Late thrombosis of the Björk-Shiley tilting disc valve in the tricuspid position

Volume 91,

Number 1

January 1986

THORACIC AND CARDIOVASCULAR SURGERY The Journal

J THORAC

CARDIOVASC SURG

of

91:1-8, 1986

Original Communications

Late thrombosis of the Bjork-Shiley tilting disc valve in the tricuspid position Thrombolytic treatment with streptokinase The main complication after implantation of a Bjork-Shiley tilting disc valve in the tricuspid position is latethrombotic obstruction. Of 28 patients with tricuspid valve replacement (16 with mitral, aortic, and tricuspid valve replacement; 12 with mitral and tricuspid valve replacement~ with a meanfollow-up of 5.2 years, seven (25%) bad thrombosis of the tricuspid prosthesis. Three patients bad a recurrent thrombotic malfunction, for a total of ten thrombotic malfunctioM in 146 patient-years, a rate of 6.8 per 100 patient-years. Thrombosis occurred late in all patients. Clinical deterioration presented with sigM of congestive heart failure. In all patients the click of the tricuspid prosthesis was not audible and Dew diastolic or systolic murmurs were. The diagnosis was confirmed with cineradiography and bidimensional echocardiography (immobile dis(; diminished opening angle of the disc). Thrombolytic treatment with streptokinase was used in all seven patients.Two patients required 12 hoursof therapy and five patieqts, 24 hours. Thrombolytic treatment was monitored by the thrombin time. Complete regression of clinical, cineradiographic, and echocardiographic sigm; of thrombosis wasseenin all seven patientsduring the fD'St 24hours of therapy.There wereno bleeding complicatiOM. In onepatient clinical sigMof mild pulmonary embolism occiuTed and were confirmed with chest radiographs. Follow-up, after successful treatment, extends from 4 to 30 months (mean 16.5 months~ In four patientslong-term results are excellent: There have been 00 clinical, cineradiographic, or echocardiographic signs of rethrombosis of the tricuspid prosthesis during the follow-up. Rethrombosis of the tricuspid prosthesis was observed in three patients4, 7, and 14 months after initial treatment with streptokinase. Repeat thrombolytic treatment with streptokinase wassuccessful in all three of these patients. Our experience with streptokinase treatment of thromI:Josis of tricuspid Bjork-Shiley prostheses indicates that this form of treatment should always be applied before surgical intervention.

Dejan Boskovic, M.D., Ph.D. (by invitation), Ivo Elezovic, M.D., Ph.D. (by invitation), Darinka Boskovic, M.D. (by invitation), Natalija Simin, M.D. (by invitation), loran Rolovic, M.D. (by invitation), and Vladan Josipovic, M.D. (by invitation), Beograd. Yugoslavia Sponsored by W. Dudley Johnson, M.D., Milwaukee. Wis. From the Departments of Cardiology and Hematology, Clinical Center, Medical Faculty, Beograd, Yugoslavia. Read at the Sixty-fifth Annual Meeting of The American Association for Thoracic Surgery, New Orleans, La., April 29-May I, 1985.

Address for reprints: Dejan Boskovic, M.D., Ph.D., Assistant Professor of Internal Medicine, Borisa Kidrica 47-49, Beograd, Yugoslavia.

2

The Journal of Thoracic and Cardiovascular Surgery

Boskovic et al.

