operated on through a median sternotomy with standard cardiopulmonary bypass, hemodilution, and general body hypothermia. Myocardial 1996 by The Society of Thoracic Surgeons /96]$15.00 Published by Elsevier Science Inc SSDI (95)

2 616 AUPART ET AL Ann Thorac Surg EDWARDS AORTIC PERICARD1AL VALVES 1996;61: Fig 1. Age distribution of patients having aortic valve replacement AGE protection was obtained with crystalloid or blood cardioplegia and topical cooling. Sizes of the valves were 19 mm (90 patients), 21 mm (172), 23 mm (160), 25 mm (155), 27 mm (14), and 29 mm (4). The postoperative anticoagulant protocol included heparin for 2 days, followed by 1 month of calcium heparin (activated partial thromboplastin time > 1.5 N) or Acenocoumarol (international normalized ratio 2.5 to 3) (Ciba-Geigy, Rueil-Malmaison, France). After 1 month, anticoagulation treatment was discontinued at the cardiologist's discretion. Follow-up information was obtained during a 3-month interval (June 1994 to August 1994) through questionnaire and telephone contact with cardiologists, family physicians, and patients. All patients but 4 were followed up for an average of 4.1 years after their operation, and total follow-up was 2,408 patient-years (0.6% lost during follow-up). More than 95% of patients underwent an echographic study at follow-up; among them, 50 patients underwent echocardiographic Doppler studies at our institution. These were performed by the same physician using the same procedure. All data were input in an IBM computer and analyzed with Sedistat (SEDIA SA, Paris, France). Standard actuarial and linearized statistical techniques were used to describe survival and the incidence of valve-related complications. Only the first event for each patient during the study was considered in the actuarial analysis, whereas all events for each patient were considered in linearized rates. Continuous data are presented as mean + standard deviation; actuarial probability and linearized rates are given as mean _+ 95% confidence limits of the mean. The guidelines for reporting mortality and morbidity after cardiac valvular operations were observed in the preparation of this report [4]. Results Patient Survival Fourteen patients died within 30 days of the operation, for a mortality rate of 2.3%. The major cause of early death was cardiac failure in 11 patients. One patient died of neurologic embolism. There were no deaths for New York Heart Association class I or II. There were 79 late deaths. The causes of these deaths were cardiac related in 17 (21%), valve related in 19 (24%), and noncardiac in 43 (54%). The actuarial survival rate including operative deaths was 71% _+ 7% after 10 years (Fig 2). At the time of the study, more than 85% of the 492 followed patients were in New York Heart Association class I or II (Fig 3), 9% received anticoagulation treatment, and 8% were in atrial fibrillation. Valve-Related Complications Nineteen patients died of valve-related causes (3 endocarditis, 7 thromboembolic complications, 1 structural failure, and 8 sudden deaths). The actuarial rate of freedom from valve-related death was 94% +_ 3% at 10 Fig 2. Actuarial survival including operative deaths. 100 i Years

3 Ann Thorac Surg AUPART ET AL ;61: EDWARDS AORTIC PERICARDIAL VALVES Fig 3. Preoperative and postoperative New York Heart Association (NYHA) class. 200 z I II III IV NYHA Class years. No postmortem examination was available for the sudden deaths. Twenty-three patients presented with a thromboembolic accident, for an actuarial rate of 93% -+ 3% of patients free of thromboembolism at 10 years. Seventeen events were neurologic, with 7 deaths, 3 permanent neurologic deficits, and 7 without sequelae. Sixteen patients were in sinus rhythm, 5 in atrial fibrillation, and 2 with a pacemaker. Six of these had anticoagulation treatment (3 in sinus rhythm, 2 in atrial fibrillation, and I with a pacemaker). Two of them underwent mitral repair at the time of operation. The linearized rate of thromboembolism was 0.9% per patient-year. No thrombosis of the bioprosthesis was observed. There were seven nonfatal cases of hemorrhagic complications meeting the definition of the guidelines. The indication for anticoagulation treatment was atrial fibrillation or cardiologist decision. Endocarditis was reported in 14 patients (1 in the first 3 months, 2 in the first year); 4 required reoperation in an elective situation, without mortality. We used another bioprosthesis for this procedure in 3. Three patients died without reoperation after being diagnosed in another institution, and 7 had