STENTED HETEROGRAFTS (PORCINE)

Comparison of Low-Pressure Versus Standard- Pressure Fixation Carpentier-Edwards Bioprosthesis Javier Fernandez, MD, Chao Chen, PhD, Jiang Gu, MD, PhD, Otto B. Brdlik, DO, Glenn W. Laub, MD, Maria M. Murphy, MS, Mark S. Adkins, MD, William A. Anderson, MD, and Lynn B. McGrath, MD Division of Cardiothoracic Surgery, Department of Surgery, Deborah Heart and Lung Center, Browns Mills, and University of Medicine and Dentistry of New Jersey, Robert Wood Johnson Medical School, New Brunswick, New Jersey.

Intermediate-phase clinical results of 51 low-pressure (LP) and 234 standard-pressure (SPl fixation porcine Carpentier-Edwards (CE) valves implanted between 1977 and 1991 were compared for valve-related events. Group similarities included New York Heart Association func- tional class, ejection fraction, and sex. Patients with SP valves were younger (mean age, 58 versus 68 years; p = 0.0001). There were 20 in-hospital deaths (8.6%) in the SF’ valve group and 5 (9.8%) in the LP valve group (p = 0.791. Follow-up was 99%, with a mean of 104 months in the SP valve group versus 55 months in the SP valve group tp = 0.0001). The actuarial survival rate was 48.2% and 22.3% at 10 and 15 years, respectively, in the SP valve group and 34.1% at 10 years in the LP valve group (p = 0.421. Freedom from events at 5,10, and 15 years in the SP valve group and at 5 years in the LP valve group was as follows:

T he first-generation Carpentier-Edwards (CE) (Baxter Edwards Division, Irvine, CA) porcine aortic valve

was introduced for clinical use in 1975 and was glutaral- dehyde fixed at a “high” or standard hydrostatic pres- sure (SF’) of 100 mm Hg. Because of concern about accelerated degeneration of these original valves, a sec- ond-generation CE porcine valve was introduced in 1981 that was fixed in the same solution at a “low” hydrostatic pressure (LP) of 2 mm Hg in an attempt to maintain leaflet geometry and a normal collagen waveform [l-4]. In the United States only the LP valve for mitral position was released for distribution.

We conducted an intermediate-phase investigation to compare the clinical performance of LP and SP Cl? porcine valves in patients undergoing isolated mitral valve replacement at our institution.

Patients and Methods

Between 1977 and 1991, 285 patients underwent isolated mitral valve replacement using the CE porcine valve: 234 received the SP bioprosthesis from 1977 to 1991, and 51

Presented at the VI International Symposium for Cardiac Bioprostheses, Vancouver, BC, Canada, July 29-31, 1994.

Address reprint requests to Dr Fernandrr, Deborah Heart and Lung Center, 200 Trenton Rd, Browns Mills, NI 08015.

for late valve-related events, 86.3%, 51.4%, and 20.2%, respectively, in the SP valve group versus 85% in the LP valve group (p = 0.44); for valve-related death, 96.4%, 93.6%, and 87.3% in the SP valve group versus 100% in the LP valve group (p = 0.20); for structural valve failure, 96%, 68%, and 35% in the SP valve group versus 100% in the LP valve group (p = 0.09); and for reoperation, 95%, 6170, and 30% in the SP valve group versus 92% in the LP valve group (p = 0.82). In conclusion, this study revealed no significant statistical difference between LP and SP valves. In the LP valve group, structural valve failure/ valve-related death was not observed, perhaps indicating a more favorable result. Absolute verification of this trend awaits long-term follow-up.

