T annulus (521 mm in diameter) is tempered by concerns

Transcription

1 Clinical and Hemodynamic Performance of the 19-mm Carpentier-Edwards Porcine Bioprosthesis Robert M. Bojar, MD, Hassan Rastegar, MD, Douglas D. Payne, MD, Charles A. Mack, MD, and Steven L. Schwartz, MD Departments of Surgery (Cardiothoracic) and Cardiology, New England Medical Center and the Tufts University School of Medicine, Boston, Massachusetts Because of concerns about the hemodynamic performance of 19-mm porcine valves, we retrospectively reviewed the clinical results and echocardiographic studies of 5 consecutive patients who received a 19-mm Carpentier-Edwards porcine bioprosthesis (model 65) for aortic valve replacement from 1986 through Nearly 87% of the patients were women, the mean age was 69 years, and the mean body surface area was 1.6 -t.7 m. Seventy-three percent of the patients had pure aortic stenosis, 96% were in New York Heart Association classes 111 and IV, and 56% underwent urgent or emergent operation. Overall hospital mortality was 7.7% with a late mortality of 8.% at a mean follow-up of 5 * 18 months. No patient experienced a valve-related compli- cation, and 95% of surviving patients were in New York Heart Association classes I and 11. Two-dimensional and Doppler echocardiography performed during the first postoperative week revealed a maximal instantaneous gradient of t 1. mm Hg. In 4 patients for whom additional data were available, the mean gradient was mm Hg with an effective orifice area of.85 *.18 cm. This study defines the normal range of postoperative gradients across the 19-mm Carpentier-Edwards porcine valve and demonstrates that patients receiving this valve can achieve significant clinical improvement despite the presence of high transvalvular gradients measured by echocardiography. (Ann Thorac Surg 199;56:1141-7) he placement of prosthetic valves in the small aortic T annulus (51 mm in diameter) is tempered by concerns over their hemodynamic performance. All prosthetic valves demonstrate so-called valve prosthesispatient mismatch, in that the valve s sewing ring reduces its effective orifice area (EOA) to one that is smaller than the native orifice [l]. This concept is commonly interpreted to mean that the prosthetic valve area is so small that the patient may be left with significant functional stenosis. Such a mismatch may be quite pronounced when the native aortic annulus can only accept a 19-mm prosthetic valve, potentially precluding clinical improvement and preventing regression of left ventricular hypertrophy. Thus, for the small aortic annulus, the surgeon must decide whether to perform a standard aortic valve replacement (AVR) using a small prosthetic valve, with possible compromise in the hemodynamic result, or consider an annulus-enlargement procedure to accommodate a larger valve. Tilting-disc valves have generally been preferred in the small aortic root because of their superior hemodynamics []. However, many patients with critical aortic stenosis are small elderly women with heavily calcified roots in whom root enlargement may be difficult, valve durability is of less significance, and anticoagulation might best be avoided. In these patients, use of a tissue valve with satisfactory hemodynamic performance is critical. Accepted for publication March 4, 199 Address reprint requests to Dr Bojar, New England Medical Center, Box 66, 75 Washington St, Boston, MA 111. Based on limited hemodynamic studies from the late 197s, many surgeons recommended that the 19-mm porcine valves should be avoided except in very small patients with limited cardiac output []. The Ionescu- Shiley pericardial valve became the tissue valve of choice in the small aortic root in the 198s because of its excellent hemodynamics [4]. After its withdrawal from the market in November 1987, the Carpentier-Edwards porcine bioprosthesis (Baxter Healthcare Corporation, Edwards CVS Division, Irvine, CA) was commonly used as an alternative in the small aortic root. Whether the 19-mm Carpentier-Edwards porcine valve can be considered a suitable prosthesis for patients with a small aortic annulus can best be determined by correlating its hemodynamic performance with the patient s clinical outcome. Two-dimensional and Doppler echocardiography is an excellent noninvasive means of assessing valve function, and its use in a large series of patients is an invaluable means of defining normal gradients for small tissue valves. These values become important when attempting to define valve dysfunction at subsequent echocardiographic evaluation. Material and Methods Pa tien t Population The hernodynamic profiles and clinical outcome of 5 consecutive patients who underwent AVR with the 19-mm Carpentier-Edwards porcine bioprosthesis from January 1986 to November 1991 at New England Medical Center were reviewed. This valve was selected in approx- 199 by The Society of Thoracic Surgeons -4975/9/$6.

