A valve was initiated at the Medical University of

Transcription

1 St. Jude Prosthesis for Aortic and Mitral Valve Replacement: A Ten-Year Experience John M. Kratz, MD, Fred A. Crawford, Jr, MD, Robert M. Sade, MD, Arthur J. Crumbley, MD, and Martha R. Stroud, MS Division of Cardiothoracic Surgery, Medical University of South Carolina, Charleston, South Carolina From January, 979 through December 99, 456 adult patients underwent isolated aortic (AVR) (254) or mitral (MVR) (22) valve replacement with the St. Jude prosthesis at the Medical University of South Carolina. Age ranged from 2 to 84 years (mean: 54 f 5 years for AVR; years for MVR). Male sex predominated in the AVR group (66%) and female sex in the MVR group (64%). Ninety-two patients (2%) had associated coronary artery bypass grafting (AVR, 25%; MVR, 4%). There were 7 deaths (3.7%) occurring during the same hospitalization or within 3 days (AVR, /254 [3.9%; MVR, 7/22 [3.5%). Follow-up is 94.5% complete and ranges from. to 3 months (mean, 55 f 37 months; total, 2,73 patient-years). In the AVR group, 53 late deaths have occurred and actuarial survival is 8% f 3% at 5 years and 47% f 9% at years. Twenty-one patients have sustained thromboembolic episodes (l.d%/patientyear), and the probability of remaining free of thromboembolism at years is 67% f 3%. The mean improvement in New York Heart Association functional class from preoperative to postoperative is 3. f.76 to.6 f.84 (p <.). In the MVR group, there have been 4 late deaths, and the actuarial survival was 8% f 3% at 5 years and 63% f 5% at years. Twenty-eight patients have sustained thromboembolic complications (2.9%/ patient-year), and the probability of remaining free of thromboembolism at years is 77% f 5%. The mean improvement in New York Heart Association functional class is from 3.4 f.63 preoperatively to postoperatively (p <.). There were no mechanical failures, but 9 patients underwent 22 replacements of a previously implanted St. Jude prosthesis for endocarditis (4), paravalvular leak (4, thrombosis (2), and hemolysis (2). This intermediate-term follow-up confirms our earlier impression that in adults in whom valve repair is not possible, the St. Jude valve is a reliable, durable prosthesis with excellent hemodynamic function and a low rate of thrombosis/thromboembolism. (Ann Thoruc Surg 993;56:462-8) clinical trial of the St. Jude Medical prosthetic heart A valve was initiated at the Medical University of South Carolina in April 979. We reported our 4-year experience with the valve in 984 [l]. This report will summarize our intermediate-term (-year) experience. In the interval April 979 through December 99, 456 patients received either isolated aortic (AVR; 254 patients) or For editorial comment, see page 43. mitral valve replacement (MVR; 22 patients) with the St. Jude prosthesis. This series is consecutive for St. Jude valves, but during this interval, approximately 4% of all adult patients undergoing valve replacement received tissue valve prostheses. We believe that patients undergoing valve replacement are best analyzed as separate groups. Accordingly, this series does not include patients undergoing combined valve procedures or children. The latter group has been the subject of previous reports from our group [2]. Presented at the Thirty-ninth Annual Meeting of the Southern Thoracic Surgical Association, Wesley Chapel, FL, Nov 5-7, 992. Address correspondence to Dr Kratz, Division of Cardiothoracic Surgery, Medical University of South Carolina, 7 Ashley Ave, Rm 49, CSB, Charleston, SC Patients and Methods In the AVR patients, age ranged from 2 to 84 years (mean, 54 years), and in the MVR patients, age ranged from 2 to 82 years (mean, 5 years). This is significantly older than the mean ages of 49 (AVR) and 44 (MVR) years in our initial report and reflects the trend toward operating on older patients as well as our expanded indications for the use of this prosthesis in that age group. The age distribution of each group is depicted in Figure. Twice as many AVR (25%) as MVR (4%) patients underwent concomitant coronary artery bypass procedures. Most patients in both groups were in New York Heart Association class or IV before operation (Figs 2, 3). Causes of disease are shown in Table. Operative techniques included standard cardiopulmonary bypass with a disposable bubble or membrane oxygenator and moderate hypothermia (26" to 3 C). Cold crystalloid or blood potassium-induced cardioplegia according to surgeon preference was used for myocardial protection. After meticulous debridement of the valve annulus, all prostheses were inserted using interrupted sutures of 2- braided Dacron reinforced with Teflon felt pledgets. Sutures were carefully placed in the annulus from above downward so that when they were tied, the valve annulus would be everted and the prostheses "in- 993 by The Society of Thoracic Surgeons /93/$6.

