- Email: [email protected]
Heart valve replacement with the Sorin tilting-disc prosthesis
Heart valve replacement with the Sorin tilting-disc prosthesis A 10-year experience From 1978 to 1988, 697 patients with a mean age of 48 ± 11 years (range 5 to 75 years) received a Sorin tilting-disc prosthesis; 358 had had aortic valve replacement, 247 mitral valve replacement, and 92 mitral and aortic valve replacement. Operative mortality rates were 7.8 %, 11.3%, and 10.8%, respectively, in the three groups. Cumulative duration of follow-up is 1650 patient-years for aortic valve replacement (maximum follow-up 11.4 years), 963 patient-years for mitral valve replacement (maximum follow-up 9.9 years) and 328 patient-years for mitral and aortic valve replacement (maximum follow-up 9.4 years). Actuarial survival at 9 years is 72 % ± 4 % after mitral valve replacement, 70 % ± 3 % after aortic valve replacement, and 50 % ± 12 % after mitral and aortic valve replacement, and actuarial freedom from valve-related deaths is 97 % ± 2 % after mitral valve replacement, 92% ± 2% after aortic valve replacement, and 62% ± 15% after mitral and aortic valve replacement. Thromboembolic events occurred in 21 patients with aortic valve replacement (1.3% ± 0.2% / pt-yr), in 12 with mitral valve replacement (1.2% ± 0.3% pt-yr), and in seven with mitral and aortic valve replacement (2.1 % ± 0.8 %), with one case of prosthetic thrombosis in each group; actuarial freedom from thromboembolism at 9 years is 92 % ± 3 % after mitral valve replacement, 91 % ± 3 % after aortic valve replacement, and 74 % ± 16 % after mitral and aortic valve replacement. Anticoagulant-related hemorrhage was observed in 15 patients after aortic valve replacement (0.9% ± 0.2%/pt-yr), in 9 after mitral valve replacement (0.9% ± 0.3%jpt-yr), and in 6 with mitral and aortic valve replacement (0.9% ± 0.5% /pt-yr); actuarial freedom from this complication at 9 years is 94 % ± 2 % after aortic valve replacement, 91 % ± 4 % after mitral valve replacement, and 68% ± 16% after mitral and aortic valve replacement. Actuarial freedom from reoperation at 9 years is 97 % ± 2 % after mitral and aortic valve replacement, 92 % ± 4 % after mitral valve replacement, and 89 % ± 3 % after aortic valve replacement, with no cases of mechanical fracture. The Sorin valve has shown a satisfactory long-term overall performance, comparable with other mechanical prostheses, and an excellent durability that renders it a reliable heart valve substitute for the mitral and aortic positions. (J THORAC CARDIOVASC SURe 1992;103:267-75)
Aldo Milano, MD," Uberto Bortolotti, MD," Alessandro Mazzucco, MD," Eugenio Mossuto, MD," Luca Testolin, MD," Gaetano Thiene, MD,b and Vincenzo Gallucci, MD," Padova, Italy
Llting-diSC mechanical prostheses have been used in the past primarily for replacement of the aortic and mitral From the Departments of Cardiovascular Surgery" and Pathology, b University of Padova Medical School, Padova, Italy. Received for publication July 6, 1990. Accepted for publication Dec. 14, 1990. Address for reprints: U. Bortolotti, MD, lstituto di Chirurgia Cardiavascolare, Universita di Padova, Via Giustiniani, 2, 35128 Padova, Italy.
12/1/28351
valves.':" Among the various models the Sorin prosthesis became available for a clinical trial at our institution in 1978. The purpose of this report is to present our experience with the Sorin prosthesis in patients with single and multiple valve replacement to evaluate its performance in the first decade of its clinical use. Patients and methods Prosthesis description. The Sorin tilting-disc prosthesis (Sorin Biomedica, Saluggia, Italy) is made of a pyrolytic carbon disc, retained and freely tilting to 60 degrees in a cage
267
268
Milano et al.
The Journal of Thoracic and Cardiovascular Surgery
Table I. Preoperative clinical characteristics AVR
Fig. 1. Viewof the Sorin tilting-disc prosthesis(Monocast) for aortic (top) and mitral (bottom) valve replacement.
obtained from a single piece of Stellite, a chrome-cobalt alloy, without welding points between the struts and the cage itself. The standard model (Monocast) (Fig. I) is provided with a knitted Teflon sewing ring; in the more recent model (Carbocast) (Fig. 2), the sewing ring has been coated, where it comes in contact with blood, by a thin film of turbostratic carbon (Carbofilm, Sorin Biomedica, Saluggia, Italy), that is, a particular crystalline structure that gives the material diamond-like characteristics. Patient population. From September 1978 to December 1988,697 patients underwent isolated aortic valve replacement (A VR = 358 patients), isolated mitral valve replacement (MVR = 247 patients), and simultaneous mitral and aortic valve replacement (MAVR = 92 patients). A total of 789 Sorin tilting-disc prostheses (692 Monocast and 97 Carbocast prostheses) were used. There were 416 male and 281 female patients, with a mean age of 48 ± II years (range 5 to 75 years). Main clinical characteristics of the three groups of patients are summarized in Table I. The majority of the patients (over 80%) were in New York Heart Association functional class III or IV at time of operation; the most frequent indication for operation was rheumatic valvular disease in the MVR and MA VR groups and either calcific or myxomatous degenerative disease in the AVR group. Approximately 30% of the patients had undergone pre-
MVR
MAVR
No. of patients 358 247 92 Male/female 281/77 82/165 53/39 Mean age (yr) 48 ± 12 48 ± 11 47 ± II Agerange (yr) 5-75 8-67 17-68 Preoperative functional class NYHA II 117 6 2 NYHA III 180 193 66 NYHA IV 61 48 21 Preoperative rhythm 48 SR 336 30 19 194 AF 62 PM 3 5 Etiology 145 Rheumatic 96 50 Myxomatous degeneration 91 29 6 Calcific degeneration 85 8 Aortic dissection 25 Congenital 10 7 I 14 4 Endocarditis 6 Ischemic 7 Prosthesis dysfunction 37 55 21 Associated procedures 131 69 20 Replacement of ascending aorta* 59 I CABG 38 15 4 Tricuspid annuloplasty 38 8 Mitralvalvuloplasty 13 Aortic valvuloplasty II Enlargement of aortic anulus 5 5 LV aneurysmectomy 2 Others 16 2 3 Previous cardiac operations 51 124 34 AYR, Aortic valve replacement; MYR, mitral valve replacement; MAYR. mitral and aortic valve replacement; NHYA, New York Heart Association; SR, sinus rhythm; AF, atrial fibrillation; PM, pacemaker; CABG, coronary artery bypass graft; LY, left ventricular. • Bentall" procedure in 51 and graft replacement ofthe aortic root in eight.
vious intracardiac procedures; 12% of the patients with AVR had had a previous operation on the aortic valve, 47% of the patients with MVR had had a previousoperation on the mitral valve,and 25%of those with MA VR had undergone a previous procedure on the aortic, mitral, or both valves.Operations associated with valve replacement were performed in 32% of cases, consisting mainly in replacement of the aortic root in AVR patients and tricuspid annuloplasty in MVR and MA VR patients. Surgical technique. All patients were operated on with the use of moderately hypothermic cardiopulmonary bypass, with topical cooling and cold potassium cardioplegic solution. Although the techniques varied among the different surgeons, prosthesis insertion was preferentially accomplished by means of mattress sutures, buttressed by Teflon felt pledgets placed in the supraannular position.Prostheses were usually oriented with their large opening corresponding to the convexity of the ascending aorta in AVR and to the posterior left ventricular wall in MVR. Patients undergoing replacement of the aortic valve and ascending aorta were treated as classically described by Bentall and De Bono.t with a Sorin valved conduit; when this
Volume 103 Number 2 February 1992
269
Sorin tilting-disc prosthesis
Fig. 2. View of the Sorin tilting-disc prosthesis (Carbocast) for mitral (left) and aortic (right) valve replacement.
Table II. Follow-up data Patients discharged Late deaths Lost to follow-up Current survivors Mean follow-up (yr) Range of follow-up (yr) Total follow-up (pt-yr) Complete (%)
AVR
MVR
MAVR
Total
330 58 2 257 5.1 ± 2.8 0.2-11.4 1650 99
219 25
82 17 3 60 4.1 ± 2.7 0.1-9.4 328 96
631 100 5 508 4.5 ± 2.6 0.1-11.4 2941 99
191 4.4 ± 2.5 0.1-9.9 963 100
Table III. Summary of major postoperative complications AVR No. Late deaths Prosthesis-related deaths Thromboemboli Fatal Prosthetic thrombosis Anticoagulant-related hemorrhage Fatal Endocarditis Para prosthetic leak Structural deterioration Reoperation
MAVR
MVR
%/pt-yr
No.