successful medical treatment without recurrence. The linearized rate of endocarditis was 0.5% per pafient-year, with an actuarial rate of 96% + 2% of patients free of endocarditis at 10 years. No hemolysis was noted in the followed patients. Nine patients required reoperation, for an actuarial rate of 97% _+ 2% of patients free of reoperation at 10 years and a linearized rate of 0.4% per patient-year. Causes of reoperation were endocardifis in 4, perivalvular leak in 3, and structural failure in 2. We used another pericardial bioprosthesis in 6 patients and a mechanical prosthesis in 3 (1 structural failure, 1 perivalvular leak, 1 endocarditis). No deaths occurred in these elective reoperations. There were only 4 patients with structural deterioration of the valve, for an actuarial rate of 96% + 4% of patients free of structural valve failure at 10 years and a linearized rate of 0.2% per patient-year. All required reoperation. One patient died of sudden death just before reoperation, whereas the other 3 had reoperations without problem (1 after the end of the study). The structural change was stenosis without regurgitation, caused by valve calcification. The explanted valve showed diffuse micromineralization of the three leaflets. It is notable that no leaflet tear was observed. The 4 patients' ages were 42, 58, 58, and 61 years, and the interval until calcification was 8, 7, 4, and 4 years, respectively. No structural change was noted in the other followed patients, but echographic studies were not performed in all patients. No deterioration was noted in patients less than 40 years of age, and only one case occurred in a patient older than 60 years. Age One hundred ten patients were less than 60 years of age. Among the 12 associated procedures, six were coronary artery bypass grafting and four were aortic arch operations. In this group, with a mean follow-up of 4.8 years and total follow-up of 454 patient-years, we observed one valve-related death (0.2%) and eight valve-related events (Table 1). Actuarial survival was 91% +_ 6% at 10 years. Actuarial freedom from valve-related death was 99% 1%, from endocarditis 96% + 4%, from thromboembolism 98% + 2%, from deterioration 92% _+ 6%, and from reoperation 96% _+ 4%. Of the total, 479 patients were older than 60 years. The major cause of the 113 associated procedures was coronary artery bypass grafting in 66. In this group, with a mean follow-up of 4 years and a total follow-up of 1,954 patient-years, we observed 18 valve-related deaths (0.9%) Table 1. Valve-Related Morbidity by Age Grouy Event <60 Years e60 Years Thromboembolism 1 (0.2%) 22 (1.1%) Hemorrhage 0 7 (0.3%) Endocarditis 3 (0.6%) 11 (0.6%) Reoperations 2 (0.4%) 6 (0.3%) Structural failure 3 (0.6%) 1 (0.05%) Valve-related death 1 (0.2%) 18 (0.9%) "~ Values are number with linearized rate in parentheses.

4 618 AUPART ET AL Ann Thorac Surg EDWARDS AORTIC PERICARDIAL VALVES 1996;61: Table 2. Hemodynamic Data Valve Size Mean Gradient Effective Orifice (ram) N (mrn Hg) Area (cn'l 2) and 47 valve-related events (see Table 1). Actuarial survival was 64% _+ 9% at 10 years. Actuarial freedom from valve-related death was 92% _+ 4%, from endocarditis 97% +_ 2%, from thromboembolism 92% -- 4%, from deterioration 99% +_ 1%, and from reoperation 98% _+ 2%. Statistical analysis showed no significant difference between the age groups in terms of valve-related morbidity. Hemodynamic Data Despite the short coaptation area of the leaflet, no major central insufficiency or structural changes were noted except in patients with structural failure. Among the 50 patients assessed by echocardiography in our institution, 8 had an aortic valve replacement with a 19-mm bioprosthesis, 14 with a 21-mm, 13 with a 23-mm, 12 with a 25-mm, and 3 with a 27-ram bioprosthesis. Hemodynamic data are presented in Table 2. Comment Using pericardium to manufacture heart valve prostheses remains controversial. Currently, valve durability is the major problem with all bioprostheses. Pericardial bioprostheses have been used clinically in humans since The first generation of this type of valve has been withdrawn from the market because of poor clinical results [5, 6]. The mode of deterioration was documented not as pericardium failure, but as design failure [7, 8] or preparation failure [9]. The Carpentier-Edwards pericardial valve is designed with an original leaflet clamping, which eliminates the need for retention suture as well as the risk of abrasion. Procurement and selection of pericardium are strictly monitored. Preservation is done with glutaraldehyde. With these