- Email: [email protected]
Ten-year experience with pericardial xenograft valves
J
THoRAc CARDIOVASC SURG
1988;95:572-6
Ten-year experience with pericardial xenograft valves Five hundred fifty-three pericardial xenografts were inserted in 520 patients during a 9-year period at the Fu Wai Hospital in Beijing. The bovine pericardial xenograft was modeled after a similar type of prosthesis manufactured by Shiley Inc. (Irvine, Calif.). The late mortality rate in this series was only 1.8 % per annum and the actuarial survival rate was 73.0 % ± 12 % at 10 years. There was a very acceptable low incidence of thromboembolism of 0.41 % per annum without the need for permanent anticoagulation. This is similar to the clinical reports with other tissue valves. The main question is the durability of the tissue prosthesis or, in other words, the freedom from valve-related clinical complications. In this series, the expected actuarial valvedurability rate was 75.0 % ± 8.8 % at 10 years. Whether this will continue to hold up over the next foUow-up period is unclear. Certainly other tissue prostheses have shown significant degeneration rates beginning after the sixth year that have risen progressively thereafter. In any case, given the relatively low probability of thromboembolic phenomenon without anticoagulation, the trade-off of a prosthesis that may not be as durable as the mechanical ones is certainly acceptable.
Zhu-Xiaodong, MD, Guo-Jiaqiang, MD, Chen-Yingchun, MD, Tang-Chengjun, MD, and Xue-Ganxing, MD, Beijing, People's Republic of China
In 1971, Ionescu,' Silverton,' and their associates developed a xenograft cardiac valve prosthesis that was constructed of glutaraldehyde-preserved bovine pericardium mounted on a flexible stent. This prosthesis has been used widely since that time and is manufactured by Shiley Inc. (Irvine, Calif.) as the Ionescu-Shiley pericardial xenograft. This valve has demonstrated excellent hemodynamic performance associated with a very low thromboembolic rate and, in addition, has exhibited exceptional durability during its 12 years of clinical implantation. With this experience in mind, we have produced a similar type of bovine pericardiaI xenograft in Beijing (Figs. 1 and 2). We have used this prosthesis exclusively since May of 1976. This report reviews our clinical experience with the Beijing pericardial xenograft in 520 patients operated on at our hospital between 1976 and 1985.
From the Cardiovascular Institute and Fu Wai Hospital, Chinese Academy of Medical Sciences, Beijing, People's Republic of China. Received for publication Sept. 25, 1986. Accepted for publication March 19, 1987. Address for reprints: Zhu-Xiaodong, MD, Cardiovascular Institute and Fu Wai Hospital, Chinese Academy of Medical Sciences, Beijing, People's Republic of China.
572
Table I. Height and orifice area of Beijing pericardial xenograft Valve size (mm)
Height (mm) Essential orifice area' (crrr') Effective orifice areat (cm)
23
25
27
29
16.5 2.83 2.41
17.5 3.46 2.91
18.5 4.15
19.5 4.91
3.77
'Secondary orifice area. t Flow rate (steady state) 12 Lyrnin.
Methods The Beijing pericardial xenograft. Bovine pericardium was harvested according to standard techniques.' The pericardium was rinsed with Hanks' solution to remove the soluble protein, and the material was fixed and preserved in a glutaraldehyde solution. The stent was constructed of titanium and covered with Dacron cloth. Various sizes of the constructed pericardial xenografts were then tested in vitro for hemodynamic patterns. The transvaIvular gradients and effective orifice areas were measured during both pulsatile and steady flows. Effective orifice area, as it was related to annular diameter and height, is shown in Table 1. Patient population. Between May 1976 and May 1985, 520 patients (276 men and 244 women) who had acquired valvular heart disease underwent valve replacement with the Beijing pericardial xenograft at the Fu Wai Hospital in Beijing. The age of the patients ranged from 10 to 63 years,
Volume 95 Number 4 April 1988
Beijing pericardial xenograft
573
Table II. Causes of valve lesion in 520 patients and operations performed No. of operations Cause
MVR
AVR
Rheumatic fever Endocarditis Marfan's syndrome Failure of valve repair Pericardial valve failure Cardiomyopathy Constrictive fibrous endocarditis Auricular myxoma Trauma Other Total
401 1
38 12 8 1
18 3
TVR
MAVR
31 1
1
1 424
-.l 61
2
-.l
33
Legend: MYR, Mitral valve replacement. AYR, Aortic valve replacement. TYR, Tricuspid valve replacement. MA YR, Mitral and aortic valve replacement.
Table
m. Operative mortality in 520 patients
No. of operations No. of deaths Mortality rate
MVR
AVR
424
61
43 10.1
6 9.8
TVR
MAVR
2
33
1
5 15.2
50.0
Fig. 1. Beijing pericardial valve (mitral): inflow. Total 520 55 10.6
(%)
Legend: MYR, Mitral valve replacement. AYR, Aortic valve replacement. TYR, Tricuspid valve replacement. MA YR, Mitral and aortic valve replacement.
with an average of 34.7 years, In this series, 424 patients had isolated mitral valve replacement, 61 had isolated aortic valve replacement, 2 underwent tricuspid valve replacement, and 33 had a combined mitral and aortic valve replacement. The causes of the valvular lesions are shown in Table II. Rheumatic fever was the primary cause of mitral valve disease, and subacute bacterial endocarditis was the most common cause of aortic valvular disease. The preoperative status of the patients shows that, significantly, 21% of the patients had congestive heart failure and that 82% of the patients in the isolated mitral valve group had atrial fibrillation as the underlying cardiac rhythm. Surgical technique. Standard cardiopulmonary bypass with a disposable bubble oxygenator and a clear fluid prime was performed in all cases. Moderate systemic hypothermia and cold crystalloid potassium cardioplegia were used. In addition, topical cooling with iced saline solution was used during the cross-clamp period, The potassium cardioplegic solution was instilled directly into the aortic root after aortic clamping or into the coronary ostia when aortic valve replacement was performed. For aortic valve replacement, interrupted 2-0 braided mattress sutures were used in most cases. A continuous suture technique was performed in only four cases of aortic valve replacement. The aortic root was enlarged with a prosthetic
Fig. 2. Beijing pericardial valve (mitral): outflow. patch when the anulus was believed to be too small for the prosthesis chosen. In most cases of mitral valve replacement, to avoid entangling the sutures around the struts, a continuous suture technique (partial or complete) was used. If an interrupted suture technique was selected, these sutures were initially inserted into the anulus of the mitral valve and were then passed through the sewing ring of the xenograft after the valve was lowered into the anulus. Fifty-nine concomitant surgical procedures were performed in this group of patients; most of these (78%) were valve reconstructive operations and 15% of the patients had the aortic root enlarged.
574
The Journal of Thoracic and Cardiovascular Surgery
Zhu-Xiaodong et al.
, l00r--...
__
80
73.0:!:12.1'
60
i
2
i i i
3
4
,
,
587
;
8
,
9
I
10
YEARS POST-OP
Fig. 3. Actuarial survival rate of 465 hospital survivors with Beijing pericardiaI valves. There were 381 mitral, 55 aortic, 1 tricuspid, and 28 mitral and aortic valve replacements.
Table IV. Follow-up data for 465 survivors Hospital survivors (No.) Late deaths (No.) Cumulative follow-up (yr) Mean follow-up (yr) Late deaths per annum (%)
MVR
AVR
381 11 979.2 2.75 1.12
55 10 185.9 3.38 5.4
TVR
MAVR
Total
0.3
28 1 60.1 2.14 1.7
465 22 1225.5 2.64 1.8
Legend: MVR, Mitral valve replacement. AVR, Aortic valve replacement. TVR, Tricuspid valve replacement. MA VR, Mitral and aortic valve replacement.
Results Early mortality. The hospital mortality rate for the entire group was 10.6% (55 deaths). There was no statistical difference between the hospital mortality rate for the group having isolated mitral valve replacement (10.1%) and that for the group having isolated aortic valve replacement (9.8%). However... the mortality rate for the group having double valve replacement was significantly higher (Table III). In 71% of the deaths, cardiac factors were believed to be the major contributors to the cause of hospital death. In 16 patients (29%), noncardiac factors, such as complications related to heart-lung bypass, were thought to contribute to the hospital mortality. Cardiac factors implicated were primary myocardial failure in 17 patients, postoperative hemorrhage in 6, ventricular dysrhythmia in 10, and endocarditis in 2. Nonfatal complications occurred in 8 patients: strut entanglement with persistent mitral regurgitation in 4 and cardiac tamponade in 4. Sternal wound infection developed in 5 patients, but there was no incidence of perivalvular leak in the early postoperative period. Anticoagulation with sodium warfarin was begun on the second day after operation and was maintained for 2 to 4 months for all patients including
those who had atrial fibrillation. No early embolic complications were encountered. Survival. Four hundred sixty-five (89.4%) patients were discharged from the hospital. The distribution of these patients according to valvular procedure is shown in Table IV. The surviving patients were followed up from 3 months to 9 years. The cumulative follow-upwas 1225.5 years, a mean of 2.6 years per patient. The late mortality rate was 4.7% (22 patients), an average of 1.8% per year. The actuarial survival rate was calculated to be 73.0% ± 12.1% at 10 years (Fig. 3). According to the New York Heart Association classification, cardiac function was significantly improved in most survivors (Fig. 4). Three hundred three patients (81.1%) who underwent isolated mitral valve replacement were found to be in functional classes I and II. Thirty-eight patients (86.4%) with isolated aortic valve replacement and 24 patients (85.7%) with double valve replacement were also found to be in functional classes I and II. Late deaths occurred in 22 patients. Causes of late death included endocarditis (4 patients), systemic embolism (2), persistent congestive heart failure (2), and renal hemorrhage (2). Four patients died as a result of
Volume 95 Number 4 April 1988
Beijing pericardia/ xenograft
AVR n44
MVR n 344
575
MAVR n 28
n m pre-op
post-op pre-op
post-op pre-op post-op
Fig. 4. Preoperative and latest postoperative New York Heart Association functional classification of long-term survivors. MVR, Mitral valve replacement. AVR, Aortic valve replacement. MAVR, Mitral and aortic valve replacement.
90 78.1tll.5~
80
.... ..
-.. ..
70
1 0
~._._._._._._-.,
"
MITRAL
_
AORTIC
i
I
;
;
3
4
8.8~
71.9t12.9~
...i
All Pl\TIENTS
2
75.0t
;
5
,
6
8
9
10
VEARS POST-OP
Fig. 5. Actuarial event-free curves for valve failure in patients with Beijing pericardial valves. The subgroups of patients with mitral and aortic valve replacement are shown with respective failure rates.
Table V. Clinical and pathologic data of valve failure
Age
Sex
Operation
Duration of implant (mo)
23 10 42 23 25
M F F F F F F M
MVR MVR MVR MVR AVR AVR AVR AVR
64 35 65 76 73 60 26
27
39 22
72
Cause of valve failure
Reoperation
Result
Calcification, cusp tear Calcification Waiting for reoperation Cusp tear Cusp tear Calcification, cusp tear Waiting for reoperation Valve thrombosis (infection)
Yes Yes No Yes Yes Yes No No
Dead Dead Alive Alive Alive Dead Alive Dead
Legend: MYR, Mitral valve replacement. AYR, Aortic valve replacement.
reoperation, and in 8 patients the cause of death was unknown. Valve dysfunction. Valve dysfunction caused by the entanglement of sutures around the struts of the prosthesis occurred in four patients undergoing isolated
mitral valve replacement. The diagnosis was made by the presence of a systolic murmur soon after operation. All patients underwent reoperation between 1 week and 50 months after the first procedure. Three patients survived. A single patient who was operated on 50
The Journal of Thoracic and Cardiovascular Surgery
5 7 6 Zhu-Xiaodong et al.
months after the first procedure died of persistent myocardial failure. In eight other patients, primary intrinsic valve failure occurred (Table V). Valve cusp tear, calcification, and infection were cited as the causes of valve dysfunction. The linearized rate of valve failure was 0.65% per annum for the total number of patients at risk. The actuarial curves of freedom from valve failure are shown in Fig. 5. The expected actuarial durability rate was 75.0% ± 8.8% at 10 years for all valves, 71.9% ± 12.9% at 10 years for aortic valves, and 78.1% ± 11.5% at 8 years for mitral valves. The linearized rate of systemic embolism was 0.41% per annum for the entire series. The actuarial rate of freedom from embolism was 94.8% at 10 years. Four cases of late endocarditis were encountered in this series, with a linearized rate of 0.34% per annum, and 98.7% of patients were free from endocarditis at 10 years. Discussion
The current study confirms the good long-term characteristics of the Beijing pericardial xenograft with respect to cardiac function and thromboembolic rate. Postoperatively, 88% of the patients had cardiac function improved one to two classifications, and the late embolic rate was only 0.41% per annum without permanent anticoagulation therapy. However, certain patients with specific problems, that is, the patients with a large left atrium or chronic atrial fibrillation, or both had anticoagulation continued indefinitely. At the beginning of our experience, four cases of suture entanglement occurred in the group having mitral valve replacement. This led us to change our suturing technique, and we now routinely use continuous sutures for mitral valve replacement. The configuration
of the Beijing pericardia 1xenograft with its narrow and more pointed struts lends itself to possible entanglement when a multiple interrupted suture technique is used. In addition, care must be taken to avoid any left ventricular damage by these prominent strut supports of the prosthesis. The valve for a patient with a small left ventricle should be carefully sized so that it does not impinge on the posterior left ventricular wall, because with vigorous systolic contraction, a strut may work its way through the endocardium. Therefore, we tend to undersize a valve prosthesis in the mitral valve position in patients with a very small left ventricle or with a hyperdynamic left ventricle. In our entire group of 465 survivors, only eight cases of intrinsic valve failure occurred, for a failure rate of only 0.65% per annum. The expected actuarial durability rate was 75.0% ± 8.8% at 10 years. We also believe, paralleling the experience with the initial Ionescu-Shiley xenograft, that better design characteristics and quality control during manufacture of the prosthesis will prolong durability. REFERENCES 1. Ionescu MI, Tandon AP, Chidambaram M, Yakirevich
VS, Silverton NP. Durability of the pericardial valve. Eur Heart J 1984;5(suppl):101-6. 2. Silverton NP, Abdulali SA, Yakirevich VS, Tandon AP, Ionescu MI. Embolism, thrombosis and anticoagulant haemorrhage in mitral valve disease: a propective study of patients having valve replacement with the pericardiaI xenograft. Eur Heart J 1984;5(suppl):19-25. 3. Guo J, Zhu X, Chen Y, et al. Clinical use of the pericardial xenograft valve prosthesis: a five-year experience. Proceedings of a Beijing symposium on cardiothoracic surgery. Beijing: John Wiley, 1982:316-22.
THoRAc CARDIOVASC SURG
1988;95:572-6
Ten-year experience with pericardial xenograft valves Five hundred fifty-three pericardial xenografts were inserted in 520 patients during a 9-year period at the Fu Wai Hospital in Beijing. The bovine pericardial xenograft was modeled after a similar type of prosthesis manufactured by Shiley Inc. (Irvine, Calif.). The late mortality rate in this series was only 1.8 % per annum and the actuarial survival rate was 73.0 % ± 12 % at 10 years. There was a very acceptable low incidence of thromboembolism of 0.41 % per annum without the need for permanent anticoagulation. This is similar to the clinical reports with other tissue valves. The main question is the durability of the tissue prosthesis or, in other words, the freedom from valve-related clinical complications. In this series, the expected actuarial valvedurability rate was 75.0 % ± 8.8 % at 10 years. Whether this will continue to hold up over the next foUow-up period is unclear. Certainly other tissue prostheses have shown significant degeneration rates beginning after the sixth year that have risen progressively thereafter. In any case, given the relatively low probability of thromboembolic phenomenon without anticoagulation, the trade-off of a prosthesis that may not be as durable as the mechanical ones is certainly acceptable.
Zhu-Xiaodong, MD, Guo-Jiaqiang, MD, Chen-Yingchun, MD, Tang-Chengjun, MD, and Xue-Ganxing, MD, Beijing, People's Republic of China
In 1971, Ionescu,' Silverton,' and their associates developed a xenograft cardiac valve prosthesis that was constructed of glutaraldehyde-preserved bovine pericardium mounted on a flexible stent. This prosthesis has been used widely since that time and is manufactured by Shiley Inc. (Irvine, Calif.) as the Ionescu-Shiley pericardial xenograft. This valve has demonstrated excellent hemodynamic performance associated with a very low thromboembolic rate and, in addition, has exhibited exceptional durability during its 12 years of clinical implantation. With this experience in mind, we have produced a similar type of bovine pericardiaI xenograft in Beijing (Figs. 1 and 2). We have used this prosthesis exclusively since May of 1976. This report reviews our clinical experience with the Beijing pericardial xenograft in 520 patients operated on at our hospital between 1976 and 1985.
From the Cardiovascular Institute and Fu Wai Hospital, Chinese Academy of Medical Sciences, Beijing, People's Republic of China. Received for publication Sept. 25, 1986. Accepted for publication March 19, 1987. Address for reprints: Zhu-Xiaodong, MD, Cardiovascular Institute and Fu Wai Hospital, Chinese Academy of Medical Sciences, Beijing, People's Republic of China.
572
Table I. Height and orifice area of Beijing pericardial xenograft Valve size (mm)
Height (mm) Essential orifice area' (crrr') Effective orifice areat (cm)
23
25
27
29
16.5 2.83 2.41
17.5 3.46 2.91
18.5 4.15
19.5 4.91
3.77
'Secondary orifice area. t Flow rate (steady state) 12 Lyrnin.
Methods The Beijing pericardial xenograft. Bovine pericardium was harvested according to standard techniques.' The pericardium was rinsed with Hanks' solution to remove the soluble protein, and the material was fixed and preserved in a glutaraldehyde solution. The stent was constructed of titanium and covered with Dacron cloth. Various sizes of the constructed pericardial xenografts were then tested in vitro for hemodynamic patterns. The transvaIvular gradients and effective orifice areas were measured during both pulsatile and steady flows. Effective orifice area, as it was related to annular diameter and height, is shown in Table 1. Patient population. Between May 1976 and May 1985, 520 patients (276 men and 244 women) who had acquired valvular heart disease underwent valve replacement with the Beijing pericardial xenograft at the Fu Wai Hospital in Beijing. The age of the patients ranged from 10 to 63 years,
Volume 95 Number 4 April 1988
Beijing pericardial xenograft
573
Table II. Causes of valve lesion in 520 patients and operations performed No. of operations Cause
MVR
AVR
Rheumatic fever Endocarditis Marfan's syndrome Failure of valve repair Pericardial valve failure Cardiomyopathy Constrictive fibrous endocarditis Auricular myxoma Trauma Other Total
401 1
38 12 8 1
18 3
TVR
MAVR
31 1
1
1 424
-.l 61
2
-.l
33
Legend: MYR, Mitral valve replacement. AYR, Aortic valve replacement. TYR, Tricuspid valve replacement. MA YR, Mitral and aortic valve replacement.
Table
m. Operative mortality in 520 patients
No. of operations No. of deaths Mortality rate
MVR
AVR
424
61
43 10.1
6 9.8
TVR
MAVR
2
33
1
5 15.2
50.0
Fig. 1. Beijing pericardial valve (mitral): inflow. Total 520 55 10.6
(%)
Legend: MYR, Mitral valve replacement. AYR, Aortic valve replacement. TYR, Tricuspid valve replacement. MA YR, Mitral and aortic valve replacement.
with an average of 34.7 years, In this series, 424 patients had isolated mitral valve replacement, 61 had isolated aortic valve replacement, 2 underwent tricuspid valve replacement, and 33 had a combined mitral and aortic valve replacement. The causes of the valvular lesions are shown in Table II. Rheumatic fever was the primary cause of mitral valve disease, and subacute bacterial endocarditis was the most common cause of aortic valvular disease. The preoperative status of the patients shows that, significantly, 21% of the patients had congestive heart failure and that 82% of the patients in the isolated mitral valve group had atrial fibrillation as the underlying cardiac rhythm. Surgical technique. Standard cardiopulmonary bypass with a disposable bubble oxygenator and a clear fluid prime was performed in all cases. Moderate systemic hypothermia and cold crystalloid potassium cardioplegia were used. In addition, topical cooling with iced saline solution was used during the cross-clamp period, The potassium cardioplegic solution was instilled directly into the aortic root after aortic clamping or into the coronary ostia when aortic valve replacement was performed. For aortic valve replacement, interrupted 2-0 braided mattress sutures were used in most cases. A continuous suture technique was performed in only four cases of aortic valve replacement. The aortic root was enlarged with a prosthetic
Fig. 2. Beijing pericardial valve (mitral): outflow. patch when the anulus was believed to be too small for the prosthesis chosen. In most cases of mitral valve replacement, to avoid entangling the sutures around the struts, a continuous suture technique (partial or complete) was used. If an interrupted suture technique was selected, these sutures were initially inserted into the anulus of the mitral valve and were then passed through the sewing ring of the xenograft after the valve was lowered into the anulus. Fifty-nine concomitant surgical procedures were performed in this group of patients; most of these (78%) were valve reconstructive operations and 15% of the patients had the aortic root enlarged.
574
The Journal of Thoracic and Cardiovascular Surgery
Zhu-Xiaodong et al.
, l00r--...
__
80
73.0:!:12.1'
60
i
2
i i i
3
4
,
,
587
;
8
,
9
I
10
YEARS POST-OP
Fig. 3. Actuarial survival rate of 465 hospital survivors with Beijing pericardiaI valves. There were 381 mitral, 55 aortic, 1 tricuspid, and 28 mitral and aortic valve replacements.
Table IV. Follow-up data for 465 survivors Hospital survivors (No.) Late deaths (No.) Cumulative follow-up (yr) Mean follow-up (yr) Late deaths per annum (%)
MVR
AVR
381 11 979.2 2.75 1.12
55 10 185.9 3.38 5.4
TVR
MAVR
Total
0.3
28 1 60.1 2.14 1.7
465 22 1225.5 2.64 1.8
Legend: MVR, Mitral valve replacement. AVR, Aortic valve replacement. TVR, Tricuspid valve replacement. MA VR, Mitral and aortic valve replacement.
Results Early mortality. The hospital mortality rate for the entire group was 10.6% (55 deaths). There was no statistical difference between the hospital mortality rate for the group having isolated mitral valve replacement (10.1%) and that for the group having isolated aortic valve replacement (9.8%). However... the mortality rate for the group having double valve replacement was significantly higher (Table III). In 71% of the deaths, cardiac factors were believed to be the major contributors to the cause of hospital death. In 16 patients (29%), noncardiac factors, such as complications related to heart-lung bypass, were thought to contribute to the hospital mortality. Cardiac factors implicated were primary myocardial failure in 17 patients, postoperative hemorrhage in 6, ventricular dysrhythmia in 10, and endocarditis in 2. Nonfatal complications occurred in 8 patients: strut entanglement with persistent mitral regurgitation in 4 and cardiac tamponade in 4. Sternal wound infection developed in 5 patients, but there was no incidence of perivalvular leak in the early postoperative period. Anticoagulation with sodium warfarin was begun on the second day after operation and was maintained for 2 to 4 months for all patients including
those who had atrial fibrillation. No early embolic complications were encountered. Survival. Four hundred sixty-five (89.4%) patients were discharged from the hospital. The distribution of these patients according to valvular procedure is shown in Table IV. The surviving patients were followed up from 3 months to 9 years. The cumulative follow-upwas 1225.5 years, a mean of 2.6 years per patient. The late mortality rate was 4.7% (22 patients), an average of 1.8% per year. The actuarial survival rate was calculated to be 73.0% ± 12.1% at 10 years (Fig. 3). According to the New York Heart Association classification, cardiac function was significantly improved in most survivors (Fig. 4). Three hundred three patients (81.1%) who underwent isolated mitral valve replacement were found to be in functional classes I and II. Thirty-eight patients (86.4%) with isolated aortic valve replacement and 24 patients (85.7%) with double valve replacement were also found to be in functional classes I and II. Late deaths occurred in 22 patients. Causes of late death included endocarditis (4 patients), systemic embolism (2), persistent congestive heart failure (2), and renal hemorrhage (2). Four patients died as a result of
Volume 95 Number 4 April 1988
Beijing pericardia/ xenograft
AVR n44
MVR n 344
575
MAVR n 28
n m pre-op
post-op pre-op
post-op pre-op post-op
Fig. 4. Preoperative and latest postoperative New York Heart Association functional classification of long-term survivors. MVR, Mitral valve replacement. AVR, Aortic valve replacement. MAVR, Mitral and aortic valve replacement.
90 78.1tll.5~
80
.... ..
-.. ..
70
1 0
~._._._._._._-.,
"
MITRAL
_
AORTIC
i
I
;
;
3
4
8.8~
71.9t12.9~
...i
All Pl\TIENTS
2
75.0t
;
5
,
6
8
9
10
VEARS POST-OP
Fig. 5. Actuarial event-free curves for valve failure in patients with Beijing pericardial valves. The subgroups of patients with mitral and aortic valve replacement are shown with respective failure rates.
Table V. Clinical and pathologic data of valve failure
Age
Sex
Operation
Duration of implant (mo)
23 10 42 23 25
M F F F F F F M
MVR MVR MVR MVR AVR AVR AVR AVR
64 35 65 76 73 60 26
27
39 22
72
Cause of valve failure
Reoperation
Result
Calcification, cusp tear Calcification Waiting for reoperation Cusp tear Cusp tear Calcification, cusp tear Waiting for reoperation Valve thrombosis (infection)
Yes Yes No Yes Yes Yes No No
Dead Dead Alive Alive Alive Dead Alive Dead
Legend: MYR, Mitral valve replacement. AYR, Aortic valve replacement.
reoperation, and in 8 patients the cause of death was unknown. Valve dysfunction. Valve dysfunction caused by the entanglement of sutures around the struts of the prosthesis occurred in four patients undergoing isolated
mitral valve replacement. The diagnosis was made by the presence of a systolic murmur soon after operation. All patients underwent reoperation between 1 week and 50 months after the first procedure. Three patients survived. A single patient who was operated on 50
The Journal of Thoracic and Cardiovascular Surgery
5 7 6 Zhu-Xiaodong et al.
months after the first procedure died of persistent myocardial failure. In eight other patients, primary intrinsic valve failure occurred (Table V). Valve cusp tear, calcification, and infection were cited as the causes of valve dysfunction. The linearized rate of valve failure was 0.65% per annum for the total number of patients at risk. The actuarial curves of freedom from valve failure are shown in Fig. 5. The expected actuarial durability rate was 75.0% ± 8.8% at 10 years for all valves, 71.9% ± 12.9% at 10 years for aortic valves, and 78.1% ± 11.5% at 8 years for mitral valves. The linearized rate of systemic embolism was 0.41% per annum for the entire series. The actuarial rate of freedom from embolism was 94.8% at 10 years. Four cases of late endocarditis were encountered in this series, with a linearized rate of 0.34% per annum, and 98.7% of patients were free from endocarditis at 10 years. Discussion
The current study confirms the good long-term characteristics of the Beijing pericardial xenograft with respect to cardiac function and thromboembolic rate. Postoperatively, 88% of the patients had cardiac function improved one to two classifications, and the late embolic rate was only 0.41% per annum without permanent anticoagulation therapy. However, certain patients with specific problems, that is, the patients with a large left atrium or chronic atrial fibrillation, or both had anticoagulation continued indefinitely. At the beginning of our experience, four cases of suture entanglement occurred in the group having mitral valve replacement. This led us to change our suturing technique, and we now routinely use continuous sutures for mitral valve replacement. The configuration
of the Beijing pericardia 1xenograft with its narrow and more pointed struts lends itself to possible entanglement when a multiple interrupted suture technique is used. In addition, care must be taken to avoid any left ventricular damage by these prominent strut supports of the prosthesis. The valve for a patient with a small left ventricle should be carefully sized so that it does not impinge on the posterior left ventricular wall, because with vigorous systolic contraction, a strut may work its way through the endocardium. Therefore, we tend to undersize a valve prosthesis in the mitral valve position in patients with a very small left ventricle or with a hyperdynamic left ventricle. In our entire group of 465 survivors, only eight cases of intrinsic valve failure occurred, for a failure rate of only 0.65% per annum. The expected actuarial durability rate was 75.0% ± 8.8% at 10 years. We also believe, paralleling the experience with the initial Ionescu-Shiley xenograft, that better design characteristics and quality control during manufacture of the prosthesis will prolong durability. REFERENCES 1. Ionescu MI, Tandon AP, Chidambaram M, Yakirevich
VS, Silverton NP. Durability of the pericardial valve. Eur Heart J 1984;5(suppl):101-6. 2. Silverton NP, Abdulali SA, Yakirevich VS, Tandon AP, Ionescu MI. Embolism, thrombosis and anticoagulant haemorrhage in mitral valve disease: a propective study of patients having valve replacement with the pericardiaI xenograft. Eur Heart J 1984;5(suppl):19-25. 3. Guo J, Zhu X, Chen Y, et al. Clinical use of the pericardial xenograft valve prosthesis: a five-year experience. Proceedings of a Beijing symposium on cardiothoracic surgery. Beijing: John Wiley, 1982:316-22.