± ± ± ± ± ± ± ± ±
58 23 21 7 I IS 5 12 21
3.5 1.4 1.3 0.4 0.1 0.9 0.3 0.7 1.3
0.4 0.3 0.2 0.1 0.1 0.2 0.1 0.2 0.2
25 8 12 2 I 9 3 2 2
24
1.4 ± 0.3
5
technique was used to repair acute dissection of the ascending aorta, the distal anastomosis of the composite graft was preferentially performed during a brief period of deep hypothermic circulatory arrest. Starting from the second postoperative day, all patients were given subcutaneous calcium heparin and sodium warfarin for
%/pt-yr 2.6 ± 0.7 ± 1.2 ± 0.2 ± 0.1 ± 0.9 ± 0.3 ± 0.2 ± 0.2 ±
No.
0.5 0.2 0.3 0.1 0.1 0.3 0.1 0.1 0.1
17 12 7 2 I 6 3 3 4
0.5 ± 0.2
2
%/pt-yr 5.1 3.6 2.1 0.6 0.3 1.8 0.9 0.9 1.2
± ± ± ± ± ± ± ± ±
1.2 1.0 0.8 0.4 0.3 0.7 0.5 0.5 0.6
0.6 ± 0.4
anticoagulation; calcium heparin was then suspended when the prothrombin activity had dropped between 20% and 30% of normal (3 to 4.5 International Nationalized Ratio), and patients were told to maintain such values after discharge. Oral anticoagulants were given indefinitely, and antiplatelet drugs were not routinely used.
The Journal of
270
Thoracic and Cardiovascular
Milano et al.
Surgery
Table IV. Causes of hospital deaths
100 90 80
~:::::~~~ -
70 0
>
.~
~
U1 ~
::::::.6.
~-.~~""'-.
•
'.6.-.6. ---.6. -~=~ ---.6.
60
<,
50
72±4 70±3
.6. 50±12
j
40 30 355297262 219180155 125 241 188165 134 97 74 58 89 65 55 44 39 31 25
20 10
85 36 14
4 9 0 - 0 AVR 17 . - . MVR 5.6.-.6. MAVR
0+-___+--+--+--+-_+_-+--_+-+----+--+--+--1
o
2
3
4
5
6
7
8
9
10
11
12
Years Pastap
Fig. 3. Actuarial survival after isolated AYR, MYR, and MAYR with the Sorin prosthesis. Numbers on the horizontal axis, in this and following figures, indicate patients at risk during each postoperative interval.
100
___ ~~-.-.-~-.-.-.-.-. ~o-o-
.6.
90
O~o-o-o-o .6.-.6.---.6._.6._.6.
~.6.,.
<,
80 Q) Q)
c:
97±2 92±2
70
.6.\1
~
.6.
60 50 40
338293256 217179 153124 223186164 133 96 74 57 85 65 55 44 38 31 24
82 35 14
62±15
49 0 - 0 AVR 1 7 . - . MVR 5 .6.-.6. MAVR
+----+-~-_+____+--+--+-_+-_+_-f--_+-+_---1
o
2
3
4
5
6
7
8
9
10
11
12
Years Pastap
Fig. 4. Actuarial freedom from valve-related deaths.
Follow-up. Data concerning current clinical status and events occurring in the postoperative period were obtained by direct visits or questionnaires; such information was supplemented by review of necropsy reports and death certificates and by direct interview of parents, relatives, or referring physicians. Follow-up data were gathered during the interval from July to December 1989 (Table II). Cumulative duration of followup in the entire series is 2941 patient-years, ranging from 0.1 to 11.4 years. Cumulative follow-up is 1650 patient-years for AYR patients (range 0.2 to /1.4 years; mean 5.1 ± 2.8 years), 963 patient-years for MYR patients (range 0.1 to 9.9 years; mean 4.4 ± 2.5 years), and 328 patient-years for MA YR patients (range 0.1 to 9.4 years; mean 4 ± 2.7 years). Three patients could not be traced, yelding a 99% complete follow-up in the entire group. Major postoperative complications and prosthesis-related events were defined according to the guidelines recently proposed for reporting morbidity and mortality after cardiac valvular operations." Prostheses explanted at reoperation or removed at postmor-
Cardiac failure Myocardial infarction Respiratory failure Cerebralembolism Intraoperative cerebral damage Postoperative hemorrhage Cerebral hemorrhage Sepsis Gastric hemorrhage Left ventricular rupture Renal failure Arrhythmia Paraprosthetic leak Prosthetic thrombosis Endocarditis Suddendeath Total
AVR
MVR
14 3 3 2 2 I I I
13 I 2
MAVR 3 2
3 I 3 1 2 I I
2 I 1
I 28
28
10
tern study underwent detailed morphologic examination to assess the causes of dysfunction . Statistical analysis. Patients contributed to the study until they died or were lost to follow-up. Living patients were removed from the study when the original prosthesis was replaced or when another valve procedure was performed. However, all patients who were alive with their original prosthesis and required another cardiac operation without removal of the former or replacement of another valve were included in the survival curves. Actuarial analysis was performed according to life-table methods.!? including only the first event for each patient. Although the rate of occurrence of postoperative complications is not constant throughout the postoperative period because of the changing of the hazard function, linearized rates, including all events, were nevertheless calculated for any valve-related complication. I I Actuarial rates are presented as mean ± I standard error of the mean (SEM), and all actuarial curves include operative deaths. Linearized rates are expressed as percentperpatient-year(%/pt-yr) ± 1SEM. Continuous data are presented as mean ± I standard deviation and categoric data as percent with their 70% confidence limits (CL).
Results (Table III) Patient survival. There were 28 operative deaths in the AVR group (7.8%; 6.3% to 9.5% CL), 28 in the MVR group (11.3%; 9.2% to 13.8% CL), and lOin the MAVR group (10.8%; 7.4% to 15.3% CL); all patients but one were in functional class III or IV preoperatively. Operative mortality after A VR was influenced by deaths occurring in patients with acute dissection of the ascending aorta (28%); in MVR and MAVR patients operative deaths occurred mainly in those operated on for endocarditis (25% and 33%, respectively) or for prosthetic dysfunction (20% and 14%, respectively). Death was
Volume 103 Number 2 February 1992
Sorin tilting-disc prosthesis
Table V. Causes of late deaths Cardiac failure Myocardial infarction Thromboembolism Anticoagulant-related hemorrhage Endocarditis Para prosthetic leak Renal failure Malignancy Sudden, unexplained death Unknown Total
100
AVR
MVR
MAVR
15
4
5
15 2 2
4
I
3
6
I I
4 3 I
5 3
12 58
""~-e-e___ ~
-A-A_A~!~~~.~~~ 92±3 91±3
90
Q) Q)
BO
c: a<1
{:,
2 I
25
2 3 2 17
mainly caused by postoperative low-output syndrome (14 patients with AVR, 13 with MVR, and three with MA VR); other causes of hospital deaths are summarized in Table IV. There were 58 late deaths after AVR (3.5% ± 0.4%/ pt-yr), 25 after MVR (2.6% ± 0.5%/pt-yr), and 17 after MA VR· (5.1% ± 1.2%/pt-yr); actuarial survival at 9 years is 72% ± 4% for MVR, 70% ± 3% for AVR, and 50% ± 12% for MA VR patients (Fig. 3). Table V shows the causes of late death. Including both hospital and late mortality, there were 23 valve-related deaths in the AVR group (1.4% ± O.3%/pt-yr), eight in the MVR group (0.7% ± 0.2%/ pt-yr), and 12in the MAVR group (3.6% ± 1.0%/pt-yr). Valve-related hospital deaths were observed in four patients after AVR (cerebral emboli in two, cerebral hemorrhage in one, and sudden death in one), in two after MVR (cerebral hemorrhage in one and gastric hemorrhage in one), and in four after MAVR (paravalvular leak of the mitral prosthesis in two, endocarditis in one, and aortic valve thrombosis in one); late postoperatively, valve-related deaths occurred in 19 patients after AVR (cerebral embolism in five, anticoagulant-related hemorrhage in four, endocarditis in four, paraprosthetic leak in three, and sudden, unexplained death in three), in six after MVR (sudden, unexplained death in two, endocarditis in one, cerebral embolism in one, valve thrombosis in one, and anticoagulant-related hemorrhage in one), and in eight after MA VR (sudden, unexplained death in three, anticoagulant-related hemorrhage in three, cerebral embolism in one, and endocarditis in one). Freedom from valve-related deaths at 9 years is 97% ± 2% for MVR, 92% ± 2% for AVR, and 62% ± 15% for MAVR patients (Fig. 4).