2 114 BOJAR ET AL Ann Thorac Surg 199; Table 1. Baseline Demographic Data PREOP POSTOP Variable Male Female Mean age (y) Body surface area (m ) Presenting symptoms Angina Heart failure Syncope Endocarditis Hemolysis More than one symptom Valve lesion Predominant stenosis Predominant regurgitation Mixed disease Valve re-replacement Value (range, -88) 1.6 t IV 111 II imately % of all patients undergoing AVR during this period. This series includes patients undergoing isolated AVR or AVR combined with other cardiac surgical procedures, and includes all emergency operations and reoperations. Baseline demographic data (Table 1) revealed 45 female and 7 male patients with a mean age of 69 years (range, to 88 years) and a body surface area of 1.6 f.7 m (range, 1. to 1.95 m ). There were 8 patients with pure aortic stenosis, in whom the preoperative peak gradient was mm Hg (range, 4 to 175 mm Hg) and the aortic valve area was.47 f.16 cm, corresponding to an aortic valve area index of.8 k.9 cm/m (Table ). Overall, 96% of the patients were in New York Heart Association (NYHA) functional classes I11 and IV (Fig 1). Heart failure and angina were the most common presenting symptoms, and more than one-third of patients experienced multiple symptoms typically associated with aortic valve disease. One-third of the patients with angina had no evidence of coronary disease. Isolated AVR was performed in 1 patients, AVR combined with coronary artery bypass grafting in 1, doubleor triple-valve replacement with or without coronary artery bypass grafting in 7, and AVR combined with other procedures in patients (Table ). The operation was elective in patients (44%), urgent in 7 (5%), and emergent in only patients. Two patients in cardiogenic Table. Baseline Catheterization Data Variable Peak gradient (mm Hg) Mean gradient (mm Hg) Aortic valve area (cm)a Aortic valve area index (cm/m)a Ejection fraction a For 8 patients with pure aortic stenosis. Value 81.6 f t t f.15 I Fig 1. Comparison of preoperative and late postoperative New York Heart Association classification. Three of the four early deaths and three of the four late deaths occurred in patients who were in preoperative class IV. Two patients were lost to follow-up. shock underwent aortic valvuloplasty, allowing them to undergo subsequent AVR in a less emergent situation. Eight procedures were reoperations, four for prosthetic aortic insufficiency, one for prosthetic aortic stenosis, and three for native aortic valve disease present at the time of other reoperative procedures. Operative Technique All patients had radial arterial lines and Swan-Ganz catheters (Abbott Laboratories, Deerfield, IL) placed before operation. Intravenous cefazolin or vancomycin was given prophylactically and continued for 48 hours postoperatively. Anesthesia was induced and maintained using fentanyl, lorazepam, isoflurane, and either pancuro- Table. Procedures Performed and Associated Mortality Procedure Number Deaths AVR AVR-C ABG A V R - M V R AVR-MVR-CABG AVR-MVR-TVR AVR-postinfarction VSD repair AVR-root enlargement AVR-root enlargement-cabg 4 (19.%) a Number in parentheses indicates number of reoperations. AVR = aortic valve replacement; CABG = coronary artery bypass grafting; MVR = mitral valve replacement; TVR = tricuspid valve replacement; VSD = ventricular septa1 defect.