specifications, this pericardial bioprosthesis has shown satisfactory intermediate results [10, 11] in both positions for rehabilitating pericardium as a valve substitute. These results, especially in the aortic position and with small sizes, promise very interesting long-term follow-up. However, few reports are available with this type of pericardial valve. The Carpentier-Edwards pericardial valve was first used in Europe and received Food and Drug Administration approval in We report here a large study with this prosthesis. We have used these pericardial valves in the aortic position since 1984 in 589 patients, representing more than half of all valve replacements during the same period. It is the only bioprosthesis used for valve replacement in our institution. Major indications were aortic stenosis and age greater than 60 years. We also used this pericardial valve in patients with a small aortic annulus, with poor ventricular function, and, of course, in cases of contraindication to anticoagulation therapy and in young women who want pregnancy. Finally, after informed consent, the patient's choice is taken into account. In this study, the overall rate of survival was 71%. The rate and causes of death were similar to those in other series of aortic valve replacement [12]. There were 19 valve-related deaths, among them eight sudden deaths. No postmortem examination was possible because these deaths occurred in another institution or in the patient's home. Despite this fact, the actuarial freedom from valve-related death is comparable to that in other studies. Morbidity with this prosthesis was low. The most important complication was thromboembolism in 23 patients. This complication was significantly more common in patients over 60 years of age. In patients younger than 60, only one thromboembolic event was noted, and freedom from thromboembolism was 98% _+ 2% at 10 years. We think that the definition of the guidelines is too large and that some thromboembolic events were not in fact valve related, especially in the elderly population, but were instead related to the vascular status of the patients [13]. We observed only four structural deteriorations, with no leaflet tears. Deterioration of the tissue was always the same: calcification of the leaflet leading to valve stenosis. Explanted valves showed intrinsic calcification of the three leaflets. The cause of this calcification is poorly understood, but we expect better results with improvement in preparation of the tissue. The hemodynamic data, as in other publications on pericardial valves [14], were satisfactory. In this study, 90 patients underwent replacement with a 19-mm valve. The mean gradient was 18 mm Hg, the peak gradient was 32 mm Hg, and the effective orifice area was 1.08 cm 2. Unlike other investigators [15], we observed no increased rate of sudden death. The hemodynamic performance is one of the major reasons to continue developing the pericardial device. Among the 589 patients, 110 were less than 60 years of age at the time of the operation. The results in this particular group were very interesting. The 10-year performance appears comparable to that obtained with mechanical valves, classically used in these patients [12, 16]. Freedom from deterioration was 96% _+ 4% at 10 years, and better quality of life (no valve-related noise and no anticoagulation therapy) was especially reported by younger patients. The behavior of this pericardial valve in young patients appears better than that obtained with porcine valves [17]. But because of the history, of pericardial valves, it is necessary to be very careful, and more follow-up is mandated in young patients. Comparison of the long-term results between different valves is difficult, because many factors vary among cardiac surgical teams. Our results at 10 years appear

5 Ann Thorac Surg AUPART ET AL ;61: EDWARDS AORTIC PERICARD1AL VALVES better than those obtained with previous pericardial bioprostheses. There is a great difference in the structural deterioration rate between the Carpentier-Edwards and the first generation of pericardial valves (Ionescu-Shiley, Shiley, Inc, Irvine, CA; Hancock, Medtronic Cardiopulmonary Division, Anaheim, CA). The reported rate is approximately 80% at 5 years for the Ionescu-Shiley [18] and 75% for the Hancock pericardial bioprosthesis [19]. Too few data are available about the Pericarbon pericardial valve (Pericarbon; Sorin Biomedica, Saluggia, Italy) to do a long-term comparison [20]. Comparison between porcine and pericardial valves was described previously [21, 22]. Cosgrove and