(Ann Tkorac Surg 1995;6O:S205-20)

received the LP bioprosthesis from 1981 to 1991. Fol- low-up was 99% complete at a mean of 97 months and a total of 2,020 patient-years. The follow-up period was longer in the SP valve group (mean, 8.7 years) than in the LP valve group (mean, 4.7 years, p = 0.0001). Mean age was 57.9 years in the SP valve group and 67.7 years in the LP valve group, with no patient younger than 30 years (p = 0.0001). Most of the remaining clinical and operative characteristics were not significantly different (Table 1). Both groups were compared for in-hospital deaths, late deaths, actuarial survival, freedom from valve-related events, structural valve dysfunction (SVD), reoperation (REOP), valve-related deaths, reoperative deaths, and late functional results. Society for Thoracic Surgeons guidelines for classification of these events were followed [5]. Events of the same category were counted once. The log-rank test was used for comparisons of actuarial freedom from events. Because of the shorter follow-up time for the LP valve group, in certain late comparisons beyond a defined time interval, extrapolations were per- formed and confidence limits determined according to the procedure of Kirklin and Blackstone [6] and Green- wood’s formula 171. Event rates were also expressed as linearized occurrence rates. Continuous variables were compared using Wilcoxon’s test and discrete variables by the Fisher’s exact test. Valve replacement was performed as previously described [8, 91.

0 1995 by The Society of Thoracic Surgeons 0003-4975/95/$9.50 0003-4975(95)00201-U

S206 CARDIAC BIOPROSTHESES FERNANDEZ ET AL LOW- VERSUS STANDARD-PRESSURE FIXATION

Table 2. Clinical Profile”

Clinical Characteristic

Age (y) <30 30-65 >65

Male Female Preop NYHA functional

classes II and III Valve lesion

Regurgitation Stenosis Mixed

Etiology Rheumatic Myxomatous Ischemic

Infective endocarditis Congenital

Previous cardiac operation

Ejection fraction (n = 209)

Cardiac index (n = 237)

Concomitant CABG Valve replacement

Cardiopulmonary bypass (min)

Aortic cross clamp (min)

Standard High-Pressure Low-Pressure

Valve Valve ’ (n = 234) (n = 51) Value

57.9

12 (5) 156 (67)

66 (28) 87 (37)

147 (63) 221(94)

90 (38) 37 (16)

107 (46)

178 (76) 34 (15)

13 (6) 7 (3) 6 (3)

61 (26)

0.625

2.6 + 0.88

47 (20) 33 (14)

95

50

67.7

0 (0) 14 (27) 37 (72) 16 (31) 35 (68)

44 (86)

27 (53)

4 (8) 20 (39)

34 (67) 15 (29)

4 (8) 1 (2) 0 (0)

10 (20)

0.630

2.2 +- 0.65

13 (26) 8 (16)

98

57

0.0001"

0.521h 0.16”

0.112”

0.214h 0.014h 0.51Rh l.OOOh 0.595h 0.09gh

0.963h

0.0015”

0.448h 0.826h 0.1428”

0.0322a

a Data presented are mean value, mean value 2 standard deviation, or number (%) of patients. “2. ’ Fischer exact test, two-tailed.

CABC = coronary artery bypass graft; NYHA = New York Heart Association; Preop = preoperative.

Results

There were 20 in-hospital deaths (8.6%) in the SP valve group and 5 (9.8%) in the LP valve group (p = 0.79), but none were valve related. During the follow-up period, there were a total of 208 late postoperative events, of which 112 (5.6%/patient-year) were considered valve related. Of these, 105 events occurred in 100 patients (5.7%/patient-year) in the SP valve group, and 7 occurred in 6 patients (3.4%lpatient-year) in the LP valve group (some patients had more than one different event). The linearized occurrence rates of these events are shown in Table 2. Overall actuarial freedom from late valve-related events was not significantly different between the two groups (Fig 1). Freedom from each event is shown in Table 3.

Structural valve dysfunction due to primary tissue failure from degeneration or calcification, or both, was detected in the SP valve group only. Overall freedom

Ann Thorac Surg 1995;60:s205-10

Table 2. Linearized Data for Late Events

Standard High-Pressure Low-Pressure

Valve Valve

Event No. %/pt-y No. %Ipt-y

Valve related TE

Major (5/O)” 16 0.87 2 0.97 Minor 6 0.33 1 0.48

ACRH Major (3/O) 4 0.21 1 0.48

Infective endocarditis 10 0.54 1 0.48 (3/O)”

Paravalvular leak 5 0.27 2 0.97 SVD (410)” 64 3.49 0 0.0

Non-valve related Heart failure 58 3.16 9 4.35 Myocardial infarction 5 0.27 0 0.0

Arrhythmia 63 0.63 7 3.38 Noncardiac 1 0.05 3 1.45 Total 182 9.9 26 12.56

a Number of nonreoperative valve-related deaths (standard high- pressure valve group/low-pressure valve group).

ACRH = anticoagulant-related hemorrhage; %/pt-y = percent per patient-year; SVD = structural valve dysfunction; TE = thrombo- emboli.

from this event is shown in Figure 2. Of the 64 SVDs in the SP valve group, only 7 occurred during the first 5 years of follow-up. By age group (Fig 3), the actuarial freedom from SVD between the SP and LP valve groups was significant only in the 30- to 65-year age group at 10 years (p < 0.01, Greenwood’s approximation). In the age group greater than 65 years, this difference was not significant.

100 -y..

20 i ;_ 20.2%

10 1211 190 17s 165 152 140 126 112 85 68 60 42 I 4s 37 35

34 18 3 36 34 15 3 2 2 2 2 2

01, ,I,, , , , , , ( 0 12 24 36 46 60 72 64 96 108 120 132 144 156 166 160

p = 0.440 MONTHS

-rHIGH (n = 100) *LOW (n = 6)

Fig 1. Actuarial freedom from valve-related events, comparing stan- dard high- and low-pressure fixation mitral Carpet&r-Edwards valves. Dashed line indicates patients at risk with no events affer the 53rd month within the low-pressure valve group. Number of patients at risk remaining is indicated at the bottom of the graph. The difference between the two groups was not signi$cant; p value is shown. Vertical bars indicate 70% confidence limits. (n = num- ber of patients in each group.)

Ann Thorac Surg CARDIAC BIOI’ROSTHESES FERNANDEZ ET AL S207 1995;60:5205-10 LOW- VERSUS STANDARD-PRESSURE FIXATION

Table 3. Freedom From Late Valve-Related Events

Event”

Standard High-Pressure Valve (%)

5Y 10 Y (mean t SD) (mean + SD)

Low-Pressure Valve (%)

5Y 10 Y (mean ? SD) (mean t SD)

p Value (Log-rank)

Thromboembolic Major (16/2) Minor (6/l)

Anticoagulant-related bleeding, major only (4/l)

Endocarditis (1011) Paravalvular leak (5/2) Structural valve dysfunction (64/O)

95.0 -t 1.7 91.0 k 2.5 94.8 t 3.7 94.8 ir 3.7 0.9803

98.4 -t 0.9 95.8 _f 1.7 97.7 i- 1.3 97.7 -c 1.3 0.7656

97.8 -+ 1.2 97.8 _f 1.2 97.4 i 2.6 97.4 k 2.6 0.7466

97.4 t 1.2 93.5 i 2.1 97.8 5 2.2 97.8 2 2.2 0.8943

97.8 -t 1.1 97.0 t 1.4 94.5 2 3.9 94.5 + 3.9 0.3271

96.1 -t 1.5 68.3 _f 4.1 0.0 0.0 0.0937

a Numbers in parentheses indicate number of events (high-pressure group/low-pressure group).

SD = standard deviation.

Survival and Late Deaths There were 161 late deaths: 146 in the SP valve group (7.9%/patient-year) and 15 in the LP valve group (7.2%1 patient-year). Of the 146 late deaths in the SF’ valve group, 15 (O.S%/patient-year) were valve related (10.2%), medically treated only (nonreoperative) and occurred out of the hospital (Table 2). Of the remaining late deaths in the SP valve group, 82 (4.5%Ipatient-year) were second- ary to cardiac-related events, and 49 (2.6%/patient-year) were due to non-cardiac-related events, mostly pulmo- nary complications. None of the 15 late LP valve group deaths were valve related: 8 (3.9%lpatient-year) were due to cardiac causes, and 7 (3.4%/patient-year) resulted from noncardiac causes. Overall actuarial patient survival for the two valve groups (Fig 4) and freedom from valve- related deaths were not significantly different (Fig 5).

Reoperation Seventy-one patients (3B%/patient-year) in the SP valve group and 3 (l.$%/patient-year) in the LP valve group underwent REOP for complications of their CE biopros- thesis. The indications and causes of reoperative deaths and the linearized occurrence rates for each are shown in

30 35.3% 1

20

10 4s 39 39 38 37 16 3 2 2 2 2 2

0, 211193 182 177 16.5 (53 137 122 ‘05 80 68 49 39 22 5 1

, { , , , , 7, , , , , I 0 12 24 36 46 60 72 84 96 106 120 132 144 156 166 160

p = O.OQ37 MONTHS

LCh’d (n = 0) *HIGH (n = 64,

Fig 2. Overall freedom from structural vulve dysfunction, standard high- versus low-pressure valve group for all age groups. Same for- mat as in Figure 1.

Table 4. Overall actuarial freedom from REOP (Fig 6) and freedom from REOP stratified according to age group (Fig 7) were not significant. No LP valve group patients were available for comparison in the age group greater than 30 years. There were 13 deaths (18.3%) between 1 and 97 days after REOP, all occurring in the SF’ valve group. Freedom from reoperative death between the SF’ and LP valve groups was significant at the lo-year inter- val only (Fig 8). Similarly, freedom from all valve-related deaths, including reoperative deaths, shown in Figure 9, was also significant at the lo-year interval, with no events in the LP valve group.

The clinical status of the 95 long-term postoperative survivors compared favorably with their preoperative New York Heart Association functional class, as can be seen in Figure 10. The two groups did not show signifi- cant differences in New York Heart Association func- tional class either preoperatively or postoperatively (p = NS).

0 12 24 36 46 60 72 64 96 108 120 132 144 156 166 160

MONTHS

LWV (ALL AGES) *HIG” z 66 *HIGH 30-65 *“IO,, -Z 30

Fig 3. Freedom from structural valve dysfunction stratijed by age group. The diference was significant only in the 30- to 65-year age group (p i 0.01) at 20 years. Dashed line represents the low-pres- sure valve group. There were no patients in the low-pressure valve group less than 30 years of age. Solid lines represenf standard high-pressure group. Age group is indicated by the graph symbols. Graph pattern similar to that of Figure 1.

s208 CARDIAC BIOPROSTHESES FERNANDEZ ET AL Ann Thorac Surg LOW- VERSUS STANDARD-PRESSURE FIXATION 1995;6D:S205-10

Tub/e 4. Late Reouerations

l- z 60 1

Y 50

s a 40

Comment

Since the 197Os, despite the low incidence of thrombo- emboli and anticoagulant-related bleeding, a high rate of primary tissue failure has been reported, with increas- ingly high rates of REOP and mortality, for first- generation CE valves in the mitral position [lo-131. Younger age was identified as the most important pre- dictor of accelerated valve deterioration [12-151. The pathogenesis of tissue valve failure was linked to the process of high-pressure fixation, causing loss of collagen waveform [l-4] and leading to the characteristic degen- erative changes predominately at the cuspal sites seen in explanted valves [4, 16-191. Glutaraldehyde was seen also seen as the cause of valve tissue stiffness [20]. Because of concern with accelerated valve dysfunction, a second-generation CE porcine valve was introduced in 1981 that was fixed in the same solution at a “low” pressure of 2 mm Hg. For the same reasons, a third- generation of porcine bioprosthesis was introduced (Medtronic Blood Systems, Inc., Minneapolis, MN) that was fixed at zero pressure and claimed an even better preservation of the collagen crimp and possible in- creased durability [21].

20 1

10 i 45 39 39 37 39 16 3 2

0 1211 195 183 177 168

2 2 2 2 156 146 136 130 108 96

4 48 23 5

I .-I1 r--17 77 q-7 0 12 24 36 46 60 72 64 96 106 120 132 144 156 166 160

p = 0.2037 MONTHS

LOW (n = 0) -HIGH (n = 15)

Fig 5. Actuarial freedom ,from valve-related dcrzths, standard k&h versus low-pressure v~zlve group, euiludin~ rcqwnztive deflths.

Standard High-Pressure Valve Low-PreSSure

Valve

Cause

Reoperations Deaths (Reoperations)

NO. %/pt-y No. wpt-y No. wpt-y

Structural valve failure

Endocarditis Paravalvular

leak Unknown

Total

58 3.1 10 0.54 0 0.0

1 0.01 1 0.05 1 0.48

3 0.16 2 0.10 2 0.97

9 0.49 0 0.0 0 0.0

71 3.8 13 0.07 3 1.45

a No deaths in the low-pressure valve group.

“,x/pt-y percent per patient-year.

There is a scarcity of clinical data regarding the rate of SVD with LP CE porcine valves in the mitral position. Vermeulen and co-workers [22] reported a 98.5% free- dom from SVD at 3 years; Jamieson and colleagues [23] reported a 98.8% t 9.7% freedom at 5 years; and David and co-workers (241, using another LP valve, the Hancock II, reported a 93 % ? 4% freedom at 8 years. These results were very close to the 100% freedom from SVD with the LP valve in the present series. Barratt-Boyes and col- leagues [21] reported a 99% + 0.92% freedom from SVD at 6 years using a zero-pressure-fixed porcine valve (the Intact valve), similar to our results during the first 5 years of follow-up. Hilbert and co-workers [20] indicated that the collagen crimp was well preserved with either zero or LP fixation in most regions of the valve cusps.

In the present series, we used the SP valve for com- parison. No significant difference was found regarding actuarial survival and freedom from valve-related events and valve-related deaths between the SF and LP valve groups during the first 5 years of follow-up. Freedom from SVD was 100% in the LP valve group within the period of observation compared with 96.1% i 1.8% at 5 years in the SF’ valve group. Although the trend was

= 60 -

8 50 -

$ 40 -

30 -

20 -

10 - 211 195 182 174 163 15, 135 121 102 76 64 47 37 I9 3 o +~%a! 38 38 35 15 3 2 2 2 2 2

I I I1 I I I I I I I, I7 0 12 24 36 46 60 72 64 96 108 120 132 144 156 166 160

p = 0.8254 MONTHS

-HIGH (n = 71) *LOW (n = 3)

Fig 6. Owrnll actuarinl freedom from reoperation for the Carpm- tiwEdwnrds valve, standard high- versus lowpressure valve group.

Ann Thorac Surg CARDIAC BIOI’ROSTHESES FERNANDEZ ET AL s209

1995:6O:S205-10 LOW- VERSUS STANDARD-PRESSURE FIXATION

*HIGH < 30 *HIGH 30-m *Low 30.65 CHIGH > 65 +Lw > 65

Fig 7. Actuarial freedom from reoperation for the Cqvcntier-Ed- wards valve, standard high- versus low-pressure valve group by IIXL group. There was no significant difference between ago groups ip .a 0.05). Graph pattern similar to that of Figure 2.

favorable for the LP valve, this difference was not statis- tically significant. Freedom from SVD with the SF valve at 15 years was 35% + 8%, similar to that reported by Jamieson and co-workers [13], 41% ? 7X, at a similar interval.

By age group, at 5 years after operation there was a significant difference between the SP and LP valve groups only for patients in the 30- to 65-year age group. These results are identical to those reported recently by Jamieson and co-workers 1251, who compared a large group of patients with SP valves with those with LP valves.

The rate of REOP in patients with the CE valve fol- lowed a similar pattern. Most of REOPs were undertaken for primary tissue failure in the SP valve group, but no patient underwent REOP for this reason in the LP valve group. Other investigators 121, 23, 241 reported a similar rate of REOP at 5 years in patients with the LP valve, comparable to the 92% + 4.5”0 freedom from REOP in the present series. The clinical results of the two bioprosthe-

MONTHS

LOW (n = 0) *HIGH (n = 13)

Fig 8. Actuarial freedom from rcoperativc death, standard high- ver- sus low-pressure valve group Cp “ 0.01, Greenwood’s nppro.Cmil- tion).

70

ki 60 rl

8 501

6 a 40

30 j

20 -.

10

$

45 39 39 39 37 16 3 2 2 2 2 2

0 21,195 183 177 168 156 146 136 130 108 96 78 69 48

IT- I I I I I I 1-T I , I I I I 0 12 24 36 46 60 72 84 96 106 120 132 144 156 166 160

MONTHS

LOW (n = 0) +HIGH (n = 26)

Fig 9. Artuarial frwdom from valve-related and reoperatizle death combined, standard high- versus low-pressure valve group (p i 0.01, Greenwood’s approximation).

ses were largely similar. Most of the survivors in the two valve groups are active and doing well with regard to New York Heart Association functional class.

In conclusion, the LP valves showed a favorable trend in terms of durability and rate of REOP. Most of the other valve-related events were not significantly different be- tween the two groups. A longer follow-up period is necessary to ascertain whether this course can be sus- tained.

This study was partially funded by a grant from the Baxter- Edwards Division, Irvine, CA. We appreciate and acknowledge the extensive data collection and work performed by Bridget Bailey and computer entry programmer Ray Ellis and the many manuscript revisions by Marcia Graybeal. Special thanks to Dr Douglas H. Joyce for reviewing the manuscript.

100

90

80

e 70

= 60 ;1 $ 50

8 40

30

20

10

0

-HIGH PRESSURE

3LOW PRESSURE

P ‘1

JL P

p1 r I II III IV I II III IV

PREOPERATIVE POSTOPERATIVE

Fig 10. Preoperative nnd postoperative New York Heart Association functional class of patients with standard high- and low-pressure z~lvcs. There is no significant diference between patients in each functiomrl class (high- versus low-pressure valve group): preopera- timely, p = O.lS8; postoperatively, p = 0.645 (2 test). Improvement is seen equally in both groups.

s210 CARDIAC BIOPROSTHESES FERNANDEZ ET AL LOW- VERSUS STANDARD-PRESSURE FIXATION

Ann Thorac Surg 1995;6O:S205-10

References

1. Broom ND. The stress/strain and fatigue behavior of glutar- aldehyde preserved heart-valve tissue. J Biomech 1977;lO: 707-24.

2. Broom ND. Fatigue-induced damage in glutaraldehyde- preserved heart valve tissue. J Thorac Cardiovasc Surg 1978;76:202-11.

3. Broom ND, Thomson FJ. Influence of fixation conditions on the performance of glutaraldehyde-treated porcine aortic valves: towards a more scientific basis. Thorax 1979;34: 166 -76.

4. Barratt-Boyes BG. Cardiothoracic surgery in the antipodes. J Thorac Cardiovasc Surg 1979;78:804-22.

5. Edmunds LH, Clark RE, Cohn LH, Miller C, Weisel RD. Guidelines for reporting morbidity and mortality after car- diac valvular operations. Ann Thorac Surg 1988;46:257-9.

6. Kirklin JW, Blackstone EH. Figures. J Thorac Cardiovasc Surg 1994;107:1175-7.

7. Kalbfleisch JD, Prentice RL. The statistical analysis of failure time data. New York: Wiley, 1980:14.

8. Fernandez J. Technical principles of valve replacement, In: Morse D, Steiner RM, Fernandez J, eds. Guide to prosthetic cardiac valves. 1st ed. New York: Springer-Verlag, 1985: 158-77.

9. Fernandez J, Laub GW, Adkins MS, et al. Early and late- phase events after valve replacement with the St. Jude medical prosthesis in 1200 patients. J Thorac Cardiovasc Surg 1994;107:394-407.

10. Magilligan DJ Jr, Lewis JW Jr, Jara FM, et al. Spontaneous degeneration of porcine bioprosthetic valves. Ann Thorac Surg 1980;30:259-66.

11. Borkon AM, McIntosh CL, Von Rueden TJ, Morrow AG. Mitral valve replacement with the Hancock bioprosthesis: five to ten-year follow up. Ann Thorac Surg 1981;32:127-37.

12. Akins CW, Carroll DL, Buckley MJ, Daggett WM, Hilgen- berg AD, Austen WG. Late results with Carpentier-Edwards porcine bioprosthesis. Circulation 1990;82(Suppl 4):65-74.

13. Jamieson WRE, Hayden RI, Miyagishma RT, et al. The Carpentier-Edwards standard porcine bioprosthesis: clinical performance to 15 years. J Cardiac Surg 1991;6(Suppl): 5550-6.

14. Jamieson WRE, Tyers GFO, Janusz MT, et al. Age as a

15.

16.

17.

18.

19.

20.

21.

22.

23.

24.

25.

determinant for selection of porcine bioprostheses for car- diac value replacement: experience with Carpentier- Edwards standard bioprosthesis. Can J Cardiol1991;7:181-8. Pupello DF, Bessone LN, Hiro SP, et al. Bioprosthetic valve durability in the elderly: the second decade. J Cardiac Surg 1991;6:575-9. Vesely I, Lozon A. Natural preload of aortic valve leaflet components during glutaraldehyde fixation: effects on tissue mechanics. Biomechanics 1992;26:121-31. Vesely I, Krucinski S, Campbell G. Micromechanics and mathematical modeling: an inside look at bioprosthetic valve function. J Cardiac Surg 1992;7:85-95. Thurbrikar MJ, Deck JD, Aouad J, Nolan SF. Role of mechan- ical stress in calcification of aortic bioprosthetic valves. J Thorac Cardiovasc Surg 1983;86:115-25. Stein I’D, Kemp SR, Riddle JM, Lee MW, Lewis JW, Magil- ligan DJ Jr. Relation of calcification to torn leaflets of spon- taneously degenerated porcine bioprosthetic valves. Ann Thorac Surg 1985;40:175-80. Hilbert SL, Barrick MK, Ferrans VJ. Porcine aortic valve bioprostheses: a morphologic comparison of the effects of fixation pressure. J Biomed Mater Res 1990;24:773-87. Barratt-Boyes BG, Ko PH, Jaffe WM. The zero pressure fixed Medtronic Intact porcine valve: clinical results over a 6-year period, including serial echocardiographic assessment. J Cardiac Surg 1991;6(Suppl):S606-12. Vermeulen F, Bennick G, Ernst S, Jaarsmu W, Chevalier PA, Lutz DV. The intact porcine bioprosthesis: early world-wide clinical experience and analysis of a single institution’s experience. Eur J Cardiothorac Surg 1992;6(Suppl l):S124-7. Jamieson WRE, Munro AI, Miyagishma RT, et al. The Carpentier-Edwards supraannular porcine bioprosthesis: new generation low pressure glutaraldehyde fixed prosthe- sis. J Cardiac Surg 1988;3:507-21. David TE, Armstrong S, Sun Z. Clinical and hemodynamic assessment of the Hancock II bioprosthesis. Ann Thorac Surg 1992;54:661-8. Jamieson WRE, Burr LH, Tyers GFO, Munro AI. Carpentier- Edwards standard and supra-annular porcine bioprostheses: 10 year comparison of structural valve deterioration. J Heart Valve Dis 1994;3:59-65.

DISCUSSION

SIR BRIAN G. BARRATT-BOYES (Moderator) (Auckland, New Zealand): These two prostheses were obviously different with regard to pressure fixation of the tissue with glutaraldehyde, but were there any other structural differences between the two prostheses or were they otherwise structurally identical?

DR FERNANDEZ: The valves were identical in all other respects including the flexible stent.

DR J. FRANCESCO NISTAL (Santander Cantabria, Spain): You indicated that you started with 46 low-pressure-fixed valves and to avoid age bias, which was very strong against the control group, you subdivided the 46 valves into several age cohorts. How many patients were enrolled in the 30- to 65-year low- pressure-fixed valve group?

DR FERNANDEZ: There were a total of 46 in the low-pressure group and 210 in the high-pressure group. There were no patients in the low-pressure group less than 30 years of age. Due to this fact we concentrated our comparison on the 30 to 65 and the 65 and over age groups. We had only 20 patients with low-pressure-fixed vaives in the 30- to 65-year age group.

SIR BRIAN G. BARRATT-BOYES: Are you still using the low-pressure-fixed Carpentier-Edwards prosthesis?

DR FERNANDEZ: The low-pressure-fixed mitral Carpentier- Edwards prosthesis is only used in the United States and has never been used in the aortic position. It is for this reason that from our total population of 500 patients we only selected isolated mitral valves with the two types of pressure fixation for comparison.

DR S. L. DZEMESHKEVICH (Moscow, Russia): Given that you have demonstrated that there is no difference in the results between high- and low-pressure-fixed biological valves, can you say that the mechanical stress is the main cause of structural deterioration in the long term?

DR FERNANDEZ: Our evaluation has only indicated a trend and not a statistically significant difference. If the trend does become significant in the following years then the only differ- ence between the two prostheses is the glutaraldehyde fixation pressure.