74±16
70 60
I
27 I
50
333 2BB 255 212173 144 116 222 1B4 161 131 93 70 54 B1 65 54 44 36 30 23
a
2
3
4
5
6
1
46 0 - 0 AVR 17 MVR 5 {:,-{:, MAVR
7B 33 13 7
e-.
B
9
10
11
12
Years Pastap
Fig. 5. Actuarial freedom from thromboembolic complications.
Table VI. Change in functional status of current survivors AVR
(n = 257)
MVR
(n = 191)
MAVR (n
= 60)
Preop. NYHA class
I II III
IV
103 119 35
5 156 30
4 47
94 88 9
23 28 9
9
Postop. NYHA class
I II III IV
146 102 8 I
Most of the current survivors show a definite improvement in clinical status, as shown in Table VI. Thromboembolic complications. Thromboemboli occurred in 21 patients with AVR (1.3% ± 0.2%/pt-yr), in 12 with MVR (1.2% ± 0.3%/pt-yr) and in seven with MAVR (2.1% ± 0.8%/pt-yr); fatal thromboembolic complications were observed in seven patients with AVR (0.4% ± 0.1%/pt-yr), in two with MVR (0.2% ± 0.1%/ pt-yr) , and in two with MAVR (0.6% ± O.4%/pt-yr). Prosthetic thrombosis occurred in one patient after AVR (0.1% ± O.1%/pt-yr) who had a successful reoperation, in one after MVR (0.1% ± O.1%/pt-yr) who died at reoperation, and in one after MAVR (0.3% ± 0.3%/ptyr) who died early postoperatively because of aortic valve thrombosis. Actuarial freedom from thromboembolic complications at 9 years is 92% ± 3% after MVR, 91% ± 3% after AVR, and 74% ± 16% after MAVR (Fig. 5).
The Journal of
272
Thoracic and Cardiovascular
Milano et at.
Surgery
1OO__~=..-----=--..-
---------,
90 80 Q) Q)
c:
~
70 60 ~o
40 30
332291257215 173149 118 222185162131 95 74 55 80 65 54 43 37 31 24
o
2
3
4
5
78 33 14
6
47 0 - 0 AVR 16 MVR 5 6 - 6 MAVR
e-e
8
9
10
11
12
Years Pastap
Fig. 6. Actuarial freedom fromanticoagulant-related hemorrhage. 100 __- - - - - - - - - - - - = - - - - - , ~~-e-e-e-e-e-e-e-. 99±1 \ --"0 T 6-6--..~-~-6-6-6-6-6 96±2 - - " 0 - 0 - 0 - 0 - 0 95±1 1
90
33429025721417314811879480-0AVR 222186164133 96 74 57 35 17 MVR 81 64 53 43 36 30 23 12 4 6 - 6 MAVR
e-e
80
+---+--+---+-~--+---+-+----+--+---+----+--------j
o
2
3
4
5
6
7
8
9
10
11
12
Years Postop
Fig. 7. Actuarial freedom from prosthetic valve endocarditis.
Anticoagulant-related hemorrhage. Hemorrhage related to anticoagulation occurred in 15 patients after AVR (0.9% ± O.2%/pt-yr), in nine after MVR (0.9% ± 0.3%/pt-yr), and in six after MAVR (1.8% ± 0.7%/ptyr). Fatal hemorrhage occurred in fivepatients with AVR (0.3% ± O.l%/pt-yr), in three with MVR (0.3% ± O.I%/pt-yr), and in three with MAVR (0.9% ± 0.5%/ pt-yr); eight died of cerebral hemorrhage and three of gastrointestinal bleeding. Actuarial freedom from anticoagulant-related hemorrhage at 9 years is 94% ± 2% for AVR, 91% ± 4% for MVR, and 68% ± 16% for MAVR'patients (Fig. 6). Endocarditis. Prosthetic valve endocarditis was observed in 12 patients after AVR (0.7% ± 0.2%/pt-yr) after a mean interval of 1.7 ± 2.0 years, in two after MVR (0.2% ± 0.1%/pt-yr) after 1 and 2 months, and in three after MAVR (0.9% ± 0.5%/pt-yr) after 20 days, 1 month, and 3 months; infection involved the mitral prosthesis in all of them. Five patients with AVR were reoperated on, with one death, and seven were managed
medically. Two of the seven managed medically died, and in five infection was cured but with persistence of a trivial residual paraprosthetic leak. One patient with MVR was reoperated on and died, and the other died without reoperation. Finally, all three patients with MAVR died before reoperation. Actuarial freedom from endocarditis at 9 years is 99% ± 1% after MVR, 96% ± 2% after MA VR, and 95% ± I% after AVR (Fig. 7). Paraprosthetic leak. Prosthetic detachment was observed in 21 patients with AVR (1.3% ± 0.2%/pt-yr) after a mean interval of 2.4 ± 2.3 years, in two with MVR(0.2% ± O.I%/pt-yr), both within 1 month,andin four with MAVR (1.2% ± 0.6%/pt-yr); of the latter, three had detachment of the mitral valve after 15 days, 20 days, and 5 months, and one had detachment of the aortic prosthesis after 3 months. In the AVR group 18 patients were reoperated on, and there were no operative deaths; three died before reoperation. Fifteen of the 18 had replacement of the prosthesis and three had direct suture of the leak. Both patients with MVR had successful prosthetic replacement at reoperation. Two patients with MAVR died before reoperation, and two had successful prosthetic replacement at reoperation. Actuarial freedom from paraprosthetic leak at 9 years is 99% ± I% after MVR, 95% ± 3% after MAVR, and 90% ± 10% after AVR (Fig. 8). Reoperation. During the follow-up period 31 patients underwent reoperation: 24 after AVR (1.4% ± 0.3%/ pt-yr), five after MVR (0.5% ± 0.2%/pt-yr), and two after MAVR (0.6% ± O.4%/pt-yr). In the AVR group reoperation was required because of paravalvular leak in 18 patients and thrombosis in one; there were no deaths. Endocarditis occurred in five, with one death (20%; 2.6% to 53.2% CL). Two patients with MVR had successful reoperation for paravalvular leak, and one patient with paravalvular leak and one with thrombosis died at reoperation. Two patients with MAVR were successfully reoperated on for paravalvular leak of the aortic and mitral prostheses, respectively. No cases of structural deterioration were observed in any patients at any valve position. Actuarial freedom from reoperation at 9 years is 97% ± 2% after MAVR, 92% ± 4% after MVR, and 89% ± 3% after AVR (Fig. 9). Overall valve performance. Freedom from valverelated mortality and reoperation at 9 years is 80% ± 5% after MVR, 75% ± 3% after AVR, and 56% ± 14% after MAYR. Freedom from valve-related mortality and permanent disability at 9 years is 83% ± 3% after AVR and MVR and 55% ± 13% after MAYR. Freedom from
Volume 103 Number 2
Sarin tilting-disc prosthesis
February 1992
100
~~~~=~=~=~=~=~=~-697±2 -0-0-0
90
-"""'0-0-0
"""T
e -"""'0 1
Q) Q)
Q) Q)
G:
273
92±4 89±3
80
G:
~
70
70
60
60
331287252208 169144 114 75 4 5 0 - 0 AVR 22218516.3 132 95 73 55 34 17 e-e MVR 80 52 52 42 35 30 23 12 5 /:,-/:, MAVR 50 +--+-+----+--+----l--+--f'-----+--+---+---t---1 o 2 3 4 5 5 7 8 9 10 11 12
50
3.33 286 251 217 168 143113 223 185 163 132 95 73 56 80 53 52 42 36 30 23 0
2
4
3
Yeors Postop
7
8
9
10
11
12
Fig. 9. Actuarial freedom from reoperation for all causes.
all valve-related morbidity and mortality at 9 years is 67% ± 6% after MYR, 63% ± 4% after A YR, and 41% ± 15% after MAYR (Fig. 10).
100 90· 80 70
Mechanical prostheses have been used in the preceding years with increasing frequency, mainly to overcome the problem .of the limited durability of tissue valves. 12, 13 Those mechanical prostheses utilizing a tilting-disc mechanism have gained worldwide acceptance, particularly because of their favorable hemodynamic characteristics.i However, complications such as systemic embolism, acute prosthetic thrombosis, and hemorrhage related to the anticoagulant treatment remain a major concern with both tilting-disc and other mechanical devices. 14, 15 In recent years, various new prosthetic models have been manufactured and used clinically to minimize such problems.l'"!? Among these, the Sorin tilting-disc prosthesis has been available in our country for more than a decade, but so far only a few studies have reported the early and mediumterm results with this device. 20, 21 Therefore we have reviewed all patients who received a Sorin prosthesis at our institution. Maximum follow-up time has been 10 years, which is a time frame long enough to assess the overall performance of a cardiac valve substitute. Although the present experience refers to patients operated on with better techniques of myocardial protection, early mortality was not dissimilar to that observed by our group in patients receiving other prosthetic valves,6,7 since also in this series most patients were in an advanced functional class at operation. It must also be considered that many of them had previously undergone one or more valvular procedures, and patients at higher risk, such as those with acute aortic dissection, were also included in the A VR group.
5
Yeors Postop
Fig. 8. Actuarial freedom from para prosthetic leak.
Discussion
5
4 5 0 - 0 AVR 18 e-e MVR 5 /:,-/:, MAVR
74 35 13
Q)
~
l •.
~
50
\-e~~-e-e --e __ --/:,__ --l::J.
-
--o~
e - e __ T
/:'-/:'-2===~-0-e --/:,
.
\1
50
/:,
40
57±5 53±4
41±15
.\0 20
351 278244 199 159 129 97 54 38 0 - 0 AVR 238182157 127 90 68 50 30 15 e-e MVR 87 52 51 41 33 27 20 11 4 /:,-/:, MAVR O+----+--+---+--+-----j--+--+---+--+---+--+----1 o 2 3 4 5 6 7 8 9 10 11 12
10
Years Postop
Fig. 10. Actuarial freedom from all valve-related morbidity and mortality.
The occurrence of thromboembolic complications was substantially low and again comparable with that reported with other mechanical prostheses.v" The total number of patients with valve occlusion by thrombus formation at initial presentation was lower than that observed by us with other tilting-disc valves/'- 7 Nevertheless, prosthetic thrombosis is still a matter of great concern, as much so as the rate of hemorrhage related to the anticoagulant treatment. Both complications, which significantly influenced the number of prosthesis-related deaths, once again clearly witness to the difficulty in maintaining adequate and safe long-term anticoagulation despite our continuing effort to observe as many patients as possible at our institution. A modification in the Sorin prosthesis, by coating the Teflon sewing ring with a thin film of turbostratic carbon, has been introduced with the aim of increasing hemocompatibility of the fabric surface, ideally reducing valve thrombogenicity. Because of the small number of patients who received this new model and their shorter
The Journal of
274
Milano et al.
follow-up, however, no significant difference in the rate of thromboembolic complications can thus far be documented in the present series. Structural deterioration, manifested by strut fracture or disc escape, has been observed with various prosthetic valves. 22- 26 This has occurred with both the spherical disc and the convexo-concave model of the Bjork-Shiley prosthesis, in which the struts are incorporated in the frame by fusion welding. The welding sites have been demonstrated to be potential sites of weakness for fatigueinduced Iesions.F:25 Sorin prostheses are manufactured by casting the whole cage in a single piece. The result is a perfectly homogeneous material structure, including those areas subjected to the greatest stress. The casting process also allows the struts to be designed with a crosssectional area optimized to obtain uniform stress conditions within these structures.i? In fact, mechanical rupture has never occurred either in our patients or, to the best of our knowledge, in those of others.P: 21 In conclusion, (I) the Sorin prosthesis shows an overall satisfactory performance in the first decade of clinical use that is substantially comparable with that of other mechanical valves; (2) thromboembolic episodes may be controlled by careful anticoagulant treatment, but hemorrhagic complications are still an important cause of postoperative morbidity; (3) the most striking advantage of this device appears to be its excellent durability without any episode of structural failure. In our opinion the Sorin prosthesis can be considered a reliable cardiac valve substitute for mitral and aortic valve positions in patients in whom insertion of a mechanical prosthesis is indicated. We thank Gian Carlo Pengo for his technical assistance. REFERENCES I. Mitha AS, Matisonn RE, le Roux BT, Chesler E. Clinical experience with the Lillehei-Kaster cardiac valve prosthesis. J THORAC CARDIOVASC SURG 1976;72:401-7. 2. Bjork VO, Henze A. Ten years' experience with the BjorkShiley tilting disc valve. J THORAC CARDIOVASC SURG 1970;78:331-42. 3. Karp RB, Cyrus RJ, Blackstone EH, Kirklin JW, Kouchoukos NT, Pacifico AD. The Bjork-Shiley valve: intermediate-term follow-up. J THORAC CARDIOVASC SURG 1981;81:602-14. 4. Lindblom D. Long-term clinical results after mitral valve replacement with the Bjork-Shiley prosthesis. J THORAC CARDIOVASC SURG 1988;95:321-33. 5. Lindblom D. Long-term clinical results after aortic valve replacement with the Bjork-Shiley prosthesis. J THORAC CARDIOVASC SURG 1988;95:658-67.
Thoracic and Cardiovascular Surgery
6. Milano A, Bortolotti U, Mazzucco A, Guerra F, Magni A, Gallucci V. Aortic valve replacement with the Hancock standard, Bjork-Shiley, and Lillehei-Kaster prostheses: a comparison based on follow-up from I to 15 years. J THORAC CARDIOVASC SURG 1989;98:37-47. 7. Milano A, Bortolotti U, Mazzucco A, et al. Mitral valve replacement with the Hancock, Bjork-Shiley and LilleheiKaster prostheses: a comparison based on a 15-year followup. Eur J Cardiothorac Surg 1989;3:312-20. 8. Bentall H, De Bono A. A technique for complete replacement of the ascending aorta. Thorax 1968;23:338-9. 9. Edmunds LH Jr, Clarke RE, Cohn LH, Miller DC, Weisel RD. Guidelines for reporting morbidity and mortality after cardiac valvular operations. J THORAC CARDIOVASC SURG 1988;96:351-3. 10. Grunkemeier GL, Starr A. Actuarial analysis of surgical results: rationale and methods. Ann Thorac Surg 1977; 24:404-8. II. Grunkemeier GL, Thorns DR, Starr A. Statistical considerations in the analysis and reporting of time-related events. Am J Cardiol 1977;39:257-8. 12. Bortolotti U, Milano A, Mazzucco A, et al. Results of reoperation for primary tissue failure of porcine bioprostheses. J THORAC CARDIOVASC SURG 1985;90: 564-9. 13. Bortolotti U, Milano A, Mazzucco A, et al. The Hancock pericardial xenograft: incidence of early mechanical failures at a medium-term follow-up. Eur J Cardiothorac Surg 1988;2:458-64. 14. Moreno-Cabral RJ, McNamara JJ, Mamiya RT, Brainard SC, Chung GKT. Acute thrombotic obstruction with Bjork-Shiley valves:diagnostic and surgical considerations. J THORAC CARDIOVASC SURG 1978;75:321-30. 15. Edmunds LH Jr. Thrombotic and bleeding complications of prosthetic heart valves. Ann Thorac Surg 1987;44:43045. 16. Emery RW, Mettler E, Nicoloff JM. A new cardiac prosthesis: the St. Jude Medical cardiac valve: in vivo results. Circulation 1979;60 {Pt 2):140-54. 17. Hashimoto A, Nagase Y, Koyanagi H. A new tilting valve prosthesis, "Omniscience valve:" clinical and hemodynamic evaluations. Artif Organs 1982;11:2-8. 18. Scotten LN, RaccaRG, NugentAH, Walker DK, Brownlee RT. New tilting disc cardiac valve prostheses: in vitro comparison of their hydrodynamic performance in the mitral position. J THORAC CARDIOVASC SURG 1981; 82:136-46. 19. Beaudet RL, Poirier NL, Doyle D, Nakhle G, Gauvin C. The Medtronic Hall cardiac valve: 7V2 years' clinical experience. Ann Thorac Surg 1986;42:644-50. 20. Colombo T, Donatelli F, Quaini E, Vitali E, Pellegrini A. Results of heart valve replacement with the Sorin prosthesis. Texas Heart Inst J 1987;14:77-87. 21. Sante P, Festa M, Renzulli A, et al. Clinical results after isolated aortic valve replacement with Sorin prosthesis: a 6-year experience. Ital J Surg Sci 1988;18:323-6.
Volume 103 Number 2 February 1992
22. Larrieu AJ, Puglia E, Allen P. Strut fracture and disc embolization of a Bjork-Shiley mitral valve prosthesis: localization of embolized disc by computerized axial tomography. Ann Thorac Surg 1982;34:192-5. 23. Hjelms E. Escape of a leaflet from a St. Jude Medical prosthesis in the mitral position. Thorac Cardiovasc Surg 1983;31:310-2. 24. Ibarra F, Gutierrez A, Martinez F, Carreras L, Lopez C, Alonso-Lej F. Fracture of the outlet strut of a Bjork-Shiley mitral prosthesis: emergency operation with survival. J THORAC CARDIOVASC SURG 1984;87:315-8. 25. Rockelein G, Breme J, von der Emde J. Lethal blockage of
Sarin tilting-disc prosthesis
275
a Bjork-Shiley artificial heart valve caused by strut fracture-the metallurgic aspect. Thorac Cardiovasc Surg 1989;37:47-5l. 26. Kepletko W, Moritz A, Mlczoch J, Schurawitzki H, Domanig E, Wolner E. Leaflet fracture in Edwards-Duromedics bileaflet valves. J THORAC CARDIOVASC SURG 1989;97:90-4. 27. Tilting disc valves: comparison between Shiley and Sorin models. The progress reports of research program TBM/ CNR. SP-CARD/3, 1977-1978 (National Research Council).
Bound volumes available to subscribers Bound volumes ofTHE JOURNAL OF THORACIC AND CARDIOVASCULAR SURGERY are available to subscribers (only) forthe 1992 issues from the Publisher, at a costof $71.00 fordomestic, $93.97 forCanadian, and $89.00 for international subscribers for Vol. 103 (JanuaryJune) and Vol. 104 (July-December). Shipping charges are included. Each bound volume contains a subjectand author index and all advertising is removed. Copies are shipped within 60 daysafter publication ofthe lastissue of the volume. The binding isdurablebuckram with the JOURNAL name, volume number, and yearstamped in goldon the spine. Payment must accompany all orders. Contact MosbyYearBook, Inc., Subscription Services, 11830 Westline Industrial Drive, St. Louis, Missouri 63146-3318, USA; phone (800) 325-4177, ext. 4351, or (314) 453-4351. Subscriptions mustbe in force to qualify. Bound volumes are not available in placeof a regular JOURNAL subscription.
Aldo Milano, MD," Uberto Bortolotti, MD," Alessandro Mazzucco, MD," Eugenio Mossuto, MD," Luca Testolin, MD," Gaetano Thiene, MD,b and Vincenzo Gallucci, MD," Padova, Italy
Llting-diSC mechanical prostheses have been used in the past primarily for replacement of the aortic and mitral From the Departments of Cardiovascular Surgery" and Pathology, b University of Padova Medical School, Padova, Italy. Received for publication July 6, 1990. Accepted for publication Dec. 14, 1990. Address for reprints: U. Bortolotti, MD, lstituto di Chirurgia Cardiavascolare, Universita di Padova, Via Giustiniani, 2, 35128 Padova, Italy.
12/1/28351
valves.':" Among the various models the Sorin prosthesis became available for a clinical trial at our institution in 1978. The purpose of this report is to present our experience with the Sorin prosthesis in patients with single and multiple valve replacement to evaluate its performance in the first decade of its clinical use. Patients and methods Prosthesis description. The Sorin tilting-disc prosthesis (Sorin Biomedica, Saluggia, Italy) is made of a pyrolytic carbon disc, retained and freely tilting to 60 degrees in a cage
267
268
Milano et al.
The Journal of Thoracic and Cardiovascular Surgery
Table I. Preoperative clinical characteristics AVR
Fig. 1. Viewof the Sorin tilting-disc prosthesis(Monocast) for aortic (top) and mitral (bottom) valve replacement.
obtained from a single piece of Stellite, a chrome-cobalt alloy, without welding points between the struts and the cage itself. The standard model (Monocast) (Fig. I) is provided with a knitted Teflon sewing ring; in the more recent model (Carbocast) (Fig. 2), the sewing ring has been coated, where it comes in contact with blood, by a thin film of turbostratic carbon (Carbofilm, Sorin Biomedica, Saluggia, Italy), that is, a particular crystalline structure that gives the material diamond-like characteristics. Patient population. From September 1978 to December 1988,697 patients underwent isolated aortic valve replacement (A VR = 358 patients), isolated mitral valve replacement (MVR = 247 patients), and simultaneous mitral and aortic valve replacement (MAVR = 92 patients). A total of 789 Sorin tilting-disc prostheses (692 Monocast and 97 Carbocast prostheses) were used. There were 416 male and 281 female patients, with a mean age of 48 ± II years (range 5 to 75 years). Main clinical characteristics of the three groups of patients are summarized in Table I. The majority of the patients (over 80%) were in New York Heart Association functional class III or IV at time of operation; the most frequent indication for operation was rheumatic valvular disease in the MVR and MA VR groups and either calcific or myxomatous degenerative disease in the AVR group. Approximately 30% of the patients had undergone pre-
MVR
MAVR
No. of patients 358 247 92 Male/female 281/77 82/165 53/39 Mean age (yr) 48 ± 12 48 ± 11 47 ± II Agerange (yr) 5-75 8-67 17-68 Preoperative functional class NYHA II 117 6 2 NYHA III 180 193 66 NYHA IV 61 48 21 Preoperative rhythm 48 SR 336 30 19 194 AF 62 PM 3 5 Etiology 145 Rheumatic 96 50 Myxomatous degeneration 91 29 6 Calcific degeneration 85 8 Aortic dissection 25 Congenital 10 7 I 14 4 Endocarditis 6 Ischemic 7 Prosthesis dysfunction 37 55 21 Associated procedures 131 69 20 Replacement of ascending aorta* 59 I CABG 38 15 4 Tricuspid annuloplasty 38 8 Mitralvalvuloplasty 13 Aortic valvuloplasty II Enlargement of aortic anulus 5 5 LV aneurysmectomy 2 Others 16 2 3 Previous cardiac operations 51 124 34 AYR, Aortic valve replacement; MYR, mitral valve replacement; MAYR. mitral and aortic valve replacement; NHYA, New York Heart Association; SR, sinus rhythm; AF, atrial fibrillation; PM, pacemaker; CABG, coronary artery bypass graft; LY, left ventricular. • Bentall" procedure in 51 and graft replacement ofthe aortic root in eight.
vious intracardiac procedures; 12% of the patients with AVR had had a previous operation on the aortic valve, 47% of the patients with MVR had had a previousoperation on the mitral valve,and 25%of those with MA VR had undergone a previous procedure on the aortic, mitral, or both valves.Operations associated with valve replacement were performed in 32% of cases, consisting mainly in replacement of the aortic root in AVR patients and tricuspid annuloplasty in MVR and MA VR patients. Surgical technique. All patients were operated on with the use of moderately hypothermic cardiopulmonary bypass, with topical cooling and cold potassium cardioplegic solution. Although the techniques varied among the different surgeons, prosthesis insertion was preferentially accomplished by means of mattress sutures, buttressed by Teflon felt pledgets placed in the supraannular position.Prostheses were usually oriented with their large opening corresponding to the convexity of the ascending aorta in AVR and to the posterior left ventricular wall in MVR. Patients undergoing replacement of the aortic valve and ascending aorta were treated as classically described by Bentall and De Bono.t with a Sorin valved conduit; when this
Volume 103 Number 2 February 1992
269
Sorin tilting-disc prosthesis
Fig. 2. View of the Sorin tilting-disc prosthesis (Carbocast) for mitral (left) and aortic (right) valve replacement.
Table II. Follow-up data Patients discharged Late deaths Lost to follow-up Current survivors Mean follow-up (yr) Range of follow-up (yr) Total follow-up (pt-yr) Complete (%)
AVR
MVR
MAVR
Total
330 58 2 257 5.1 ± 2.8 0.2-11.4 1650 99
219 25
82 17 3 60 4.1 ± 2.7 0.1-9.4 328 96
631 100 5 508 4.5 ± 2.6 0.1-11.4 2941 99
191 4.4 ± 2.5 0.1-9.9 963 100
Table III. Summary of major postoperative complications AVR No. Late deaths Prosthesis-related deaths Thromboemboli Fatal Prosthetic thrombosis Anticoagulant-related hemorrhage Fatal Endocarditis Para prosthetic leak Structural deterioration Reoperation
MAVR
MVR
%/pt-yr
No.
± ± ± ± ± ± ± ± ±
58 23 21 7 I IS 5 12 21
3.5 1.4 1.3 0.4 0.1 0.9 0.3 0.7 1.3
0.4 0.3 0.2 0.1 0.1 0.2 0.1 0.2 0.2
25 8 12 2 I 9 3 2 2
24
1.4 ± 0.3
5
technique was used to repair acute dissection of the ascending aorta, the distal anastomosis of the composite graft was preferentially performed during a brief period of deep hypothermic circulatory arrest. Starting from the second postoperative day, all patients were given subcutaneous calcium heparin and sodium warfarin for
%/pt-yr 2.6 ± 0.7 ± 1.2 ± 0.2 ± 0.1 ± 0.9 ± 0.3 ± 0.2 ± 0.2 ±
No.
0.5 0.2 0.3 0.1 0.1 0.3 0.1 0.1 0.1
17 12 7 2 I 6 3 3 4
0.5 ± 0.2
2
%/pt-yr 5.1 3.6 2.1 0.6 0.3 1.8 0.9 0.9 1.2
± ± ± ± ± ± ± ± ±
1.2 1.0 0.8 0.4 0.3 0.7 0.5 0.5 0.6
0.6 ± 0.4
anticoagulation; calcium heparin was then suspended when the prothrombin activity had dropped between 20% and 30% of normal (3 to 4.5 International Nationalized Ratio), and patients were told to maintain such values after discharge. Oral anticoagulants were given indefinitely, and antiplatelet drugs were not routinely used.
The Journal of
270
Thoracic and Cardiovascular
Milano et al.
Surgery
Table IV. Causes of hospital deaths
100 90 80
~:::::~~~ -
70 0
>
.~
~
U1 ~
::::::.6.
~-.~~""'-.
•
'.6.-.6. ---.6. -~=~ ---.6.
60
<,
50
72±4 70±3
.6. 50±12
j
40 30 355297262 219180155 125 241 188165 134 97 74 58 89 65 55 44 39 31 25
20 10
85 36 14
4 9 0 - 0 AVR 17 . - . MVR 5.6.-.6. MAVR
0+-___+--+--+--+-_+_-+--_+-+----+--+--+--1
o
2
3
4
5
6
7
8
9
10
11
12
Years Pastap
Fig. 3. Actuarial survival after isolated AYR, MYR, and MAYR with the Sorin prosthesis. Numbers on the horizontal axis, in this and following figures, indicate patients at risk during each postoperative interval.
100
___ ~~-.-.-~-.-.-.-.-. ~o-o-
.6.
90
O~o-o-o-o .6.-.6.---.6._.6._.6.
~.6.,.
<,
80 Q) Q)
c:
97±2 92±2
70
.6.\1
~
.6.
60 50 40
338293256 217179 153124 223186164 133 96 74 57 85 65 55 44 38 31 24
82 35 14
62±15
49 0 - 0 AVR 1 7 . - . MVR 5 .6.-.6. MAVR
+----+-~-_+____+--+--+-_+-_+_-f--_+-+_---1
o
2
3
4
5
6
7
8
9
10
11
12
Years Pastap
Fig. 4. Actuarial freedom from valve-related deaths.
Follow-up. Data concerning current clinical status and events occurring in the postoperative period were obtained by direct visits or questionnaires; such information was supplemented by review of necropsy reports and death certificates and by direct interview of parents, relatives, or referring physicians. Follow-up data were gathered during the interval from July to December 1989 (Table II). Cumulative duration of followup in the entire series is 2941 patient-years, ranging from 0.1 to 11.4 years. Cumulative follow-up is 1650 patient-years for AYR patients (range 0.2 to /1.4 years; mean 5.1 ± 2.8 years), 963 patient-years for MYR patients (range 0.1 to 9.9 years; mean 4.4 ± 2.5 years), and 328 patient-years for MA YR patients (range 0.1 to 9.4 years; mean 4 ± 2.7 years). Three patients could not be traced, yelding a 99% complete follow-up in the entire group. Major postoperative complications and prosthesis-related events were defined according to the guidelines recently proposed for reporting morbidity and mortality after cardiac valvular operations." Prostheses explanted at reoperation or removed at postmor-
Cardiac failure Myocardial infarction Respiratory failure Cerebralembolism Intraoperative cerebral damage Postoperative hemorrhage Cerebral hemorrhage Sepsis Gastric hemorrhage Left ventricular rupture Renal failure Arrhythmia Paraprosthetic leak Prosthetic thrombosis Endocarditis Suddendeath Total
AVR
MVR
14 3 3 2 2 I I I
13 I 2
MAVR 3 2
3 I 3 1 2 I I
2 I 1
I 28
28
10
tern study underwent detailed morphologic examination to assess the causes of dysfunction . Statistical analysis. Patients contributed to the study until they died or were lost to follow-up. Living patients were removed from the study when the original prosthesis was replaced or when another valve procedure was performed. However, all patients who were alive with their original prosthesis and required another cardiac operation without removal of the former or replacement of another valve were included in the survival curves. Actuarial analysis was performed according to life-table methods.!? including only the first event for each patient. Although the rate of occurrence of postoperative complications is not constant throughout the postoperative period because of the changing of the hazard function, linearized rates, including all events, were nevertheless calculated for any valve-related complication. I I Actuarial rates are presented as mean ± I standard error of the mean (SEM), and all actuarial curves include operative deaths. Linearized rates are expressed as percentperpatient-year(%/pt-yr) ± 1SEM. Continuous data are presented as mean ± I standard deviation and categoric data as percent with their 70% confidence limits (CL).
Results (Table III) Patient survival. There were 28 operative deaths in the AVR group (7.8%; 6.3% to 9.5% CL), 28 in the MVR group (11.3%; 9.2% to 13.8% CL), and lOin the MAVR group (10.8%; 7.4% to 15.3% CL); all patients but one were in functional class III or IV preoperatively. Operative mortality after A VR was influenced by deaths occurring in patients with acute dissection of the ascending aorta (28%); in MVR and MAVR patients operative deaths occurred mainly in those operated on for endocarditis (25% and 33%, respectively) or for prosthetic dysfunction (20% and 14%, respectively). Death was
Volume 103 Number 2 February 1992
Sorin tilting-disc prosthesis
Table V. Causes of late deaths Cardiac failure Myocardial infarction Thromboembolism Anticoagulant-related hemorrhage Endocarditis Para prosthetic leak Renal failure Malignancy Sudden, unexplained death Unknown Total
100
AVR
MVR
MAVR
15
4
5
15 2 2
4
I
3
6
I I
4 3 I
5 3
12 58
""~-e-e___ ~
-A-A_A~!~~~.~~~ 92±3 91±3
90
Q) Q)
BO
c: a<1
{:,
2 I
25
2 3 2 17
mainly caused by postoperative low-output syndrome (14 patients with AVR, 13 with MVR, and three with MA VR); other causes of hospital deaths are summarized in Table IV. There were 58 late deaths after AVR (3.5% ± 0.4%/ pt-yr), 25 after MVR (2.6% ± 0.5%/pt-yr), and 17 after MA VR· (5.1% ± 1.2%/pt-yr); actuarial survival at 9 years is 72% ± 4% for MVR, 70% ± 3% for AVR, and 50% ± 12% for MA VR patients (Fig. 3). Table V shows the causes of late death. Including both hospital and late mortality, there were 23 valve-related deaths in the AVR group (1.4% ± O.3%/pt-yr), eight in the MVR group (0.7% ± 0.2%/ pt-yr), and 12in the MAVR group (3.6% ± 1.0%/pt-yr). Valve-related hospital deaths were observed in four patients after AVR (cerebral emboli in two, cerebral hemorrhage in one, and sudden death in one), in two after MVR (cerebral hemorrhage in one and gastric hemorrhage in one), and in four after MAVR (paravalvular leak of the mitral prosthesis in two, endocarditis in one, and aortic valve thrombosis in one); late postoperatively, valve-related deaths occurred in 19 patients after AVR (cerebral embolism in five, anticoagulant-related hemorrhage in four, endocarditis in four, paraprosthetic leak in three, and sudden, unexplained death in three), in six after MVR (sudden, unexplained death in two, endocarditis in one, cerebral embolism in one, valve thrombosis in one, and anticoagulant-related hemorrhage in one), and in eight after MA VR (sudden, unexplained death in three, anticoagulant-related hemorrhage in three, cerebral embolism in one, and endocarditis in one). Freedom from valve-related deaths at 9 years is 97% ± 2% for MVR, 92% ± 2% for AVR, and 62% ± 15% for MAVR patients (Fig. 4).
74±16
70 60
I
27 I
50
333 2BB 255 212173 144 116 222 1B4 161 131 93 70 54 B1 65 54 44 36 30 23
a
2
3
4
5
6
1
46 0 - 0 AVR 17 MVR 5 {:,-{:, MAVR
7B 33 13 7
e-.
B
9
10
11
12
Years Pastap
Fig. 5. Actuarial freedom from thromboembolic complications.
Table VI. Change in functional status of current survivors AVR
(n = 257)
MVR
(n = 191)
MAVR (n
= 60)
Preop. NYHA class
I II III
IV
103 119 35
5 156 30
4 47
94 88 9
23 28 9
9
Postop. NYHA class
I II III IV
146 102 8 I
Most of the current survivors show a definite improvement in clinical status, as shown in Table VI. Thromboembolic complications. Thromboemboli occurred in 21 patients with AVR (1.3% ± 0.2%/pt-yr), in 12 with MVR (1.2% ± 0.3%/pt-yr) and in seven with MAVR (2.1% ± 0.8%/pt-yr); fatal thromboembolic complications were observed in seven patients with AVR (0.4% ± 0.1%/pt-yr), in two with MVR (0.2% ± 0.1%/ pt-yr) , and in two with MAVR (0.6% ± O.4%/pt-yr). Prosthetic thrombosis occurred in one patient after AVR (0.1% ± O.1%/pt-yr) who had a successful reoperation, in one after MVR (0.1% ± O.1%/pt-yr) who died at reoperation, and in one after MAVR (0.3% ± 0.3%/ptyr) who died early postoperatively because of aortic valve thrombosis. Actuarial freedom from thromboembolic complications at 9 years is 92% ± 3% after MVR, 91% ± 3% after AVR, and 74% ± 16% after MAVR (Fig. 5).
The Journal of
272
Thoracic and Cardiovascular
Milano et at.
Surgery
1OO__~=..-----=--..-
---------,
90 80 Q) Q)
c:
~
70 60 ~o
40 30
332291257215 173149 118 222185162131 95 74 55 80 65 54 43 37 31 24
o
2
3
4
5
78 33 14
6
47 0 - 0 AVR 16 MVR 5 6 - 6 MAVR
e-e
8
9
10
11
12
Years Pastap
Fig. 6. Actuarial freedom fromanticoagulant-related hemorrhage. 100 __- - - - - - - - - - - - = - - - - - , ~~-e-e-e-e-e-e-e-. 99±1 \ --"0 T 6-6--..~-~-6-6-6-6-6 96±2 - - " 0 - 0 - 0 - 0 - 0 95±1 1
90
33429025721417314811879480-0AVR 222186164133 96 74 57 35 17 MVR 81 64 53 43 36 30 23 12 4 6 - 6 MAVR
e-e
80
+---+--+---+-~--+---+-+----+--+---+----+--------j
o
2
3
4
5
6
7
8
9
10
11
12
Years Postop
Fig. 7. Actuarial freedom from prosthetic valve endocarditis.
Anticoagulant-related hemorrhage. Hemorrhage related to anticoagulation occurred in 15 patients after AVR (0.9% ± O.2%/pt-yr), in nine after MVR (0.9% ± 0.3%/pt-yr), and in six after MAVR (1.8% ± 0.7%/ptyr). Fatal hemorrhage occurred in fivepatients with AVR (0.3% ± O.l%/pt-yr), in three with MVR (0.3% ± O.I%/pt-yr), and in three with MAVR (0.9% ± 0.5%/ pt-yr); eight died of cerebral hemorrhage and three of gastrointestinal bleeding. Actuarial freedom from anticoagulant-related hemorrhage at 9 years is 94% ± 2% for AVR, 91% ± 4% for MVR, and 68% ± 16% for MAVR'patients (Fig. 6). Endocarditis. Prosthetic valve endocarditis was observed in 12 patients after AVR (0.7% ± 0.2%/pt-yr) after a mean interval of 1.7 ± 2.0 years, in two after MVR (0.2% ± 0.1%/pt-yr) after 1 and 2 months, and in three after MAVR (0.9% ± 0.5%/pt-yr) after 20 days, 1 month, and 3 months; infection involved the mitral prosthesis in all of them. Five patients with AVR were reoperated on, with one death, and seven were managed
medically. Two of the seven managed medically died, and in five infection was cured but with persistence of a trivial residual paraprosthetic leak. One patient with MVR was reoperated on and died, and the other died without reoperation. Finally, all three patients with MAVR died before reoperation. Actuarial freedom from endocarditis at 9 years is 99% ± 1% after MVR, 96% ± 2% after MA VR, and 95% ± I% after AVR (Fig. 7). Paraprosthetic leak. Prosthetic detachment was observed in 21 patients with AVR (1.3% ± 0.2%/pt-yr) after a mean interval of 2.4 ± 2.3 years, in two with MVR(0.2% ± O.I%/pt-yr), both within 1 month,andin four with MAVR (1.2% ± 0.6%/pt-yr); of the latter, three had detachment of the mitral valve after 15 days, 20 days, and 5 months, and one had detachment of the aortic prosthesis after 3 months. In the AVR group 18 patients were reoperated on, and there were no operative deaths; three died before reoperation. Fifteen of the 18 had replacement of the prosthesis and three had direct suture of the leak. Both patients with MVR had successful prosthetic replacement at reoperation. Two patients with MAVR died before reoperation, and two had successful prosthetic replacement at reoperation. Actuarial freedom from paraprosthetic leak at 9 years is 99% ± I% after MVR, 95% ± 3% after MAVR, and 90% ± 10% after AVR (Fig. 8). Reoperation. During the follow-up period 31 patients underwent reoperation: 24 after AVR (1.4% ± 0.3%/ pt-yr), five after MVR (0.5% ± 0.2%/pt-yr), and two after MAVR (0.6% ± O.4%/pt-yr). In the AVR group reoperation was required because of paravalvular leak in 18 patients and thrombosis in one; there were no deaths. Endocarditis occurred in five, with one death (20%; 2.6% to 53.2% CL). Two patients with MVR had successful reoperation for paravalvular leak, and one patient with paravalvular leak and one with thrombosis died at reoperation. Two patients with MAVR were successfully reoperated on for paravalvular leak of the aortic and mitral prostheses, respectively. No cases of structural deterioration were observed in any patients at any valve position. Actuarial freedom from reoperation at 9 years is 97% ± 2% after MAVR, 92% ± 4% after MVR, and 89% ± 3% after AVR (Fig. 9). Overall valve performance. Freedom from valverelated mortality and reoperation at 9 years is 80% ± 5% after MVR, 75% ± 3% after AVR, and 56% ± 14% after MAYR. Freedom from valve-related mortality and permanent disability at 9 years is 83% ± 3% after AVR and MVR and 55% ± 13% after MAYR. Freedom from
Volume 103 Number 2
Sarin tilting-disc prosthesis
February 1992
100
~~~~=~=~=~=~=~=~-697±2 -0-0-0
90
-"""'0-0-0
"""T
e -"""'0 1
Q) Q)
Q) Q)
G:
273
92±4 89±3
80
G:
~
70
70
60
60
331287252208 169144 114 75 4 5 0 - 0 AVR 22218516.3 132 95 73 55 34 17 e-e MVR 80 52 52 42 35 30 23 12 5 /:,-/:, MAVR 50 +--+-+----+--+----l--+--f'-----+--+---+---t---1 o 2 3 4 5 5 7 8 9 10 11 12
50
3.33 286 251 217 168 143113 223 185 163 132 95 73 56 80 53 52 42 36 30 23 0
2
4
3
Yeors Postop
7
8
9
10
11
12
Fig. 9. Actuarial freedom from reoperation for all causes.
all valve-related morbidity and mortality at 9 years is 67% ± 6% after MYR, 63% ± 4% after A YR, and 41% ± 15% after MAYR (Fig. 10).
100 90· 80 70
Mechanical prostheses have been used in the preceding years with increasing frequency, mainly to overcome the problem .of the limited durability of tissue valves. 12, 13 Those mechanical prostheses utilizing a tilting-disc mechanism have gained worldwide acceptance, particularly because of their favorable hemodynamic characteristics.i However, complications such as systemic embolism, acute prosthetic thrombosis, and hemorrhage related to the anticoagulant treatment remain a major concern with both tilting-disc and other mechanical devices. 14, 15 In recent years, various new prosthetic models have been manufactured and used clinically to minimize such problems.l'"!? Among these, the Sorin tilting-disc prosthesis has been available in our country for more than a decade, but so far only a few studies have reported the early and mediumterm results with this device. 20, 21 Therefore we have reviewed all patients who received a Sorin prosthesis at our institution. Maximum follow-up time has been 10 years, which is a time frame long enough to assess the overall performance of a cardiac valve substitute. Although the present experience refers to patients operated on with better techniques of myocardial protection, early mortality was not dissimilar to that observed by our group in patients receiving other prosthetic valves,6,7 since also in this series most patients were in an advanced functional class at operation. It must also be considered that many of them had previously undergone one or more valvular procedures, and patients at higher risk, such as those with acute aortic dissection, were also included in the A VR group.
5
Yeors Postop
Fig. 8. Actuarial freedom from para prosthetic leak.
Discussion
5
4 5 0 - 0 AVR 18 e-e MVR 5 /:,-/:, MAVR
74 35 13
Q)
~
l •.
~
50
\-e~~-e-e --e __ --/:,__ --l::J.
-
--o~
e - e __ T
/:'-/:'-2===~-0-e --/:,
.
\1
50
/:,
40
57±5 53±4
41±15
.\0 20
351 278244 199 159 129 97 54 38 0 - 0 AVR 238182157 127 90 68 50 30 15 e-e MVR 87 52 51 41 33 27 20 11 4 /:,-/:, MAVR O+----+--+---+--+-----j--+--+---+--+---+--+----1 o 2 3 4 5 6 7 8 9 10 11 12
10
Years Postop
Fig. 10. Actuarial freedom from all valve-related morbidity and mortality.
The occurrence of thromboembolic complications was substantially low and again comparable with that reported with other mechanical prostheses.v" The total number of patients with valve occlusion by thrombus formation at initial presentation was lower than that observed by us with other tilting-disc valves/'- 7 Nevertheless, prosthetic thrombosis is still a matter of great concern, as much so as the rate of hemorrhage related to the anticoagulant treatment. Both complications, which significantly influenced the number of prosthesis-related deaths, once again clearly witness to the difficulty in maintaining adequate and safe long-term anticoagulation despite our continuing effort to observe as many patients as possible at our institution. A modification in the Sorin prosthesis, by coating the Teflon sewing ring with a thin film of turbostratic carbon, has been introduced with the aim of increasing hemocompatibility of the fabric surface, ideally reducing valve thrombogenicity. Because of the small number of patients who received this new model and their shorter
The Journal of
274
Milano et al.
follow-up, however, no significant difference in the rate of thromboembolic complications can thus far be documented in the present series. Structural deterioration, manifested by strut fracture or disc escape, has been observed with various prosthetic valves. 22- 26 This has occurred with both the spherical disc and the convexo-concave model of the Bjork-Shiley prosthesis, in which the struts are incorporated in the frame by fusion welding. The welding sites have been demonstrated to be potential sites of weakness for fatigueinduced Iesions.F:25 Sorin prostheses are manufactured by casting the whole cage in a single piece. The result is a perfectly homogeneous material structure, including those areas subjected to the greatest stress. The casting process also allows the struts to be designed with a crosssectional area optimized to obtain uniform stress conditions within these structures.i? In fact, mechanical rupture has never occurred either in our patients or, to the best of our knowledge, in those of others.P: 21 In conclusion, (I) the Sorin prosthesis shows an overall satisfactory performance in the first decade of clinical use that is substantially comparable with that of other mechanical valves; (2) thromboembolic episodes may be controlled by careful anticoagulant treatment, but hemorrhagic complications are still an important cause of postoperative morbidity; (3) the most striking advantage of this device appears to be its excellent durability without any episode of structural failure. In our opinion the Sorin prosthesis can be considered a reliable cardiac valve substitute for mitral and aortic valve positions in patients in whom insertion of a mechanical prosthesis is indicated. We thank Gian Carlo Pengo for his technical assistance. REFERENCES I. Mitha AS, Matisonn RE, le Roux BT, Chesler E. Clinical experience with the Lillehei-Kaster cardiac valve prosthesis. J THORAC CARDIOVASC SURG 1976;72:401-7. 2. Bjork VO, Henze A. Ten years' experience with the BjorkShiley tilting disc valve. J THORAC CARDIOVASC SURG 1970;78:331-42. 3. Karp RB, Cyrus RJ, Blackstone EH, Kirklin JW, Kouchoukos NT, Pacifico AD. The Bjork-Shiley valve: intermediate-term follow-up. J THORAC CARDIOVASC SURG 1981;81:602-14. 4. Lindblom D. Long-term clinical results after mitral valve replacement with the Bjork-Shiley prosthesis. J THORAC CARDIOVASC SURG 1988;95:321-33. 5. Lindblom D. Long-term clinical results after aortic valve replacement with the Bjork-Shiley prosthesis. J THORAC CARDIOVASC SURG 1988;95:658-67.
Thoracic and Cardiovascular Surgery
6. Milano A, Bortolotti U, Mazzucco A, Guerra F, Magni A, Gallucci V. Aortic valve replacement with the Hancock standard, Bjork-Shiley, and Lillehei-Kaster prostheses: a comparison based on follow-up from I to 15 years. J THORAC CARDIOVASC SURG 1989;98:37-47. 7. Milano A, Bortolotti U, Mazzucco A, et al. Mitral valve replacement with the Hancock, Bjork-Shiley and LilleheiKaster prostheses: a comparison based on a 15-year followup. Eur J Cardiothorac Surg 1989;3:312-20. 8. Bentall H, De Bono A. A technique for complete replacement of the ascending aorta. Thorax 1968;23:338-9. 9. Edmunds LH Jr, Clarke RE, Cohn LH, Miller DC, Weisel RD. Guidelines for reporting morbidity and mortality after cardiac valvular operations. J THORAC CARDIOVASC SURG 1988;96:351-3. 10. Grunkemeier GL, Starr A. Actuarial analysis of surgical results: rationale and methods. Ann Thorac Surg 1977; 24:404-8. II. Grunkemeier GL, Thorns DR, Starr A. Statistical considerations in the analysis and reporting of time-related events. Am J Cardiol 1977;39:257-8. 12. Bortolotti U, Milano A, Mazzucco A, et al. Results of reoperation for primary tissue failure of porcine bioprostheses. J THORAC CARDIOVASC SURG 1985;90: 564-9. 13. Bortolotti U, Milano A, Mazzucco A, et al. The Hancock pericardial xenograft: incidence of early mechanical failures at a medium-term follow-up. Eur J Cardiothorac Surg 1988;2:458-64. 14. Moreno-Cabral RJ, McNamara JJ, Mamiya RT, Brainard SC, Chung GKT. Acute thrombotic obstruction with Bjork-Shiley valves:diagnostic and surgical considerations. J THORAC CARDIOVASC SURG 1978;75:321-30. 15. Edmunds LH Jr. Thrombotic and bleeding complications of prosthetic heart valves. Ann Thorac Surg 1987;44:43045. 16. Emery RW, Mettler E, Nicoloff JM. A new cardiac prosthesis: the St. Jude Medical cardiac valve: in vivo results. Circulation 1979;60 {Pt 2):140-54. 17. Hashimoto A, Nagase Y, Koyanagi H. A new tilting valve prosthesis, "Omniscience valve:" clinical and hemodynamic evaluations. Artif Organs 1982;11:2-8. 18. Scotten LN, RaccaRG, NugentAH, Walker DK, Brownlee RT. New tilting disc cardiac valve prostheses: in vitro comparison of their hydrodynamic performance in the mitral position. J THORAC CARDIOVASC SURG 1981; 82:136-46. 19. Beaudet RL, Poirier NL, Doyle D, Nakhle G, Gauvin C. The Medtronic Hall cardiac valve: 7V2 years' clinical experience. Ann Thorac Surg 1986;42:644-50. 20. Colombo T, Donatelli F, Quaini E, Vitali E, Pellegrini A. Results of heart valve replacement with the Sorin prosthesis. Texas Heart Inst J 1987;14:77-87. 21. Sante P, Festa M, Renzulli A, et al. Clinical results after isolated aortic valve replacement with Sorin prosthesis: a 6-year experience. Ital J Surg Sci 1988;18:323-6.
Volume 103 Number 2 February 1992
22. Larrieu AJ, Puglia E, Allen P. Strut fracture and disc embolization of a Bjork-Shiley mitral valve prosthesis: localization of embolized disc by computerized axial tomography. Ann Thorac Surg 1982;34:192-5. 23. Hjelms E. Escape of a leaflet from a St. Jude Medical prosthesis in the mitral position. Thorac Cardiovasc Surg 1983;31:310-2. 24. Ibarra F, Gutierrez A, Martinez F, Carreras L, Lopez C, Alonso-Lej F. Fracture of the outlet strut of a Bjork-Shiley mitral prosthesis: emergency operation with survival. J THORAC CARDIOVASC SURG 1984;87:315-8. 25. Rockelein G, Breme J, von der Emde J. Lethal blockage of
Sarin tilting-disc prosthesis
275
a Bjork-Shiley artificial heart valve caused by strut fracture-the metallurgic aspect. Thorac Cardiovasc Surg 1989;37:47-5l. 26. Kepletko W, Moritz A, Mlczoch J, Schurawitzki H, Domanig E, Wolner E. Leaflet fracture in Edwards-Duromedics bileaflet valves. J THORAC CARDIOVASC SURG 1989;97:90-4. 27. Tilting disc valves: comparison between Shiley and Sorin models. The progress reports of research program TBM/ CNR. SP-CARD/3, 1977-1978 (National Research Council).
Bound volumes available to subscribers Bound volumes ofTHE JOURNAL OF THORACIC AND CARDIOVASCULAR SURGERY are available to subscribers (only) forthe 1992 issues from the Publisher, at a costof $71.00 fordomestic, $93.97 forCanadian, and $89.00 for international subscribers for Vol. 103 (JanuaryJune) and Vol. 104 (July-December). Shipping charges are included. Each bound volume contains a subjectand author index and all advertising is removed. Copies are shipped within 60 daysafter publication ofthe lastissue of the volume. The binding isdurablebuckram with the JOURNAL name, volume number, and yearstamped in goldon the spine. Payment must accompany all orders. Contact MosbyYearBook, Inc., Subscription Services, 11830 Westline Industrial Drive, St. Louis, Missouri 63146-3318, USA; phone (800) 325-4177, ext. 4351, or (314) 453-4351. Subscriptions mustbe in force to qualify. Bound volumes are not available in placeof a regular JOURNAL subscription.