3 Ann Thorac Surg 199;56: BOJAR ET AL 114 nium or vecuronium. The aorta was cannulated with a 6.5-mm Sarns cannula (M Health Care, Ann Arbor, MI), and venous drainage was provided by a single two-stage cavoatrial cannula (Research Medical, Midvale, UT). Cardiopulmonary bypass was established using systemic hypothermia to 8 C with flow rates of to. L * min- - rn- and a mean arterial pressure of 65 to 7 mm Hg. The left ventricle was vented through the right superior pulmonary vein. Blood cardioplegia was used to maintain temperatures below 15 C in the interventricular septum during the cross-clamp period. Until mid-1989, cardioplegia was given antegrade into the aortic root or directly into the coronary ostia, or into both. Subsequently, retrograde coronary sinus cardioplegia using a self-inflating catheter (Research Medical) was used for reinfusions of cardioplegia and was used nearly exclusively for patients with aortic insufficiency. A topical cooling device was used in all patients. Retrograde warm blood cardioplegia (5 ml) was administered before removing the aortic cross clamp. Distal coronary anastomoses were performed first if coronary bypass grafting was indicated. Mitral valve replacement was performed before AVR, and the valve was kept incompetent with a Foley catheter. After thorough debridement of the aortic annulus, the 19-mm Carpentier-Edwards valve was placed using interrupted mattress sutures of - Ticron (Davis + Geck, Danbury, CT) with pledgets on the ventricular side. Proximal vein graft anastomoses were performed either during the cross-clamp period or after removal of the cross-clamp. The patient was weaned from bypass when the systemic temperature reached 6 C. Intravenous administration of heparin was started on the second postoperative day, and administration of warfarin sodium (Coumadin; DuPont Pharmaceuticals, Wilmington, DE) was begun to achieve a prothrombin time of 18 to seconds. Aspirin and dipyridamole were generally substituted for warfarin after months unless the patient remained in atrial fibrillation. If there were contraindications to Coumadin, aspirin and dipyridamole were given at the time of discharge from the hospital. Patient Follow-up Postoperative complications and causes of death were ascertained from a review of the patient s medical record. Operative mortality was defined as death occurring within days of the operation or during the initial hospitalization. Follow-up data were obtained from telephone contact with referring physicians and surviving patients. Valve-related morbidity followed the guidelines set forth by The Society of Thoracic Surgeons [5]. An assessment of NYHA classification was based on the patient s symptomatic and functional capacities both before operation and at the time of follow-up interview. Hernodynamic Follow-up Postoperative two-dimensional Doppler echocardiography was performed in all patients during the first postoperative week. Patients were studied using the Hewlett-Packard 77A ultrasound imaging system (Hewlett-Packard Company, Palo Alto, CA) with a.5- or.5-mhz phased array transducer. The velocity in the left ventricular outflow tract was assessed using pulsed-wave Doppler. The maximal instantaneous gradient (MIG) was calculated from the peak transvalvular flow velocity obtained by continuous wave Doppler using the modified Bernoulli s equation (MIG = 4 x V, where V is the maximal velocity in meters per second). The mean gradient was determined by a computerized integration of the flow velocity profile. The prosthetic valve EOA was calculated using the continuity equation of Skjaerpe and colleagues [6], where EOA = stroke volume divided by the velocity time integral of flow across the prosthesis. Statistical Analysis Values were expressed as mean f standard error of the mean. Results Postoperative Morbidity and Mortality The overall hospital mortality rate was 7.7% for the 5 patients receiving 19-mm Carpentier-Edwards valves (Table ). There was no mortality among patients undergoing isolated AVR or double- and triple-valve replacements. No patients undergoing an elective operation or reoperation died. All four deaths occurred in patients undergoing urgent AVR-coronary artery bypass grafting (4/1 = 4%), and all but one death occurred in patients in NYHA class IV (/ = 15%). Two of the four early deaths occurred in women with severe underlying cirrhosis as a result of progressive hepatic and multisystem organ failure. The third death occurred in an 8-year-old man as a result of aspiration pneumonia and pulmonary sepsis. The fourth death occurred in an 8-year-old man who suffered a cardiac arrest just before hospital discharge after an uneventful AVR-coronary artery bypass grafting x. His echocardiographic MIG was 6 mm Hg. Morbidity was relatively common in this patient population but usually was of minor consequence: Atrial fibrillation Infections Urinary tract Pneumonia Sternal Reexploration for bleeding Gastrointestinal problems Multisystem organ failure Respiratory failure not associated with multisystem organ failure Renal failure not associated with multisystem organ failure Stroke Low cardiac output syndrome not associated with multisystem organ failure Requirement for pacemaker Only about one-third of the patients did not experience any complications. Atrial fibrillation was the most com- 1

4 1144 BOJAR ET AL Ann Thorac Surg 199;56: mon complication, occurring in 18 patients, but resulted in no morbidity. Echocardiographic Follow-up Two-dimensional and Doppler echocardiographic evaluation was performed in all patients within the first postoperative week to provide a baseline for future assessment of valvular function. The MIG for all 5 patients was 44.7 f 1. mm Hg. Sufficient echocardiographic data for the calculation of mean gradients and EOAs were available in 4 patients. These revealed a mean gradient of 6.4 f 8. mm Hg and an EOA of.85 f.18 cm'. Late Follow-up Follow-up as of November 1991 was 96% complete with the status of only patients unknown. The mean follow-up period was 5 18 months (range, 1 to 61 months). Late deaths occurred in 4 patients. Two patients experienced sudden cardiac death at 9 and 8 months, respectively. Both had been in NYHA class I before their deaths. A third patient had undergone AVR for severe aortic insufficiency present at the time of emergency operation for a postinfarction ventricular septa1 defect. She had experienced extensive myocardial damage and died 6 months after operation of congestive heart failure. The fourth patient had critical aortic stenosis and an ejection fraction of.. She suffered a stroke and respiratory failure after operation but was later discharged from the hospital. She died 6 months postoperatively of pneumonia. The early postoperative prosthetic valve MIG for the 4 patients experiencing late deaths ranged from.7 to 44 mm Hg, values comparable to those noted in surviving patients. All surviving patients improved by at least one NYHA classification (Fig 1). Excluding the eight deaths and the patients lost to follow-up, 95% of the surviving patients were in classes I to 11. Two patients who had undergone AVR-coronary artery bypass grafting noted only slight improvement in congestive symptoms and improved from NYHA class IV to 111. The postoperative MIGs in these patients were and 5 mm Hg. None of the known surviving 4 patients at the time of follow-up had experienced any valve-related morbidity, including prosthetic valve endocarditis, thromboembolic events, anticoagulant-related hemorrhage, or intrinsic valve failure. Comment When a small aortic root is encountered during aortic valve operation, the surgeon must decide whether to perform an annulus-enlargement procedure or perform a standard AVR with placement of a small prosthetic valve. In active patients with a body surface area greater than. m, many surgeons would enlarge the aortic root to accommodate a valve that is at least 1 or mm in diameter to minimize the postoperative gradient []. If it is elected to perform a standard AVR, it is important to select a prosthetic valve with excellent hemodynamic function to minimize the degree of "valve prosthesispatient mismatch" that might offset the benefits of the procedure [l]. In patients under the age of 65 years with no contraindication to anticoagulation, it is generally accepted that a