2 Ann Thorac Surg 993;56:462-8 KRATZETAL 463 % Patients 3 2 Aortic Valves Mitral Valves % Patients B% NYHACI~~~I NYHA Class II NYHA class III NYHA ClaSS IV Age by Decades Fig I. St. Jude valve replacement patients' age range. traannular" in location. No annulus enlarging procedures were performed. Valve sizes ranged from 9 to 33 mm (Fig 4). Heparin administration was started (5, units subcutaneously every 6 hours) on the first postoperative day and continued until the prothrombin time could be regulated by the administration of warfarin sodium. During the first few years of this report, we attempted to maintain the prothrombin time at approximately twice normal control. However, over the last 8 years the prothrombin time has only been prolonged to approximately % times normal. FO~~OW-UP Patients were followed up prospectively on a yearly basis at the Medical University of South Carolina and by their family physicians. All patients were contacted through their referring physician and by written questionnaire (additional data were collected by telephone or personal interview in some cases). Twenty-five patients (6 AVR, 9 MVR) could not be located and are lost to follow-up. Follow-up is thus 94.5% complete overall (95.5% MVR; 93.7% AVR). Follow-up to death or last assessment for AVR ranges from to 3 months (mean, 52? 37 months; total,, patient-years). Follow-up for MVR ranges from to 29 months (mean, 57 * 36 months; total, 962 % Patients W NYHACI~~~I NYHA Class NYHA Class NYHA Class IV 47% PREOP YEAR YEAR 5 YEAR a Fig 2. Aortic St. Jude valve replacement patients' New York Heart Association (NYHA) class at preoperative, -year, 5-year, and 8-year evaluation. 2 PREOP YEAR YEAR 5 YEAR 8 Fig 3. Mitral St. lude valve replacement patients' New York Heart Association (NYHA) class at preoperative, -year, 5-year, and 8-year evaluation. patient-years). Deaths and complications were stringently defined according to published guidelines of The Society of Thoracic Surgeons and the American Association for Thoracic Surgery [3]. Statistical Analysis Continuous variables are reported as the mean? standard deviation, and categorical variables are pre- Table. Patient Data Variable Total number Age (Y) Range Mean? SD Sex Male Female Cause Rheumatic Endocarditis Calcific Malfunctioning prosthesis Ischemic Degenerative Congenital Other Lesion Stenosis Insufficiency Combined Previous valve replacement Associated CABG Preoperative functional class (mean? SD) Operative mortality CABG = coronary artery bypass grafting; Group I (Aortic Valve Replacement) & (66%) 86 (34%) 59 (23%) 32 (3%) 5 (45%) 9 (7%) (%) 7 (7%) 5 (2%) 7 (3%) 2 (47%) 3 (4%) 3 (2%) 9 (7%) 63 (25%) (3.9%) Group I (Mitral Valve Replacement) ? (36%) 3 (64%) (5%) 3 (6%) 6 (8%) 25 (2%) 6 (8%) 25 (2%) 4 (2%) 3 (%) 79 (39%) (5%) 22 (%) 25 (2%) 29 (4%) (3.5%) SD = standard deviation.

3 464 KRATZETAL Ann Thorac Surg 993;56:462-8 % Values % Surviving 3 2 Aortic Valves Mitral Valves n Valve Sizes Fig 4. St. Jude valve replacement patients valve size range *2.9% Valve Related Deaths 46.7 * 8.7 X All Deaths Yearshstop #Patients A % Surviving Valve Related Deaths sented as percentages. Comparisons of changes in New York Heart Association functional class were made with the Wilcoxon signed-rank test [4]. Actuarial curves were constructed to describe mortality and the incidence of valve-related complications. Individual event-free curves were calculated for valve-related mortality and overall mortality, as well as for the incidence of reoperation, endocarditis, thromboembolic events, and anticoagulantrelated hemorrhage. Fivefold decremental curves were calculated to describe valve-related mortality and morbidity jointly [5,6]. For these composite functions, each valve was considered to be at risk for only one event. Valverelated deaths were given precedence over nonfatal complications in cases in which both were experienced. Actuarial estimates were calculated using the Kaplan-Meier [7] technique and are reported with the standard error of the estimate. Comparisons of these estimates among subgroups were performed with the Mantel-Cox (log-rank) test [8]. All statistical analyses were calculated with the BMDP statistical software package, edition 988. Results Group (Aortic Valve Replacement) HOSPITAL MORTALITY. Ten patients died in the postoperative period, resulting in an operative mortality rate of 3.9%. LATE MORTALITY. There have been 53 late deaths, and the overall actuarial survival at years is 46.7% (Fig 5A). The causes of late death are listed in Table 2. Of the 53 late deaths 28 were possibly the result of valve-related complications. Twenty-five deaths were not valve related. The majority of patients (3) died as the result of progressive heart failure in spite of proper prosthetic valve function. Cancer (8 patients) was the second most frequent cause of death. VALVE-RELATED COMPLICATIONS. No cases of structural deterioration were encountered. Two cases of noninfectious paravalvular leak occurred. No patient had leaflet malfunction or clinically important hemolysis. Twenty AVR patients have had thromboembolic events (8 embolic, 2 valve thrombosis) during the follow-up 6 4 I All Deaths *O I OYearspOstop #Patients B Fig 5. (A) Probability of survival for St. Jude aortic valve patients. (B) Probability of survival for St. Jude mitral valve patients. period. In our earlier report all thromboembolic episodes occurred within the first 2 months after operation. This has not held up with longer follow-up, however, and thromboembolic events have occurred at a relatively constant rate of.8%/patient-year. Freedom from thrombo- Table 2. Late Deaths Aortic Valves Mitral Valves Cause of Death (n = 53) (n = 4) Valve related Unknown sudden Thrombosed valve Cerebral embolism Cerebral bleed Cardiac tamponade Myocardial infarct Aortic dissection Endocarditis Repeat heart catheterization-perivalvular leak Not valve related Congestive heart failure Cancer Arrhythmia Seizure/alcohol Hepatitis Automobile accident Renal failure

4 Ann Thorac Surg 993: KRATZETAL 465 % Free from Thromboembolic Events 8 X Free from Endocarditis Mitral valves Mitral Valves % \ p<.5 Aortic Valves 93.5 *.7% Linearized rates: Aortic =.S%/patient-year Mitral = 2.9%/patient-year % Aortic valves 8 6 Linearized rates: Aortic =.I%/patient-year Mitral =.4%/patient-year Oy~l OStOP YearsPostOp #Patients #Patients Fig 6. Freedom from thromboembolic event for aortic and mitral St. Iude valve patients. Fig 8. Freedom from prosthetic valve endocarditis for aortic and mitral valve patients. embolism at years is 67.? 3% (Fig 6). Overall, 53 of 6 (95%) surviving AVR patients remain on long-term warfarin anticoagulation. Six of the 2 patients in whom thromboembolic events occurred were not anticoagulated at the time. Anticoagulant-related complications occurred in 39 patients (3.5%/patient-year) and five (4 cerebral hemorrhages, cardiac tamponade) were fatal (Fig 7). Prosthetic valve endocarditis involving 6 St. Jude valves developed in 2 AVR patients (.4%/patient-year). Native valve endocarditis was the reason for the initial valve replacement in 3 of these patients. Of these 3,2 died with recurrent endocarditis ( at reoperation and the other awaiting reoperation). The third underwent successful re-replacement with a homograft (Fig 8). CONSEQUENCES OF MORBID EVENTS. Twelve of the 254 AVR patients underwent 5 reoperations for replacement of their previously implanted St. Jude valve. This resulted in an incidence of reoperation of.4%/patient-year. Freedom from reoperation at years was 92.% * 2.% (Fig 9). Indications for re-replacement were endocarditis (2), paravalvular leak (2), and valve thrombosis (). Of the 28 patients considered to have sustained valverelated deaths, died suddenly and without known cause after having recovered satisfactorily from AVR (see Table 2). Permanent valve-related impairment occurred in 7 of the AVR patients who experienced thromboembolic events. The impairments experienced are peripheral weakness and numbness (2), residual paralysis (2), speech deficits (2), and partial blindness (). Permanent impairment has occurred as the result of anticoagulantrelated hemorrhage in 2 patients. One experienced blindness from a retinal hemorrhage, and the other has residual paralysis from a cerebral bleed (Fig ). OVERALL FREEDOM FROM COMPLICATION. Figure demonstrates the overall freedom from all complications. Excluding deaths from non-valve-related causes, the freedom from all complications at years was 35.% 2 7.9%. However, if non-valve-related deaths are included, only 27.6% % were free of death (all causes) or valverelated complication. FUNCTIONAL IMPROVEMENT. Of the 38 long-term survivors at 8 years, 86% were in New York Heart Association class I or I (see Fig 2). New York Heart Association functional class improved from a mean of 3. preoperatively to.3 (p <.) at year, and this improvement was largely maintained at.8 at 8 years. % Anticoagulant Complication Free 8 6 Mitral Valves 8.O* 5.% a * 3.- p<.5 73.*4.5% Aortic Valves % Free from Reoperation Mitral valves 94. k i2.96 Aortic valves p = ns 4 2 Linearized rates: Aortic = 3.5 %/patient-year Mitral = Z.Z%/patient-year Yearshstop i I #Patients 4 2 Linearized rates: Aortic =.4%/patient-year Mitral = Ol%/patient-year Years Fbstop 4% #Patients Fig 7. Freedom from anticoagulant-related complications for aortic and mitral valve patients. Fig 9. Freedom from reoperation for valve-related complications for aortic and mitral St. Jude valve patients. (ns = not significant.)

5 466 KRATZETAL Ann Thorac Surg 993; % Impairment Free % Event Free 8 i 2 4 Linearized rates: Aortic = OS%/patient-year Mitral =.Z%/patient-year p = ns Living, Complication Free 5.3 f 5.6%. 39. f 5.3%*' ' YearsPostop #Patients Fig. Freedom from valve-related permanent impairment for aortic and mitral St. Jude valve patients. (ns = not significant.) Group (Mitral Valve Replacemenf) HOSPITAL MORTALITY. Seven patients died in the postoperative period, for an operative mortality rate of 3.5%. LATE MORTALITY. There have been 4 late deaths, and the actuarial survival rate at years is % (Fig 5B). Twenty-five of the late deaths were valve related. VALVE-RELATED COMPLICATIONS. No case of structural deterioration occurred. In no case did implantation of the valve result in entrapment of the leaflets. Clinically significant hemolysis developed in 2 patients postoperatively. Twenty-eight of the 22 mitral valve patients have sustained thromboembolic (27 embolic, valve thrombosis) complications (2.9%/patient-year). Freedom from thromboembolism was 76.9% * 4.9% at years (see Fig 6). All five of the fatal thromboembolic events were strokes. In three instances, warfarin administration had been stopped before the event. The two remaining patients were reported to be receiving warfarin, but l had a normal prothrombin time/partial thromboplastin time on admission to the emergency room. Anticoagulant-related complications occurred in 2 patients (2.2%/patient-year), and three were fatal. % Event Flee f 7.9%* 27.6 f 6.5%** 2 Living, Complication Free Prosthetic endocarditis developed in 4 patients (.4%/ patient-year), and died. Endocarditis did not develop in any patient after the fourth year. Freedom from endocarditis at 4 years was 97.6% Ifr.2% and persisted through the following years (see Fig 8). CONSEQUENCES OF MORBID EVENTS. Seven of 254 patients underwent reoperation for replacement of their valve (.7%/patient-year). Freedom from reoperation was 94.% * 2.4% at years (see Fig 9). Reoperation was performed for endocarditis (3), paravalvular leak (l), hemolysis (2), and valve thrombosis (). Of the 25 patients with valve-related deaths, 9 died suddenly with a previously properly functioning valve. Other causes of death are listed in Table 2. Permanent valve-related impairment has occurred in 2 MVR patients as the result of thromboembolic events (see Fig ). OVERALL FREEDOM FROM COMPLICATION. Freedom from all types of complications was 39.% * 5.3% at years. If non-valve-related deaths are excluded, freedom from complications was 5.3% * 5.6% at years (Fig 2). FUNCTIONAL IMPROVEMENT. Of the 33 MVR patients surviving 8 years, 78% were in either New York Heart Association class I or I (see Fig 3). New York Heart Association functional class improved from a mean of 3.4 preoperatively to a mean of.5 ( p <.) at year and a mean of.7 at 5 years. Comment We began using the St. Jude valve in 979 while it was still an investigational device because of our concern about the durability of porcine valves and because of structural complications that were beginning to be reported with other mechanical valves. We also thought that its design offered the possibility of improved hemodynamic performance and perhaps a decrease in the rate of thromboembolic complications. In 984 we reported our early experience with the valve. At that time, we thought it demonstrated excellent durability (no structural failures),

6 Ann Thorac Surg 993;56:462-8 KRATZETAL 467 excellent hemodynamics, and a low incidence of thromboembolism. We are now able to examine our -year experience with the valve and compare it with the two other series with similar follow-up that have been reported by Arom and associates [9] and Czer and colleagues [. Operative mortality may be more strongly influenced by preoperative patient characteristics than by valve performance. Our overall operative mortality (AVR, 3.9%; MVR, 3.5%) is not significantly different from that in our earlier report and compares favorably with that reported by Arom and associates (AVR, 3.2%; MVR, 9.9%) and Czer and colleagues (AVR, 5.2%; MVR,.9%). It also compares favorably with other contemporary series using porcine valves (AVR, 6.2%; MVR, 9.2%) [ll, 2 and the other most commonly used mechanical prosthesis (Medtronic) (AVR, 6.%; MVR, 7.2%) [3, 4. Long-term survival is related to both underlying cardiac function and age as well as to valve Performance. Tenyear survival of 47% (AVR) and 63% (MVR) again is similar to that reported by Czer and colleagues (AVR, 4%; MVR, 69%) and Arom and associates (AVR, 75%; MVR, 57%). Survival at 7% years with the Medtronic valve has been reported to be 69% (AVR) and 8% (MVR) [3]. Longer term follow-up extending to 5 years is now available with porcine heterografts, but at years survival has been reported to be 64% (AVR) and 64% (MVR) [2]. Although occasional reports of structural failure of the St. Jude valve have appeared, these incidents have been extremely rare. To our knowledge, none of our patients has sustained this complication. This lack of structural failure continues to represent a significant advantage of the St. Jude valve over porcine and pericardial xenografts as well as earlier mechanical prostheses. Thromboembolic complications remain the Achilles heel of all mechanical valves. The guidelines for reporting this complication include any neurologic complication not specifically explained. In a population with a mean age of 5 years, undoubtedly many of these neurologic events are not valve related. However, if reporting guidelines are consistently used, a much more meaningful comparison between groups can be made. Although the rate of thromboembolic complications in this series (AVR,.8%/ patient-year; MVR, 2.9%/patient-year) has increased slightly from our earlier report, it still compares favorably with the other series of St. Jude valves (Arom and associates [9]: AVR,.%; MVR,.8%; Czer and colleagues [lo]: AVR, 2.5%; MVR,.6%). These thromboembolic rates also compare favorably with those reported by Daenen and co-workers [5] for AVR with ball valve prostheses (2.8%), tilting-disc prostheses (2.5%), and porcine prostheses (3.8%) as well as for MVR with ball valve prostheses (3.2%), tilting-disc prostheses (3.5%) and porcine prostheses (3.8%). The low incidence of valve thrombosis (two events occurring in patients in whom warfarin administration had been discontinued) is also reassuringly low but does reflect the continuing need for longterm anticoagulation. The incidence of anticoagulant-related complications has decreased slightly to 3.5% (AVR) and 2.2% (MVR) from our earlier report (AVR, 6.9%; MVR, 2.4%). AS we gained experience with the St. Jude valve and became favorably impressed with its low thromboembolic rate, we as others [6] have gradually decreased the level of anticoagulation that we thought to be acceptable from approximately twice normal control to the current approximately.5 times normal. It is likely that the decreased rate of anticoagulant-related complications as well as the slight increase in thromboembolic complications from our earlier report reflects this change. The incidence of prosthetic valve endocarditis is similar to that of Czer and colleagues [lo] and Arom and associates [9], and is quite low. Of interest is the total freedom from acquiring endocarditis after the fourth year for both aortic and mitral valves. Presumably the cloth sewing ring has undergone complete tissue ingrowth by this time and is protected to some extent from bacterial invasion. However, our data confirm those of others indicating that once prosthetic valve endocarditis is acquired, the prognosis is quite poor, even with reoperation [7]. An examination of Figures and 3 is quite revealing. Freedom from all valve-related complications at years was 35.% for aortic valves and 5.3% for mitral valves. Despite the overall excellent performance of the St. Jude valve in each category, this low incidence of freedom from all complications at years serves as a reminder that patients who undergo prosthetic valve replacement do not have a normal outlook but instead have undergone substitution of one disease (prosthetic valve) for the original disease (native valve). Accordingly, repair of the mitral valve should be performed when possible, and continued efforts should be made to develop more reliable methods of aortic valve repair. Gradients obtained by Doppler echocardiography may not accurately reflect true gradients across the St. Jude valve [MI. However, the data that are available indicate hemodynamic performance of the St. Jude valve that is at least as good as that of any other prosthetic valve [9-22. We have no objective hemodynamic data available from this series. However, a large proportion (5/254) of our AVR patients received small (9 or 2 mm) prostheses and the significant, sustained, functional improvement noted in them as well as in the MVR group attests to the overall excellent hemodynamic performance of the St. Jude valve. In summary, this series of patients undergoing isolated AVR and MVR with the St. Jude prosthesis now with follow-up extending to years confirms the excellent performance of this valve as documented in our earlier report as well as that reported by other institutions. Accordingly, it remains our prosthesis of choice when valve replacement with a mechanical prosthesis is indicated. References. Crawford FA Jr, Kratz JM, Sade RM, Stroud MR, Bartles DM. Aortic and mitral valve replacement with the St. Jude Medical prosthesis. Ann Surg 984;99: Sade RM, Crawford FA Jr, Fyfe DA, Stroud MR. Valve prosthesis in children: reassessment of anticoagulation. J Thorac Cardiovasc Surg 988;95:553-6.

7 468 KRATZETAL Ann Thorac Surg 993;56: Edmunds LH Jr, Clark RE, Cohn LH, Miller DC, Weisel RD. Guidelines for reporting morbidity and mortality after cardiac valvular operations. Ann Thorac Surg 988;46: Hettmansperger TP. Statistical inference based on ranks. New York: Wiley, Grunkemeier GL, Lambert LE, Bonchek LI, Starr A. An improved statistical method for assessing the results of operation. Ann Thorac Surg 975;2: Bodnar E, Wain WH, Haberman S. Assessment and comparison of the performance of cardiac valves. Ann Thorac Surg 982;34: Kaplan EL, Meier P. Non-parametric estimation from incomplete observations. J Am Stat Assoc 958;53: Mantel N. Evaluation of survival data and two new rank order statistics arising in its consideration. Cancer Chemother Rep 966;5: Arom KV, Nicoloff DM, Kersten TE, Northrup WF, Lindsay WG, Emery RW. Ten years experience with the St. Jude Medical valve prosthesis. Ann Thorac Surg 989; Czer LSC, Chaux A, Matloff JM, et. al. Ten-year experience with the St. Jude Medical valve for primary valve replacement. J Thorac Cardiovasc Surg 99;: Louagie Y, Noirhomme P, Aranguis E, Schoevaerdts JR, Chalant CH. Use of the Carpentier-Edwards porcine bioprosthesis: assessment of a patient selection policy. J Thorac Cardiovasc Surg 992;4: Jones EL, Weintraub WS, Craver JM, et al. Ten-year experience with the porcine bioprosthetic valve: interrelationship of valve survival and patient survival in,5 valve replacement. Ann Thorac Surg 99;49:37(? Beaudet RL, Poirier NL, Doyle D, Nakhle G, Gauvin C. The Medtronic-Hall cardiac valve: 7% years clinical experience. Ann Thorac Surg 986;42:64&5. 4. Vallejo JL, Gonzalez-Santos JM, Albertos J, et al. Eight years experience with the Medtronic-Hall valve prosthesis. Ann Thorac Surg 99;5: Daenen W, Nevelsteen A, van Cauwelaert P, de Maesschalk E, Willems J, Stalpaert G. Nine years experience with the Bjork-Shiley prosthetic valve: early and late results of 932 valve replacements. Ann Thorac Surg 983;35: DiSesa WJ, Collins JJ Jr, Cohn LH. Hematological complications with the St. Jude valve and reduced-dose Coumadin. Ann Thorac Surg 989;48: Cowgill LD, Addonizio VP, Hopeman AR, Harken AH. A practical approach to prosthetic valve endocarditis. Ann Thorac Surg 987;43: Chambers J, Deverall P. Limitations and pitfalls in the assessment of prosthetic valves with Doppler ultrasonography. J Thorac Cardiovasc Surg 992;4: Horstkotte D, Haerten K, Seipel L, et al. Central hemodynamics at rest and during exercise after mitral valve replacement with different prostheses. Circulation 983;68(Suppl 2): Gray RJ, Chaux A, Matloff JM, et al. Bileaflet, tilting disc and porcine aortic valve substitutes: in vivo hydrodynamic characteristics. J Am Coll Cardiol 984;3: Wortham DC, Tri TB, Bowen TE. Hemodynamic evaluation of the St. Jude Medical valve prosthesis in the small aortic anulus. J Thorac Cardiovasc Surg 982;8: Rashtiart MY, Stevenson DM, Allen DT, et al. Flow characteristics of four commonly used mechanical heart valves. Am J Cardiol 986;58: DISCUSSION DR KIT V. AROM (Minneapolis, MN): I congratulate Dr Kratz and associates, and concur with them about good long-term results with the St. Jude Medical prosthesis. We at the Minneapolis Heart Institute are now working on a 5-year follow-up, and the results should become available soon. As Dr Kratz mentioned, at the end of years, the incidence of freedom from complications is near 9%. Our data support his findings in that at years, the incidences of freedom from thromboembolism, anticoagulant-related hemorrhage, and valve thrombosis were 88% * 4%, 94% t 6%, and 99%, respectively. The freedom from reoperation and freedom from prosthetic endocarditis were also high (99% and 99%). Also at years, the survival rate in the aortic position is about 75% if early death is included, or about 78% when early death is excluded. As far as the mitral valve is concerned, the survival rate was about 58% and 64%, respectively. I noticed in your report, Dr Kratz, that at years the mitral valve group has a better survival rate than the aortic valve group, which differs from our study. Do you think the size of the prosthesis in the aortic position makes any difference in longterm survival? DR KRATZ: Doctor Arom s group and Dr Czer s group are two groups that have recently published -year data on this valve also, and we did look at our data compared with those; in general, I think all three reports are similar. As Dr Arom has shown, our survival rates are a bit lower than theirs. I think that there are a lot of things that go into this survival number. Even though we call it valve-related survival, a great deal of it is going to relate to the age of the patients who go into the group and the preoperative cardiac status of the patients as they go in. Our aortic group was somewhat older, 4 years older, than our mitral group, and I think that contributed some to the difference between those two in mortality. I think it is going to be very difficult to compare long-term survivals, meaningfully, with different groups from different parts of the country because of inherently different types of entry groups.