Despite anticoagulant therapy and improvements in the materials and design of prosthetic valves, thrombosis of the prosthesis remains a continuing problem. In patients with Bjork-Shiley tilting disc valves, the risk of thrombosis is higher in the tricuspid position than in the mitral and aortic positions.J Relief of this serious complication can be obtained only by thrombolytic treatment or surgical replacement of the clotted prosthesis. This communication reports our experience with streptokinase treatment of thrombosed Bjork-Shiley prostheses in the tricuspid position. Patients From January, 1976, through December, 1982, 28 patients survived multiple Bjork-Shiley valve replacements including replacement of the tricuspid valve. (All tricuspid prostheses were Bjork-Shiley, standard, flat disc valves.) Sixteen of the 28 had required triple valve replacement and 12 had required double (mitral and tricuspid) valve replacement. The follow-up period ranged from 4 months to 8 years, averaging 5.2 years. Thrombotic malfunction of the tricuspid prosthesis developed in seven patients (25%). Three patients had a recurrent thrombotic malfunction, for a total of ten thrombotic malfunctions in 146 patient-years, an incidence of 6.8 per 100 patient-years. All seven Patients were women, with a mean age of 34.5 years (range 22 to 51 years). Five patients had triple and two patients double valve replacement. In all patients the large standard Bjork-Shiley tricuspid T-31 valve was implanted. Five patients had cardiomegaly (cardiothoracic ratio 0.51 to 0.55). Three patients presented with atrial fibrillation, two with sinus rhythm, one with atrioventricular junctional rhythm, and one with a pacemaker. All patients were receiving anticoagulants. The criterion for adequate anticoagulation was defmed as maintenance of the prothrombin time 1.5 to 2 times the control level on at least 75% of the occasions tested.' Five patients were receiving long-term anticoagulants alone and two were receiving anticoagulants in combination with antiplatelet therapy (dipyridamole in one and aspirin in the other). Anticoagulation was deemed inadequate in six of seven patients (86%). The time interval between tricuspid valve replacement and the diagnosis of first prosthetic valve thrombosis is presented in Fig. 1. Thrombosis occurred late in all patients, in the third (two patients), fourth (two patients), sixth, seventh, and eighth postoperative years. In all seven cases clinical presentation of thrombotic valve malfunction was in the form of mild or moderate right-sided congestive heart failure. On auscultation the

POSTOPERATIVE YEARS

o

2

3

4

5

,6

7

8

9

10

CASE 1. CASE 2. CASE 3. CASE 4. CASE 5. CASE 6. CASE 7.

Fig. 1. Time interval between tricuspid valve replacement and the diagnosis of prosthetic valve thrombosis.

tricuspid prosthetic click was absent and a long diastolic murmur was heard at the lower left sternal border in all patients. A high-frequency systolic murmur, increasing on inspiration, was present at the same location in six patients. Bidimensional echocardiograms were performed on all patients. Disc motion of the tricuspid prosthesis was studied from the apical four-chamber view.An M-mode echocardiogram through the prosthetic tricuspid valve was simultaneously recorded from the same position. The apical four-chamber view demonstrated an immobile disc in six patients and very restricted disc movement in the remaining patient. All malfunctioning valve discs had a radiopaque marker. Therefore, cineradiography was the most effective method of assessing valve function. Cineradiographic studies were performed in all seven patients. The disc was observed to be immobile, in the semi-open position, in six patients and to have restricted opening movements and incomplete closure in the remaining patient. Fibrinolytic treatment The fibrinolytic agent was streptokinase in all ten episodes of valve thrombosis. The time interval between confirmation of initial tricuspid valve thrombosis and the application of streptokinase varied (Fig. 2). In two patients the period was 3 months; in one, I month; in two, 2 weeks; and in two patients, 1 week. For the two patients with a 3 month interval, the delay was purposeful. One patient was pregnant and refused treatment until after delivery and the other had uterine bleeding. Streptokinase was administered continuously via a peripheral vein with an infusion pump. Initial dosage was determined with a streptokinase resistance test. An initial loading dosage of 250,000 international units

Volume 91 Number 1 January, 1986

Thrombosis of Bjork-Shiley valve 3

WEE K S

o

2

3

I

"

4

5

6

7 I

e

9

10

11

12

, .

CASE 1. CASE 2 CASE 3.

CASE 4. CASE 5. CASE

e.

CASE 7.

I---t H

Fig. 2. Time interval between diagnosis of tricuspid Bjork-Shiley valve thrombosis and treatment with streptokinase.

(IV) together with 80 mg of hydrocortisone was administered over 30 minutes in six patients. In the remaining patient the initial dosage was 750,000 IV. The maintenance dosage in six patients averaged 150,000 Il.I/hr, and in one patient, 100,000 Il.J/hr. Two patients required 12 hours of therapy and five patients, 24