mechanical tilting-disc valve is preferable to a tissue valve because of its superior durability and hemodynamics []. In elderly patients, a tissue valve is often selected to avoid anticoagulation because of an increased risk of hemorrhagic events. In addition, the limited durability of tissue valves is less important in elderly patients because of their shorter anticipated life span. Despite concern that these patients may return in their late 7s or early 8s for a reoperation at higher risk, it has been reported that structural deterioration of porcine valves is "essentially nonexistent" in patients older than age 7 years at 1-year follow-up [7]. The concern over the hemodynamic function of the 19-mm porcine valves has been a potential limitation to their use. Catheterization studies of patients receiving the 19-mm Hancock standard orifice porcine valve in the late 197s demonstrated such significant gradients that many surgeons recommended that they should be avoided []. As a result, 19-mm porcine valves have not been widely used, and very few studies have been performed that examine both postoperative hemodynamics and clinical outcome. Without such information, it is difficult to differentiate among the various types of 19-mm valves to determine which is preferable in the small aortic annulus. The most accurate means of directly comparing the hernodynamics of various prostheses is with in vitro studies that use controlled flow rates to eliminate variables encountered with in vivo studies. Hemodynamic measurements include the transvalvular gradients, the EOA, and the performance index (the ratio of the EOA to the sewing ring area). Although direct catheter measurements have traditionally been used to obtain such hemodynamic data, echocardiography is becoming the standard for the evaluation of normal and abnormal valve function because it is a noninvasive, sensitive, and readily available diagnostic modality. It is therefore important to define the terminology used in the assessment of hemodynamic performance and to compare the measurements obtained by catheter studies and echocardiography for various types of prosthetic valves. The echocardiographic peak or MIG is a value obtained from the transvalvular flow velocity that is often used to assess the severity of prosthetic valve stenosis. There are several reasons to be cautious when using this measurement to evaluate valve hemodynamics. First, the MIG may detect nonhomogeneous flow through a prosthetic valve that leads to a higher gradient than is measured by catheter studies. Although in vitro studies have shown an excellent correlation of catheter and Doppler peak instantaneous gradients for most valves, Doppler studies may overestimate the gradient for certain valve designs, such as the bileaflet tilting-disc valves (St. Jude Medical), because of the detection of high-velocity jets between the leaflets [8]. Second, the MIG for all prosthetic valves generally exceeds the peak-to-peak gradient obtained by postoperative catheterization by about 1 to mm Hg because the latter does not measure an instantaneous gradient. Last, the flow velocity from which the MIG is

5 Ann Thorac Surg 199; BOJAR ET AL 1145 CAWENTIER-EDWARDS PORCINE VALVES Table 4. Postoperative Evaluation of Patients With 19-mm Valves by Echocardiography Peak Mean No. of Gradient" Gradient EOA Mean Interval Author [Ref. No.] Patients (mm Hi4 (mm Hg) (cm) at Study St. Jude Medical valve Panidis [ll] Not stated Cooper [1] days Perin [1] Variable Chafizadeh [14] days CarboMedics bileaflet valve Globits [15] mo Chambers [16] wk Bjork-Shiley Monostrut valve Aris [17] 1 4.6b Early postop Ionescu-Shiley pericardial valve Bojar [4] wk Cooper [1] days Teoh [] mo Lesbre [4] mo Carpentier-Edwards pericardial valve Lesbre [4] mo Frater [7] N/A y Carpentier-Edwards porcine (model 65) Bojar [present report] wk a Also referred to in text as the maximal instantaneous gradient (MIG). N/A = not available. derived is directly related to the stroke volume, which in turn is influenced by the loading conditions of the heart and by left ventricular function. As these parameters will vary between patients, echocardiography