Nineteen-Millimeter Aortic St. Jude Medical Heart Valve Prosthesis: Up to Sixteen Years Follow-up

Transcription

1 Nineteen-Millimeter Aortic St. Jude Medical Heart Valve Prosthesis: Up to Sixteen Years Follow-up Dilip Sawant, FRCS, Arun K. Singh, MD, William C. Feng, MD, Arthur A. Bert, MD, and Fred Rotenberg, MD Division of Cardiothoracic Surgery and Department of Anesthesiology, Rhode Island Hospital, Brown University Medical School, Providence, Rhode Island Background. Prosthetic valve replacement in a small aortic root without annulus enlargement raises concern about its long-term benefits. Methods. Between July 1979 and June 1994, 104 (18%) of 593 patients underwent aortic valve replacement using the 19-mm St. Jude Medical heart valve prosthesis. There were 93 women and 11 men, with a mean age of years. Forty-four patients (42%) were 70 or more years old. The mean body surface area was m 2 (range, 1.2 to 2.1 m 2 ). Forty-nine patients (47%) underwent concomitant procedures; 23 patients (22%) required coronary artery bypass grafts and 25 patients (24%), mitral valve replacement. Ninety-eight patients (94%) presented in New York Heart Association class III and IV. Results. The operative mortality was 7.6% (8 patients). Follow-up was 100% with a mean of years (range, 1 to 16 years) and a total of 708 patient-years. There were 18 late deaths, with a mortality of 2.5% patient-years. The incidence of thromboembolism was 0.4% patient-years (3 patients) and anticoagulant-related morbidity was 0.85% patient-years (6 patients). Longterm survival in the two groups with a body surface area of less than 1.7 m 2 and 1.7 m 2 or more was not statistically different (p 0.30). The univariate analysis with body surface area as a predictor of mortality showed that a larger body surface area had no effect on the long-term mortality ( 2 p value 0.36). Survival for 5 and 10 years with the 95% confidence interval was 80.6% 8.3% and 61.6% 15%. Freedom from thromboembolism was 96.3% 4.2% and anticoagulant-related hemorrhage was 91.8% 6.8% at the end of 16 years. Cox proportional hazards model, with time-dependent covariates, showed that events of thromboembolism, anticoagulant-related hemorrhage, and myocardial infarction during follow-up increased the risk of late death (risk ratio, 9.5, 10.3, and 32.8, respectively). The age at operation was an independent risk factor, with decreased survival with age 70 or more years (p ). However, body surface area (p 0.97) and concomitant cardiac procedures (p 0.86) were not statistically significant predictors of death. Conclusions. The long-term performance of the 19-mm St. Jude Medical heart valve prosthesis in the small aortic root is satisfactory irrespective of the body surface area, and it is a viable alternative for such patients. (Ann Thorac Surg 1997;63:964 70) 1997 by The Society of Thoracic Surgeons The use of small aortic valve prostheses raises concerns about the potential for significant hemodynamic limitations resulting from residual obstruction to the left ventricular outflow tract. The St. Jude Medical (SJM) medical heart prostheses are designed to have minimal gradient at rest and after exercise. Wortham [1] and Gray [2] and their colleagues demonstrated that the gradients across 19-mm SJM valves in patients with a body surface area (BSA) of less than 1.7 m 2 is small and For editorial comment, see page 933. acceptable. Rashtian and colleagues [3] confirmed this optimistic finding by demonstrating the mean resting gradient of 5 mm Hg and peak gradient of less than 15 mm Hg in 19-mm SJM valves. Since June 1979 at Rhode Island Hospital small size SJM aortic prostheses have been used without root enlarging procedures. We Accepted for publication Aug 26, Address reprint requests to Dr Singh, Division of Cardiothoracic Surgery, Medical Office Center, Suite 470, Two Dudley St, Providence, RI believe that the good hemodynamic quality of this prosthesis surpasses the fear of using small prosthesis in the native aortic annulus. Excellent hemodynamic performance of the SJM valves has been reported by many investigators [4 7]. This study reports our experience with 19-mm SJM prostheses in the native aortic root in an attempt to investigate the determinants of their longterm morbidity and mortality. Material and Methods Between July 1979 and June 1994, 104 (18%) of 593 patients underwent aortic valve replacement with 19-mm SJM heart valve prostheses. There were 93 (89%) women and 11 (11%) men with a mean age of years (range, 32 to 85 years). Forty-four (42%) patients were 70 or more years old. The BSA ranged from 1.2 to 2.1 m 2 (mean, m 2 ). Sixteen patients (15%) had previous cardiac operations. There were 49 patients (47%) requiring concomitant cardiac procedures: 23 patients (22%) required coronary artery bypass grafting (CABG); 25 patients (24%) mitral valve replacement; and 1 patient 1997 by The Society of Thoracic Surgeons /97/$17.00 Published by Elsevier Science Inc PII S (96)

2 Ann Thorac Surg SAWANT ET AL 1997;63: mm ST. JUDE AORTIC PROSTHESIS 965 Table 1. Patient Demographics Characteristic Mean SD No. of Patients % Age (y) BSA (m 2 ) Sex Female Male History of hypertension Family history of CAD Diabetes Preoperative atrial fibrillation Duration of follow-up (y) NYHA II III IV Concomitant CABG Concomitant MVR Elective operation Emergent operation Previous cardiac operation Valve pathology Stenosis Regurgitation 9 8 Mixed lesion Cause of pathology Bicuspid Rheumatic Myxomatous Infective 1 1 Operative mortality Late mortality %/pt-y BSA body surface area; CABG coronary artery bypass grafting; CAD coronary artery disease; MVR mitral valve replacement; NYHA New York Heart Association; pt-y patient-year; SD standard deviation. required atrial septal defect closure. Preoperative New York Heart Association (NYHA) functional class presentation revealed 98 patients (94%) in classes III and IV. Eighty-four patients (81%) underwent operation electively. Table 1 shows patient demographics. Operative Technique Before 1987, moderate hypothermic (25 C to 28 C) cardiopulmonary bypass was used, including bicaval cannulation with snares and a membrane oxygenator. The left ventricle was vented through the right superior pulmonary vein. Myocardial protection was achieved using cold crystalloid cardioplegia followed by multidose cold blood cardioplegia. Topical myocardial cooling was done with ice cold saline slush. After 1987, normothermic (37 C) cardiopulmonary bypass was used [8]. The aortic annulus was debrided meticulously and measured with a snugly fitting sizer. The SJM valve prosthesis was implanted intraannularly using everting interrupted and pledgeted mattress sutures. Fifty-nine patients (57%) were found to have pure aortic stenosis, 36 patients (35%), mixed aortic valve disease, and only 9 patients (8%), pure aortic regurgitation. Pathologic examination of the explanted valves studied in our series showed bicuspid valve in 65 patients (63%), rheumatic in 29 (28%), myxomatous degeneration in 9, and infective in 1 patient (Table 1). Anticoagulation All patients were anticoagulated with warfarin sodium from the second postoperative day. Before 1992 prothrombin time was used to guide the dose. Before 1985 it was kept between 2 and 2.5 times the control and after 1985 it was kept between 1.5 and 2. Since 1992 international normalized ratio has been used as control and kept between 3.0 and 3.5. All patients since 1992 have been followed up to receive warfarin sodium to keep the mean international normalized ratio 3.0. Follow-up All patients were followed up annually (100% follow-up) by their referring physicians or cardiologist. Their follow-up information was obtained and analyzed. In the fall of 1995 all surviving patients were asked to complete a standard questionnaire by trained personal via the telephone (100% compliance). All hospital records for the patient admissions were analyzed to enumerate their valve-related problems. Follow-up ranged from 1 to 16 years, with mean of years and cumulative patient survival was 708 patient-years. Mortality and morbidity is stringently defined as per guidelines published by The Society of Thoracic Surgeons and The American Association for Thoracic Surgery [9]. All data are stored in a Fox-pro data base specially written for our use. Statistical Methods Statistical analyses were performed using SAS statistical software package. Discrete variables are presented as counts and percentages. Continuous variables are presented as mean standard deviation. Late events were also calculated as linearized rates and presented as percent per patient-year. Statistical significance was at the nominal level of For the graphs of freedom from an event, the Kaplan- Meier product limit method was used [10]. The number of patients at risk for interval is shown, along with the 95% confidence limits for the estimates. Variables considered to be risk factors (age at operation, BSA, NYHA class at implantion, and concomitant cardiac procedures) were examined by the Kaplan-Meier product limit method and the stratified log-rank test was used to determine statistical significance. A Cox proportional hazards model with time-dependent covariates [11] was performed to determine the model that best predicts time to mortality. The importances of the covariates were evaluated singly and in combination using stepwise procedures. All patients were included in the analysis no matter what length of follow-up; hence, all deaths were included. Age at im-

3 966 SAWANT ET AL Ann Thorac Surg 19-mm ST. JUDE AORTIC PROSTHESIS 1997;63: plantation and BSA were used as continuous variables, and the remaining variables were categoric. The model that best fits the data includes age at implantation and the late events of anticoagulation complication, thromboembolism, and myocardial infarction. Variables that did not make it into final model were BSA, NYHA class, concomitant CABG, and late events of reoperation, endocarditis, and paravalvular leak. Results Operative Mortality The operative mortality was 7.6% (eight deaths); however, during the past 10 years it was 4.4% (four deaths among 90 patients). The majority of deaths were among patients more than 70 years of age (six deaths; mortality 13%). Six deaths were among 73 patients with a BSA of less than 1.7 m 2, and the remaining two among those with a BSA of 1.7 m 2 or more (31 patients). There were three deaths among 38 patients in NYHA class IV (mortality, 7.9%), whereas five deaths were in the remaining 66 patients (mortality, 7.6%). The mortality for those with previous cardiac operations was 18% (3 deaths among 16 patients), whereas mortality for patients having concomitant cardiac procedure was 10% (4 deaths among 49 patients). There was strong association between operative mortality and operation done emergently (5 deaths among 20 patients). The relative risk of operation done emergently was 1.2 (p using 2 test). Functional Class All patients were assessed for their functional performance as per the NYHA class during their follow-up. Among the survivors in NYHA class IV 35 patients (83%) improved to class I or II and the remaining to class III, whereas patients in NYHA class III (55 patients, 91%) improved to class I and II, and the rest remained in class III (Fig 1). The late survival as per the preoperative NYHA class was not statistically significant (p 0.35). Late Mortality Eighteen patients died during follow-up, giving a linearized rate of late mortality of 2.5%/patient-year. There were 13 patients with cardiac death, with a linearized rate of 1.8% patient-years. Among the cardiac deaths, 3 patients died of sudden death (0.4%/patient-year); in 10 patients the cause of death was myocardial infarction. The remaining five deaths were attributable to causes unrelated to their operation (0.7%/patient-year). The Kaplan-Meier life table (Fig 2) showed survival at 5 and 10 years as 80% 8.3% and 61.6% 15.% with 95% confidence interval, respectively. Figure 3 shows survival curves for freedom from cardiac deaths. Table 2 shows univariate analysis of risk factors for mortality. Influence of age on the long-term survival was investigated by dividing patients into two groups as per their age at operation (group 1, 69 years; group 2 70 years). There were 44 patients (42%) 70 or more years old. The age at operation was found to be an independent risk factor decreasing long-term survival with age 70 years or more (Fig 4) with p Fig 1. Preoperative and follow-up New York Heart Association (NYHA) class of patients, excluding the eight operative deaths. Influence of BSA on the long-term survival was studied. Figure 5 shows the survival curves. A BSA of 1.7 m 2 as a risk factor for the long-term results was chosen for two reasons; calculated from the published data the valve area index of the 19-mm SJM prosthesis for a patient with a BSA of 1.7 m 2 remains only 0.96 cm 2 /m 2 and in our study group 30% of the patients had a BSA of 1.7 m 2 or more. Therefore, the study group was divided into group 1 ( 1.7 m 2 ) and group 2 ( 1.7 m 2 ). There was no statistical difference in survival (p 0.30) in the two groups. Univariate analysis with BSA as the predictor of late mortality showed that larger BSA did not increase the risk of late mortality ( 2 p 0.36). When BSA was added to the Cox proportional hazards model as a continuous variable it was found to be an insignificant predictor for late mortality (p 0.97) (Table 4). The influence of concomitant procedures on late mortality was analyzed by dividing the study population into three groups: isolated aortic valve replacement, aortic valve replacement with CABG, and aortic valve replacement with mitral valve replacement. There was no statistically significant difference in the survival of these groups (p 0.86). Late Morbidity There was no structural malfunction noted in this series. Postoperative endocarditis did not occur in any patient.