hours. In all three cases of valve rethrombosis the resistance to streptokinase was increased. The initial dosage in two patients was 500,000 IV and in the third it was 750,000 IV. In two patients with valve rethrombosis the maintenance dosage was 150,000 IV /hr and in one patient, 185,000 IV/hr. Duration of streptokinase treatment in these patients was 24 hours in two cases and 48 hours in one case. Streptokinase therapy was monitored and controlled with thrombin time. Treatment was considered adequate if the thrombin time was 2 to 4 times the normal control value. The first test was performed 4 hours after the initial dosage and every 4 hours for the duration of treatment. As soon as the tricuspid prosthetic click was heard and normal valve function was confirmed with echocardiography or cinefluoroscopy, the streptokinase treatment was discontinued. Heparin was then administered in a dosage of 10,000 units every 6 hours for 24 hours and then 5,000 units every 6 hours for 24 more hours. Warfarin was started during the second 24 hours, as were platelet inhibitors. Three patients received dipyridamole and aspirin, three dipyridamole, and one

aspirin.

Results There was prompt and complete therapeutic success in all seven cases. Clinical improvement was dramatic, .

with regression of all signs of congestive heart failure, disappearance of the prolonged diastolic murmur as well as the systolic murmur, and recovery of the normal prosthetic click of the tricuspid prosthesis. In all patients the motility of the radiopaque tilting disc was restored to normal on cineradiographic examination (Figs. 3 and 4). Bidimensional echocardiographic examination showed a normal disc motion pattern in all patients (Fig. 5). Clinical course. The clinical course is shown in Fig. 6. Follow-up after initially successful treatment with streptokinase extends from 4 to 30 months (mean 16.5 months). Four patients have required no further treatment, and there have been no clinical, cineradiographic, or echocardiographic signs of rethrombosis of the tricuspid prosthesis during the follow-up period. Anticoagulant therapy has been adequate. In three patients rethrombosis occurred 4, 7, and 14 months after initial treatment with streptokinase. In two of these patients anticoagulant therapy was adequate, but one of these patients was using contraceptive pills. In the third patient anticoagulant therapy was inadequate. Repeat thrombolytic treatment with streptokinase was performed with success in all three patients. During the follow-up of these three patients, which is relatively short (4 to 6 months), there have been no signs of recurrence. In all three patients anticoagulation therapy has been adequate after the second thrombolytic treatment. Complications. There were no hemorrhagic complications. Febrile reactions during the later course of streptokinase (between 20 and 24 hours) occurred in four patients and were self-limited. Psychic disturbances occurred in four patients. Early reactions were muscular

The Journal of

4

Boskovic et al.

Thoracic and Cardiovascular Surgery

Fig. 3. Selected frames from cineradiography of a patient with double (mitral and tricuspid) Bjork-Shiley valve replacement before and after treatment with streptokinase. M, Mitral prosthesis. T, Tricuspid prosthesis. A, Pretreatment, systole, mitral prosthetic disc is in normal, closed position, while tricuspid prosthetic disc is immobile and fixed in semi-open position. B, Post-treatment, diastole, mitral and tricuspid prosthetic discs are both fully open. C. Post-treatment, systole, mitral and tricuspid discs are both fully closed.

pain in two patients and headache in one. In one patient allergic reaction in the form of urticaria, caused by the quickly given initial dosage of streptokinase, was successfully treated with corticosteroids without discontinuation of treatment. In another patient a clinical picture of mild pulmonary embolism (dyspnea, cough, and pleuritic pain) developed and was confirmed by chest radiograph (a lung scan was not performed). In three patients without clinical signs of pulmonary embolism, the absence of embolism was confirmed by lung scintigraphy. Discussion Several authors have reported and discussed the problem of thrombotic malfunction of the tricuspid Bjork-Shiley valve."? However, data on the time-related incidence are sparse. We found only four reports that

provided data on the time-related incidence of this complication in a reasonably large group of patients (Table I). The first successful thrombolytic treatment of tricuspid Bjork-Shiley valve thrombosis was reported by Inberg and associates" in 1977. Since then, various authors have reported treatment with streptokinase in relatively small numbers of patients,"? II Guidelines for patient selection for thrombolytic therapy commonly state that there should be evidence that the thrombus is of recent origin (a few days to 1 week at the most). Peterffy and colleagues' showed in their experimental studies and clinical observation that thrombolytic treatment with streptokinase was effective provided that prosthetic thrombosis was due to relatively recent red clot. This suggests that the diagnosis of valve thrombosis must be made as early as possible so that

Volume 91 Number 1 January, 1986

Thrombosis of Bjork-Shiley valve

5

Fig. 4. Selected frames from cineradiography of a patient with triple (mitral, aortic, and tricuspid) Bjork-Shiley valve replacement before and after application of streptokinase. T. Tricuspid prosthesis. A, Pretreatment, systole, tricuspid prosthetic disc is immobile and fixed in semi-open position. B, Post-treatment, diastole, tricuspid prosthetic disc is fully open. C. Post-treatment, systole, tricuspid disc is fully closed.

treatment with streptokinase can be initiated before organization and development of fibrosis of the throm-

bus. In our group of patients, valve function was promptly restored even though the valves had been malfunctioningfor as long as 3 months in two patients and 1 month in one patient. However, incomplete resolution of the prosthetic valve clot may have been a causative factor in the three recurrences. In two of the patients with rethrombosis, the time interval between diagnosis of initial valvethrombosis and streptokinase treatment was long (12 weeks in one and 4 weeks in the other patient). Although better results can be expected from early therapy, our experience suggests that thrombolytic therapy should be considered even in cases of prolonged thrombosis. In one of our patients a mild pulmonary embolism occurred. Embolic episodes originating from the tricus-

Table I. Incidence of Bjork-Shiley tricuspid valve thrombosis First author

Ryder'?

Tornburn" Peterffy" Wellens" Boskovic

No. of patients

Mean follow-up (yr)

Thrombosisfl 00

25 32 52 19 28

6.3 5.3 4.9 6.0 5.2

0.63 3.0 3.2 4.3 6.8

pt-yr

pid prosthetic valve are not easy to recognize because they do not always produce detectable signs and symptoms. Thus, pulmonary embolism may have occurred without being detected in our group of patients (in only three patients were lung scintiscans performed). Inadequate anticoagulation seems to be the most

The Journal of Thoracic and Cardiovascular Surgery

6 Boskovic et al.

Fig. 5. Bidimensional echocardiograms (apical four-chamber view) of a patient with double (mitral and tricuspid) Bjork-Shiley valve replacement before and after the treatment with streptokinase. A, Pretreatment, mitral prosthetic disc (M) is moving normally in diastole while tricuspid prosthetic disc (T) is immobile. B, Post-treatment, normal motion pattern of mitral and tricuspid prosthetic disc in diastole. MONTHS

o

4

8 ,

12

16

20

24

28 ,

32 I

36 ,

40 ,

CASE 1. CASE 2. CASE 3. CASE 4. CASE 5.

I I I I

RT

II-RT RT

I~

CASE 6.

HI--

CASE 7.

~

Fig. 6. Follow-up of patients after successful treatment with streptokinase. RT. Rethrombosis of tricuspid Bjork-Shiley valve.

important factor in the development of tricuspid valve thrombosis. Nevertheless, there are cases in which prosthetic thrombosis developed despite "adequate" anticoagulation. We suggest that, after successful treatment with streptokinase, oral anticoagulants should be supplemented with thrombocyte inhibitors to minimize the chances of rethrombosis. Streptokinase has now been administered to many patients according to only standardized dose schedules. However, the most serious complication with streptokinase treatment is hemorrhage. Therefore, we prefer careful clinical and laboratory management of thrombolytic therapy. Several tests can be used to monitor the maintenance dose of streptokinase. We believe that

thrombin time is the most useful test for this purpose. 14 We had no bleeding complications in our ten treatments. It is our opinion that streptokinase treatment for more than 48 hours is inappropriate. Failure of streptokinase treatment to resolve the thrombus after 48 hours indicates that tissue overgrowth or fibrous, organized thrombus is the cause of prosthetic malfunction. If streptokinase treatment fails, surgical intervention becomes necessary. It may be desirable to replace a thrombosed tricuspid prosthesis with a tissue valve in order to decrease the risk of further valve thrombosis. Urokinase is a more expensive alternative to streptokinase. It can be used in patients allergic to streptokinase, and it is recommended for rethrombosis." Potentially, the use of tissue plasminogen activator would be even more effective. We did not use urokinase or tissue plasminogen activator because they were not available to us. We recommend close follow-up, since rethrombosis of the tricuspid prosthesis may occur despite effective anticoagulant therapy, as demonstrated by two of our patients (one of these was taking contraceptive pills). Besides clinical examination, we are performing noninvasive examinations (cinefluoroscopy, cineradiography, or echocardiography) every 3 months for early detection of recurrent thrombus formation. Conclusion This series demonstrates a very high success rate for streptokinase treatment of prosthetic tricuspid valve

Volume 91 Number 1

Thrombosis of Bjork-Shiley valve 7

January. 1986

thrombosis and rethrombosis. Minimal complications occurred, perhaps because of careful titration of doses. Clot at least 3 months old was successfully lysed. Surgical intervention for this problem should be considered only after thrombolytic therapy has been tried and has failed. We acknowledge the assistance of Mr. Kenneth L. Kayser and Mrs. Susan M. Kender in the preparation of the manuscript.

for thrombotic complication of valve prosthesis after tricuspid valve replacement. Scand J Thorac Cardiovasc Surg 11:195-198, 1977 14 Kakkar VV, Scully MF: Thrombolytic therapy. Br Med Bull 2:191-199, 1978 15 Witchitz S, Veyrat C, Moisson P, Scheinman N, Rozenstajn L: Fibrinolytic treatment of thrombus on prosthetic heart valves. Br Heart J 44:545-554, 1980

Discussion DR. VIKING 0. BJORK

REFERENCES Bjork VO, Henze H: Ten years' experience with the Bjork-Shiley tilting disc valve. J THORAC CARDIOVASC SURG 78:331-342, 1979 2 Bamhorst DA, Oxman HA, Connolly DC, Pluth JR, Danielson GK, Wallace RB, McGoon DC: Long-term follow-up of isolated replacement of the aortic or mitral valve with the Starr-Edwards prosthesis. Am J Cardiol 35:228-233, 1975 3 Azpitarte J, De Vaga NG, Santalla A, Rabago P, Rabago G: Thrombotic obstruction of Bjork-Shiley tricuspid valve prosthesis. Acta Cardiol 6:419-426, 1975 4 Raj MV, Srinivas V, Evans DW: Thrombotic jamming of tricuspid prosthesis. Br Heart J 38:1355-1358, 1976 5 BourdillonPD, Sharratt GP: Malfunction of Bjork-Shiley prosthesis in tricuspid position. Br Heart J 38: 1149-1153, 1976 6 Schoevaerdts JC, Jaumin P, Piret L, Kremer R, Ponlot R, Chalant CH: Tricuspid valve surgery. J Cardiovasc Surg 18:397-399, 1977 7 Bjork VO, Henze A, Peterffy A: Can a mechanical heart valve be used in the tricuspid position? Experience with the Bjork-Shileytilting disc valve in 70 patients. Eur Heart J 1:55-61, 1980 8 Peterffy A, Henze A, Savidge GF, Landou C, Bjork VO: Late thrombotic malfunction of the Bjork-Shiley tilting disc valve in the tricuspid position. Scand J Thorac Cardiovasc Surg 14:33-41, 1980 9 Beeuwsaert R, Denef B, De Geest H: Diagnosis and treatment of obstruction of a tricuspid Bjork-Shiley prosthesis. Acta Cardiol 1:13-25, 1983 10 RyderSJ, Bradley H, Brannan JJ, Turner MA, Bain WH: Thrombotic obstruction of the Bjork-Shiley valve. The Glasgow experience. Thorax 39:487-482, 1984 II Tornburn CW, Morgan JJ, Shanahan MX, Chang VP: Long-term results of tricuspid valve replacement and the problem of prosthetic valve thrombosis. Am J Cardiol 51:1128-1132, 1983 12 Wellens F, Van Dale P, Deuvaert FE, Leclerc JL, Primo G: The role of porcine heterografts in a l.4-year experience with tricuspid valve replacement, Cardiac Bioprostheses, Proceedings of the Second International Symposium, LH Cohn, V Gallucci, eds., New York, 1982, Yorke Medical Books, pp 502-515 13 Inberg MV, Havia T, Arstila M: Thrombolytic treatment

Stockholm, Sweden

In Stockholm we used streptokinase in four of seven patients with thrombosis in the tricuspid area. They have been followed up for 9 years. Two of them had Ebstein's disease, one was receiving aspirin and dipyridamole but not warfarin sodium, and one had anticoagulant therapy withdrawn after a traffic accident. Streptokinase restored normal movements in three cases with good long-term follow-up. However, in one case with incomplete resolution of the clot and an opening angle only reaching 49 degrees after 24 hours of streptokinase, the valve rethrombosed and reoperation was necessary. The flat disc was responsible for thrombosis in three cases. That is the same valve that the authors used. The disc comes all the way out to the ring in the fully opened position, and if anticoagulation therapy is not perfect thrombosis will start. We changed to the convexo-concave disc, in which there is a space between the disc and the ring. The monostrut version is now used all over the world except in the United States. Eight hundred such valves have been used at the Karolinska. Instantaneous thrombosis was much more pronounced in the flat disc than in the convexo-concave disc valve during the 5 year follow-up in two series comprising more than 350 flat and 350 convexo-concave disc valves. Also, the 7 year follow-up showed a 9% greater actuarial survival rate with the convexoconcave discs than with the flat disc valves. DR. RENE G. FAVALORO Buenos Aires. Argentina

It is an excellent idea to introduce streptokinase to combat this kind of problem, but I think it is much more important to prevent the problem. I imagine the majority of us are using biological valves in the tricuspid position. Nevertheless, last year I was able to present to this Association a valve that was totally constructed in Argentina. The characteristics of the valve are outlined in the November, 1984, issue of the JOURNAL (88:714, 1984). We have inserted 470 valves. I will report the long-term results-that was my promise last year---of the first 288 patients with at least I year of follow-up. The total mortality (operative and late) is 10.4%. A significant number of these cases were complicated: 28 patients had subacute bacterial endocarditis, 61 had coronary artery disease with concomitant revascularization, 13 had dissecting aneurysm, and 40 patients had previous operations. The majority of the valves, 180, were placed in the aortic

8 Boskovic et al.

position, 70 in the mitral position, and 38 were double or triple valve replacements. There have been only three cases of late embolization, all of them occurring in patients who were not receiving anticoagulants. There have been only two cases of bleeding due to the anticoagulant therapy, and there has been no instance of prosthetic failure. The majority of the patients were in Class III or Class IV preoperatively, and now the majority of them are totally asymptomatic. The characteristics of this valve are noteworthy. The design is simple but, more than that, there are only three points of contact between the disc and the anulus. As a consequence, there is less trauma for the blood element and less tendency to produce embolization. I really think that at this moment we have a mechanical valve that is much better than the others on the market.

The Journal of Thoracic and Cardiovascular Surgery

concave Bjork-Shiley tilting disc valve. Recently, in another hospital we had a patient with two episodes of thrombosis of the tricuspid convexo-concave Bjork-Shiley valve. The patient was reoperated on twice (the first time for thrombectomy and the second time for reimplantation of the tricuspid BjorkShiley valve). We have a little experience with biological valves in the tricuspid position, which does not allow us to compare long-term results with this type of valve and the Bjork-Shiley valve in the tricuspid position. The most important factor in preventing thrombosis of mechanical valves in the tricuspid position is adequate anticoagulant therapy, which was difficult to achieve in most of our patients. I think that oral anticoagulant therapy should be supplemented with antiplatelet therapy in all patients with a tricuspid Bjork-Shiley tilting disc valve.

DR. BOSKOVIC (Closing) I want to thank Dr. Bjork and Dr. Favaloro for their comments. We have no experience with the tricuspid convexo-

Bound volumes available to subscribers Bound volumes of THEJOURNAL OF THORACIC AND CARDIOVASCULAR SURGERY are available to subscribers (only) for the 1986 issues from the Publisher, at a cost of $49.00 ($66.00 international) for Vol. 91 (January-June) and Vol. 92 (July-December). Shipping charges are included. Each bound volume contains a subject and author index and all advertising is removed. Copies are shipped within 60 days after publication of the last issue of the volume. The binding is durable buckram with the JOURNAL name, volume number, and year stamped in gold on the spine. Payment must accompany all orders. Contact The C. V. Mosby Company. Circulation Department, 11830 Westline Industrial Drive, St. Louis, Missouri 63146, USA; phone (800) 325-4177, ext. 351. Subscriptions must be in force to qualify. Bound volumes are not available in place of a regular JOURNAL subscription.