will measure a wide range of gradients for identical, similar-sized valves in different patients. The echocardiographic mean gradient, although still flow-dependent, is probably a more reliable measure of prosthetic valve hemodynamics than the MIG because it assesses the mean pressure drop across the valve. This gradient is determined by planimetry of multiple instantaneous gradients and has been found to correlate well with the mean gradient of native valves determined by direct catheterization [9]. For prosthetic valves, the correlation of Doppler and catheter mean gradients is similar to that noted for peak gradients, with excellent correlation noted for porcine valves [8]. Because of the dependence of echocardiographic gradient calculations on flow phenomena, it has been suggested that the EOA, which is based on both flow rate and pressure gradients, may be a better method of assessing prosthetic valve function. Although Doppler and catheter EOAs are comparable for porcine valves, the Doppler EOA tends to be lower in the bileaflet valves because Doppler detects higher gradients than does catheterization [lo]. Of note is the finding of in vitro studies that the EOA of the mechanical valves is relatively independent of flow, whereas that of the porcine bioprostheses is lower at low flow rates due to suboptimal leaflet opening [8, 11. This finding accounts in part for the wide spectrum of EOA measurements obtained in our study because the flow rates were variable and not controlled. Calculated from peak flow velocity (V) by modified Bernoulli equation (4V'). The performance index has also been used as a measurement of the hemodynamic efficiency of a prosthetic valve. It reflects the relation between the effective orifice of the prosthesis and its overall diameter. In vitro studies have shown a significantly greater performance index for the tilting-disc valves (performance index of.65 to.7) than for the porcine valves (performance index of. to.4), suggesting the superiority of the tilting-disc valves in the small aortic root []. Analysis of available postoperative hemodynamic data, however, has not demonstrated a significant difference between the peak gradients and EOAs of mechanical and porcine valves measured by echocardiography (Table 4) [lo-171. This may reflect the tendency of Doppler to overestimate gradients in mechanical valves compared with catheter studies. Furthermore, there have been no reports of any clinical advantage to use of mechanical valves in patients with a small aortic annulus. Laboratory studies of porcine valves have documented that they have the least efficient hemodynamics of all prosthetic valves. Porcine valves have been shown to have suboptimal leaflet opening with high pressure drops and narrow jet-type flows [Z]. The Hancock standard orifice valve (model 4) had an obstructive muscular shelf in the right coronary cusp and was changed to a "modified orifice" valve by replacing this cusp with another noncoronary cusp. In three studies of only 8 patients receiving the 19-mm Hancock modified orifice valve (Medtronic, Minneapolis, MN), the peak gradient measured by catheterization ranged from 19 to mm Hg with an estimated EOA of.9 to 1.1 cm [18-1. In vitro echocardiographic studies of this valve have shown an

6 1146 BOJAR ET AL Ann Thorac Surg 199;56: EOA of only.74 to.8 cm [lo]. There are no postoperative echocardiographic studies of patients receiving the 19-mm Hancock modified orifice valve. Although the Carpentier-Edwards porcine valve has been the most commonly implanted valve in the world during the past few years, there have been very few reports of its hemodynamics in the 19-mm size. This valve was designed with a flexible stent and a slightly asymmetric annulus to minimize the contribution of the right coronary leaflet with its muscular shelf to the orifice. Bove and associates [1] studied patients with 19-mm valves and reported an average mean gradient of mm Hg and an EOA of 1.1 cm during catheterization. A comparative intraoperative catheterization study of Carpentier-Edwards and Hancock modified orifice porcine valves performed by Khan and associates [] reported an average mean gradient of 16.9 mm Hg and an EOA of.85 cm in 4 patients with Hancock valves and an average mean gradient of 1.7 mm Hg and an EOA of.77 cm in 9 patients receiving Carpentier-Edwards valves. Our current study of 5 patients evaluated by Doppler echocardiography within the first week after operation showed an MIG of 44.7 f 1. mm Hg, a mean gradient of 6.4 f 8. mm Hg, and an EOA of.85?.18 cm. The pericardial valves have demonstrated excellent hemodynamics in the small aortic root due to nearly complete leaflet opening at low flow rates. The 19-mm Ionescu-Shiley valve demonstrated excellent clinical and hemodynamic performance with peak gradients of 8 to 4 mm Hg by echocardiography [4, 1,, 41. We had considered this our tissue valve of choice in the aortic position until its withdrawal in We then decided to use the Carpentier-Edwards porcine valve. The Carpentier-Edwards pericardial valve became available in the United States in late 1991 after nearly 1 years of clinical investigation. This valve was designed with a flexible stent and similar leaflet design to the Ionescu-Shiley valve, but it was constructed with different mounting techniques to improve its durability. Early clinical reports have shown it to have excellent hemodynamics. Postoperative catheterization studies of patients receiving the 19-mm valve have shown an average peak-to-peak gradient of.4 mm Hg, a mean gradient of to.8 mm Hg, and an EOA of 1. to 1.1 cm [5, 61. Echocardiographic studies have shown an average MIG of to mm Hg with an EOA of 1. cm [4, 71. Although there are no echocardiographic studies available that have directly compared the hemodynamics of the Carpentier-Edwards porcine valve with any of the other available prosthetic valves, review of the literature does suggest that the maximal instantaneous gradient measured across this valve exceeds that across the pericardial valves by about 1 to 15 mm Hg. Because of the curvilinear relation between mean systolic gradient and aortic valve area, any improvement in valve area to a value approaching 1 cm can result in a substantial reduction in the transvalvular gradient. Most studies have shown that the 19-mm Carpentier-Edwards porcine valve has an EOA of about.8 to 1. cm, which should suffice to minimize the degree of prosthesis- patient mismatch and produce significant clinical improvement in patients with critical aortic stenosis. The excellent clinical results achieved in our series are in part attributable to our patient population, with the typical patient being a relatively inactive 7-year-old woman with a body surface area of 1.6 m. We did, however, note a wide range of gradients in our postoperative patients with normally functioning valves, which reflects the dependence of the gradient calculation on transvalvular flow rates. Although several patients had an MIG that exceeded the mean value of 44.7 mm Hg, we did not note any correlation of high gradients with adverse clinical outcome. This study demonstrates that patients receiving the 19-mm Carpentier-Edwards porcine valve can achieve an excellent clinical outcome with essentially no intermediate-term morbidity and with demonstrable improvement in NYHA classification despite the presence of significant postoperative transvalvular gradients. These data clearly justify selective use of this valve in the elderly patient with a body surface area less than 1.9 m. This study also documents the range of postoperative gradients measured across this prosthesis that can be used as a benchmark when using echocardiography to evaluate valvular function at follow-up examination. We gratefully acknowledge the assistance of Janet Simonetti, BS, for her analysis of echocardiographic data and Celeste Lynch, RN, for her help in obtaining follow-up from referring physicians, patients, and their families. References 1. Rahimtoola SH. The problem of valve prosthesis-patient mismatch. Circulation 1978;58:-4.. Yaganathan 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;9: Jones EL, Craver JM, Morris DC, et al. Hernodynamic and clinical evaluation of the Hancock xenograft bioprosthesis for aortic valve replacement (with emphasis on management of the small aortic root). J Thorac Cardiovasc Surg 1978;75: Bojar RM, Diehl JT, Moten M, et al. Clinical and hemodynamic performance of the Ionescu-Shiley valve in the small aortic root. Results in 117 patients with 17 and 19 mm valves. J Thorac Cardiovasc Surg 1989; 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; Skjaerpe T, Hegrenaes L, Hatle L. Noninvasive estimation of valve area in patients with aortic stenosis by Doppler ultrasound and two-dimensional echocardiography. Circulation 1985;781W. 7. Jamieson WRE, Burr LH, Munro AI, Miyagishima RT, Gerein AN. Cardiac valve replacement in the elderly: clinical performance of biologic prostheses. Ann Thorac Surg 1989;48: Baumgartner H, Khan S, DeRobertis M, Czer L, Maurer G. Effect of prosthetic aortic valve design on the Dopplercatheter gradient correlation: an in vitro study of normal St. Jude, Medtronic-Hall, Starr-Edwards and Hancock valves. J Am Coll Cardiol 199;19:&. 9. Oh JK, Taliercio CP, Holmes DR Jr, et al. Prediction of the severity of aortic stenosis by Doppler aortic valve area

7 Ann Thorac Surg 199;56: BOJAR ET AL 1147 determination: prospective Doppler-catheterization correlation in 1 patients. J Am Coll Cardiol 1988;11: Baumgartner H, Khan SS, DeRobertis M, Czer LS, Maurer G. Doppler assessment of prosthetic valve orifice area. An in vitro study. Circulation 199;85: Panidis IP, Ross J, Mintz GS. Normal and abnormal prosthetic valve function as assessed by Doppler echocardiography. J Am Coll Cardiol 1986;8: Cooper DM, Stewart WJ, Schiavone WA, et al. Evaluation of normal prosthetic valve function by Doppler echocardiography. Am Heart J 1987;114: Perin EC, Jin BS, de Castro CM, Ferguson JJ, Hall RJ. Doppler echocardiography in 18 normally functioning St. Jude Medical aortic valve prostheses. Early and late postoperative assessments. Chest 1991;1: Chafizadeh ER, Zoghbi WA. Doppler echocardiographic assessment of the St. Jude Medical prosthetic valve in the aortic position using the continuity equation. Circulation 1991;8: Globits S, Rodler S, Mayr H, et al. Doppler sonographic evaluation of the CarboMedics bileaflet valve prosthesis: one-year experience. J Cardiac Surg 199; Chambers J, Cross J, Deverall P, Sowton E. Echocardiographic description of the CarboMedics bileaflet prosthetic heart valve. J Am Coll Cardiol 199;1: Aris A, Padro JM, Camara ML, et al. The monostrut Bjork- Shiley valve. Seven years experience. J Thorac Cardiovasc Surg 199;1: Foster AH, Tracy CM, Greenberg GJ, McIntosh CL, Clark RE. Valve replacement in narrow aortic roots: serial hemodynamics and long-term clinical outcome. Ann Thorac Surg 1986; 4: Rossiter SJ, Miller DC, Stinson EB, et al. Hemodynamic and clinical comparison of the Hancock modified orifice and standard orifice bioprostheses in the aortic position. J Thorac Cardiovasc Surg 198;8:544.. Craver JM, King SB 111, Douglas JS, et al. Late hemodynamic evaluation of Hancock modified orifice aortic bioprosthesis. Circulation 1979;6O(Suppl 1): Bove EL, Marvasti MA, Potts JL, et al. Rest and exercise hemodynamics following aortic valve replacement. A comparison between 19 and 1 mm Ionescu-Shiley pericardial and Carpentier-Edwards porcine valves. J Thorac eardiovasc Sura 1985;9: Khgn SS, Mitchell RS, Derby GC, Oyer PE, Miller DC. Differences in Hancock and Carpentier-Edwards porcine xenograft aortic valve hemodynamics. Effect of valve size. Circulation 199;8( Suppl 4): Teoh KH, Fulop JC, Weisel RD, et al. Aortic valve replacement with a small prosthesis. Circulation 1987;76(Suppl ):1. 4. Lesbre JP, Chassat C, Lesperance J, et al. Evaluation des nouvelles bioprostheses pericardiques par Doppler pulse et continu. Arch Ma1 Coeur 1986;79: Cosgrove DM, Lytle BW, Williams GW. Hemodynamic performance of the Carpentier-Edwards pericardial valve in the aortic position in vivo. Circulation 1985;7(Suppl): Pelletier LC, Leclerc Y, Bonan R, Crepeau S, Dyrda I. Aortic valve replacement with the Carpentier-Edwards pericardial bioprosthesis: clinical and hemodynamic results. J Cardiac Surg 1988;: Frater RWM, Salomon NW, Rainer WG, Cosgrove DM 111, Wickham E. The Carpentier-Edwards pericardial aortic valve: intermediate results. Ann Thorac Surg 199;5: