- Email: [email protected]
Valve replacement in children under twenty years of age
J
THORAC CARDIOVASC SURG
88:217-224, 1984
Valve replacement in children under twenty years of age Experience with the Bjork-Shiley prosthesis Prosthetic valve replacement in young patients bas been reported to be associated with a high mortality and morbidity because of valve-related problems. Of 549 patients tmdergoing valve replacement with the Bjork-5hi1ey valve prosthesis, 136 were under the ageof 20 years. Sixty-four patients wereunder 16 years of age, the youngest being 6 years old. Of the 136 patients, 61 underwent mitral valve replacemen~ 50 received an aortic valvt\ and 25 received both aortic and mitral valves. Overall operative mortality was 10.3%. Late mortality overa follow-up period of 6 montm to 8 years was 4.4 %. Actuarial survival curves up to 8 years of follow-up are presented.Results obtained in this group are compared with those obtained in 413 patients over 20 years of age operated during the same period. Valve thrombosis was not seen in any patient under 20 years of agt\ but it occurred in 4.13 % of the patients over 20 years of age. The incidence of thromboembotism and anticoagulant-related hemorrhage was very low. There bas been no imtance of structural failure of the valve. l.ong-term results are excellent, with 90 % of the survivors returning to New York Heart Association Functional Class I. The Bjork-Sbileyvalve gives excellent and durable long-term palliation in young patients requiring valve replacement.
Krishna Subramony Iyer, M.D., K. Srinath Reddy, M.D., I. Mritunjaya Rao, M.D., P. Venugopal, M.D., M. L. Bhatia, M.D., and N. Gopinath, M.D.,
New Delhi, India
h e controversy over the choice of ideal valve substitutes for replacing diseased valves in children continues to surface frequently. The introduction of glutaraldehyde-treated porcine heterografts appeared at one time to provide the solution to this problem. However, the alarming incidence of early valve failure because of degenerative calcification' has raised doubts regarding the continued use of these valves in young patients, and current trends seem to favor mechanical valves. Reports on large series of children undergoing valve repiacement with a particular valve are scant in Western literature, owing possibly to the paucity of acquired valvular heart disease in young patients in the West. In India, however, From the Department of Cardio-Thoracic and Vascular Surgery and the Department of Cardiology, All India Institute of Medical Sciences, Ansari Nagar, New Delhi, India. Received for publication Aug. 22, 1983. Accepted for publication Oct. 11, 1983. Address for reprints: Professor P. Venugopal, Head, Department of Cardio-Thoracic and Vascular Surgery, All India Institute of Medical Sciences, Ansari Nagar. New Delhi-l 10029, India.
rheumatic heart disease (RHO) continues to be a major contributor to valvular heart disease, especially in the young. RHD is diagnosed in 30% to 40% of the patients attending the cardiac clinic at the All India Institute of Medical Sciences Hospital. One fourth of these patients are under 20 years of age. The entities of "juvenile mitral stenosis" and "juvenile RHO," first described by Roy and associates'< in 1963, continue to influence the clinical profile of RHD in this country. Vast numbers of patients with RHO require surgical intervention. Though valve reconstruction theoretically would seem to be the ideal surgical procedure in these patients, the extensive valvular deformities that frequently occur in juvenile RHO preclude such reparative procedures and necessitate valve replacement. In a developing country like India, where cardiac operations impose a major fmancial burden on the patient, a valve with a low reoperation rate due to valve failure is needed. It has been our policy, therefore, to use a mechanical valve in all young patients. Since 1975, a fairly large number of patients under the age of 20 years 217
The Journal of Thoracic and Cardiovascular
2 1 8 Iyer et al.
Surgery
Table I. Site ofprosthetic valve replacement
Table IV. Comparison of valve sizes
No. of patients
Patients
Site
Less than 20 years
More than 20 years
Aortic Mitral Aortic + mitral
50 61 25
116 231 66
136
413
Total*
'Total No. of patients = 549; total No. of valves implanted = 640.
Table D. Associated procedures performed (under 20 years)
VSD closure OMC OMC + De Vega repair of tricuspid valve De Vega repair of tricuspid valve ASD closure Aortic root enlargement Others
Legend: AVR, Aortic valve replacement. MVR, Mitral valve replacement. DVR, Double valve replacement. VSD, Ventricular septal defect. OMC, Open mitral commissurotomy. ASD, Atrial septal defect.
Table ill. Pathology of valvular lesion (under 20 years) AVR
MVR
DVR
Pathology RHD Pure AR AS+AR AR+MS AR+MS+TR CHD AS VSD+AR RHD MS (restenosis) MS+MR Pure MR CHD ASD+MVPS Myxomatous mitral valve RHD Total
Less than 20 years
More than 20 years
Aortic 19 21 23 25 27
21 (28%) 35 (46.7%) 15 (20.0%) 3 (4.0%) I (1.3%)
23 (12.63%) 76 (41.76%) 50 (27.5%) 29 (15,93%) 4 (2.1%)
Total
75
182
Mitral 25 27 29 Total
19122.01%) 57 (66.2%) 10 (11.6%) 86
82 (27.6%) 185 (62.2%) 30 (10.1%) 297
have undergone valve replacement, and this report analyzes our experience with the Bjork-Shiley valve in this age group. The operative results and long-term function of this valve in the younger age group are compared with results obtained in adult patients over the same period of time to highlight the excellent results obtained with the Bjork-Shiley valve in children.
Total
Group
Size (mm)
No. of cases 31 20 3 6 2 19 II 8
58 2 45 II 3 2 I 25 RHD: 114 (83.3%) CHD: 22 (16.2%)
Legend: RHD, Rheumatic heart disease. CHD, Congenital heart disease. AR, Aortic regurgitation. AS, Aortic stenosis. MS, Mitral stenosis. TR, Tricuspid regurgitation. MR, Mitral regurgitation. MVPS, Mitral valve prolapse syndrome. For other abbreviations see Table II.