associates [21] showed that pericardial valves are less obstructive than porcine, and Pelletier and colleagues [22] found a higher rate of structural failure with pericardial devices over 6 years. However, the prosthesis used in the latter study was the Ionescu-Shiley valve, which presents a high rate of premature failure. At 10 years, our low deterioration rate with Carpentier-Edwards valves compares favorably with the best published results on porcine prostheses [23, 24]. The difference with our study may be due to the better stress behavior of pericardium than of porcine tissue over time. The comparison between mechanical and biologic valves was established by Bloomfield and co-workers [25] and Hammermeister and associates [26]. These studies involved the first generation of pericardial valves or porcine valves. The better results of the last generation of pericardial valves, especially in younger patients, will certainly change the results of the comparison. In many studies, homograft is the procedure of choice in terms of valve-related morbidity. The problems with homograft involve technical procedures and procurement. Structural deterioration rates were variable and sometimes included substantial regurgitation. Nevertheless, at 10 years the reported rates of deterioration and reoperation with cryopreserved [27] and 4 C antibioticstored valves [28] are identical to those in our study, although the patients are younger. The coming years will determine the long-term behavior of this bioprosthesis compared with homografts, especially in young patients. Despite their hemodynamic results and their interesting intermediate morbidity rates, no comparison can be made with stentless valves, which have had a shorter follow-up [29]. We conclude that this aortic pericardial valve appears as good and perhaps better than available porcine valves, and is at least as good as the results with homografts at 10 years. The results with the stentless valves and the new porcine valves in the coming years must be compared with this very good tissue valve. For us, this is the prosthesis of choice when a tissue valve is needed, especially in patients older than 60 years and in those with a small aortic annulus. Nevertheless, calcification appears to be unavoidable with time, and longer follow-up and further studies are necessary to provide definitive conclusions. References 1. Frater RW, Salomon NW, Rainer WG, Cosgrove DM, Wickman E. The Carpentier Edwards pericardial valve: intermediate results. Ann Thorac Surg 1992;53: Perier P, Mihaileanu S, Fabiani JN, et al. Long term evaluation of the Carpentier-Edwards pericardial valve in the aortic position. J Cardiac Surg 1991;6: Aupart M, Neville P, Dreyfus X, Meurisse Y, Sirinelli A, Marchand M. The Carpentier Edwards pericardial aortic valve: intermediate results in 420 patients. Eur J Cardiothorac Surg 1994;8: Edmunds LH Jr, Clark RE, Cohn LH, Miller DC, Weisel RD. Guidelines for reporting morbidity and mortality after cardiac valvular operations. Ann Thorac Surg 1988;46: Gallo I, Nistal F, Revuelta JM, Garcia-Statue E, Artinano E, Duran CG. Incidence of primary, valve failure with the Ionescu-Shiley pericardial valve. J Thorac Cardiovasc Surg 1985;90: Bortolotti U, Milano A, Guerra F, et al. Failure of Hancock pericardial xenografts: is prophylactic bioprosthetic replacement justified? Ann Thorac Surg 1991;51: Walley VM, Keon WJ. Patterns of failure in lonescu-shiley bovine pericardial valves. J Thorac Cardiovasc Surg 1987;93: Walley VM, Rubens FD, Campagna M, Pipe AL, Keon WJ. Patterns of failure in Hancock pericardial bioprostheses. J Thorac Cardiovasc Surg 1991;102: Leandri J, Bertrand P, Mazzucotelli JP, Loisance D. Mode of failure of the Mitroflow pericardial valve. J Heart Valve Dis 1992;1: Meyns B, Szecsi J, Flameng W, Daenen W. Aortic and mitral valve replacement with the Carpentier-Edwards pericardial bioprosthesis: mid-term clinical results. J Heart Valve Dis 1993;2: Aupart M, Neville P, Meurisse Y, Dreyfus X, Sirinelli A, Marchand M. The Carpentier-Edwards pericardial mitral bioprosthesis: intermediate follow up in 121 patients. In: Gabbay S, Frater R, eds. New horizon and the future of heart valve bioprostheses. Proceedings of the third meeting of the International Association for Cardiac Biological Implants. Austin: Silent Partners, 1994: Czer LSC, Chaux A, Matloff JM, et al. Ten year experience with the St Jude Medical valve for prima~ valve replacement. J Thorac Cardiovasc Surg 1990;100: Butchard EG, Moreno de la Santa P, Rooney SJ, Lewis PA. Arterial risk factors and ischemic cerebrovascular events after aortic replacement. J Heart Valve Dis 1995;4: Cosgrove DM, Lytle BW, Williams W. Hemodynamic performance of the Carpentier Edwards pericardial valve in the aortic position in vivo. Circulation 1985;72(Suppl 2): Kratz JM, Sade RM, Crawford FA, Crumbley BAJ, Stroud MR. The risk of small St Jude aortic valve prostheses. Ann Thorac Surg 1994;57: Nitter-Hauger S, Abdelnoor M. Ten year experience with the Medtronic Hall valvular prosthesis. Circulation 1989;80 (Suppl 1): Al-Khaja N, Belboul A, Rashid M, et al. The influence of age on the durability of Carpentier-Edwards biological valves. Eur J Cardiothorac Surg 1991;5: Daenen W, Noyez L, Lesaffre E. The Ionescu-Shiley pericardial valve: results in 473 patients. Ann Thorac Surg 1988;46: Scully H, Goldman B, Fulop J, et al. Five year follow up of Hancock pericardial valves: management of premature failure. J Cardiac Surg 1988;3: Wallberger E, Jault F, Fontanel M, et al. Bioprostheses p6ricardiques de Sorin-Pericarbon. Arch Mal Coeur 1994;87: Cosgrove DM, Lytle BW, Gill CC, et al. In vivo hemodynamic comparison of porcine and pericardial valves. J Thorac Cardiovasc Surg 1985;89: Pelletier CL, Carrier M, Leclerc Y, Lepage G, DeGuise P,

6 620 AUPART ET AL Ann Thorac Surg EDWARDS AORTIC PERICARDIAL VALVES 1996;61: Dyrda I. Porcine versus pericardial bioprostheses: a comparison of late results in 1593 patients. Ann Thorac Surg 1989;47: Burdon TA, Miller DC, Oyer PE, et al. Durability of porcine valve at fifteen years in a representative North American patient population. J Thorac Cardiovasc Surg 1992;103: Barratt-Boyes BG, Jaffe WM, Hong Ko P, Whitlock RML. The zero pressure fixed Medtronic Intact porcine valve: an 8.5 year review. J Heart Valve Dis 1993;2: Bloomfield P, Wheatley DJ, Prescott RJ, Miller HC. Twelve year comparison of a Bj6rk-Shiley mechanical valve with porcine bioprosthesis. N Engl J Med 1991;324: Hammermeister KE, Sethi GK, Henderson WG, et al. A comparison of outcomes in men 11 years after heart valve replacement with a mechanical valve or bioprosthesis. N Engl J Med 1993;328: Kirklin JK, Smith D, Novik W, et al. Long term function of cryopreserved aortic homografts. J Thorac Cardiovasc Surg 1993;106: Barratt-Boyes BG, Roche AHG, Subramanyan R, Pemberton JR, Whitlock RML. Long term follow up with the antibiotic sterilized aortic homograft valve inserted freehand in the aortic position. Circulation 1987;75: David TE, Feindel CM, Bos J, Sun Z, Scully HE, Rakowski H. Aortic valve replacement with a stentless porcine aortic valve. J Thorac Cardiovasc Surg 1994;108: INVITED COMMENTARY Aupart and colleagues have served an important purpose by putting second-generation pericardial valves into proper perspective. Although the Ionescu-Shiley pericardial valve showed excellent hemodynamics [1], its ultimate failure to receive widespread acceptance was based on (1) a design problem using a rigid stent and a fixation suture at the top of each commissure, creating a "stress point" that was responsible for the starting point of most leaflet disruptions, and (2) implantation of the valve into patients of nearly all ages. The resultant difficulties at reoperation were magnified by rigid fixation of a Dacron patch on the back of each strut, which formed a firm fibrous bond to the adjacent aortic wall making removal frustrating, difficult, and associated with higher risk. The ingenious and innovative design by an Edwards Laboratory engineer whereby the tissue leaflets are firmly attached behind the struts without the need for anchoring sutures has nearly eliminated the problem of leaflet disruption. This achievement, along with the use of a flexible stent, highlights the importance of close collaboration between the clinical cardiac surgeon and the biomedical engineer. Contemporaneously, based on several clinical studies, it is now apparent that tissue valve longevity and performance are far better in patients in the older age groups (whether this be 60, 65, or 70 years and older). Aupart and colleagues have added further evidence to the thesis that pericardial valves in the aortic position function very well and over an extended period of time if placed in patients selected on the basis of age and if the valve has the improvements incorporated in the second-generation design. W. Gerald Rainer, MD Department of Surgery University of Colorado Health Sciences Center 2005 Franklin St, Suite 380 Denver, CO Reference 1. Rainer WG, Christopher RA, Sadler TR Jr, Hilgenberg AD. Dynamic behavior of prosthetic aortic tissue valves as viewed by high-speed cinematography. Ann Thorac Surg 1979;28: