Selective annuloplasty of the tricuspid valve

J THORAC CARDIOVASC SURG 1990;99:846-51

Selective annuloplasty of the tricuspid valve Two-year experience Between July 1986 and March 1989,48 patients whose ages averaged 60 ± 6 years were subjected to a new technique for tricuspid valve annuloplasty. The preoperative New York Heart Association functional class averaged 3.7. There were 32 mitral-tricuspid procedures, 13 aortic-mitral-tricuspid procedures, and three repeat tricuspid annuloplasties. The hospital mortality rate was 6.3 % with three deaths resulting from cardiac or respiratory failure, or both. Follow-up averaged 14 ± 8 months. One patient (2.2 %) died suddenly, 7 months postoperatively. AU patients whose foUow-up period lasted 6 months or more improved to class I or D. Twenty-six patients were reinvestigated by catheter or echocardiographic methods, or both. The gradient over the tricuspid valve averaged 1.4 ± 0.6 mm Hg, and a moderate regurgitant murmur (2/3) was detected in four cases. The diameter of the tricuspidal anulus in the apical four-chamber view decreased from 23.7 ± 3.9 mm/m2 body surface area preoperatively to 15.7 ± 1.9 mm/m 2 body surface area at late foUow-up examination. The present technique aUows an anatomic and functional reconstruction of the tricuspid valve with a good compromise between stenosis and regurgitation. The overall mortality rate is low and functional improvement is . striking.

C. Minale, MDa (by invitation), H. Lambertz, MDb (by invitation), S. Nikol, MDa (by invitation), N. Gerich, MDb (by invitation), and B. J. Messmer, MD,a

Aachen, Federal Republic of Germany

Experiences of recent years indicate that tricuspid valvuloplasty is the treatment of choice for acquired tricuspid insufficiency. Opinions still differ on the best way to render the valve competent without causing stenosis.!" All competent annuloplasties result in a significant reduction of the valve orifice and in about half ofthe cases 5-8 in a transvalvular gradient. The present study was undertaken to examine early and medium-term results after our method of tricuspid valve repair.

Methods Patient data. Between July 1986 and March 1989', a new method of tricuspid valve repair was attempted in 48 patients with multiple valve disease. There were 14 men and 34 women

From the Departments of Thoracicand Cardiovascular Surgery" and Internal Medicine I,b University Hospital of R.W.T.H., Aachen, Federal Republic of Germany. Read at the Sixty-ninth Annual Meeting of The American Association for ThoracicSurgery, Boston, Mass., May 8-10, 1989. Address for reprints: Prof. Dr. C. Minale, Department of Thoracicand . Cardiovascular Surgery,Klinikum RWTH Aachen, 51DO-Aachen, Federal Republic of Germany. 12/6/18198

846

with an average age of 60 ± 6 years (29 to 73 years). Preoperatively, 20 patients were considered to be in class III and 28 in class IV (New York Heart Association). Twenty-eight patients (59%) had had a previous cardiac operation. One patient had pure organic tricuspid insufficiency, 34 had functional insufficiency, and 11 patients had both. Three further patients had had previous repair of the tricuspid valve; the Carpentier ring technique was used in two and the De Vega annuloplasty in the third. In the first two patients the ring was broken and in the third the leaflets were incompetent. Preoperatively, right atrial v-wave and pulmonary mean pressures averaged 19 ± 10 and 38 ± 13 mm Hg, respectively. Surgical technique and operative data. Cardioplegic arrest with standard cardiopulmonary bypass technique and moderate general hypothermia were used for the operation. Tricuspid valve repair was concomitant to mitral valve replacement in 32 patients and to aortic-mitral valve replacement in 13. Additional coronary artery bypass grafting was done in six patients. Details of the tricuspid valve repair are illustrated in Fig. 1.Anterior and posterior leaflets of the valve were separated from the anulus over a length of 5.1 ± 1.3 cm (2.5 to 8.0). An average of 3.6 ± 0.9 em (1.5 to 6.0) of the isolated anulus was excluded within the annulorrhaphy, which accounts for 69% of the whole length. The longitudinal diameter of the obturator, which fit before and on completion of the annulorrhaphy, averaged 45.6 ± 7.5 mm (34.0 to 62.0) and 32.9 ± 3.5 mm (28.0 to 38.0), respectively. The tricuspid valve was sized before and on completion of the procedure with obturators similar to those of the Carpentier ring.

Volume 99 Number 5 May 1990

Tricuspid valve annuloplasty 8 4 7

I

.iI.;)

\

\

\

Fig. 1. Top left, Schematic representation of a tricuspid valve. The leaflets are slightly retracted, the anulus is enlarged, and the valve is incompetent. Top right, The anterior and posterior leaflets are separated completely from the anulus along the lateral commissure. The three leaflets show a good coaptation in the middle. Bottom left, Two thirds of the free part of the anulus has been excluded by a continuous suture (3-0 Ti-Cron suture, Davis & Geck, Danbury, Conn.). Bottom right, Starting from the middle, the cut edge of the tricuspid leaflets has been anchored to the remaining third ofthe anulus by a continuous suture (4-0 suture, Ethicon, Inc., Somerville, N.J.). The shapes of the tricuspid valve before (continuous black line) and on completion of the repair are compared. Leaflet surface is unmodified with respect to the preoperative state; anterior and posterior leaflets have been shifted centrally toward the septal leaflet and the anulus circumference has been reduced.

Echocardiography. M-mode and two-dimensional echocardiography were performed before the operation, before discharge from the hospital, and at a late postoperative stage. From the apical transducer position, both atrioventricular valvescould be visualized simultaneously. The color-coded flow-mapping technique was the method applied for assessing the degree of tricuspid regurgitation. The extent of regurgitant flow into the right atrium was recorded from the parasternal and apical windows. Tricuspid insufficiency was graded from 1 to 3 for minimal, moderate, and severe regurgitation. The diameter of the tricuspid anulus was measured between the insertion of the anterior and septal leaflets. Inasmuch as the diameter changes during the heart cycles, only the smallest diameter measured at the mesosystole (at a mean 160 ± 15 msec after the Q wave) was considered and related to the body surface area (BSA).10 Pulsed Doppler study with the sample volume just distal to the

tricuspid orifice demonstrated the flow velocity through the valve. Mean instantaneous gradients were calculated by means of the modified Bernoulli equation.s-!' Only tricuspid valve gradients measured during expiration were considered. The values of three consecutive beats were averaged. Hemodynamic measurements. Catheter measurements were done preoperatively in all patients and in the first 22 survivors before discharge. Gradients across the tricuspid valve were gauged by a double-tipped micromanometer with simultaneous pressure measurement in the right atrium and ventricle. Follow-up. All survivors but one, who was lost to follow-up, were examined in our outpatient clinic in March 1988. Clinical assessment, routine laboratory work-up, and electrocardiographic and echocardiographic study were done. Statistics. Continuous variables are presented with one

The Journal of Thoracic and Cardiovascular

84 8 Minale et al.

Surgery

V> V>

:5

II

u

PREOP.

POSTOP.

o

16 (44%)

0

20 (56%)

-c

:r: >-

z

III

15 (42%)

0

IV

21 (58%)

o

3.7

AVERAGE

1.7

Fig. 2. Comparison of preoperative and postoperative functional state according to the New York Heart Association (NYHA) classification.

Table I. Hemodynamic results Preop. No. of patients RA(v) (mm Hg) PA(s) (mm Hg) PA(m) (mm Hg) Gradient (mm Hg)

48

19.6 55.5 37.8 0.3

± 10.0 ± 18.9 ± 13.3 ± 1.2

Postop. day 17 ± 10

p Value

22 10.3 ± 6.8 43.1 ± 12.1 27.2 ± 8.8 1.3 ± 1.0

<0.001 <0.001 <0.001 <0.001

RA(v). Right atrial v-wave amplitude; PAls). systolic pulmonary artery pressure; PA(m). mean pulmonary artery pressure.

standard deviation of the mean and are compared by means of t tests.

Results The overall hospital mortality rate was 6.3% with three deaths. Two patients died because of cardiac failure and one because of bronchopneumonia followed by multiple organ failure. Two ofthe three deaths occurred in patients undergoing reoperation. The mortality rate was 5.9% for double and 8.3% for triple valve operations. All survivors had an uneventful recovery and could be discharged, on average, 17 ± 7 days postoperatively. Follow-up averaged 14 ± 8 months (2 to 32 months). One patient was lost to follow-up and one patient died suddenly 7 months after the operation despite a good clinical result up to that time. None of the survivors had va-lve-related complications. For functional evaluation only patients whose follow-up period lasted 6 months or

more were considered. All of them improved to class I or II postoperatively (Fig. 2). Hemodynamic results. Twenty-two patients were reinvestigated by cardiac catheterization before discharge. Results are listed in Table I. Postoperatively, right atrial v-wave amplitude was significantly decreased despite persistent moderate-to-severe pulmonary hypertension in the majority of the patients. In three patients the gradient over the tricuspid valve was more than 2 mm Hg (i.e, 2.6, 2.6, and 3.3 mm Hg). Echocardiographic results. Because of the excellent correlation of the hemodynamic and echocardiographic measurements (R = 0.96),12 an echocardiographic study was preferred for late follow-up of 26 patients at an average of 466 ± 72 days postoperatively (Table II). A moderate regurgitation was detected in four patients. The diameter of the tricuspid valve in the apical four-chamber view decreased from 23.7 ± 3.9 mm/rn? BSA (range 17 to 36) to 18.2 ± 2.6 mm/rn? BSA (range 12 to 23) before discharge and to 15.7 ± 1.9 mm/rn? BSA (range 12 to 19) at late follow-up examination. The new dimensions of the anulus late postoperatively (Fig. 3) are at the upper limits of normal. 10. 13 The compromise achieved is a minimal stenosis with a gradient of 1.4 ± 0.6 mm Hg and a regurgitation grade averaging 1.7 ± 0.7. Discussion Untreated relative or organic tricuspid valve incompetence generally persists or progresses even though the

Volume 99 Number 5

Tricuspid valve annuloplasty 8 4 9

May 1990

28

26

24

~ 22

en

E' 20 0-

'- 18 E 16

E

14 12

-

-..- --_. ----- -------- -- -----_. ---_:::::: ::

....... --- ..... ,..:-::::-:--::::::::::::.-:::::: -----------_. ---------- _. -----------.-

----_._._._._.----_._._-_._-_.~-----~~~-~~~~~~~~--_____ ••

._ ••

. _ . __ •

._._ •• _.

0_ • • - - - - . - . - - - - _ · · _ - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - . _ . __ •

•

__

10 +-----r----.,------r----.,-------.

o

100

200

300

POSTOP. DAY

500

400

Fig. 3. The solid line depicts the modifications ofthetricuspid diameter with respect to itspreoperative, early, and latepostoperative dimensions. Thedash-dot-dash line isthedimension ofthetricuspid diameter inthegeneral health population. Thedot lines enclose thestandard deviation ofboth curves. Measurements were done bytwo-dimensional echocardiography with the apical four-chamber view. Results were standardized byrelating them to the BSA ofthe patients. left-sided heart valves have been repaired. Cardiac failure'< 15 and cardiac-related liver cirrhosis developed later in several of these patients. Operative repair of tricuspid regurgitation has been controversial for many years. The mortality rate for tricuspid valve replacement is high-12% to 45%1. 14. 16. l 7 -and the survival rate at 5 years is about 36%.17 Undoubtedly, the results of tricuspid valveannulorrhaphy have been better than those of prosthetic replacementof the valve. Applying hisring technique,Carpentier and hisgroup' reported hospital mortality rates of 9.5% for double and 14% for triple valve operations. Duran and colleagues' had an average hospitalmortality rate of 8.4%in a series of 359 patients, most of whom had the tricuspid valverepaired with a flexible ring. De Vega18 reported an early mortality rate of 9% applying his own technique. Boyd and associates 19 had a 14%mortality rate in the group in which the tricuspid valvewas repaired by the Kay annuloplasty. With the present technique, the early mortality rate averaged 6.3%. Among the major concerns of tricuspid repair, however,are the hemodynamicresults. Recent studies in animals have shown that a significant reduction of the tricuspidvalve orificecauses a marked decrease of cardiac output.-" Duran and colleagues' observeda gradient averaging 3.0 ± 2.0 mm Hg in 38%of the cases postoperatively. The De Vega annuloplasty leads in 50% to a

Table II. Echocardiographic results Preop. No. of patients RA(v) (mm Hg) PA(s) (mm Hg) Gradient (mm Hg) Regurgitation 3.8 ± 0.4 (score 0-3) AD (mm/m? BSA) 27.7 ± 3.9

Postop. day Postop. day 17 ± 10 466 ± 270 p Value 22

1.4 ± 0.7 18.2 ± 2.6

6.5 30.5 1.4 1.7

26 ± 4.0 ± 13.0 ± 0.6 ± 0.7 <0.001

15.7 ± 1.9

<0.001

RA(v). Right atrial v-wave amplitude; PA(s). systolic pulmonary artery pressure; AD, tricuspid anulus diameter.

transvalvular gradient between 2 and 8 mm Hg, as reportedby Haerten and associates." With the Carpentier ring, Hanania and associates7 had a gradient between 3 and 5 mm Hg in 12 of 25 patients and a light incompetence in five of 10 angiographic follow-up studies. By comparing the Carpentier, De Vega, and Kay annuloplastiesof the tricuspid valve, Grondin and co-workers-' found light or moderate mitral incompetence in 22%, 23%, and 91% of the patients. In a series of 25 patients reinvestigated hemodynamically after ring annuloplasty, Carpentier and co-workers 1 found normal function of the tricuspidvalvein 17patients, light regurgitation in seven, and moderate regurgitation in one patient. Nine patients had a residual gradient of less than 4 mm Hg. Delocheand colleagues'?demonstrated the asymmet-

The Journal of Thoracic and Cardiovascular

8 5 0 Minale et al.

ric dilatation of the anulus in a case of relative incompetence affecting mostlyits anterior and evenmore its posterior portion, which increases up to 80% of its original length. Accordingly, with our method the length of the anulus was reduced selectively on its anterior and posterior portions without excluding the active surface of the valveleaflets. In this way the natural relationship of the tricuspidvalveorificewasrestored.As a rule, the anterior and posteriorleafletswere completely detached from the anulus. An average of two thirds to three fourths of the free anulus was excluded. By following this simple rule, we obtained an optimal compromise between gradient and regurgitationgrade (Fig. 3). On completion of repair, direct examination of the tricuspid valve with the heart beating was, in our experience, the best method of judging the competenceof the valve. Light regurgitation that was detected by flushing the right ventricle with saline solutionduring cardiac arrest disappearedas soonas the papillary muscles were functioning. In three cases in which the valvedid not closeperfectlybecauseof a moderate organicdiseaseof the leaflets, the 4-0 Prolenesuture (Ethicon Inc.,Somerville, N.J.) was removed and the anulus was isolated and excluded more extensively. In conclusion, the present technique allows an anatomicand functionalreconstruction of the tricuspidvalve. The anulus remains pliable and physiologic reductionof its diameter up to 15%during systole is stillpossible. Furthermore, because of avoidance of any prostheticdevice, a progressive decrease of the tricuspiddiameter could be observed over time. These aspects allowa good hemodynamic compromisebetweenresidualgradient and incompetence, as shown earlier. No or only physiologic regurgitation is seen in the majority of the patients, and no or minimal gradient is presentoverthe valve. Functionalresults reflect the hemodynamicfindings; all patients with a follow-up of 6 months or more are free of symptoms. Both with other techniquesof repair and with the present one, early and late survival can be considered excellent with respect to the severityof the congestive heart disease preoperatively and to the high percentage of reintervention among these patients. I.

2. 3.

• 4.

REFERENCES Carpentier A, Deloche A, Hanania G, et al. Surgical management of acquired tricuspid valve disease. J THORAC CARDIOVASC SURG 1974;67:53-65. Kay JH, Maselli-Campagna G, Tsuji HK. Surgical treatment of tricuspid insufficiency. Ann Surg 1965;162:53-8. De Vega NG. La anuloplastia selectiva, regulable y permanente. Una technica original para el tratamiento de la insuficiencia tricuspide. Rev Esp Cardiol 1972;25:555-60. Duran CMG, Ubago JL. Clinical and hemodynamic performance of a totally flexible prosthetic ring for atrioven-

Surgery

5.

6.

7.

8.

9.

10.

II.

12.

13.

14.

IS.

16.

17.

18. 19.

20.

tricular valve reconstruction. Ann Thorac Surg 1976; 22:458-63. Duran CMG, Pomar JL, Colman T, Figueroa A, Revuelta 1M, Ubago 1L. Is tricuspid valve repair necessary? J THORAC CARDIOVASC SURG 1980;80:849-60. Haerten K, Seipel L, Loogen F, Herzer J. Hemodynamic studies after De Vega's tricuspid annuloplasty. Circulation I978;58(Pt 2):128-33. Hanania G, Sellier P, Deloche A, et al. Resultats a moyen terme de l'annuloplastie tricuspide reconstructive de Carpentier. A propos de 25 cas avec catheterisme post-operatoire. Arch Mal Coeur 1974;67:895-909. Hatle L, Angelsen B, ed. Doppler ultrasound in cardiology. Physical principles and clinical applications. 2nd ed. Philadelphia: Lea & Febiger, 1985:97-205. Minale C, Lambertz H, Messmer BJ. New developments for reconstruction of the tricuspid valve. J THORAC CARDIOVASC SURG 1987;94:626-31. Lambertz H, Sechtem U, Soeding S, Kemrnen HP, Krebs W. Zur Pathophysiologie der Trikuspidalinsuffizienz. Bewegungsanalyse des Trikuspidalanulus mittels zweidimensionaler Echokardiographie. Z Kardiol 1985;74:662-9. Holen J, Aaslid R, Landmark K, Simonsen S. Determination of pressure gradient in mitral stenosis with non-invasive ultrasound Doppler technique. Acta Med Scand 1976; 199:455-60. Chapman JV, Sgalambro A, ed. Basic concepts in Doppler echocardiography. Methods of clinical applications based on a multi-modality Doppler approach. 1st ed. Dordrecht: Martinus Nijhoff, 1987:164-5. Lambertz H. Kontrastechokardiographie zur Verlaufbeurteilung der Trikuspidalklappendysfunktion bei Herzinsuffizienz. In: Erhel R, Meyer J, Brenneke R, eds. Fortschritte der Echokardiographie. 1st ed. Heidelberg: Springer Verlag, 1985:158-67. Breyer RH, McClenathan JH, Michaelis LL, McIntosh CL, Morrow AG. Tricuspid regurgitation: a comparison of nonoperative management, tricuspid annuloplasty, and tricuspid valve replacement. J THORAC CARDIOVASC SURG 1976;72:867-74. . Pluth JR, Ellis FH Jr. Tricuspid insufficiency in patients undergoing mitral valve replacement: conservative management, annuloplasty, or replacement. J THORAC CARDIOVASC SURG 1969;58:484-91. Sanfelippo PM, Giuliani ER, Danielson GK, Wallace RB, Pluth JR, McGoon De. Tricuspid valve prosthetic replacement: early and late results with the Starr-Edwards prosthesis. J THORAC CARDIOVASC SURG 1976;71:441-5. Bodh Ij, Fraser RS, Lee SJK, Rossall RE, Callagan rc. Long-term survival after tricuspid valve replacement: results with seven different prostheses. J THORAC CARDIOVASC SURG 1977;74:20-7. De Vega NG. Discussion of Breyer et al. 14 Boyd AD, Engelman RM, Isom OW, Reed GE, Spencer Fe. Tricuspid annuloplasty. J THORAC CARDIOVASC SURG 1974;68:344-51. Higashidate M, Tamiya K, Kurasawa H, Takanshi Y, Imai

Volume 99 Number 5 May 1990

Y. Real-time measurement of tricuspid valve annular area for annuloplasty: experimental study. J THORAC CARDIOV ASC SURG 1988;96:88-91. 21. Grondin P, Meere C, Limet R, Lopez- Bescos L, Declan JL, Rivera R. Carpentier's annulus and De Vegas' annuloplasty: the end of the tricuspid challenge. J THORAC CARDIOVASC SURG 1975;70:852-61. 22. Deloche A, Guerinon J, Fabiani IN, et al. Etude anatomique desvalvulopathies rhumatismales tricuspidiennes. Ann Chir Thorac Cardiovasc 1973;4:32-7.

Discussion Dr. Frank C. Spencer (New York. N.y.). I compliment the authors for their ingenious approach. As they indicate, their follow-up is short but the impressive results are encouraging. Their approach is similar to the Carpentier reconstructive techniques of annuloplasty for the mitral valve except that no segment of leaflet is excised. At New York University over the past 10 years we have performed about 200 tricuspid operations, almost all in association with a mitral valve operation. An annuloplasty was done in about 165 and a replacement in the others. A number of these patients were heroin addicts with endocarditis. In the majority of the tricuspid reconstructions, a posterior leaflet annuloplasty developed by Jerry Kay and George Reed was used. This operation leaves about 8 cm of functioning tricuspid anulus, which represents an orifice diameternear 27 mm. Dr. Minale, if I understand the manuscript correctly, the reconstructed tricuspid orifice has a diameter of 15 to 17 mm. How do you arrive at the decision to make the diameter this size? Dr. Minale. The approach of excluding the posterior leaflets of the tricuspid valve isgood, as well. We have used it in the past. With the present method, however, no portion of leaflet tissue need be excluded, as commonly occurs with the Kay technique: Only the stretched portion of the anulus is excluded. This detail isimportant to avoid an overcorrection of the valve insufficiency, which can lead to stenosis. With regard to your question concerning anulus diameter, I ~ould like to clarify that the mentioned 15 to 17 mm/rn? BSA IS the length of the cross diameter of the tricuspid anulus observed postoperatively in the four-chamber view by echocardiographic methods. Intraoperatively, we used only the empiric rule of excluding two thirds of the length of the anterior and posterior portions of the anulus. From a retrospective analysis of data collected during the operations, we noted that the valve area index after annuloplasty was similar in all cases and large enough to avoid stenosis. Dr. Spencer. The annuloplasty method used at New York University has been very durable. Reoperation for recurrent insufficiency is rarely necessary. My other question concerns leaflet reconstruction. The leaflet appears ruffled, or pleated, in your diagrams. Have you excised any leaflet tissue or have you had any dehiscence of the suture line? Dr. Minale. We have had no difficulty in anchoring the cut edges of the leaflets to the shortened anulus. For this reason it was not necessary to excise any leaflet tissue. We also had no breakdown of the suture line. I believe that during systole most

Tricuspid valve annuloplasty 8 5 1

of the stress occurs on the free edges of the leaflets at the insertions of the chordae tendineae because of the maximal deceleration at that level. In my opinion, the only critical point is the meeting point between the anterior and posterior portions of the anulus. However, the continuous double-roof 4-0 Prolene suture sh?uld help to distribute the longitudinal stress at this single pomt over the whole length of the anterior and posterior portions of the anulus. Dr. Roland Hetzer (Berlin. Federal Republic a/Germany). Since 1983 I have used a modified suture technique that allows precise gauging of the anulus and that has not resulted in annuloplasty breakdown or tricuspid reoperation. ~~e frequent use of tricuspid annuloplasty has been greatly facilitated by the transseptal approach tothe mitral valve, which has been applied in almost all mitral operations. With this approach the tricuspid valve lies on the way, and an additional tricuspid annuloplasty adds only 5 minutes to the operating time. In Berlin, during the past 3 years, we have used this type of combined procedure in more than one third of all mitral operations. I would like to ask Dr. Minale to comment on his approach in the combined operations. Second, what do you think about intraoperative echocardiography to detect valve competence? We have been using this technique routinely for a year. Dr. Minale. In our clinic we routinely use the interatrial groove approach for mitral valve operations. I only use the transseptal approach in case of mitral reoperation with or without concomitant tricuspid lesions. With regard to intraoperative echocardiography, I believe it should be routinely used in case of valve reconstruction. Unfortunately I had not the opportunity to evaluate the results of our procedure by intraoperative echocardiography. I am sure it would help us to get even better results. Dr. Yasunaru Kawashima (Osaka. Japan). Our method of choice has been to do the bicuspidalization annuloplasty initially described by Kay and his associates in 1965 for tricuspid regurgitation. However, in patients with a mild degree of annular dilatation, it was difficult to obtain a competent valve with this procedure without leading to stenosis. Therefore, in such a patient, we have placed several interrupted sutures on the leaflets adjacent to the commissural zone between the anterior and posterior leaflets to avoid overcorrection leading to stenosis. Thus, in a recent series of 60 patients, there were three operative deaths (5%) and no late deaths in the follow-up periods up to 5 years. On the basis of these results, I presume the most important point is in the surgical treatment of tricuspid valve disease is to avoid replacement even in the presence of a mild degree of tricuspid stenosis. In our hospital tricuspid valve replacement was performed only in 2% of the patients who needed the surgical procedure for tricuspid valve disease. In what percentage of patients with tricuspid valve disease have you replaced the tricuspid valve since you began using this procedure? Dr. Minale. I have the same reluctance as you in replacing the tricuspid valve. Since introduction of the present technique I have had to replace the valve in only two cases because of severe organic lesions. Both patients died in the perioperative period, as a result of cardiac failure and multiple lung abscesses, respectively.