Patient groups and methods Between January, 1975, and December, 1982, 549 patients underwent valve replacement with the BjorkShiley valve for various valvular disorders. Of these patients, 136 were below 20 years of age. The youngest patient was 6 years old and 64 patients were under 16 years of age. Sites of valve replacement are indicated in Table I. There were proportionately more patients undergoing isolated mitral valve replacement in the adult age group. All patients received a Bjork-Shiley valve with a pyrolytic carbon disc, the standard model with a flat disc before 1979 and the currently used convexo-concave model since then. Valve replacement was performed with routine cardiopulmonary bypass and moderate hypothermia. A modified cold cardioplegic perfusate (composition: NaHC0 3 50 mEq, KCI 15 mEq, and dexamethasone 8 mg in 500 ml of perfusate) is routinely used currently. Valve replacement alone was performed in 107 of the younger patients. In 29 patients, procedures in addition to valve replacement were done (Table II). As indicated in Table III, RHD (83.8%) predominated as the underlying etiologic factor, with congenital heart disease (16.2%) contributing to the rest. Valve sizes used in the two groups were tabulated and compared and are presented in Table IV. The most
Volume 88 Number 2 August, 1984
Valve replacement in children
2 19
Table V. Early mortality Patients Group
Less than 20 years
MVR AVR DVR
9/61 (14.75% ± 4.54%) 3/50 ( 6.0% ± 3.36%) 2/25 ( 8.0% ± 5.43%)
Total
14/136 (10.29% ± 2.60%)
I
More than 20 years
x'
25/231 (10.82% ± 2.04%) 8/116 ( 6.89% ± 2.35%) 17/66 (25.75% ± 5.38%)
0.39 (NS) 0.02 (NS) 2.47 (NS)
50/413 (12.10% ± 1.6%)
0.17 (NS)
Legend: NS, Not significant. For other abbreviationssee Table II.
Table VI. Late mortality Group
Less than 20 years
More than 20 years
MVR AVR DVR
2/61 (3.27% ± 2.27%) 2/50 (4.0% ± 2.77%) 2/25 (8.0% ± 5.43%)
12/231 (5.19% ± 1.45%) 2/116 (1.72% ± 1.2%) 4/66 (6.06% ± 2.94%)
0.08 (NS) 0.11 (NS) 0.02 (NS)
Total
6/136 (4.41% ± 1.76%)
18/413 (4.35% ± 1.0%)
0.05 (NS)
Legend: NS, Not significant. For other abbreviationssee Table II.
Table VII. Valve-related complications Group Less than 20 years Total No. of survivors Total follow-up Range Valve-related complication Valve thrombosis Thromboembolism Fatal
122 409 pt-yr
3 mo-B yr
15 (4.13% ± 1.04%) (1.2% PPY)
o
1 (0.27% (0.08% 3 (0.82% (0.24% 3 (0.82% (0.24% 3 (0.82% (0.24% 3 (0.82% (0.24% 27 (7.43% (2.15%
Prosthetic valve endocarditis
o
± 0.81 %) PPY) ± 0.81%) PPY)
Peri prosthetic leak
1 (0.82% (0.24% 3 (2.45% (0.73%
± 0.81%) PPY) ± 1.33%) PPY)
Anticoagulant-related hemorrhage
Total
363 1,253 pt-yr 3 mo-S yr
o
1 (0.82% (0.24% 1 (0.82% (0.24%
Nonfatal
More than 20 years
± 0.27%) PPY) ± 0.47%) PPY) ± 0.47%) PPY) ± 0.47%) PPY) ± 0.47%) PPY) ± 1.38%) PPY)
Legend: PPY. Per patient-year.
common valvesize used in the mitral position was the 27 mln valve in both groups, whereas in the aortic position the 21 mm valve was most frequently used. In the majority of young patients, the cardiomegaly produced by the underlying disease process permitted use of adult-sized valves. At the time of referral for operation, most patients were in advanced states of cardiac disability. One hundred eighteen (86.7%) patients were in New York
Heart Association (NYHA) Class III or IV before the operation. Following the operation, all patients were placed on a strict anticoagulant regimen with acenocoumarin (Sintrom), starting on the first postoperative day. Individual dosages were titrated and prothrombin time was monitored at frequent intervals. For the past 3 years we have used an antiplatelet agent, dipyridamole, in addition to acenocoumarin.
The Journal of Thoracic and Cardiovascular Surgery
2 2 0 Iyer et al.
-AVR . - .. MVR •.••• , OVR
0/0
100
i ± 25.0
90
........ 89·2 t
---.t:;:------+..----~=-----
6· 6 %
81·8 ± 6'9 % 81'0 ± 5· 9 %
80 70 60
47 52 23
0
32 33 14
39 38 17
26 27 12
12 15 4
20 21 8
3 4 5 6 YEARS FOLLOW-UP
2
4 AVR 3 MVR o bVR.
10 8 1 7
8
Fig. 1. Actuarial survival curves inclusive of hospital mortality for patients under the age of 20 years. Cross-bars represent ± 2 standard deviations from the mean. Figures along the base indicate number of survivors entering each year of follow-up. AVR. Aortic valve replacement. MVR. Mitral valve replacement. DVR. Double valve replacement. AVR - -.. MVR •·.. ·... DVR
90
g---y-----4------+-----+----490.,:t 3· '0/.
----.-----+---------+----.--- -+-----.---- -1"'" ,."" ~ ~ + ·····1 '0".
: + 80
a
108 84 206 164 49 41
70 130 35 2
+-
+
63 106 29
48 85 21
33 64 14
3 4 5 YEARS FOLLOW-UP
AVR MVR DVR
7 11 3
20 37 7 6
o
7
8
Fig. 2. Actuarial survival curves inclusive of hospital mortality for patients over the age of 20 years. Cross-bars represent ± 2 standard deviations from the mean. Figures along the base indicate number of survivors entering each year of follow-up. For abbreviations see Fig. 1.
Patients have been followed up by clinical evaluation, chest roentgenogram, and electrocardiogram periodically at an average interval of 6 months. For the past 3 years all patients have been subjected to periodic cinefluoroscopic and echocardiographic assessments to evaluate valve function and improvement in cardiac status.
Results Overall operative mortality for the entire group was 11.65% ± 1.36%. As indicated in Table V, there was no
significant difference in the operative mortality between the two groups of patients. The majority of the deaths were due to low cardiac output in the perioperative period.
Volume 88
Valve replacement in children 2 2 1
Number 2 August, 1984
Over a follow-up period of 3 months to 8 years, there were six late deaths in the younger age group (4.41%) and 18 late deaths in the adult patients (4.35%). Late mortality did not significantly differ in the two groups (Table VI). None of the late deaths in the younger age group could be attributed to valve-related causes, whereas in the older age group two deaths resulted from valve-related causes--one cerebral embolus and one thrombosed mitral valve prosthesis. Actuarial survival curves up to 8 years of follow-up are depicted in Figs. 1 and 2. Long-term survival was comparable in both groups. Most of the late deaths occurred in the first 2 years after operation, with the actuarial survival curves reaching a plateau by the third year of follow-up. Of the 116 survivors in the <20 age group, 104 were in NYHA Class I and 12 in NYHA Class II (Fig. 3). No patient remained in Class III or IV. This subjective improvement was accompanied by a marked reduction in cardiomegaly in all patients (Fig. 4) and echocardiographic evidence of normal left ventricular function in the majority. Valve-related complications. Valve-related complications were compared in the two groups and are given in Table VII. There were no instances of valve thrombosis in the younger age group but 15 instances in the adult patients. This difference was strikingly significant. No case of structural valve failure occurred in either group. There were no instances of fatal embolization in the <20 age group, whereas one adult patient had a fatal cerebral embolism 2 years after operation. Nonfatal embolism, anticoagulant-related hemorrhage, and periprosthetic leak occurred with equal frequency in both groups. Prosthetic valve endocarditis, however, was not observed in the young. At 8 years, the embolus-free survival rate was 98.8% ± 1.1% for the young patients and 98% ± 0.72% for the adults (Fig. 5). The reoperation-free survival rates at 8 years were 99% ± 0.88% and 92.8% ± 1.9%, respectively (Fig. 6).
Discussion Continuing disenchantment with the durability of tissue valves in children has again brought to the fore the once exhaustively debated subject of prosthetic valve replacement in the young. The excellent early results initially reported with the use of tissue valvesv' and the freedom from anticoagulation, which was a significant advantage, prompted the widespread use of these prostheses in young patients. However, instances of tissue valve failure owing to calcific degeneration are being
AVR
n=50
[D-:~ II
III IV
[!Y'~
~,
[II
~4~
[if ~ Yr; [£]
[If 0
PRE-OP.
DVR
MVR
n=45
POST-OP.
,
36 10
PRE-OP.
3
[I] [II POST-OP.
n=21
19
~:~
~,'~
PRE-OP.
POST-OP.
Fig. 3. Preoperative and postoperative New York Heart Association functional class of patients under 20 years of age.
reported in increasing numbers'" and have necessitated a reevaluation of the various valve substitutes being used in children. Data on the use of individual prostheses in large groups of young patients are scarce. It is in this context that this report on valve replacement with the Bjork-Shiley valve in 136 patients below the age of 20 years assumes relevance. With improving expertise in cardiopulmonary bypass and cardiac intensive care, the mortality of valve replacement in children is no greater than in adults, as reported by many centers"!' and also substantiated in this report. The early mortality for the younger patients was 10.29% compared to 12.10% for the adult patients. This observation becomes more significant when viewed in the context of the fact that cardiologists often recommend valve replacement in children at a stage of greater clinical disability or hemodynamic severity than in adults with comparable lesions. The reason is their apprehension regarding the long-term durability of valves implanted in children, management of anticoagulation, and other prosthesis-related complications. The Bjork-Shiley valve has proved its long-term durability. Instances of valve failure because of strut fracture or disc wear are rare. Thrombotic obstruction of the valve and thromboembolism remain the most dreaded complications of this valve. Thrombotic obstruction did not occur in any of the young patients, whereas in the adult patients this complication occurred at a rate of 1.2 events per 100 patient-years. The reason for this observation has not been elucidated as yet. Low rates of thrombotic obstruction of this valve in younger patients have also been reported by other workers.v" but no satisfactory explanation has been offered. Higher mean resting heart rates with more frequent disc motion in the younger patients and a greater degree of thera-
2 2 2 Iyer et a/.
The Journal of Thoracic and Cardiovascular Surgery
Fig. 4. A, Preoperative (left) and postoperative (right) x-ray films of a IO-year-oldgirl who underwent mitral valve replacement for rheumatic heart disease. There is marked reduction in cardiomegaly I year after operation. B, Preoperative (left) and postoperative (right) x-ray films of a 15-year-old boy who underwent aortic and mitral valve replacement. There is marked reduction in cardiomegaly I year after operation.
peutic compliance because of parental supervision are two possible but speculative causes. Anticoagulation control was not difficult to achieve in children, and anticoagulation-related hemorrhage occurred in only one patient, who had hematuria. The bleeding stopped soon after temporary discontinuation
of anticoagulation, and the prothrombin time was regulated more strictly thereafter. The fear that children respond erratically to anticoagulation seems largely unfounded, in our experience. We believe that the problem of patient-prosthesis mismatch has been more discussed than actually
Volume 88
Valve replacement in children 223
Number 2 August. 1984
0/0
100
----__ ~ __ 98· 8 ± 1.1 / 0 98:!: 0.72 0 / 0 0
~-..::..:::t=_=~~:::.:.:.:~:..:.:~:.:.:=+:...:.:~~.:.:::t
90 - - ..... :<20YRS _ : > 20YRS ± 1S.D.
I :
80
1
12 2 363
95
66 198
81 235
289
O-__r----r--"'T"""--.,...---r----.--~____r 2
3
49 1S4
4
19 64
31 111
5
6
7
21
7
8
YEARS FOLLOW UP
Fig. 5. Actuarial embolus-free survival in the two groups. Cross-bars represent ± I standard deviation from the mean.
0
/0
100 - -..... - - - - 0 - - - -
-0--- - -1-- ---0---- -I- ---
-0-- -
-199.0
--------t
± 0'88 0 / 0
92'8 :!:1.9
0/0
90 20 YRS -:')20YRS ± 1SD
_---o:l.
I:
8 Fig. 6. Actuarial reoperation-free survival in the two groups. Cross-bars represent ± 1 standard deviation from the mean.
observed. Contrary to the observations of LeVY,13 Smeloff," Friedman, 15 and their associates, none of our patients have required reoperation for this problem. Like others, 11.16 we have found that cardiomegaly resulting from the underlying valvular disease has allowed the use of adult-sized valves in majority of patients. Comparison of the valve sizes used in both groups of patients showed no significant difference in the proportions of the various valve sizes used (Table IV). This could explain why no instance of patient-prosthesis mismatch has so far been
documented on periodic follow-up with chest roentgenograms, electrocardiograms, and echocardiograms. Although the tissue valves assure a certain degree of freedom from anticoagulation, they are not altogether free from the risk of thromboembolism. Even with tissue valves, routine anticoagulation is advocated for periods of 6 weeks to 3 months postoperatively, and patients with atrial fibrillation or a large left atrium need more prolonged anticoagulation. However, their long-term durability is currently the greatest cause of concern.
2 2 4 Iyer et al.
Early calcific degeneration is being reported at prohibitive rates. Williams and associates' have reported a valve failure rate of 7.1% per patient-year with a predicted replacement-free survival rate of only 58.5% ± 15.7%at 5 years. Since then, they have discontinued the use of porcine valve heterografts in children. Similarly, Geha,' Wada," and their associates reported valve failure rates of 20% and 30%, respectively. This problem has been observed frequently at the Texas Heart Institute, which advocates only mechanical valves for use in the younger age group." While the causes of early calcification are being actively researched, the problem continues to cause concern. While the questionable long-term durability of tissue valves has considerably dampened their popularity as the preferred valve substitutes in young patients, our gratifying experience with the Bjork-Shiley valve, both in terms of operative mortality and valverelated complications, suggests that this valve is a viable alternative for cardiac valve replacement in young patients.
2 3
4
5
6
7
REFERENCES Roy SB, Bhatia ML, Lazaro ET, Ramalingaswami V: Juvenile mitral stenosis in India. Lancet 2:1193-1195, 1963 Roy SB, Gopinath N: Mitral stenosis. Circulation 37, 38:Suppl 5:68, 1968 Stinson EB, Griepp RB, Oyer PE, Shumway NE: Longterm experience with porcine aortic valve xenografts. J THoMc CARDIOVASC SURG 73:54-63, 1977 Cohn LH, Sanders JH Jr, Collins 11 Jr: Aortic valve replacement with the Hancock porcine xenograft. Ann Thorac Surg 22:221-227, 1976 Williams DB, Danielson GK, McGoon DC, Puga FJ, Mair DD, Edwards WD: Porcine heterograft valve replacement in children. J THORAC CARDIOVASC SURG 84:446-450, 1982 Wada J, Yokoyama M, Hashimoto A, Imai Y, Kitamura N, Takao A, Momma K: Long-term follow-up of artificial valves in patients under 15 years old. Ann Thorac Surg 29:519-521, 1980 Geha AS, Laks H, Stansel HC, Cornhill JF, Kelman JR,
The Journal of Thoracic and Cardiovascular Surgery
Buckley MJ, Roberts WC: Late failure of porcine heterografts in children. J THORAC CARDIOVASC SURG 78:351364, 1979 8 Silver MM, Pollock J, Silver MD, Williams WG, Trusler GA: Calcification in porcine xenograft valves in children. Am J Cardiol 45:685-689, 1980 9 Rufilanchas 11, Juffe A, Miranda AL, Tellez G, Arosti J, Maronas JM, Figuera D: Cardiac valve replacement with the Bjork-Shiley valve in young patients. Scand J Thorac Cardiovasc Surg 11:11-14, 1977 10 Freed MD, Bernhard WF: Prosthetic valve replacement in children. Prog Cardiovasc Dis 7:475-487, 1975 11 Berry BE, Ritter DG, Wallace RB, McGoon DC, Danielson GK: Cardiac yalve replacement in children. J THoRAc CARDIOVASC SURG 68:705-710, 1974 12 John S, Munsi SC, Gupta RP, Ramachandran V, Milledge JS, Sukumar IP, Cherian G: Results of mitral valve replacement in young patients with rheumatic heart disease. J THORAC CARDIOVASC SURG 66255, 1973 13 Levy MJ, Vidne B: Rheumatic valve disease in childhood and adolescence. Surgical treatment. Chest 63:933-936, 1973 14 Smeloff EA, Cayler GG, Smith DF: The use of valve prosthesis in childhood. J THoRAc CARDIOVASC SURG 51:839-843, 1966 15 Friedman S, Edmunds LH Jr, Cugso CC: Long-term mitral valve replacement in young children. Influence of somatic growth on prosthetic valve adequacy. Circulation 57:981-986, 1978 16 Klint R, Hernandez A, Weldon C, Hartmann AF Jr, Goldring D: Replacement of cardiac valves in children. J Pediatr 80:980-987, 1972 17 Cooley DA: Results with Bjork-Shiley tilting disc and Ionescu-Shiley bioprosthesis. Bull Texas Heart Inst 9:467471, 1982 18 Ferrans VJ, Spray TL, Billingham ME, Roberts WC: Structural changes in glutaraldehyde-treated porcine heterografts used as substitute cardiac valves. Transmission and scanning electron microscopic observation in 12 patients. Am J Cardiol 41:1159-1184, 1978 19 Clark RE, Swanson WM, Kardos JL, Hagen RW, Beauchamp RH: Durability of prosthetic heart valves. Ann Thorac Surg 26:323-335, 1978
THORAC CARDIOVASC SURG
88:217-224, 1984
Valve replacement in children under twenty years of age Experience with the Bjork-Shiley prosthesis Prosthetic valve replacement in young patients bas been reported to be associated with a high mortality and morbidity because of valve-related problems. Of 549 patients tmdergoing valve replacement with the Bjork-5hi1ey valve prosthesis, 136 were under the ageof 20 years. Sixty-four patients wereunder 16 years of age, the youngest being 6 years old. Of the 136 patients, 61 underwent mitral valve replacemen~ 50 received an aortic valvt\ and 25 received both aortic and mitral valves. Overall operative mortality was 10.3%. Late mortality overa follow-up period of 6 montm to 8 years was 4.4 %. Actuarial survival curves up to 8 years of follow-up are presented.Results obtained in this group are compared with those obtained in 413 patients over 20 years of age operated during the same period. Valve thrombosis was not seen in any patient under 20 years of agt\ but it occurred in 4.13 % of the patients over 20 years of age. The incidence of thromboembotism and anticoagulant-related hemorrhage was very low. There bas been no imtance of structural failure of the valve. l.ong-term results are excellent, with 90 % of the survivors returning to New York Heart Association Functional Class I. The Bjork-Sbileyvalve gives excellent and durable long-term palliation in young patients requiring valve replacement.
Krishna Subramony Iyer, M.D., K. Srinath Reddy, M.D., I. Mritunjaya Rao, M.D., P. Venugopal, M.D., M. L. Bhatia, M.D., and N. Gopinath, M.D.,
New Delhi, India
h e controversy over the choice of ideal valve substitutes for replacing diseased valves in children continues to surface frequently. The introduction of glutaraldehyde-treated porcine heterografts appeared at one time to provide the solution to this problem. However, the alarming incidence of early valve failure because of degenerative calcification' has raised doubts regarding the continued use of these valves in young patients, and current trends seem to favor mechanical valves. Reports on large series of children undergoing valve repiacement with a particular valve are scant in Western literature, owing possibly to the paucity of acquired valvular heart disease in young patients in the West. In India, however, From the Department of Cardio-Thoracic and Vascular Surgery and the Department of Cardiology, All India Institute of Medical Sciences, Ansari Nagar, New Delhi, India. Received for publication Aug. 22, 1983. Accepted for publication Oct. 11, 1983. Address for reprints: Professor P. Venugopal, Head, Department of Cardio-Thoracic and Vascular Surgery, All India Institute of Medical Sciences, Ansari Nagar. New Delhi-l 10029, India.
rheumatic heart disease (RHO) continues to be a major contributor to valvular heart disease, especially in the young. RHD is diagnosed in 30% to 40% of the patients attending the cardiac clinic at the All India Institute of Medical Sciences Hospital. One fourth of these patients are under 20 years of age. The entities of "juvenile mitral stenosis" and "juvenile RHO," first described by Roy and associates'< in 1963, continue to influence the clinical profile of RHD in this country. Vast numbers of patients with RHO require surgical intervention. Though valve reconstruction theoretically would seem to be the ideal surgical procedure in these patients, the extensive valvular deformities that frequently occur in juvenile RHO preclude such reparative procedures and necessitate valve replacement. In a developing country like India, where cardiac operations impose a major fmancial burden on the patient, a valve with a low reoperation rate due to valve failure is needed. It has been our policy, therefore, to use a mechanical valve in all young patients. Since 1975, a fairly large number of patients under the age of 20 years 217
The Journal of Thoracic and Cardiovascular
2 1 8 Iyer et al.
Surgery
Table I. Site ofprosthetic valve replacement
Table IV. Comparison of valve sizes
No. of patients
Patients
Site
Less than 20 years
More than 20 years
Aortic Mitral Aortic + mitral
50 61 25
116 231 66
136
413
Total*
'Total No. of patients = 549; total No. of valves implanted = 640.
Table D. Associated procedures performed (under 20 years)
VSD closure OMC OMC + De Vega repair of tricuspid valve De Vega repair of tricuspid valve ASD closure Aortic root enlargement Others
Legend: AVR, Aortic valve replacement. MVR, Mitral valve replacement. DVR, Double valve replacement. VSD, Ventricular septal defect. OMC, Open mitral commissurotomy. ASD, Atrial septal defect.
Table ill. Pathology of valvular lesion (under 20 years) AVR
MVR
DVR
Pathology RHD Pure AR AS+AR AR+MS AR+MS+TR CHD AS VSD+AR RHD MS (restenosis) MS+MR Pure MR CHD ASD+MVPS Myxomatous mitral valve RHD Total
Less than 20 years
More than 20 years
Aortic 19 21 23 25 27
21 (28%) 35 (46.7%) 15 (20.0%) 3 (4.0%) I (1.3%)
23 (12.63%) 76 (41.76%) 50 (27.5%) 29 (15,93%) 4 (2.1%)
Total
75
182
Mitral 25 27 29 Total
19122.01%) 57 (66.2%) 10 (11.6%) 86
82 (27.6%) 185 (62.2%) 30 (10.1%) 297
have undergone valve replacement, and this report analyzes our experience with the Bjork-Shiley valve in this age group. The operative results and long-term function of this valve in the younger age group are compared with results obtained in adult patients over the same period of time to highlight the excellent results obtained with the Bjork-Shiley valve in children.
Total
Group
Size (mm)
No. of cases 31 20 3 6 2 19 II 8
58 2 45 II 3 2 I 25 RHD: 114 (83.3%) CHD: 22 (16.2%)
Legend: RHD, Rheumatic heart disease. CHD, Congenital heart disease. AR, Aortic regurgitation. AS, Aortic stenosis. MS, Mitral stenosis. TR, Tricuspid regurgitation. MR, Mitral regurgitation. MVPS, Mitral valve prolapse syndrome. For other abbreviations see Table II.
Patient groups and methods Between January, 1975, and December, 1982, 549 patients underwent valve replacement with the BjorkShiley valve for various valvular disorders. Of these patients, 136 were below 20 years of age. The youngest patient was 6 years old and 64 patients were under 16 years of age. Sites of valve replacement are indicated in Table I. There were proportionately more patients undergoing isolated mitral valve replacement in the adult age group. All patients received a Bjork-Shiley valve with a pyrolytic carbon disc, the standard model with a flat disc before 1979 and the currently used convexo-concave model since then. Valve replacement was performed with routine cardiopulmonary bypass and moderate hypothermia. A modified cold cardioplegic perfusate (composition: NaHC0 3 50 mEq, KCI 15 mEq, and dexamethasone 8 mg in 500 ml of perfusate) is routinely used currently. Valve replacement alone was performed in 107 of the younger patients. In 29 patients, procedures in addition to valve replacement were done (Table II). As indicated in Table III, RHD (83.8%) predominated as the underlying etiologic factor, with congenital heart disease (16.2%) contributing to the rest. Valve sizes used in the two groups were tabulated and compared and are presented in Table IV. The most
Volume 88 Number 2 August, 1984
Valve replacement in children
2 19
Table V. Early mortality Patients Group
Less than 20 years
MVR AVR DVR
9/61 (14.75% ± 4.54%) 3/50 ( 6.0% ± 3.36%) 2/25 ( 8.0% ± 5.43%)
Total
14/136 (10.29% ± 2.60%)
I
More than 20 years
x'
25/231 (10.82% ± 2.04%) 8/116 ( 6.89% ± 2.35%) 17/66 (25.75% ± 5.38%)
0.39 (NS) 0.02 (NS) 2.47 (NS)
50/413 (12.10% ± 1.6%)
0.17 (NS)
Legend: NS, Not significant. For other abbreviationssee Table II.
Table VI. Late mortality Group
Less than 20 years
More than 20 years
MVR AVR DVR
2/61 (3.27% ± 2.27%) 2/50 (4.0% ± 2.77%) 2/25 (8.0% ± 5.43%)
12/231 (5.19% ± 1.45%) 2/116 (1.72% ± 1.2%) 4/66 (6.06% ± 2.94%)
0.08 (NS) 0.11 (NS) 0.02 (NS)
Total
6/136 (4.41% ± 1.76%)
18/413 (4.35% ± 1.0%)
0.05 (NS)
Legend: NS, Not significant. For other abbreviationssee Table II.
Table VII. Valve-related complications Group Less than 20 years Total No. of survivors Total follow-up Range Valve-related complication Valve thrombosis Thromboembolism Fatal
122 409 pt-yr
3 mo-B yr
15 (4.13% ± 1.04%) (1.2% PPY)
o
1 (0.27% (0.08% 3 (0.82% (0.24% 3 (0.82% (0.24% 3 (0.82% (0.24% 3 (0.82% (0.24% 27 (7.43% (2.15%
Prosthetic valve endocarditis
o
± 0.81 %) PPY) ± 0.81%) PPY)
Peri prosthetic leak
1 (0.82% (0.24% 3 (2.45% (0.73%
± 0.81%) PPY) ± 1.33%) PPY)
Anticoagulant-related hemorrhage
Total
363 1,253 pt-yr 3 mo-S yr
o
1 (0.82% (0.24% 1 (0.82% (0.24%
Nonfatal
More than 20 years
± 0.27%) PPY) ± 0.47%) PPY) ± 0.47%) PPY) ± 0.47%) PPY) ± 0.47%) PPY) ± 1.38%) PPY)
Legend: PPY. Per patient-year.
common valvesize used in the mitral position was the 27 mln valve in both groups, whereas in the aortic position the 21 mm valve was most frequently used. In the majority of young patients, the cardiomegaly produced by the underlying disease process permitted use of adult-sized valves. At the time of referral for operation, most patients were in advanced states of cardiac disability. One hundred eighteen (86.7%) patients were in New York
Heart Association (NYHA) Class III or IV before the operation. Following the operation, all patients were placed on a strict anticoagulant regimen with acenocoumarin (Sintrom), starting on the first postoperative day. Individual dosages were titrated and prothrombin time was monitored at frequent intervals. For the past 3 years we have used an antiplatelet agent, dipyridamole, in addition to acenocoumarin.
The Journal of Thoracic and Cardiovascular Surgery
2 2 0 Iyer et al.
-AVR . - .. MVR •.••• , OVR
0/0
100
i ± 25.0
90
........ 89·2 t
---.t:;:------+..----~=-----
6· 6 %
81·8 ± 6'9 % 81'0 ± 5· 9 %
80 70 60
47 52 23
0
32 33 14
39 38 17
26 27 12
12 15 4
20 21 8
3 4 5 6 YEARS FOLLOW-UP
2
4 AVR 3 MVR o bVR.
10 8 1 7
8
Fig. 1. Actuarial survival curves inclusive of hospital mortality for patients under the age of 20 years. Cross-bars represent ± 2 standard deviations from the mean. Figures along the base indicate number of survivors entering each year of follow-up. AVR. Aortic valve replacement. MVR. Mitral valve replacement. DVR. Double valve replacement. AVR - -.. MVR •·.. ·... DVR
90
g---y-----4------+-----+----490.,:t 3· '0/.
----.-----+---------+----.--- -+-----.---- -1"'" ,."" ~ ~ + ·····1 '0".
: + 80
a
108 84 206 164 49 41
70 130 35 2
+-
+
63 106 29
48 85 21
33 64 14
3 4 5 YEARS FOLLOW-UP
AVR MVR DVR
7 11 3
20 37 7 6
o
7
8
Fig. 2. Actuarial survival curves inclusive of hospital mortality for patients over the age of 20 years. Cross-bars represent ± 2 standard deviations from the mean. Figures along the base indicate number of survivors entering each year of follow-up. For abbreviations see Fig. 1.
Patients have been followed up by clinical evaluation, chest roentgenogram, and electrocardiogram periodically at an average interval of 6 months. For the past 3 years all patients have been subjected to periodic cinefluoroscopic and echocardiographic assessments to evaluate valve function and improvement in cardiac status.
Results Overall operative mortality for the entire group was 11.65% ± 1.36%. As indicated in Table V, there was no
significant difference in the operative mortality between the two groups of patients. The majority of the deaths were due to low cardiac output in the perioperative period.
Volume 88
Valve replacement in children 2 2 1
Number 2 August, 1984
Over a follow-up period of 3 months to 8 years, there were six late deaths in the younger age group (4.41%) and 18 late deaths in the adult patients (4.35%). Late mortality did not significantly differ in the two groups (Table VI). None of the late deaths in the younger age group could be attributed to valve-related causes, whereas in the older age group two deaths resulted from valve-related causes--one cerebral embolus and one thrombosed mitral valve prosthesis. Actuarial survival curves up to 8 years of follow-up are depicted in Figs. 1 and 2. Long-term survival was comparable in both groups. Most of the late deaths occurred in the first 2 years after operation, with the actuarial survival curves reaching a plateau by the third year of follow-up. Of the 116 survivors in the <20 age group, 104 were in NYHA Class I and 12 in NYHA Class II (Fig. 3). No patient remained in Class III or IV. This subjective improvement was accompanied by a marked reduction in cardiomegaly in all patients (Fig. 4) and echocardiographic evidence of normal left ventricular function in the majority. Valve-related complications. Valve-related complications were compared in the two groups and are given in Table VII. There were no instances of valve thrombosis in the younger age group but 15 instances in the adult patients. This difference was strikingly significant. No case of structural valve failure occurred in either group. There were no instances of fatal embolization in the <20 age group, whereas one adult patient had a fatal cerebral embolism 2 years after operation. Nonfatal embolism, anticoagulant-related hemorrhage, and periprosthetic leak occurred with equal frequency in both groups. Prosthetic valve endocarditis, however, was not observed in the young. At 8 years, the embolus-free survival rate was 98.8% ± 1.1% for the young patients and 98% ± 0.72% for the adults (Fig. 5). The reoperation-free survival rates at 8 years were 99% ± 0.88% and 92.8% ± 1.9%, respectively (Fig. 6).
Discussion Continuing disenchantment with the durability of tissue valves in children has again brought to the fore the once exhaustively debated subject of prosthetic valve replacement in the young. The excellent early results initially reported with the use of tissue valvesv' and the freedom from anticoagulation, which was a significant advantage, prompted the widespread use of these prostheses in young patients. However, instances of tissue valve failure owing to calcific degeneration are being
AVR
n=50
[D-:~ II
III IV
[!Y'~
~,
[II
~4~
[if ~ Yr; [£]
[If 0
PRE-OP.
DVR
MVR
n=45
POST-OP.
,
36 10
PRE-OP.
3
[I] [II POST-OP.
n=21
19
~:~
~,'~
PRE-OP.
POST-OP.
Fig. 3. Preoperative and postoperative New York Heart Association functional class of patients under 20 years of age.
reported in increasing numbers'" and have necessitated a reevaluation of the various valve substitutes being used in children. Data on the use of individual prostheses in large groups of young patients are scarce. It is in this context that this report on valve replacement with the Bjork-Shiley valve in 136 patients below the age of 20 years assumes relevance. With improving expertise in cardiopulmonary bypass and cardiac intensive care, the mortality of valve replacement in children is no greater than in adults, as reported by many centers"!' and also substantiated in this report. The early mortality for the younger patients was 10.29% compared to 12.10% for the adult patients. This observation becomes more significant when viewed in the context of the fact that cardiologists often recommend valve replacement in children at a stage of greater clinical disability or hemodynamic severity than in adults with comparable lesions. The reason is their apprehension regarding the long-term durability of valves implanted in children, management of anticoagulation, and other prosthesis-related complications. The Bjork-Shiley valve has proved its long-term durability. Instances of valve failure because of strut fracture or disc wear are rare. Thrombotic obstruction of the valve and thromboembolism remain the most dreaded complications of this valve. Thrombotic obstruction did not occur in any of the young patients, whereas in the adult patients this complication occurred at a rate of 1.2 events per 100 patient-years. The reason for this observation has not been elucidated as yet. Low rates of thrombotic obstruction of this valve in younger patients have also been reported by other workers.v" but no satisfactory explanation has been offered. Higher mean resting heart rates with more frequent disc motion in the younger patients and a greater degree of thera-
2 2 2 Iyer et a/.
The Journal of Thoracic and Cardiovascular Surgery
Fig. 4. A, Preoperative (left) and postoperative (right) x-ray films of a IO-year-oldgirl who underwent mitral valve replacement for rheumatic heart disease. There is marked reduction in cardiomegaly I year after operation. B, Preoperative (left) and postoperative (right) x-ray films of a 15-year-old boy who underwent aortic and mitral valve replacement. There is marked reduction in cardiomegaly I year after operation.
peutic compliance because of parental supervision are two possible but speculative causes. Anticoagulation control was not difficult to achieve in children, and anticoagulation-related hemorrhage occurred in only one patient, who had hematuria. The bleeding stopped soon after temporary discontinuation
of anticoagulation, and the prothrombin time was regulated more strictly thereafter. The fear that children respond erratically to anticoagulation seems largely unfounded, in our experience. We believe that the problem of patient-prosthesis mismatch has been more discussed than actually
Volume 88
Valve replacement in children 223
Number 2 August. 1984
0/0
100
----__ ~ __ 98· 8 ± 1.1 / 0 98:!: 0.72 0 / 0 0
~-..::..:::t=_=~~:::.:.:.:~:..:.:~:.:.:=+:...:.:~~.:.:::t
90 - - ..... :<20YRS _ : > 20YRS ± 1S.D.
I :
80
1
12 2 363
95
66 198
81 235
289
O-__r----r--"'T"""--.,...---r----.--~____r 2
3
49 1S4
4
19 64
31 111
5
6
7
21
7
8
YEARS FOLLOW UP
Fig. 5. Actuarial embolus-free survival in the two groups. Cross-bars represent ± I standard deviation from the mean.
0
/0
100 - -..... - - - - 0 - - - -
-0--- - -1-- ---0---- -I- ---
-0-- -
-199.0
--------t
± 0'88 0 / 0
92'8 :!:1.9
0/0
90 20 YRS -:')20YRS ± 1SD
_---o:l.
I:
8 Fig. 6. Actuarial reoperation-free survival in the two groups. Cross-bars represent ± 1 standard deviation from the mean.
observed. Contrary to the observations of LeVY,13 Smeloff," Friedman, 15 and their associates, none of our patients have required reoperation for this problem. Like others, 11.16 we have found that cardiomegaly resulting from the underlying valvular disease has allowed the use of adult-sized valves in majority of patients. Comparison of the valve sizes used in both groups of patients showed no significant difference in the proportions of the various valve sizes used (Table IV). This could explain why no instance of patient-prosthesis mismatch has so far been
documented on periodic follow-up with chest roentgenograms, electrocardiograms, and echocardiograms. Although the tissue valves assure a certain degree of freedom from anticoagulation, they are not altogether free from the risk of thromboembolism. Even with tissue valves, routine anticoagulation is advocated for periods of 6 weeks to 3 months postoperatively, and patients with atrial fibrillation or a large left atrium need more prolonged anticoagulation. However, their long-term durability is currently the greatest cause of concern.
2 2 4 Iyer et al.
Early calcific degeneration is being reported at prohibitive rates. Williams and associates' have reported a valve failure rate of 7.1% per patient-year with a predicted replacement-free survival rate of only 58.5% ± 15.7%at 5 years. Since then, they have discontinued the use of porcine valve heterografts in children. Similarly, Geha,' Wada," and their associates reported valve failure rates of 20% and 30%, respectively. This problem has been observed frequently at the Texas Heart Institute, which advocates only mechanical valves for use in the younger age group." While the causes of early calcification are being actively researched, the problem continues to cause concern. While the questionable long-term durability of tissue valves has considerably dampened their popularity as the preferred valve substitutes in young patients, our gratifying experience with the Bjork-Shiley valve, both in terms of operative mortality and valverelated complications, suggests that this valve is a viable alternative for cardiac valve replacement in young patients.
2 3
4
5
6
7
REFERENCES Roy SB, Bhatia ML, Lazaro ET, Ramalingaswami V: Juvenile mitral stenosis in India. Lancet 2:1193-1195, 1963 Roy SB, Gopinath N: Mitral stenosis. Circulation 37, 38:Suppl 5:68, 1968 Stinson EB, Griepp RB, Oyer PE, Shumway NE: Longterm experience with porcine aortic valve xenografts. J THoMc CARDIOVASC SURG 73:54-63, 1977 Cohn LH, Sanders JH Jr, Collins 11 Jr: Aortic valve replacement with the Hancock porcine xenograft. Ann Thorac Surg 22:221-227, 1976 Williams DB, Danielson GK, McGoon DC, Puga FJ, Mair DD, Edwards WD: Porcine heterograft valve replacement in children. J THORAC CARDIOVASC SURG 84:446-450, 1982 Wada J, Yokoyama M, Hashimoto A, Imai Y, Kitamura N, Takao A, Momma K: Long-term follow-up of artificial valves in patients under 15 years old. Ann Thorac Surg 29:519-521, 1980 Geha AS, Laks H, Stansel HC, Cornhill JF, Kelman JR,
The Journal of Thoracic and Cardiovascular Surgery
Buckley MJ, Roberts WC: Late failure of porcine heterografts in children. J THORAC CARDIOVASC SURG 78:351364, 1979 8 Silver MM, Pollock J, Silver MD, Williams WG, Trusler GA: Calcification in porcine xenograft valves in children. Am J Cardiol 45:685-689, 1980 9 Rufilanchas 11, Juffe A, Miranda AL, Tellez G, Arosti J, Maronas JM, Figuera D: Cardiac valve replacement with the Bjork-Shiley valve in young patients. Scand J Thorac Cardiovasc Surg 11:11-14, 1977 10 Freed MD, Bernhard WF: Prosthetic valve replacement in children. Prog Cardiovasc Dis 7:475-487, 1975 11 Berry BE, Ritter DG, Wallace RB, McGoon DC, Danielson GK: Cardiac yalve replacement in children. J THoRAc CARDIOVASC SURG 68:705-710, 1974 12 John S, Munsi SC, Gupta RP, Ramachandran V, Milledge JS, Sukumar IP, Cherian G: Results of mitral valve replacement in young patients with rheumatic heart disease. J THORAC CARDIOVASC SURG 66255, 1973 13 Levy MJ, Vidne B: Rheumatic valve disease in childhood and adolescence. Surgical treatment. Chest 63:933-936, 1973 14 Smeloff EA, Cayler GG, Smith DF: The use of valve prosthesis in childhood. J THoRAc CARDIOVASC SURG 51:839-843, 1966 15 Friedman S, Edmunds LH Jr, Cugso CC: Long-term mitral valve replacement in young children. Influence of somatic growth on prosthetic valve adequacy. Circulation 57:981-986, 1978 16 Klint R, Hernandez A, Weldon C, Hartmann AF Jr, Goldring D: Replacement of cardiac valves in children. J Pediatr 80:980-987, 1972 17 Cooley DA: Results with Bjork-Shiley tilting disc and Ionescu-Shiley bioprosthesis. Bull Texas Heart Inst 9:467471, 1982 18 Ferrans VJ, Spray TL, Billingham ME, Roberts WC: Structural changes in glutaraldehyde-treated porcine heterografts used as substitute cardiac valves. Transmission and scanning electron microscopic observation in 12 patients. Am J Cardiol 41:1159-1184, 1978 19 Clark RE, Swanson WM, Kardos JL, Hagen RW, Beauchamp RH: Durability of prosthetic heart valves. Ann Thorac Surg 26:323-335, 1978