4 Ann Thorac Surg SAWANT ET AL 1997;63: mm ST. JUDE AORTIC PROSTHESIS 967 Fig 2. Kaplan-Meier survival curve of freedom from late mortality. The dashed lines represent the 95% confidence intervals. Paravalvular leak occurred in 2 patients, with a linearized rate of 0.3%/patient-year; 1 died long after reoperation due to metastatic cancer, and the other remains asymptomatic. Freedom from the paravalvular leak with 95% confidence interval was 98.1% 2.7% and 95.3% 6% at 5 and 10 years, respectively. Valve thrombosis did not develop in any patient. Three patients had a stroke, with a linearized rate of thromboembolism of 0.4%/patient-years. Although the dose of warfarin sodium and the level of anticoagulation was varied for the different time periods, the patients presenting with this complication were so few that their calculated statistical significance will be invalid. Table 3 shows the freedom from thromboembolism. Hemorrhage was noted in 6 patients requiring hospitalization and transfusion; 4 patients were 70 or more years of age. None of the patients required surgical intervention. The linearized rate was 0.85%/patient-year. Freedom from anticoagulant-related hemorrhage at 5 and 10 years with 95% confidence interval was 94.3% 4.9% and 91.8% 6.8%, respectively. Comment Management of the small aortic root is a challenge to the surgeon with regard to operative technique and selection of prosthesis. These study patients are predominantly older aged women. In this series there are 89% women Fig 3. Kaplan-Meier survival curve of freedom from cardiac-related death. The dashed lines represent the 95% confidence intervals.

5 968 SAWANT ET AL Ann Thorac Surg 19-mm ST. JUDE AORTIC PROSTHESIS 1997;63: Table 2. Cox s Model Univariate Analysis of Mortality Variable p Value Risk Ratio 95% Confidence Interval Age at operation BSA Preimplantation NYHA Concomitant CABG Concomitant MVR BSA body surface area; CABG coronary artery bypass grafting; MVR mitral valve replacement; NYHA New York Heart Association. with mean age of 66.2 years. Only 30% of patients had a BSA of more than 1.7 m 2 (mean BSA, 1.61 m 2 ). Experience with small prostheses [12 15] with a narrow aortic root has been unfavorable. Many techniques have been described [16 18] to accommodate a larger prosthesis to circumvent this problem. These techniques potentially increase the risks of injury to the coronary arteries and conduction bundle, and of surgical hemorrhage. David and Uden [19] described techniques to accommodate a larger prosthesis with satisfactory short-term results; as yet the long-term results are not available. Homograft replacement of the aortic valve as a functional unit or root replacement [20] has an added advantage of freedom from anticoagulation. Replacement of the aortic valve with a pulmonary autograft may have some growth potential [21]. However, these techniques require specialized training and experience and are not currently widely used. Toronto SPV prostheses have early encouraging results [22], with a decrease in systolic gradient and good hemodynamics. Very few valves smaller than 21 mm have been used. The present study demonstrates a standard and reproducible technique with the 19-mm SJM prosthesis with satisfactory long-term results. Although 47% of patients underwent concomitant cardiac procedures, in this series the operative mortality and the actual survival at 5 and 10 years (80.6% 8.3%%, 61.6% 15.0%, respectively) compare well with other reported series [5, 7, 23 25]. Czer [25] and Arom [5] and their colleagues reported a 9-year survival of 51% and 75.5%, respectively, when larger size valves were used. Age at operation is also a relevant factor as average age in their series was less than 60 years, compared with our average of 66.2 years. The principal concern of the mechanical valve is its thrombogenic potential and the need for anticoagulation. In this series there was no incidence of valve thrombosis and the linearized rate of thromboembolism and anticoagulant-related hemorrhage were comparable with all reported series. The freedom from thromboembolism at 10 years was 96.3% 4.2% and for anticoagulant-related hemorrhage it was 91.8% 6.8%. The review by Akins [26] of four common mechanical prostheses revealed a linearized rate of thromboembolism at 1.6% patientyears (range, 0.7% to 2.8%) for the SJM prosthesis, and our results are consistent with those (Table 3). There are very few data available showing the ideal level of anticoagulation for minimum thromboembolic episodes and hemorrhagic complications. Kopf and co-workers [27] used a prothrombin level of 1.3 to 1.5 times the control and found the linearized rate of thromboembolism at 0.67% patient-years; the anticoagulant-related hemorrhage at 1.3% patient-years; and freedom from these events at the end of 5 years as 97.4% and 94.4%, respectively. Our series compares well with this finding (Table 3). Akins [28] recently suggested a composite thromboembolism and bleeding index. In this series it is 1.27%/ patient-year, which compares well with all available mechanical prostheses. These findings confirm the durability and minimum thrombogenicity of the SJM prostheses. Fig 4. Cumulative survival curves for the two age groups: group 1 ( 69 years) and group 2 ( 70 years).

6 Ann Thorac Surg SAWANT ET AL 1997;63: mm ST. JUDE AORTIC PROSTHESIS 969 Fig 5. Cumulative survival curves for two body surface area (BSA) categories: group 1 (body surface area 1.7 m 2 ) and group 2 (body surface area 1.7 m 2 ). Hemodynamic Performance The hemodynamic performance of the 19-mm prosthesis in the native aortic root is a genuine concern, especially for patients with larger BSA. A high gradient across the small aortic prosthesis may adversely affect long-term results. Although the gradients across small Hancock [14, 15] porcine valves and Björk-Shiley tilted-disc prostheses [12] are reported to be high, Schaff and colleagues [23] reported good hemodynamic performance, satisfactory long-term results, and functional improvement with the 19-mm Björk-Shiley prosthesis. Rahimtoola [29] identified the valve area index at 0.75 cm 2 /m 2 or less as severe aortic stenosis in an unoperated patient. Calculated from the published data for the orifice area of standard SJM 19-mm prosthesis, valve area index of the 19-mm valve in patients with a BSA of 1.7 m 2 is calculated at 0.96 cm 2 /m 2. This postimplantation valve area index would suggest significant residual aortic stenosis, generating concerns of poor long-term cardiac performance. Nevertheless, moderate improvement in the left ventricular orifice reduces the gradient exponentially, as the relationship of gradient to valve area is curvilinear [29] and improvement in the cardiac performance is noted. Our series Table 3. Freedom From Late Events Events 5 Years 10 Years Estimate (%) 95% CI Estimate (%) 95% CI Late death Cardiac death Sudden death Thromboembolism Anticoagulant related Paravalvular leak CI confidence interval. demonstrated long-term functional improvement. Hayashi and co-workers [30] reported favorable results with small prostheses; the mean BSA in their patients was only 1.45 m 2. Kratz and colleagues [7] found that a BSA of 1.9 m 2 was an independent risk factor, but they do not mention the number of patients with sizes 19-mm and 21-mm prostheses. In this series 2 patients had a BSA of 1.9 m 2 or more and 31 had a BSA of 1.7 m 2 or more. Our analysis shows that a BSA of 1.7 m 2 or more was not a risk factor for long-term results (p 0.3). We found that a larger BSA did not increase late mortality by multivariate analysis (p 0.97). There was no statistical difference in long-term survival in the patients with a BSA of 1.7 m 2 or more, compared with the group with a BSA of less than 1.7 m 2 (Fig 5). Prosthesis patient mismatch is associated with higher incidence of adverse clinical events [31]. Nevertheless, this was not so in our series, as we found the incidence of angina and congestive cardiac failure was small and long-term survival was good. It is possible that because the majority of our patients are elderly women they may limit their physical activity. Unfortunately, we do not have serial preoperative and postoperative echocardiograms to compare the left ventricular mass index. Table 4. Cox s Proportional Hazards Model Variable p Value Risk Ratio Beta Coefficient 95% CI Age at implant Anticoagulation complication a Thromboembolism a Myocardial infarction a Body surface area a Time-dependent variables. CI confidence intervals.

7 970 SAWANT ET AL Ann Thorac Surg 19-mm ST. JUDE AORTIC PROSTHESIS 1997;63: Risk Factors for Long-Term Survival Age at operation is the only independent risk factor for the long-term survival in our series (age 70 years increased the late mortality, p ). This study shows survival in two groups (age 69 years and age 70 years) was 90.5% 8.1%, 84.4% 13.7%, and 67.2% 15.0%, 36.6% 21.9%, respectively, at 5 and 10 years with a 95% confidence interval. Despite the relative disadvantage older patients long-term survival remains substantial. He and colleagues [32] reported that the small prosthesis with concomitant CABG was the only negative determinant of long-term survival. We did not find that BSA or concomitant CABG affected the long-term survival. Age at operation, thromboembolism, anticoagulant-related hemorrhage, and myocardial infarction during follow-up increased the risk of death, and the Cox proportional hazards model with time-dependent covariates showed these events decreased the time to death. In conclusion, the long-term performance of aortic valve replacement using the SJM 19-mm prosthesis without annulus enlargement, as judged by improvement in the functional class, survival statistics, durability of the prosthesis, and comparable valve-related morbidity, is satisfactory. Age at operation of 70 years or more, stroke, anticoagulant-related hemorrhage, and myocardial infarction were risk factors affecting long-term survival. Body surface area, preoperative functional class, and concomitant cardiac procedures do not influence longterm survival. Thus, we conclude that the use of the SJM 19-mm prosthesis in the native aortic annulus is a reasonable alternative. The effective orifice of the new high-performance SJM valve is the same as the next larger standard valve, and it is reasonable to assume that the wide use of these valves may improve the results further. We gratefully acknowledge the assistance of Sharla Williamson who provided statistical consultations, and Merry Meyers for her assistance in the data collection. References 1. Wortham DC, Major, MC, Tri TB, Bowen TE. Hemodynamic evaluation of St. Jude Medical valve prosthesis in the small aortic annulus. J Thorac Cardiovasc Surg 1981; 81: Gray RJ, Chaux A, Matloff JM, et al. Bileaflet, tilting disc and porcine aortic valve substitutes: in vivo hydrodynamic characteristics. J Am Coll Cardiol 1984;3: Rashtian MY, Stevenson DM, Allen DT, et al. Flow characteristics of four commonly used mechanical heart valves. Am J Cardiol 1986;58: Chaux A, Gray RJ, Matloff JM, Feldman H, Sustaita H. An appreciation of the new St. Jude valvular prosthesis. J. Thorac Cardiovasc Surg 1981;81: Arom KV, Nicoloff DM, Kersten TE, Northrup WF III, Lindsay WG, Emery RW. Ten years experience with the St. Jude Medical valve prosthesis. Ann Thorac Surg 1989;47: Arom KV. St. Jude Medical prosthesis: another 10 year follow up report. Ann Thorac Surg 1993;56: Kratz JM, Sade RM, Crawford FA Jr, Crumbley AJ, Stroud MR. The risk of small St. Jude aortic valve prostheses. Ann Thorac Surg 1994;57: Singh AK, Feng WC, Bert AA, Rotenberg F. Warm body, cold heart surgery : clinical experience in 1817 patients. Eur J Cardiothorac Surg 1993;7: 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: Kaplan E, Meier P. Non-parametric estimation from incomplete observations. J Am Stat Assoc 1958;53: Cox DR. Regression models and life-tables. J R Stat Soc B 1972;34: Björk VO, Henze A, Holmgren A. Central hemodynamics at rest and during exercise before and after aortic valve replacement with the Björk-Shiley tilting disc valve in patients with isolated aortic stenosis. Scand J Thorac Cardiovasc Surg 1973;7: Hannah H, Reis RL. Current status of porcine heterograft prostheses: a 5-year appraisal. Circulation 1975;52(Suppl 2): Cohn LH, Sanders JH Jr, Collins JJ Jr. Aortic valve replacement with the Hancock porcine xenograft. Ann Thorac Surg 1976;22: Jones EL, Craver JM, Morris DC, et al. Hemodynamic 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: Manouguian S, Seybold-Epting W. Patch enlargement of the aortic valve ring by extending the aortic incision in to the anterior mitral leaflet. J Thorac Cardiovasc Surg 1979;78: Blank RH, Pupello DF, Bessone LN, Harrison EE, Sbar S. Method of managing the small aortic annulus during valve replacement. Ann Thorac Surg 1976;22: Konno S, Imai Y, Iida Y, Nakajima M, Tatsuno K. A new method for prosthetic valve replacement in congenital aortic stenosis associated with hypoplasia of the aortic valve ring. J Thorac Cardiovasc Surg 1975;70: David TE, Uden DE. Aortic valve replacement in adult patients with small aortic annuli. Ann Thorac Surg 1983;36: Knott-Craig CJ, Elkins RC, Stelzer PL, et al. Homograft replacement of the aortic valve and root as a functional unit. Ann Thorac Surg 1994;57: Ross DN. Replacement of the aortic valve with a pulmonary autograft: the switch operation. Ann Thorac Surg 1991;52: Del Rizzo DF, Goldman BS, David TE. Aortic valve replacement with a stentless porcine bioprosthesis: multicentre trial. Canadian Investigators of the Toronto SPV valve trial. Can J Cardiol 1995;11: Schaff HV, Borkon AM, Hughs C, et al. Clinical and hemodynamic evaluation of the 19 mm Björk-Shiley aortic valve prosthesis. Ann Thorac Surg 1981;32: He GW, Acuff TE, Ryan WH, et al. Aortic valve replacement: determinants of operative mortality. Ann Thorac Surg 1994; 57: Czer LS, Chaux A, Matloff GM, et al. Ten-year experience with the St. Jude Medical valve for primary valve replacement. J Thorac Cardiovasc Surg 1990;100: Akins CW. Mechanical cardiac valvular prostheses. Ann Thorac Surg 1991;52: Kopf GS, Hammond GL, Geha AS, Elefteriades J, Hashim SW. Long-term performance of the St. Jude Medical valve: low incidence of thromboembolism and hemorrhagic complications with modest doses of warfarin. Circulation 1987; 76(Suppl 3): Akins CW. Results with mechanical cardiac valvular prostheses. Ann Thorac Surg 1995;60: Rahimtoola SH. The problem of valve prosthesis patient mismatch. Circulation 1978;58: Hayashi J, Nakazawa S, Okazaki H, et al. Long term clinical results with 19 mm or 21 mm standard aortic St. Jude Medical prosthesis. J Heart Valve Dis 1994;3: He GW, Grunkemeier GL, Gately HL, Furnary AP, Starr A. Up to thirty-year survival after aortic valve replacement in the small aortic root. Ann Thorac Surg 1995;59: